© H. Liu et al., published by IFP Energies nouvelles, 2020

Nomenclature

C_r : Specific heat of the rock (J/(kg °C))

H : Heat accumulation amount (J/m³)

k : Permeability (m²)

k_rl : Relative liquid permeability (–)

k_rg : Relative gas permeability (–)

P : Pore pressure (Pa)

S : Saturation

T : Temperature (°C)

X : Mass fraction (–)

u_l : Specific internal energy in liquid phase (J/kg)

u_g : Specific internal energy in gas phase (J/kg)

ϕ : Porosity (–)

ρ : Density (kg/m³)

ρ_r : Rock grain density (kg/m³)

μ : Viscosity (Pa s)

w : Superscript water component

c : Superscript CO₂ component

l : Subscript liquid phase

g : Subscript gas phase

s : Subscript solid phase

1 Introduction

Since the winter of 2013, more than 30 provinces in China have experienced very serious air pollution termed “haze”. To control this air pollution, the following measures have been implemented: modifying the energy consumption structure; reducing the consumption of coal, which releases large amounts of CO₂, SO_x and dust into the atmosphere; and greatly increasing the proportion of renewable energy in the energy consumption structure [1]. The enthusiasm for the extraction of low-medium enthalpy geothermal resources is motivated by the successful geothermal source district heating system in Xiongxian, which is located 130 km from Beijing. More than 92% of the district heating systems are driven by geothermal hot water with temperatures less than 90 °C [2]. China has abundant low-medium enthalpy hydrothermal reservoirs at shallow depths. If the type of low-medium enthalpy geothermal source district heating system applied in Xiongxian and Tianjin can be copied and widely applied in other regions of China, the air quality can be improved because of the decrease in traditional coal-based thermal power plants.

The overexploitation of hot water from subsurface reservoirs may result in land subsidence and thermal and chemical pollution if the cooled geothermal tail-water is not treated properly and discharged directly into the ground [3–5]. The reinjection of recycled geothermal water is an important strategy for maintaining reservoir pressures, prolonging the depletion of geothermal reservoirs, and efficiently avoiding problems, such as ground subsidence and heat and chemical pollution of the environment. Considering the history of cooled geothermal reinjection worldwide, France generated the first doublet system with one reinjection well and one production well in the geothermal reservoir located at Melun l’Almont geothermal field in 1969. The geyser is known for having not only the largest geothermal power station worldwide but also a mature reinjection system that has been reinjecting the waste water near the geothermal field into the subsurface high-temperature geothermal reservoir to maintain its pore pressure since 1997. Germany also has many successful experiences with reinjecting waste water into geothermal reservoirs, e.g., Waren and Neubrandenburg, with a reinjection rate of 50 m³/h [6]. Reinjecting cooled geothermal water into fractured carbonates, volcanic debris, and consolidated sandstones is easier than reinjecting this water into loose porous sandstone reservoirs because of the clogging problems caused by physical and chemical effects [7, 8]. Many numerical and experimental studies focus on the reinjection efficiency, which is affected by factors, including the characteristics of the geological formations at the macroscale, such as the stratigraphic configuration and fracture distribution system, and at the microscale, such as the porosity, permeability, and pore and grain distributions [6]. In addition, the reinjection strategy, including the pressure, temperature, and reinjection rate, has a great impact on the efficiency of the reinjection operation [9, 10].

In addition to the reinjection of pure water to maintain stable reservoir pressure, CO₂ has been considered for injection into geothermal reservoirs in both H₂O-based and CO₂-based strategies in recent studies [11–24]. CO₂-Aided Geothermal Systems (CO₂-AGES) are helpful for maintaining the reservoir pressure and present much higher flow rates than water systems, thus extracting more heat [9, 19]. Some geothermal reservoirs are initially rich in CO₂ dissolved in geothermal fluids [21]. For instance, to generate 1 kWh electricity, the CO₂ emission amount can reach 0.9–1.8 kg in most geothermal fields in Turkey. If CO₂ is emitted into the atmosphere, the geothermal reservoir will be depleted in the long term. The injection of CO₂ into the geothermal reservoir can maintain its pressure, generate an artesian flow of brine and increase the storage capacity of CO₂ by generating a large amount of space induced by hot water extraction. This type of geothermal energy extraction with the help of CO₂ injection can actually be regarded as energy storage in a sedimentary basin [22–24]. A pilot-scale injection of CO₂ combined with cooled degassed injectate lasted for a period of two months in the Umurlu geothermal field in Turkey between October and December of 2017, highlighting the profound effect of injection on the behavior of reservoir pressure surrounding the production wells close to the injection wells [21]. In addition, the injection of CO₂ into a geothermal reservoir can increase the porosity and permeability by strengthening the water–rock interactions [24], thus potentially improving the geothermal reinjection efficiency of cooled geothermal return water in sandstone reservoirs.

A negative impact of fluid reinjection is the associated obvious temperature decline or cooling effect in the reservoir [25, 26]. Furthermore, the temperature decline may last for a long time before heat recovery, strongly affecting the long-term production of geothermal energy. The injection of CO₂ may cause a much more complex cooling effect in the geothermal reservoir because thermal breakthrough highly depends on the geological structure of the geothermal reservoir, which controls the mode of fluid transport and flow-channel/fracture-space characteristics [27, 28]. However, few studies showing the cooling effects in geothermal reservoirs by comparing the simultaneous or separate injection of cold CO₂ and cooled geothermal return water are available. The injection of CO₂ and water into either the same or different layers of aquifers may cause a cooling effect surrounding the production region if poorly organized. Thus, the careful design of the injection–production–reinjection well configuration should be tested for the development of CO₂-aided geothermal energy extraction systems.

In this paper, a small-scale simplified, idealistic geological model is established based on the Guantao formation and its overlying and underlying formations in Binhai district in Tianjin. A comparison of the cooling effects in the geothermal reservoir (e.g., Guantao formation) caused by the injection of CO₂ and cooled geothermal return water is performed by numerical methods. The goal of the preliminary study reported in this paper is to provide a general understanding of the pressure and temperature response of the geothermal reservoir when using this type of injection–production system. Sensitivity studies are carried out while considering different injection/production rates, temperatures, boundary conditions, well configurations, etc. A basic understanding of how to control the cooling effect in a multilayered reservoir–caprock system is essential before an optimized injection–production–reinjection system is established in a large-scale geothermal reservoir.

2 Exploitation of geothermal energy in Tianjin

The Bohaiwan basin, especially the Huabei plain, is characterized by rich hydrothermal resources [29] with a geothermal water reserve of approximately 1.3 × 10¹³ m³. The successful utilization of hydrothermal resources has been widely applied in district heating systems in regions, including Tianjin, Beijing, and parts of Hebei province (e.g., Xiongxian). Hot water at a low-medium temperature is extracted from several formations (Tab. 1) in Bohaiwan basin from top to bottom, including the Minghuazhen formation (N_m), Guantao formation (N_g), Dongying formation (E_d), and Wumishan formation (J_w). Most hot water-bearing formations are also rich in oil. The main productive layer of the Gudao oilfield is the Guantao formation, while the main productive layers of the Shuguang and Shengli oilfields are $E_S^{1+3}$ and $E_S^{2+3}$, respectively (Tab. 2). In addition to the rich geothermal resources in Bohaiwan basin, many other oilfields have been discovered and exploited, especially in the Jizhong depression zone (Fig. 1), including the buried hill oil reservoir in Wumishan carbonates represented by the Renqiu oilfield. The temperature of the Renqiu oil reservoir ranges from 110 °C to 125 °C with a geothermal gradient of 1.7–1.8 °C/100 m. In addition, the Chaheji oilfield, which is characterized as an $E_d^3-E_s^1$ fluvial sandstone reservoir, is the largest Tertiary oilfield.

Geothermal energy has been utilized in district heating systems in China since 1990, and the covered area greatly increased to 88.75 million m² by 2016 (Fig. 2). Tianjin is famous for its successful geothermal utilization, especially through the geothermal direct heating system, currently covering a total area of 22.33 million m². At the end of 2013, its geothermal direct heating system reached an area of 20 million m² with 474 geothermal production wells [30]. Its proportion was approximately 50% of China’s total geothermal direct heating systems in 2009 with a heating area of 12.13 million m², including 264 production wells and 53 reinjection wells with a reinjection ratio of only 24.5%. The sandstone aquifers of the Guantao formation (N₁ g), karst/fractured carbonate reservoirs of the Ordovician and Wumishan (Jxw), and Cambrian formations are used to extract hot water in Tianjin [27, 28]. The maximum drilling depth is more than 4000 m, and the maximum discharge rate of a single well is more than 100 m³/h [28]. However, the long-term extraction of subsurface hot water has resulted in a serious decrease in the groundwater level and subsidence in Tianjin. The reinjection of cooled geothermal return water into production aquifers controls subsidence to some extent, although most parts of Tianjin still have a subsidence rate of 10–40 mm/year [31].

As a successful model of geothermal energy utilization in China, studies investigating the reinjection technology of Tianjin began in the 1980s, and many valuable improvements, such as in the equipment used in the reinjection technology, well completion technology, etc., have been achieved. Reinjection has efficiently inhibited the water level drawdown and subsidence in the Tianjin geothermal system to some degree [31, 32]. The maximum reinjection rate in Tianjin in the sandstones of the Guantao formation reaches 100–120 m³/h. The reinjection of cooled geothermal water is useful for maintaining the reservoir pressure. However, if the reinjection wells are located close to production wells or the reinjection amount is greatly increased, the cooled reinjection fluid will cause a cooling effect on the geothermal reservoir, and the water from the production wells will be rapidly cooled and impacted by the thermal breakthrough [31, 33]. This phenomenon is common not only in deep hydrothermal reservoirs but also hot dry rocks [34].

The reinjection ratio (i.e., the ratio between the reinjection and production amounts of geothermal water) in Tianjin greatly varies. Based on statistical data, the reinjection ratio was as high as 63% in 2000, while it was as low as 24.5% in 2009. Additionally, large differences exist in the reinjection ratios of different geothermal reservoirs [30]. The karstic carbonate reservoirs of the Ordovician have the largest reinjection ratio of 114.87%, followed by the Cambrian geothermal reservoirs at 66.61% and the Wumishan formation at 47.4%, while the sandstone reservoirs of the Guantao formation have a much lower reinjection ratio of 9.9%. Minghuazhen (N₂ m) had a very limited reinjection amount of return water, and there was no reinjection activity in the Dongying (E₃ d) formation (see Fig. 3). The Guantao formation plays a great role in producing 19% of Tianjin’s hot geothermal water. A lower reinjection ratio of geothermal water may result in the depletion of reservoir pressure over the long term. Therefore, more fluid needs to be injected into its production layer to increase the reservoir pressure and inhibit continuous subsidence [35].

Tianjin is characterized by very thick Cenozoic group formations (more than 4000 m) in some regions, e.g., in the Binhai district, which belongs to the Huanghua depression zone (Fig. 4). The bottom depth of the Guantao formation is located at a depth of 990–2660 m. In the Cangxian uplift zone, the thickness of the Guantao formation is thin, and the buried depth is shallow. Far from the uplift zone, the buried depth is deep, and the thickness of the formation increases. The top depth of the Guantao formation in the Huanghua depression ranges between 1152 and 1788 m with a thickness of 126–1059 m [6]. The detailed reservoir thermal properties of the Huanghua depression and Cangxian uplift zone in the Guantao formation are listed in Table 2.

In Binhai district, the geothermal reserve is 6.28 × 10¹⁹ J, and the annual reserves of hot water in the Guantao formation can reach 1.15 × 10⁷ m³ [36]. The fluid temperature at the wellhead, which is produced from the Guantao formation, is 45–78 °C, and the water chemistry shows that the water is HCO₃–Na and HCO₃–Cl–Na with TDS (total dissolved solids) values ranging from 0.8 g/L to 1.9 g/L. The temperature gradient ranges from 2.5 to 3.0 °C/100 m [30]. The permeability is less than 50 mD, the permeability coefficient is 0.3–2.2 m/d, the transmission coefficient is 40–212 m²/d, and the porosity is 18–36.6%. Based on production and reinjection data from Binhai district in Tianjin in 2013, there are 85 production wells with a total annual hot water production of 3.866 million m³ (i.e., 1.44 kg/s/well) and 14 reinjection wells with an annual reinjection amount of 28.8 × 10⁴ m³ (i.e., 0.65 kg/s/well).

3 Numerical model setup and parameterizations

3.1 Mathematical equations

Generally, the injection of CO₂ and the reinjection of cooled geothermal water into geothermal reservoirs are multiphase, multicomponent, and nonisothermal problems. Therefore, a mass conservation function involving two phases (gas and liquid) is used to describe the fluid flow in a porous or fractured medium. Using the water component as an example, the governing equation is stated as follows:$$\frac{\mathrm{\partial }}{\mathrm{\partial }x}\left(\varphi \left(X_l^w{S}_l{\rho }_l+X_g^w{S}_g{\rho }_g\right)\right)=-\nabla \times \left\{X_l^w{\rho }_l\left(-k\frac{k_{\mathrm{rl}}}{{\mu }_l}\left(\nabla P_l-{\rho }_{l}g\right)\right)+X_g^w{\rho }_g\left(-K\frac{k_{\mathrm{rg}}}{{\mu }_g}\left(\nabla P_g-{\rho }_{g}g\right)\right)\right\}+\left(q_l^w+q_g^w\right),$$(1)where S represents the saturation (–), X is the mass fraction (–), ϕ is the porosity (–), ρ is the density (kg/m³), μ is the viscosity (Pa s), k is the permeability (m²), k_rl represents the relative liquid permeability (–), k_rg represents the relative gas permeability (–), P is the pore pressure (Pa), superscript w represents water, l is the liquid phase, and g is the gas phase.

Regarding the CO₂ component, the governing equations of mass conservation are rewritten as follows:$$\frac{\mathrm{\partial }}{\mathrm{\partial }x}\left(\varphi \left(X_l^c{S}_l{\rho }_l+X_g^c{S}_g{\rho }_g\right)\right)=-\nabla \times \left\{X_l^c{\rho }_l\left(-k\frac{k_{\mathrm{rl}}}{{\mu }_l}\left(\nabla P_l-{\rho }_{l}g\right)\right)+X_g^c{\rho }_g\left(-k\frac{k_{\mathrm{rg}}}{{\mu }_g}\left(\nabla P_g-{\rho }_{g}g\right)\right)\right\}+\left(q_l^c+q_g^c+q_s^c\right),$$(2)where superscript c indicates CO_2, and subscript s indicates the solid phase.

Similarly, the Heat (H) accumulation in the two-phase system is described as follows:$$H=\left(1-\varphi \right){\rho }_r{C}_{r}T+\varphi \left(S_l{\rho }_l{u}_l+S_g{\rho }_g{u}_g\right),$$(3)where ρ_r represents the rock grain density (kg/m³), C_r is the specific heat of the rock (J/(kg °C)), T is the temperature (°C), u_l is the specific internal energy in the liquid phase (J/kg), and u_g is the specific internal energy in the gas phase (J/kg).

3.2 Geological model setup and mesh discretization

A small-scale, simplified geological model is established based on the geology of the Guantao formation and its overlying and underlying layers in Binhai district in Tianjin (see the detailed description in Sect. 2). The geometry of this model is used as the geometry (with dimensions of 3000 m, 1500 m, 700 m in the x, y, and z directions, respectively) of the numerical model, which aims to analyze the pressure and temperature responses caused by injection of cooled geothermal water and CO₂ to realize the optimization of the well configuration and strategies concerning the injection and production rates, pressure, temperature, etc. The scheme of the XZ geological section along with the multilayered reservoir–caprock configuration is shown in Figure 5. The top of the model is set at a depth of 1800 m, and the bottom depth is 2500 m. The total thickness of the model is 540 m, while the thickness of the caprock is 160 m. Based on the geological description of Binhai district in Tianjin, aquifer 2, which has a thickness of 500 m, can be used to represent the Guantao formation. Considering that cooled tail water can be injected into a different layer far from the production formation to avoid early thermal breakthrough in many geothermal engineering situations (e.g., Mori, Bardy, etc.) [37], a hetero-layered “two production wells and two reinjection wells” configuration mode is applied in this study, and another well is designed for CO₂ injection. The geothermal production wells are completed in highly permeable aquifers, and two cooled geothermal water reinjection wells are placed in the upper layer as shown in Figure 5. This type of well configuration is used to avoid thermal breakthrough caused by cooled geothermal water injection. Thus, the impacts of CO₂ injection on the reservoir pressure and temperature can be studied in detail. The well spacing is assumed to be 500 m in the X direction. The discretization of the meshes is fine near the wells, and the meshes become progressively coarser as the distance from the wells increases. In total, there are 52 500 meshes (N_x = 50, N_y = 30, N_z = 35) in the numerical model.

3.3 Input parameters in the numerical model

The ECO2N module of the parallel simulator TOUGH2MP is applied to simulate the cooling effect in the geothermal reservoir induced by geothermal water production and fluid injection [9]. Three components (water, NaCl, and CO₂) are considered, and the pressure range is less than 60 MPa with temperatures ranging from 10 °C to 110 °C. The reservoir pressure and temperature in low-medium enthalpy geothermal reservoirs in Binhai district in Tianjin are in the applicable range of the ECO2N module. The capillary pressure function is described by Van Genuchten [38], and the relative permeabilities of liquid and gas are described by Van Genuchten [38] and Corey [39], respectively.

3.3.1 Hydrogeological properties of geothermal reservoirs

Table 3 summarizes the parameters, including the hydrogeological properties of the aquifer and aquitard formations (e.g., porosity, permeability, thickness, density, and thermal conductivity), initial and boundary conditions of the numerical model, and the strategies of injection, production, reinjection, etc., involved in the simulation study of the base case representing low rates of geothermal water production and reinjection. The porosity is set as 0.3 and 0.1 for the aquifer and aquitard, respectively. Accordingly, the permeability of the aquifer is assumed to be 50 mD, while that of the aquitard is assumed to be 0.001 mD. It is assumed that the siltstone reservoir is used to store the cooled geothermal reinjection water and that its porosity and permeability are low. The parameters of the geological formations used in the base case are determined based on data obtained from the Guantao formation in Tianjin.

3.3.2 Initial and boundary conditions

The initial pore pressure is set as the hydrostatic pressure. The conservative estimation of the geothermal gradient is 25 °C/km, and the salinity of the fluid in the reservoir is defined as 1.8 g/L. The reservoir is assumed to be water-saturated before the injection of CO₂ and hot water production. The top boundary is set as having no heat or mass flow with an average annual surface temperature of 12 °C. The bottom boundary is assumed to have no heat or mass flow (BB1), and the lateral boundaries are set as open with heat and mass flows.

3.3.3 Injection, production, and reinjection strategies

In the base case, considering the feasibility of the injectivity and engineering operation, the injection pressure and temperature at the well bottom are 30 MPa (approximately 5.2 MPa higher than the initial reservoir pore pressure) and 10 °C, respectively. In engineering operations, CO₂ is often transported by a tank truck or pipeline after compression and cooled to as low as −20 °C. A booster pump and an electric heater are often employed on the ground at the injection location to increase the pressure and temperature (e.g., to approximately 0 °C) for the CO₂ injection. Thus, 10 °C in the base case is a very conservative estimation of the CO₂ temperature at the well bottom. Much higher CO₂ injection temperatures at the well bottom may be possible when the CO₂ temperature is increased at the surface before injection. Thus, different injection temperatures are studied in the sensitivity analysis. The hot water production rate and cooled geothermal return water reinjection rate are set as 2 kg/s, which is comparable to the average geothermal production rate in Binhai district, while the enthalpy of the reinjection water is assumed to be 42 kJ/kg in the base case.

4 Results

4.1 Propagation of pressure and CO₂ plume

When CO₂ is injected into aquifers to maintain the reservoir pressure, which is helpful for the production of hot water, the propagation of the pressure plume mainly concentrates at the periphery of the injection well [21]. The affected region is greatly controlled by the lateral migration of CO₂ and increases over time (Figs. 6 and 7). At the top of aquifer 2, the pressure buildup substantially increases with the upward migration of the injected CO₂. In addition to the aquifers holding the injected CO₂, the pore pressure in the overlying impermeable caprock also substantially increases. At the production well, the pressure at the well bottom increases over time before the CO₂ breakthrough. The pressure sharply decreases when two phases (i.e., CO₂ and hot water) are produced for several months. Then, the pressure rebounds to a stable state (Fig. 8).

4.2 Evolution of CO₂ flowing rates and dissolution in the aqueous phase

In different directions (e.g., downward and outward from the injection point), the difference in the CO₂ flow rate increases over time with a maximum difference of 2 kg/s after 35 years of geothermal hot water production (Fig. 9). The downward CO₂ flow plays a dominant role compared with that in the lateral direction. This result may be driven by gravity effects because the dense water containing dissolved CO₂ also strengthens the downward CO₂ flow.

The spatial evolution of the CO₂ mass fraction in the aqueous phase shows that more CO₂ is dissolved into the aqueous phase over time (Fig. 10), which is consistent with the variation trend of the CO₂ plume migration. The CO₂ dissolved fluid, which has a maximum CO₂ mass fraction of 0.05–0.06 (Fig. 11), migrates to the production well much earlier than the thermal breakthrough (Figs. 10 and 11). In most geothermal reservoirs, the CO₂ in the geothermal fluid is enriched; thus, anti-corrosion geothermal pipelines, which can withstand the corrosion induced by fluid with dissolved CO₂ for a long period, are applied.

4.3 Vertical temperature profile along the hot water production well

The temperature distribution trend along the vertical profile of the hot water production well is shown in Figure 12. After the CO₂ breakthrough at the production well, the temperature in sections shallower than 2190 m increases during the 30-year geothermal production period. At points deeper than the threshold depth (i.e., 2190 m), three stages of temperature changes occur. Stage 1 is characterized by a sharp temperature decrease in depth between −2190 m and −2330 m, stage 2 is characterized by a general temperature decrease to a stable state in depths ranging from −2330 m to −2430 m, and a sharp increase in temperature occurs in stage 3 at the bottom layer of the reservoir. Figure 13 shows that the temperature of the production fluid slightly increases after 5 years until the end of 10 years, which is caused by the CO₂ breakthrough, and the flowing enthalpy increases accordingly when both CO₂ and hot water are extracted at the production well.

5 Discussion

To better understand the cooling effects on geothermal reservoirs induced by both the injection of cold CO₂ and cooled geothermal return water over the long term and understand the temperature recovery attributable to the heating of neighboring non-affected rock, several scenarios are simulated (see details in Tab. 4) while considering various parameters, including the well configuration (base case, case 1, and case 2), CO₂ injection temperature (case 2 vs. cases 3, and 4), CO₂ injection-caused reservoir pressure and temperature responses (case 4 vs. case 5), production rate of hot water (case 5 vs. case 6), injection pressure and temperature of CO₂ (case 2 vs. case 7), bottom boundary condition (base case vs. base case 2), and thermal conductivity of the saturated aquifer (case 6 vs. case 8). The designs of the scenarios are developed based on many related studies [40–43]. Sensitivity studies are applied to provide an optimized injection–production–reinjection strategy for use in hydrothermal reservoirs. The well configurations in the different scenarios are shown in Figure 14. The simulations in these case studies last for 50 years with continuous injection of CO₂/cooled geothermal water reinjection.

5.1 Cooling effect in a geothermal reservoir with low injection and extraction rates

5.1.1 Temperature changes with and without CO₂ injection

Compared with the traditional geothermal reservoir strategy (i.e., only cooled geothermal water injection), CO₂ injection into a geothermal reservoir can increase the reservoir pressure and decrease the CO₂ emission into the atmosphere; thus, this approach may be regarded as a more attractive method. The simulation results show that after 10 years of geothermal production, the reservoir temperature surrounding the production wells increases by 2–3 °C with CO₂ injection compared with that under the traditional water injection condition (Fig. 15), although CO₂ breakthrough is found at the production well. However, this positive impact on the reservoir temperature only lasts for several years, and a negative effect on the reservoir temperature dominates in the long term. This phenomenon is also supported by numerical studies conducted by other researchers [44].

However, reasonable well configurations and injection-production strategies may be helpful for prolonging geothermal production by controlling the negative impacts of thermal breakthrough at the production well. Under low production and reinjection rates of geothermal water (i.e., 2 kg/s), CO₂ injection is the main source of thermal breakthrough at the production well over the 50 years of operation. Under the well configuration of type III, different reinjection enthalpies (see cases 2–4) of the cooled geothermal water do not result in differences in the temperature at the production well, indicating that no thermal breakthrough is caused by cooled water reinjection.

5.1.2 Location of water reinjection wells and the cooling effect of the geothermal reservoir

The thermal breakthrough at the production well can be controlled by the configuration of the water reinjection wells, which is the most important issue in the design of a traditional water reinjection system [41, 42]. In the geothermal field, water reinjection wells can be located between production wells or far from production wells in the outfield region [43]. Generally, to maintain stable reservoir pressure, cooled geothermal water is often injected into the same formation of low-medium enthalpy geothermal reservoirs. However, serious thermal and chemical breakthrough render the location of reinjection wells at a different layer far from the production formation appealing [37]. In this paper, when the CO₂ injection mainly controls the thermal breakthrough, although the water reinjection wells are located at the upper aquifer, which is separated from the production aquifer by a low impermeable caprock, their impact on the thermal breakthrough is negligible (Fig. 16a) compared with the case of water reinjection in the same layer of geothermal reservoirs (Fig. 16c). In contrast, when the CO₂ injection wells are completed at the upper parts of the geothermal reservoirs, thermal breakthrough occurs much earlier because more CO₂ is injected into the geothermal reservoir under the same injection pressure and temperature conditions (Fig. 16b).

After 20 years of CO₂ injection and cooled water reinjection, the small pressure buildup induced by CO₂ injection at the same layer pushes the hot water and CO₂ flow towards the production well, and the fluid temperature at the production well is significantly decreased (Fig. 17). The upper configuration of the CO₂ injection well is helpful for the production of fluid at the lower part of the geothermal reservoir, producing a 4 °C temperature difference compared with that of the lower configuration of the CO₂ injection well. There is no temperature difference at the production well among case 2, case 3, and case 4 based on the numerical simulations; thus, these cases are not listed here.

5.1.3 CO₂ injection state and temperature evolution

When the injection pressure is set as 30 MPa, the annual CO₂ storage amount is larger than 2 million tons, and the lateral flowing rate of CO₂ increases over time from 7 kg/s to 13 kg/s. The lateral flowing rate of CO₂ highly depends on the injection pressure and increases over time [44]. When the injection pressure is decreased to 25.8 MPa (base case 1), i.e., 4.2 MPa less than that in the base case, the lateral migration rate of CO₂ in the reservoir obviously declines. Under this injection condition, the annual storage amount of CO₂ is approximately 0.4 million tons. The CO₂ flow rate in the lateral direction is 4–10 kg/s less than that in the base case (Fig. 18). The minor difference in injection pressure (i.e., 0.2 MPa) between base case 1 and case 7 results in an obvious difference in the CO₂ flow rate of 1 kg/s. When CO₂ is injected at the well bottom located at the upper part of the aquifer (cases 1–1), a much larger CO₂ flowing rate (i.e., by more than two times) than that in case 7 occurs under the same injection pressure and temperature conditions (i.e., 26 MPa and 25 °C).

By comparing the temperature changes in the extracted geothermal fluid at the production wells under different production strategies, it can be observed that a decreased injection pressure (see base case 1 and case 7) can greatly inhibit the cooling effect surrounding the production wells (Fig. 19). This inhibition is due to the late CO₂ breakthrough and cooled geothermal water at the production well. The CO₂ breakthrough occurs much earlier than the thermal breakthrough under high CO₂ injection rates, and the flowing temperature at the production well increases over the 12 years of production. Under the same injection pressure, when the CO₂ injection well is located at the upper part of the reservoir, colder CO₂ is injected, which is unfavorable compared to the geothermal production caused by much earlier thermal breakthrough. However, when the injection pressure decreases and the distance between production wells increases, the thermal breakthrough is greatly inhibited, and the reservoir temperature declines by less than 5 °C even though much more CO₂ is injected compared with that in case 7. Under low reinjection rates (2 kg/s), the cooled geothermal water reinjected at the same layer does not reach the production well. Thus, a similar variation trend of the fluid temperature at the production well occurs in the base case and case 2.

5.1.4 Impact of the bottom boundary on the geothermal field

When a constant heat flux (i.e., 54 mW/m²) is set at the bottom boundary of the geothermal reservoir, the production fluid temperature only shows minor changes (Fig. 20), indicating that its contribution to the temperature field is negligible.

5.2 Cooling effect on the geothermal reservoir by commercial-scale hot water extraction and reinjection

5.2.1 Pressure response

In most geothermal fields, the reinjected mass includes geothermal waste fluid (i.e., brine and condensates) and additional water, such as river water [45], treated waste water [46], and supplementary water [47]. CO₂ is not injected into geothermal reservoirs with a low proportion of CO₂ production. When the production rate is increased to a commercial scale (100 m³/h), the reservoir pressure buildup at the reinjection wells and pressure decrease at the production wells are very obvious (Fig. 21). The pressure response to the injection of cooled return water is rapid, which has been proven by many injection tests, including that at Well HE-04 of Reykjanes in Iceland with a pore pressure increase of 0.1 MPa after 2 h of water injection and an injection rate increase of 50% [48]. A water level increase also occurred in an Otake geothermal field within a short period after water reinjection into the reservoir [49]. The water level will obviously increase in Tianjin due to the commercial-scale reinjection of cold water [50].

5.2.2 Temperature changes in the geothermal reservoir

Because of the reinjection of cooled geothermal return water, the reservoir is substantially cooled at the regions surrounding the reinjection well, and the cold region increases over time (Fig. 22). However, the short-term monitoring results (less than 1 month) imply that there is no decrease in the reservoir temperature at the production well [51] because no thermal breakthrough occurs at the production well within several days. Thermal breakthrough may occur in the production well (500 m in distance) after a period of production (months or years) depending on the well configuration and reinjection and production rates. The reinjection of cooled geothermal water in the Xiaotangshan geothermal field in China showed a temperature decrease of 7 °C in the geothermal reservoir after 10 years of reinjection. Here, in the numerical simulation, after 50 years of geothermal production, the maximum temperature decrease in the reservoir at the production well (pro1) is approximately 20 °C during the full cooled water reinjection treatment. The vertical temperature profile along the hot water production well is shown in Figure 23.

5.2.3 Injection rate effects on the temperature evolution

The comparison between the low and high reinjection rates of cooled geothermal water (case 5 vs. case 6) demonstrates that the reservoir temperature is not negatively affected in the short term (less than 5 years). A minor temperature decrease of 2 °C appears after 10 years, while the maximum temperature difference is approximately 8 °C after 20 years of water reinjection (Fig. 24).

5.2.4 Thermal conductivity effect on the temperature evolution

The increased thermal conductivity in the production aquifer plays a very minor role in the temperature profile along the production well. The simulation shows that less than 1 °C of a temperature difference occurs after 30 years of geothermal production (Fig. 25). This finding implies that compared with other factors, including the fluid injection and production rates, fluid injection temperature, well configuration, etc., the contribution of thermal conductivity to the temperature is negligible.

Therefore, considering that this information includes reservoir properties, production and injection conditions, reinjection strategies, and the responses of geothermal reservoirs from past experience with reinjection practices [41], the comprehensive comparison results are listed in the above sections. When the water production and water reinjection rates are low (2 kg/s), their impact on the fluid temperature at the production well is negligible, and thermal breakthrough is mainly caused by the injection of cold CO₂ (see the results of the base case, base case 1, case 1, etc.). A comparison of these cases (with and without CO₂ injection in case 5 and case 6) shows that the temperature decrease is very limited (<0.5 °C) when the production rate is very low (2 kg/s) and that the reinjection enthalpy is 125.75 kJ/kg. However, when the production and reinjection rates are high (27.8 kg/s), even without CO₂ reinjection into the reservoir, the thermal breakthrough caused by the water reinjection is very strong. The temperature decrease in the production water reaches 5.4 °C, 9.2 °C and 11.3 °C after 20, 30, and 50 years of geothermal production, respectively. When the injection temperature of the CO₂ increases to 25 °C, the produced fluid temperature increases compared with that in the initial state because of the much higher flowing enthalpy. The impact of the reservoir thermal conductivity on the temperature of the extracted fluid is negligible when the production rate is high.

5.3 Methods inhibiting early CO₂ breakthrough

From an operational perspective, the main purpose of CO₂ injection into the geothermal reservoir is to increase the efficiency of hot water production. Therefore, the CO₂ injection pressure can be decreased to inhibit early CO₂ breakthrough (e.g., base case 1 and case 7) as shown in Figures 26 and 27. Thus, the win–win effect of CO₂ injection into the geothermal reservoir can be achieved by enabling the storage of CO₂.

If the CO₂ injection amount is not decreased, the lateral distance of the production wells can be increased to prolong the CO₂ breakthrough. We performed a simulation test at a lateral distance of 1 km between the CO₂ injection point and the hot water production point while maintaining the other conditions the same as in the base case. Compared with the base case (Fig. 7), it is found that no CO₂ breakthrough occurs after 30 years of CO₂ injection when the lateral distance between the CO₂ injection point and the hot water production point is increased to 1 km (Fig. 28).

6 Summary and conclusion

Serious air pollution has accelerated the need to change the energy consumption structure in China, providing a good opportunity for the development of renewable energy, especially deep geothermal energy. Win–win CO₂-aided geothermal energy strategies can be applied in low-medium enthalpy hydrothermal reservoirs to simultaneously maintain the pressure balance, inhibit ground subsidence to some extent, and reduce CO₂ emissions.

A new mode consisting of a CO₂ injection, hot water production, and cooled geothermal return water reinjection system is presented in this paper. Based on numerical studies applying CO₂-AGES to the low-medium enthalpy hydrothermal reservoirs of the Guantao formation using the conceptual geological model of a small region in Binhai district in Tianjin, the following main conclusions can be drawn:

1. CO₂ injection is feasible for maintaining the reservoir pressure. However, a suitable injection strategy (i.e., injection pressure and temperature, well configuration, etc.) should be applied to prolong hot water production. CO₂ breakthrough occurs much earlier than thermal breakthrough, demonstrating that the production temperature increases even after CO₂ breakthrough occurs.

2. When the injection pressure decreases from 30 MPa to 26 MPa, the CO₂ flow rate rapidly decreases, which is beneficial for hot water extraction at the production well (late CO₂ breakthrough).

3. Pore pressure buildup occurs in the overlying caprocks, but the CO₂ front stops at the bottom of the caprock layer. The pressure front is much larger than the CO₂ plume front.

4. After CO₂ breakthrough at the production well, during the two phase fluid (i.e., CO₂ + water) flow period, the temperature of the extracted fluid increases compared with that under the pure hot water production condition. This phenomenon lasts for several years before a sharp decrease in the fluid temperature at the production well is induced by the production of pure CO₂.

5. The thermal breakthrough induced by the reinjection of cooled geothermal water is very limited and caused by a low water reinjection rate.

6. Water production and reinjection rates greatly control the degree of the cooling effect in geothermal reservoirs. The extracted water temperature greatly decreases after long-term geothermal production when the hot water production and cooled water reinjection rates increase from 2 kg/s to 27.8 kg/s.

CO₂-aided geothermal extraction technology is only at the early research stage, and more work is required before its commercial application. The cooling effects of a geothermal reservoir characterized by multiple thin layers segmented by complex fault or fracture systems require further investigation. Tracer tests are also required to understand the flow channels of reinjected return water. These issues will be considered in future works.

Acknowledgments

This research was funded by the CAS Pioneer Hundred Talents Program and National Natural Science Foundation of China (Nos. 51809259 and 51774056).

References

All Tables

All Figures