Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2021 Dec 6;6(50):34832–34841. doi: 10.1021/acsomega.1c05475

Experimental Analysis and Numerical Simulation of the Stability of Geological Storage of CO2: A Case Study of Transforming a Depleted Gas Reservoir into a Carbon Sink Carrier

Yang Sun 1,*, Renyi Lin 1, Yi Pan 1, Lei Sun 1, Yong Tang 1
PMCID: PMC8697398  PMID: 34963966

Abstract

graphic file with name ao1c05475_0017.jpg

Geological storage of CO2 is one of the most economical, feasible, and effective measures to slow down global warming. In this study, a combined long core model was designed to study the seepage characteristics of supercritical CO2 displacement. Moreover, the stability of permanent storage of cushion gas layers formed by supercritical CO2 injection has been systematically studied. The research results showed that in the supercritical temperature and pressure range of CO2, the front edge of CO2 displacement can form a relatively stable seepage zone. Supercritical CO2 displacement can achieve a high gas-storage rate and stable CO2 storage. At the same time, the recovery rate of remaining natural gas has been significantly improved. As the injection pressure increased, supercritical CO2 inhibited the reverse diffusion of natural gas molecules. Therefore, the breakthrough of the supercritical CO2 displacement front under a high pressure lagged behind. However, due to the increase in the density difference of gas molecules, the forward diffusion of supercritical CO2 has been enhanced. Temperature will not significantly affect the displacement and storage effects of supercritical CO2 in gas reservoirs. The increase in injection pressure and reasonable control of the injection rate can delay the breakthrough of supercritical CO2 displacement. These measures are conducive to the stable storage of CO2 and the improvement of remaining natural gas recovery. The implementation of CO2 geological storage is suitable for the later stage of gas reservoir depletion development. The high-density gravitational heterogeneity of supercritical CO2 enables the injected CO2 to form a stable high-density cushion gas layer in the gas reservoir, which can achieve stable CO2 storage for more than 100 years.

1. Introduction

CO2 storages include ocean storage, chemical storage, and geological storage.1−10 Geological storage of CO2 is one of the most economical, feasible, and effective measures to slow down global warming.1−4 The concept “carbon neutral” that promotes low-carbon development has been widely accepted by people.5−7 Carbon neutral is an ecosystem concept proposed to achieve the goal of zero net greenhouse gas emissions. The ways to achieve carbon neutral include afforestation, energy saving and emission reduction, and the development of low-carbon energy.8−14 These measures can be used to balance the amount of carbon released and absorbed back to the earth, so that the “zero emission” of CO2 can be achieved.15−19

Broadly speaking, all geological structures with good sealing properties can be used as CO2 storage sites,1−3,10−24 such as oil reservoirs, gas reservoirs, brine beds, coal beds, and unconventional resource reservoirs (methane hydrate reservoirs, etc.). The previous studies have successfully conducted some pilot experiments and demonstration projects for improving oil recovery through CO2 injection. These studies provided valuable experience for the implementation of CO2 geological storage. For example, the United Nations Agency for International Development has implemented a project in the Weyburn Oilfield to improve the recovery of remaining oil through CO2 injection and it realized the geological storage of CO2.1,2 China’s Qinshui Basin used the CO2 injection technology to improve coal bed methane recovery and achieve CO2 storage. For CO2 storage in saltwater aquifers, the famous Norwegian Sleipner offshore CO2 capture and storage project was employed. In addition, some other research studies have also explored the storage of CO2 in unconventional reservoirs such as natural gas hydrates.8 Some studies have explored the prospects of using CO2 to recover natural gas from methane hydrate reservoirs.25−28 On the whole, there were relatively few studies on geological storage of CO2 in gas reservoirs. Moreover, the existing research studies on CO2 geological storage in gas reservoirs mainly focused on the storage efficiency, and there were few discussions on the storage stabilities.1−3,29,30

In this study, a candidate gas reservoir that was compatible with the CO2 supercritical temperature and pressure range was selected, so the depleted gas reservoir selected was an independent gas reservoir. The independent gas reservoir has good sealing properties, and the formation temperature and pressure can enable CO2 to be stored in the form of supercritical fluid. A combined long core model was designed to study the seepage characteristics of supercritical CO2 displacement. Moreover, the stability of permanent storage of cushion gas layers formed by supercritical CO2 injection has been systematically studied. This study can provide a reference for the implementation of CO2 geological storage in late-development stage or depleted mid-shallow buried gas reservoirs.

2. Materials and Methods

2.1. Natural Gas Displacement Experiments by Supercritical CO2 (SCCO2)

The one-dimensional composite long core model was composed of 14 gas reservoir cores with a diameter of 2.5 cm and a length of 5.0–7.0 cm with a total length of 92.42 cm. In theory, the core of the core displacement experiment can be as long as possible. However, the long core displacement device developed by the American company Ruska used in this experiment was limited by the size of the oven and the length of the long core holder, and the total length of the combined long core cannot exceed 100 cm. All cores were taken from actual depleted gas reservoirs. The combination order of each short core was determined by the length-weighted average; and two adjacent cores were connected with a filter paper to eliminate “end effects.”

The long core model has a total pore volume of 65.674 cm3, an average porosity of 14.72%, and an average rock permeability of 11.05 × 10–3 μm2. The device for the long core gas displacement experiment is shown in Figure 1. A constant pressure and constant speed pump was used for gas injection and displacement, while pressure sensors and back pressure valves were used for production fluid control and metering.

Figure 1.

Figure 1

Open in a new tab

Experimental device for long core gas displacement under a high temperature and a high pressure. (1) Four-way shut off valve; (2) oil piston cylinder; (3) SCCO2 piston cylinder; (4) natural gas piston cylinder; (5) four-way shut off valve; (6) advection pump; (7) three-way shut off valve; (8) ring pressure valve; (9) long core holder; (10) constant temperature system; (11) three-way shut off valve; (12) gas flow meter; (13) gas–liquid separation device; (14) electronic balance instrument; (15) gas flow meter; (16) computer; and (17) vacuum pump.

The natural gas used in the displacement experiment was taken from the middle-shallow buried gas reservoir M in the Sichuan Basin, China. The original parameters of the target reservoir were as follows: a buried depth of 1000 m, a formation temperature of 35 °C, an initial formation pressure of 10 MPa, a water saturation of 30%, and an original natural gas reserve of 20.45 × 108 m3. In addition, the mol % composition of CO2, N2, C1, C2, C3, IC4, NC4, IC5, NC5, and FC6 in natural gas was 4.528, 4.303, 90.014, 1.057, 0.083, 0.007, 0.008, 0.004, 0.002, and 0.004, respectively. The experimental CO2 was industrially pure CO2.

The procedures of long core gas displacement experiments are as follows:

  • (1)

    The temperature set in the experiment was consistent with the actual temperature of the formation, which was 35 °C. In addition, the formation pressure of the depleted gas reservoir was 10 MPa, and the overlying formation pressure was 25 MPa. Afterward, the confining pressure of the core was set to be consistent with the overlying formation pressure. In order to compare the relationship between supercritical CO2 and natural gas storage in actual gas reservoirs in the early and late stages of gas injection, the experimental pressures were set to 10 and 20 MPa, respectively. When the injection pressure was 10 MPa, the outlet pressure was controlled at about 8 MPa; and when the injection pressure was 20 MPa, the outlet pressure was controlled at about 18 MPa.

  • (2)

    Supercritical CO2 injection and continuous dry gas displacement were performed under the conditions of a temperature of 35 °C and pressures of 10 and 20 MPa, respectively. Furthermore, the extent and stability of the front-edge displacement of supercritical CO2 injection were evaluated by testing the composition of CO2 and its displacement efficiency changes at the outlet.

  • (3)

    First, 0.2 PV supercritical CO2 was injected for dry gas displacement at a temperature of 35 °C and pressures of 10 and 20 MPa. Then, the production end valve was closed and the constant pressure “simmering well” was maintained for 8–12 h. Finally, dry gas displacement with supercritical CO2 injection under a constant pressure was conducted, and the composition of CO2 and its displacement efficiency changes were measured at the outlet end. The test results can be used to evaluate the degree of convection and diffusion of the displacement front in the “simmering” stage. In addition, the test can also provide basic data (data on changes in CO2 composition, breakthrough time, and displacement efficiency at the outlet) for the numerical simulation of the stability of CO2 geological storage.

In actual operation, the pipelines used to connect different devices were very short, thereby reducing the influence of dead volume and ineffective volume.

2.2. One-Dimensional Interwell Gas Displacement and Seepage Model Considering Convection and Diffusion

In order to further understand the seepage mechanism of supercritical CO2 displacement, the Computer Modelling Group (CMG) component diffusion equation of the multiphase multicomponent system suitable for describing the multiphase flow and viscous fingering phenomenon of CO2 displacement was introduced.20−22 This method was used to evaluate the percolation properties and storage stability of supercritical CO2 displacement in interwell reservoirs. The constructed one-dimensional interwell reservoir core model had a total length of 500 m, and the grid was divided into I × J × K = 30 × 1 × 1. Since there is no drastic phase change in the supercritical CO2 displacement process, therefore, the mesh size is not sensitive to the results. That is, the solution of equally dividing the grids is reasonable. The average porosity was 14.72%, the horizontal permeability was 11.05 × 10–3 μm2, and the rock compressibility was 10–6 MPa–1. The natural gas used in the model was collected from the M natural gas reservoir in the Sichuan Basin, and the injection medium was industrially pure CO2. A constant pressure injection well and a constant pressure production well were arranged at the first grid block and the last grid block, respectively. They were used to simulate a supercritical CO2 injection well and a natural gas production well in the gas displacement experiment.

2.3. Numerical Simulation of the Stability of Supercritical CO2 Storage

In order to further verify the reliability of implementing supercritical CO2 stable storage in actual gas reservoirs, in this study, taking the M gas reservoir in the Sichuan Basin as an example, a three-dimensional well group model of the gas reservoir was constructed. In the model, two gas injection wells, three gas production wells, and a nonuniform 5-point well pattern were used, and all the wells were vertical wells. The selected gas reservoir was located in a complete anticline structure, which was covered by tight mudstone cap rocks. The average reservoir depth was 1000 m, the formation temperature was 35 °C, and the initial formation pressure was 10 MPa; the water saturation was 30% and the original geological reserve of natural gas was 20.45 × 108 m3. The pressure parameters were automatically adjusted with the gas density according to the designed base surface pressure. In addition, the constructed heterogeneous geological model used corner grids, the total number of grids was 50 × 60 × 6 = 18,000, and the plane grid size was 50 m × 50 m.

3. Results and Discussion

3.1. Seepage Stability of the Front Edge of Supercritical CO2 Displacement

Supercritical CO2 has high-density properties, which makes both the diffusion of supercritical CO2 in natural gas and the diffusion of natural gas in supercritical CO2 weakened. Furthermore, this enables supercritical CO2 to have strong piston-type gas displacement and stable gas-storage capabilities.

3.1.1. Comparison of Seepage Characteristics of Displacement Front by Continuous Supercritical CO2 Injection

Figure 2a shows the relationship between the concentration of CO2 and the injection volume multiples at the outlet in the first experimental procedure. When the displacement pressure was 10 MPa, the gas injection volume at the CO2 breakthrough point was close to 0.8 PV, and when the gas injection volume reached 1.1 PV, the concentration of CO2 in the outlet gas reached 100%. Therefore, the transition zone formed by the displacement front edge was about 0.3 PV. Similarly, when the displacement pressure was 20 MPa, the breakthrough point of continuous natural gas displacement by supercritical CO2 was slightly lagging behind 10 MPa, which was about 0.85 PV. When the supercritical CO2 injection volume reached 1.2 PV, the outlet gas CO2 concentration reached 100% and the displacement front transition zone was close to 0.3 PV.

Figure 2.

Figure 2

Open in a new tab

Relationship between supercritical CO2 mol %, displacement efficiency, and injection volume PV in the produced gas under different displacement pressures. (a) Change of CO2 content in the produced gas in the outlet and (b) cumulative recovery degree of natural gas by CO2 displacement.

According to the above results, a stable displacement front and a transition zone can be formed in the supercritical CO2 displacement process. Although the shapes of the transition zone curves after the breakthrough of supercritical CO2 under different displacement pressures were different, the widths of the transition zones did not change significantly. This showed that the CO2 displacement process was similar to the displacement of oil by water, both of which had relatively stable displacement fronts. In addition, it can be seen from Figure 2b that when the supercritical CO2 injection reached 1.2 PV, the natural gas production in the two groups of experiments reached 100%. This showed that supercritical CO2 displacement can achieve high gas-storage rate and stable CO2 storage. At the same time, the recovery rate of the remaining natural gas has been significantly improved.

3.1.2. Convection and Diffusion Degree of the Displacement Front

Figure 3 shows the relationship between the CO2 concentration in the produced gas and the injection volume multiples of CO2 at different injection pressures in the second experimental procedure. In the experiment, first, 0.2 PV of supercritical CO2 was injected into the core at 35 °C and two injection pressures of 10 and 20 MPa. Then, the valve at the production end of the long core was closed, the injection pressure was kept constant, and the well was simmered for more than 8 h. During this process, convection and diffusion occurred in the core of the supercritical CO2 displacement front. After that, the production end valve was opened, and the natural gas displacement was continued through supercritical CO2 injection. Finally, the CO2 gas concentration at the outlet end of the long core and the CO2 injection volume multiples were detected.

Figure 3.

Figure 3

Open in a new tab

Relationship between the CO2 concentration in the produced gas and the injection volume multiples of CO2 at different injection pressures.

According to the experimental results, under the condition of 10 MPa injection pressure, the supercritical CO2 displacement front achieved a breakthrough at 0.75 PV and reached a stable state at 1.05 PV. The width of the formed transition zone was about 0.3 PV (Figure 3). Under the condition of 20 MPa, the breakthrough of the supercritical CO2 displacement front was about 0.1 PV behind that under the condition of 10 MPa, and the width of the transition zone was about 0.35 PV (Figure 3). This was because as the injection pressure increased, the density, van der Waals force, and viscosity of supercritical CO2 increased, and supercritical CO2 inhibited the reverse diffusion of natural gas molecules. Therefore, the breakthrough of the supercritical CO2 displacement front under a high pressure lagged behind. However, due to the increase in the density difference of gas molecules, the forward diffusion of supercritical CO2 has been enhanced.

Combining the analysis of gas displacement experiment data and previous studies, it can be known that supercritical CO2 displacement could be regarded as “miscible displacement,” and the relative permeability curves satisfy the diagonal relationship of ideal miscible flooding.31−37

3.2. Analysis of Controlling Factors for Stable Storage of Supercritical CO2

When CO2 geological storage was implemented in an abandoned or late-development stage gas reservoir, the main factors affecting the stability of CO2 storage could be the temperature and pressure of the gas reservoir and the migration speed of supercritical CO2 in the reservoir. Changes in temperature and pressure may affect the supercritical properties of CO2 and in turn affect the percolation behavior of the process of CO2 displacement. However, the rate of CO2 injection may affect the dynamic dispersion process between the supercritical CO2 zone and the natural gas zone.

The basic conditions for calculation of the 500 m one-dimensional displacement model are listed in Table 1.

Table 1. Basic Conditions for Calculation of the 500 m One-Dimensional Displacement Model.

calculation model design conditions goal
500 m one-dimensional displacement model simulation corrected by displacement experimental data (the calculation results were magnified 108 times). 20, 35, and 60 °C under 10 MPa. influence of temperature on the gas migration process under the current formation pressure.
3, 5, 10, and 20 MPa at a temperature of 35 °C. influence of pressure on the gas migration process under the formation temperature.
1.16 and 0.92 m3/d displacement rates at a temperature of 35 °C. influence of different injection rates on the displacement process under the formation temperature.
Open in a new tab

In Table 1, the first case was conducted at the current formation pressure of 10 MPa. The influence of temperature changes on underground gas migration was studied by changing the temperature to 20, 35, and 60 °C. Changing the temperature is not to heat the formation but to consider that the conditions for underground supercritical CO2 storage are within the temperature and pressure range. This study is an expanded study for the selection of other possible potential storage sites.

The second case was conducted by changing the pressure at the formation temperature. The purpose is to study the impact of formation pressure recovery on gas migration as CO2 continues to be injected.

In the third case, the influence of the injection rate on gas migration in the formation was studied by continuously changing the CO2 injection rate under the formation pressure and temperature. The simulation data in this study were obtained based on the gas displacement experiments. The displacement experiments adopted the method of constant pressure injection, and the injection rates of supercritical CO2 were uncertain values. After CMG software was used to correct the 92 cm one-dimensional displacement model with the same length of the core according to the core displacement experimental data, the constant pressure average injection rates were calculated: 4.8 mL/h without holding pressure and 3.8 mL/h with holding pressure. There is a certain simulation calculation error here. Then, all the data were enlarged to 108 times, and the gas displacement rates of the 500 m one-dimensional model were obtained, which were 1.16 and 0.92 m3/d, respectively.

3.2.1. Gas Reservoir Temperature

Figures 4 and 5 show the stability characteristics of the supercritical CO2 displacement front at three temperature points of 20, 35, and 60 °C under a 10 MPa injection pressure. It could be found that the breakthrough point of liquid CO2 at 20 °C was 0.9 PV, while the breakthrough point of supercritical CO2 displacement at 35 and 60 °C was 0.8 PV (Figure 4). The range of the transition zone tested at different temperatures was similar, which was 0.3 PV. However, the changing trends of the CO2 content in the produced gas in different tests had a certain difference (Figures 4 and 5). On the whole, temperature will not significantly affect the displacement and storage effects of supercritical CO2 in gas reservoirs.

Figure 4.

Figure 4

Open in a new tab

Relationship between the CO2 concentration in the produced gas and the injection volume multiples of CO2 at different temperatures. The length of the one-dimensional test core model was 500 m, and the injection pressure was 10 MPa.

Figure 5.

Figure 5

Open in a new tab

Relationship between the cumulative gas-storage rate of SCCO2 and the injection volume multiples of CO2 at different temperatures. The length of the one-dimensional test core model was 500 m, and the injection pressure was 10 MPa.

3.2.2. Gas Reservoir Pressure

Figures 6 and 7 show the stability characteristics of the supercritical CO2 displacement front at four injection pressure points of 3, 5, 10, and 20 MPa at a temperature of 35 °C. From the CO2 concentration curves in the produced gas, it can be found that the injection pressure had a certain impact on the stable seepage effect of CO2 displacement (Figure 6). In the low pressure range of 3 and 5 MPa in the nonsupercritical region, the breakthrough point of CO2 displacement with gaseous characteristics was about 0.65 PV, and the transition zone from gaseous CO2 to natural gas was long. Especially when the injection pressure was 3 MPa and only when the CO2 injection reached 1.2 PV, the CO2 concentration in the produced gas reached 99%. This showed that the gaseous CO2 at the displacement front diffused violently in natural gas, and its seepage stability was weakened. However, in the injection pressure range of 5–20 MPa, that is, in the near-critical region of CO2, especially in the supercritical region, the emergence of the CO2 breakthrough point will lag significantly, and the transition zone will also become shorter. This indicated that the diffusion degree of supercritical CO2 in natural gas was weakened, while its seepage stability was enhanced. The gas-storage rates of supercritical CO2 in different injection pressure tests were high and there were little differences. This showed that even if long-distance displacement was implemented in the gas reservoir, CO2 can still effectively displace natural gas.

Figure 6.

Figure 6

Open in a new tab

Relationship between the CO2 concentration in the produced gas and the injection volume multiples of CO2 at different injection pressures. The length of the one-dimensional test core model was 500 m, and the test temperature was 35 °C.

Figure 7.

Figure 7

Open in a new tab

Relationship between the cumulative gas-storage rate of SCCO2 and the injection volume multiples of CO2 at different injection pressures. The length of the one-dimensional test core model was 500 m, and the test temperature was 35 °C.

3.2.3. Gas Injection Speed

Figures 8 and 9 show the supercritical CO2 displacement curves of a 500 m-long one-dimensional homogeneous model at 35 °C and different displacement speeds.

Figure 8.

Figure 8

Open in a new tab

Relationship between the CO2 concentration in the produced gas and the injection volume multiples of CO2 at different injection speeds. The length of the one-dimensional test core model was 500 m, and the test temperature was 35 °C.

Figure 9.

Figure 9

Open in a new tab

Relationship between the cumulative gas-storage rate of SCCO2 and the injection volume multiples of CO2 at different injection speeds. The length of the one-dimensional test core model was 500 m, and the test temperature was 35 °C.

The comparison results showed that as the injection rate increased, the CO2 breakthrough point appeared in advance, and the transition zone became wider (Figure 8). In addition, as the injection rate increases, the displacement efficiency and the gas-storage rate of supercritical CO2 will also decrease (Figure 9). However, for the one-dimensional homogeneous model, the effect of speed changes on the percolation stability of supercritical CO2 displacement was not particularly obvious.

The numerical simulation model has been corrected based on experimental data. According to the simulation results, it can be known that in the supercritical temperature and pressure range, the displacement front of supercritical CO2 can form relatively stable seepage, which can meet the requirements of stable storage of supercritical CO2. Moreover, the increase in the injection pressure and reasonable control of the injection rate can delay the breakthrough of supercritical CO2 displacement. These measures can strengthen the stability of the supercritical CO2–natural gas transition zone and are conducive to the stable storage of CO2 and the improvement of remaining natural gas recovery.

3.3. Stability Analysis of Supercritical CO2 Storage Based on Injection–Production Well Group Simulation

The gas reservoir used in this study had a complete anticline structure, and the upper and lower parts of the gas reservoir were sealed by tight layers. It is a closed antirhythm gas reservoir, and this study ignores the possibility of gas leakage in the periphery of the gas reservoir.

3.3.1. Injection–Production Design Model for Supercritical CO2 Storage in Gas Reservoirs

In this study, CMG software was used for the modeling of the target layer and the numerical simulation of supercritical carbon dioxide displacement. The porosity and permeability distributions of gas reservoirs are shown in Figure 10. The central part of the gas reservoir structure has good porosity and permeability, while the porosity and permeability at the edges become poor. The target layer has an antirhythmic geological structure with fine grains in the upper part and coarse grains in the lower part. Figure 11 shows the porosity and permeability profiles of the three-dimensional geological model with antirhythmic structures of the target layer. Supercritical CO2 displacement is similar to piston displacement, and its relative permeability curve satisfies a diagonal relationship.

Figure 10.

Figure 10

Open in a new tab

Planar distribution of porosity and permeability of the target layer of the M gas reservoir. (a) Porosity distribution and (b) permeability distribution.

Figure 11.

Figure 11

Open in a new tab

Porosity and permeability profiles in the target layer of the M gas reservoir. The target layer has an antirhythmic geological structure with fine grains in the upper part and coarse grains in the lower part. (a) Porosity profile and (b) permeability profile.

In the gas reservoir model, two CO2 injection wells were designed at the lower part on both sides of the long-axis direction, and three gas production wells were designed along the short-axis direction at the upper part of the gas reservoir structure (Figure 12). The lower quarter of the production horizon in the long-axis direction was opened for CO2 injection, while the upper quarter of the production horizon in the short-axis direction was opened for natural gas production. The adoption of this method of well deployment can reduce the mixing degree of CO2 and natural gas, thereby better improving the displacement effect of supercritical CO2.

Figure 12.

Figure 12

Open in a new tab

Geological model of the M gas reservoir and its 5-point well pattern. IN1 and IN2 are the injection wells, and PR1, PR2, and PR3 are the production wells.

In this study, the M gas reservoir used a depletion development method from the original pressure of 10 to 3 MPa, and then CO2 was injected to increase the recovery rate of the remaining natural gas and achieve CO2 storage. In the depletion development stage, all five wells produced gas and all layers were opened. The initial single well gas production rate was 15 × 104 m3/d. When the formation pressure was reduced to 5 MPa using the depletion development method, the two production wells on the side of the long axis were subjected to critical CO2 injection, which opened the lower part of the production horizon. The results showed that the initial daily CO2 injection volume of the injection well was 15 × 104 m3, and the initial daily gas production volume of the production well was 10 × 104 m3. When CO2 breakthroughs (the CO2 concentration of the production well reached 10%), the production wells were closed. Then, the daily CO2 injection volume of the two injection wells was increased to 50 × 104 m3. When the formation pressure returned to the original formation pressure, the supercritical CO2 injection from the injection wells was stopped. Thus, permanent storage of supercritical CO2 can be realized. The calculations in this simulation were continued for 100 years after the wells were closed. The continuous injection time of supercritical CO2 was 13.5 years, and the total duration of storage stability prediction was 100 years.

3.3.2. Enhanced Oil Recovery, Breakthrough Time, and Storage Effect of Supercritical CO2 Injection

This study showed that the time from the start of supercritical CO2 injection to the breakthrough of CO2 in the production well was 3 years and 8 months. This indicated that after supercritical CO2 was injected for the formation, it will not rapidly mix with the formation natural gas to form a supercritical CO2 channeling effect. The prediction results of the natural gas recovery factor during the supercritical CO2 displacement process are shown in Table 2. The results showed that when supercritical CO2 storage was implemented in depleted gas reservoirs, the remaining natural gas recovery rate could reach 12.65% and the CO2 storage capacity could reach 9.02 million tons.

Table 2. Natural Gas Recovery of Supercritical CO2 Displacement in Different Stages.
injection–production scheme recovery factor in depletion stage recovery factor of CO2 displacement total recovery of depletion + CO2 displacement final storage of supercritical CO2 (t)
CO2 injection implementation at a depletion pressure of 3 MPa 72.83% 12.65% 85.48% 9,021,276
Open in a new tab

For depleted gas reservoirs, most of the natural gas in the formation was recovered, and the potential effective remaining pore space for supercritical CO2 storage was increased. Therefore, the effect of CO2 storage in depleted gas reservoirs will be enhanced, and the total amount of CO2 storage will be large. The lower the abandonment pressure of the gas reservoir, the greater the total amount of supercritical CO2 stored.

3.3.3. Stability Analysis after 100 years of Supercritical CO2 Storage

In order to better understand the stability of supercritical CO2 storage in the long-term storage process, the influences of the gravity differentiation and density difference of the gas reservoir on the migration of supercritical CO2 at the end of supercritical CO2 injection and at the end of 100 years were analyzed. Figures 13–15 show the simulation results of the CO2 molar fraction, formation gas density, and fluid flow vector after the supercritical CO2 injection was completed and after 100 years of CO2 storage, respectively.

Figure 13.

Figure 13

Open in a new tab

CO2 mole fraction and formation gas density distribution in the target layer at the end of supercritical CO2 injection. (a) CO2 mole fraction and (b) formation gas density.

Figure 15.

Figure 15

Open in a new tab

Formation fluid flow vector profiles at the end of supercritical CO2 injection and when supercritical CO2 has been stored for 100 years. (a) At the end of supercritical CO2 injection and (b) after supercritical CO2 has been stored in a gas reservoir for 100 years.

It can be seen from Figure 13 that the gravity differentiation and convection diffusion between supercritical CO2 and natural gas occurred under the influence of density difference. Furthermore, supercritical CO2 migrated to the lower and bottom of the gas reservoir, spontaneously and gradually, to form a “cushion gas layer”. At the same time, the remaining natural gas migrated to the top of the upper central zone of the gas reservoir, forming a relatively stable “gas cap.” It can be seen from Figure 13b that the supercritical CO2 cushion layer and the natural gas top layer were formed due to gravity differentiation.

According to the calculation results in Figures 14 and 15, when CO2 has been stored for 100 years, the supercritical CO2 and remaining natural gas in the formation will not reach a completely miscible mixed state. The cushion gas layer formed by supercritical CO2 at the edge and bottom of the gas reservoir can still be very stable, and the fluids in the middle and lower parts of the cushion gas layer do not flow to the top of the gas reservoir. In addition, the transition zone between the supercritical CO2 and the remaining natural gas in the upper and middle parts of the gas reservoir also remain stable. This showed that the gas flow caused by gravity differentiation and convection diffusion between supercritical CO2 and natural gas will became slower and slower as the storage time increased. Therefore, the migration of supercritical CO2 to the top cover of the structure was finally transformed into molecular diffusion, which made the diffusion much more slower. Then, the possibility that the supercritical CO2 cushion gas layer “impacts” upward into the gas reservoir became less. Ultimately, the time for the stable storage of supercritical CO2 will be extended to a great extent.

Figure 14.

Figure 14

Open in a new tab

CO2 mole fraction and formation gas density distribution in the target layer when supercritical CO2 was stored in a gas reservoir for 100 years. (a) CO2 mole fraction and (b) formation gas density.

4. Conclusions

  • (1)

    In this study, a combined long core model was designed to study the seepage characteristics of supercritical CO2 displacement. The results showed that in the supercritical temperature and pressure range of CO2, the front edge of CO2 displacement can form a relatively stable seepage zone.

  • (2)

    Supercritical CO2 displacement can achieve a high gas-storage rate and stable CO2 storage. At the same time, the recovery rate of the remaining natural gas has been significantly improved. As the injection pressure increased, supercritical CO2 inhibited the reverse diffusion of natural gas molecules. Therefore, the breakthrough of the supercritical CO2 displacement front under a high pressure lagged behind. However, due to the increase in the density difference of gas molecules, the forward diffusion of supercritical CO2 has been enhanced.

  • (3)

    Temperature will not significantly affect the displacement and storage effects of supercritical CO2 in gas reservoirs. The increase in the injection pressure and reasonable control of the injection rate can delay the breakthrough of supercritical CO2 displacement. These measures are conducive to the stable storage of CO2 and the improvement of the remaining natural gas recovery.

  • (4)

    The implementation of the CO2 geological storage is suitable for the later stage of gas reservoir depletion development. The high-density gravitational heterogeneity of supercritical CO2 enables the injected CO2 to form a stable high-density cushion gas layer in the gas reservoir, which can achieve stable CO2 storage for more than 100 years.

Acknowledgments

This research was supported by the 13th Five-Year National Major Special Project of China (No.2016ZX05048-004-005).

The authors declare no competing financial interest.

References

  1. Wikipedia . Carbon capture and storage. http://en.wikipedia.org/wiki/Carboncaptureandstorage, 2012.
  2. Wikipedia . Carbon sequestration. http://en.wikipedia.org/wiki//Carbonsequestration, 2012.
  3. Fan C. H.; Li H.; Qin Q. R.; He S.; Zhong C. Geological conditions and exploration potential of shale gas reservoir in Wufeng and Longmaxi Formation of southeastern Sichuan Basin, China. J. Pet. Sci. Eng. 2020, 191, 107138 10.1016/j.petrol.2020.107138. [DOI] [Google Scholar]
  4. Zhang G. Y.; Yu Q. G.; Zhao G. B. Three-dimensional simulation of soft overburden ″two-zone″ height due to high-intensity mining in Shajihai Coal Mine. J. Min. Strata Control Eng. 2021, 3, 023519. [Google Scholar]
  5. Fan Z.; Cheng L. Visualization of polymer microsphere profile control and plugging seepage field and mine field test. Sci. Technol. Eng. 2012, 12, 7543–7546. [Google Scholar]
  6. Li X.; Fang Z.; Wei N. Discussion on the technical route of CO2 capture and storage in China. Rock Soil Mech. 2009, 9, 2674–2696. [Google Scholar]
  7. Diao Y.; Zhu G.; Jin X. Theoretical CO2 geological utilization and storage potential evaluation of the Sichuan Basin. Geol. Bull. 2017, 36, 1088–1095. [Google Scholar]
  8. Gao H.; Li H. Determination of movable fluid percentage and movable fluid porosity in ultra-low permeability sandstone using nuclear magnetic resonance (NMR) technique. J. Pet. Sci. Eng. 2015, 133, 258–267. 10.1016/j.petrol.2015.06.017. [DOI] [Google Scholar]
  9. Wen D.; Guo J.; Zhang S. Research progress on geological storage of carbon dioxide in China. Geol. China 2014, 10, 1716–1722. [Google Scholar]
  10. Chen G. B.; Li T.; Yang L.; Zhang G. H.; Li J. W.; Dong H. J. Mechanical properties and failure mechanism of combined bodies with different coal-rock ratios and combinations. J. Min. Strata Control Eng. 2021, 3, 023522. [Google Scholar]
  11. Li S. L.; Liu G. J.; Jia R. F.; Xue F.; Wang K. Y. Study on friction effect and damage evolution of end face in uniaxial compression test. J. Min. Strata Control Eng. 2021, 3, 033014. [Google Scholar]
  12. Shi Y.; Jia Y.; Rigorous J.. CO2 injection selection and numerical simulation study of Daniudi tight and low-permeability gas reservoir. In 2016 Natural Gas Academic Annual Conference (C) 1994–2019; China Academic Journal Electronic Publishing House. All rights reserved. http://www.cnki.net. [Google Scholar]
  13. Guo P.; Li X.; Sun Z. Numerical simulation of CO2 flooding and storage in low-permeability gas reservoirs. Sci. Technol. Eng. 2019, 19, 68–76. [Google Scholar]
  14. Zuo J.; Yu M.; Hu S. Experimental investigation on fracture mode of different thick rock strata. J. Min. Strata Control Eng. 2019, 1, 013007. [Google Scholar]
  15. Sun Y.; Du Z.; Sun L. The phase behavior of CO2 storage and enhancement of natural gas recovery. Nat. Gas Ind. 2012, 32, 39–42. [Google Scholar]
  16. Kempers L. J. T. M.; Haas H. The dispersion zone between fluids with different density and viscosity in a heterogeneous porous medium. J. Fluid Mech. 1994, 267, 299–324. 10.1017/S0022112094001199. [DOI] [Google Scholar]
  17. Wang J.Modern Thermodynamics: Based on the Extended Carnot Theorem; Fudan University Press: 2010, pp. 45–48. [Google Scholar]
  18. Mchugh M. A.; Krukonis V. J.. Supercritical Fluid Extraction: Principles and Practice; Butterworths: Boston, MA, 1986 [Google Scholar]
  19. Zhu Z.Supercritical fluid technology: principles and applications; Chemical Industry Press: 2000, pp. 56–59. [Google Scholar]
  20. Chen F.; Yao R. Carbon dioxide capture and storage technology. Energy Res. Inform. 2011, 27, 193–196. [Google Scholar]
  21. Gunter W. D.; Bachu S.; Law D. H.; Marwaha V.; Drysdale D. L.; Macdonald D. E.; McCann T. J. Technical and economic feasibility of CO2 disposal in aquifers within the Alberta sedimentary basin, Canada. Energy Convers. Manag. 1996, 37, 1135–1142. 10.1016/0196-8904(95)00311-8. [DOI] [Google Scholar]
  22. Xu Z.; Chen D.; Zeng R. Principles and conditions of CO2 underground geological storage. J. Southwest Pet. Univ. 2009, 31, 117–123. [Google Scholar]
  23. Zhang H.; Wen D.; Li Y. Analysis of CO2 geological storage conditions in China and relevant suggestions. Geol. Bull. 2005, 24, 29–32. [Google Scholar]
  24. Li X.; Feng Q.; Liu B. Research status and application prospects of CO2 geological storage in China. Clean Coal Technol. 2009, 13, 11–15. [Google Scholar]
  25. Zheng J.; Chong Z.; Qureshi M. F.; Linga P. Carbon Dioxide Sequestration via Gas Hydrates: A Potential Pathway toward Decarbonization. Energy Fuels 2020, 34, 10529–10546. 10.1021/acs.energyfuels.0c02309. [DOI] [Google Scholar]
  26. Jadhawar P.; Yang J.; Chapoy A.; Tohidi B. Subsurface Carbon Dioxide Sequestration and Storage in Methane Hydrate Reservoirs Combined with Clean Methane Energy Recovery. Energy Fuels 2021, 35, 1567–1579. 10.1021/acs.energyfuels.0c02839. [DOI] [Google Scholar]
  27. Niu M.; Wu G.; Yin Z.; Sun Y.; Liu K.; Chen D. Effectiveness of CO2-N2 injection for synergistic CH4 recovery and CO2 sequestration at marine gas hydrates condition. Chem. Eng. J. 2021, 420, 129615 10.1016/j.cej.2021.129615. [DOI] [Google Scholar]
  28. Yin Z.; Linga P. Methane hydrates: A future clean energy resource. Chin. J. Chem. Eng. 2019, 27, 2026–2036. 10.1016/j.cjche.2019.01.005. [DOI] [Google Scholar]
  29. Guo Y. CO2 injection enhanced oil recovery technology for edge-bottom water gas reservoirs. Sino-Foreign Energy 2019, 53–61. [Google Scholar]
  30. Du J.; Zhao Y.; Guo P. Experimental study on enhanced oil recovery by CO2 injection in tight gas reservoirs in northern Hubei. Sci. Technol. Eng. 2020, 20, 6459–6465. [Google Scholar]
  31. Hemmati-Sarapardeh A.; Dabir B.; Ahmadi M.; Mohammadi A. H.; Husein M. M. Toward mechanistic understanding of asphaltene aggregation behavior in toluene: The roles of asphaltene structure, aging time, temperature, and ultrasonic radiation. J. Mol. Liq. 2018, 264, 410–424. 10.1016/j.molliq.2018.04.061. [DOI] [Google Scholar]
  32. Hua Z.; Lin M.; Guo J.; Xu F.; Li Z.; Li M. Study on plugging performance of cross-linked polymer microspheres with reservoir pores. J. Pet. Sci. Eng. 2013, 105, 70–75. 10.1016/j.petrol.2013.03.008. [DOI] [Google Scholar]
  33. Pollet P.; Eckert C. A.; Liotta C. L. Switchable solvents. Chem. Sci. 2011, 2, 609–614. 10.1039/c0sc00568a. [DOI] [Google Scholar]
  34. Zhou Y.; Yin D.; Chen W.; Liu B.; Zhang X. A. A comprehensive review of emulsion and its field application for enhanced oil recovery. Energy Sci. Eng. 2019, 7, 1046–1058. 10.1002/ese3.354. [DOI] [Google Scholar]
  35. Clark K. A.; Pasternack D. S. Hot Water Seperation of Bitumen from Alberta Bituminous Sand. Ind. Eng. Chem. 1932, 24, 1410–1416. 10.1021/ie50276a016. [DOI] [Google Scholar]
  36. Sitnov S. A.; Vakhin A. V.; Mukhamatdinov I. I.; Onishchenko Y. V.; Feoktistov D. A. Effects of calcite and dolomite on conversion of heavy oil under subcritical condition. Pet. Sci. Technol. 2019, 37, 687–693. 10.1080/10916466.2018.1564766. [DOI] [Google Scholar]
  37. Jessen K.; Stenby E.; Orr F. Interplay of Phase Behavior and Numerical Dispersion in Finite-Difference Compositional Simulation. SPE J. 2004, 9, 193–201. 10.2118/88362-PA. [DOI] [Google Scholar]

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES