Molecular Dynamics Simulation Study on the Mechanism of CO₂-CH₄ Synergistic Enhanced Oil Recovery in Tight Oil Reservoirs

Abstract

Tight oil reservoirs are currently a hot topic in petroleum exploration and development. However, due to the low porosity and low permeability of reservoirs and the lack of external energy supplementation, there is a significant mismatch between resources and production in tight oil. Mining and experimental studies have shown that CO₂ and gaseous hydrocarbons have a high injectivity and effective oil displacement effect in tight oil reservoirs. Currently, research is mostly focused on a single energy supplement medium, and whether CO₂ hydrocarbon mixtures can more effectively improve oil recovery needs to be further studied. The features of crude oil expansion capacity and interaction energy changes following various fluid interactions were investigated via molecular dynamics simulation techniques in response to the ambitious comprehension of the mechanistic changes underlying the CO₂–CH₄ synergistic effect during the development of tight oil reservoirs. The research results indicate that the expansion and diffusion abilities of crude oil are improved after being treated with pure CO₂, CO₂-CH₄ (9:1), CO₂-CH₄ (7:3), and CO₂-CH₄ (1:1), and enhanced with increasing CO₂ content in the injected fluid.

1. Introduction

The huge demand for oil and gas energy in China has driven the steady growth of the global oil and gas industry [1,2,3]. Unconventional oil and gas have shown a huge resource endowment and broad development prospects [4,5,6]. Tight oil, as an important unconventional resource, is a crucial direction in today’s petroleum industry and a powerful support for rising oil and gas reserves and production, while also replacing traditional oil and gas energy sources [7]. The practice of developing tight oil reservoirs has shown that the overall utilization degree of depleted reservoirs is low, and they face the challenge of further elevating single well production alongside the utilization rate. Implementing effective energy replenishment measures is a key means of improving the recovery rate of tight oil reservoirs (TOR) [8,9,10]. A TOR is an oil and gas reservoir with a low porosity alongside low permeability, with a porosity generally below 10% and a permeability generally below 1 × 10⁻⁴ μm² [11,12,13,14]. Water injection and gas injection are the main methods for reservoir energy replenishment [15,16,17]. However, the pore structure of tight reservoirs is complex, and the permeability of water in the reservoir is poor [18]. Water injection usually requires a high injection pressure; it is hard to establish an effective injection production system [19,20,21]. The water injection development of TOR can effectively supplement the energy of the formation, decrease the viscosity of CO₂, and improve the flow capacity of crude oil (CO₂), thereby enhancing the recovery rate of CO₂ and avoiding the sudden decrease in production capacity caused by rapid pressure drops [22,23,24]. Meanwhile, the economic cost of water injection development is relatively low, making it one of the main means for stable production in TOR [25,26,27]. Yang et al. quantitatively [28] described the remaining oil distribution characteristics during the water injection development process of TOR via nuclear magnetic resonance technology. Research has shown that the remaining oil distributed in 10–100 μm after water flooding is 29% to 64%, while the remaining oil in pores smaller than 10 μm is 34% to 69%. Scholars such as Wei [29] conducted experiments on the water flooding efficiency of dense oil reservoir cores, and the main research results showed that the remaining oil in a free state is used during the water flooding process. After the completion of water flooding, the remaining oil content in a thin film on the pore surface increases by more than 2.5%. The biggest challenge for hydraulic fracturing horizontal wells in TOR during the mid-to-late stages of depleted development is how to increase single well productivity and further improve the overall recovery rate of the reservoir. An efficient and economical increase in oil recovery is the only way for the sustainable development of tight oil, and injecting energy replenishment media is an effective means of improving development efficiency. Taking into account factors such as the injection capacity and oil displacement effect, most TOR and mining sites both domestically and internationally tend to focus on gas injection development. Song et al. proposed [30] a novel minimum mixed phase pressure calculation model combining the phase equilibrium calculation model considering nanoscale confinement effects with the interfacial tension disappearance method. The simulation calculation results show that, as the aperture is below 100 nm, the pressure of the oil–gas mixture decreases significantly. Tao et al. conducted [31] research on the characteristics of minimum miscible pressure changes in nanoscale space through the visualization of nanofluid displacement technology. The study found that the pore size reduction and increase in temperature exacerbated the nanoscale effect, resulting in a decrease in minimum miscible pressure. Li et al. innovatively [32] proposed a technique for measuring the diffusion coefficient of gas in tight rock cores saturated with oil, and deeply explored the influence of factors such as core permeability, pore radius, and tortuosity on the diffusion behavior of gas in micro–nanoporous media. Research has shown that the diffusion coefficient of CO₂ in tight reservoirs is on the order of 10⁻⁹ m² s⁻¹. As the core permeability and average pore throat radius increase, the diffusion coefficient of CO₂ also shows a rising trend, but the magnitude of this increase gradually decreases and eventually stabilizes. With the continuous improvement of computer performance and the development of theoretical chemistry, molecular dynamics simulation technology has received increasing attention. More and more researchers are using molecular dynamics simulation techniques to research the microscopic migration characteristics of injected fluids in CO₂ molecular systems. Liu et al. studied [33] the solubility and adsorption capacity of CO₂ surfactant on different mineral walls. The research results suggest that the degree of improvement in the solubility of CO₂ in CO by surfactants is as follows: CFP > SF > SDS. Santos and other scholars [34] researched the adsorption behavior of CO₂ and alkanes in calcite pores. Research has shown that CO₂ is more easily adsorbed on pore walls, and most n-butane and n-octane are adsorbed parallel to the surface of calcite. At present, the use of mixed gases can to some extent alleviate the shortage of a single gas source and the problem of gas purification faced by mining sites. There are remarkable differences in the mechanism of action between gases and formation CO₂. Relevant studies have shown that mixed gases can integrate the advantages of different gases under specific conditions, exert synergistic oil displacement effects, and further improve the degree of CO₂ recovery. Therefore, this article leverages molecular dynamics simulations to study the changes in micro-properties such as the expansion coefficient and interaction energy of CO after being treated with pure CO₂, CO₂-CH₄ (9:1), CO₂-CH₄ (7:3), and CO₂-CH₄ (1:1).

2. Models and Methods

Molecular Models

Molecular dynamics simulation (MDS) studies the dynamic behavior of molecules or atomic systems through computer simulations [35,36,37]. This simulation can accurately depict the interaction forces between molecules and their potential energy changes. To research the macroscopic properties of the system through microscopic analysis, the thermodynamic ensemble was introduced. There are four common ensembles. An ensemble expresses all the microscopic states that a system possesses under certain macroscopic conditions. This microscopic state includes the structural morphology, dynamic behavior, and other characteristics of various particles. These four ensembles are NVE ensembles, NVT ensembles, NPT ensembles, and NPH ensembles. LAMMPS, as an 2022 MDS software, has the characteristics of efficient simulation and accurate simulation results, and can simulate the interaction characteristics of molecules and atoms under different conditions [38,39,40,41,42,43]. The simulated temperature is 80 °C.

Taking the tight oil of the Chang 6 reservoir in the Ordos Basin as the research subject, this study first identified the variation characteristics of CO₂ components by employing an oil–gas component analysis apparatus. To accurately reflect the influence of the bulk environment on the oil–gas interface properties, based on the dead oil components obtained from experiments, alkanes (C1, C2, C3, C4, C6, C10, C19, C30) were selected to simulate tight oil and construct an oil–gas two-phase interface model. The molecular force field parameters for different alkanes are shown in

Table 1. Simulated non-bonding potential energy parameters of fluids.

Molecular Type	Label	Atomic Type	Electric Charge/e	σ/Å	ε/(kcal·moL−1)
Methane/Ethane/Propane/Butane/Hexane	C	Methane carbon	−0.24	3.5	0.066
	C	Methyl carbon	−0.18	3.5	0.066
	C	Methylene carbon	−0.12	3.5	0.066
	H	Alkane hydrogen	0.06	2.5	0.030
Decane/Nonadecane/Tricontane	C	Methyl carbon	−0.222	3.5	0.066
	C	Methylene carbon	−0.148	3.5	0.066
	H	Methyl hydrogen	0.074	2.5	0.030
	H	Methylene hydrogen	0.074	2.5	0.0263
Carbon Dioxide	C	Carbon dioxide—carbon	0.6512	2.757	0.0559
	O	Carbon dioxide—oxygen	−0.3256	3.033	0.1599

. Owing to the limitations of the experimental conditions, it is not feasible to determine the detailed molecular structure of CO molecules with a high carbon number. Meanwhile, to reduce computational time, this study exclusively employs linear alkanes for the simulation research. This article uses 2022 LAMMPS software to conduct MDS calculations, with a truncation radius of 1.2 nm for Lennard Jones potential interactions and a time step of 1 fs. Firstly, simulate 3 ns in the NPT ensemble, with a fixed number of simulated molecules and a size of 6 nm in the x and y directions. Simulate the molecular distribution of the system under different pressure conditions by varying the pressure. Next, simulate 1 ns in the NVT ensemble to conduct data analysis and research. The molecular dynamics model of the interaction between bulk CO₂ and CO is displayed in Figure 1.

3. Results and Discussion

3.1. Coefficient of Expansion

The expansion coefficient of CO₂ is one of the key parameters affecting its flow capacity. The ratio of the volume at the end of the simulation to the initial volume represents the expansion coefficient of crude oil. Therefore, this article calculates the system expansion coefficient of multi-component CO₂ using formula (1). The calculation results of the expansion coefficient of CO₂ after different injection fluids are displayed in Figure 2 and

Table 2. Data table of CO₂ expansion coefficient under different pressures.

P/MPa	CO2	CO2-Methane (9:1)	CO2-Methane (7:3)	CO2-Methane (1:1)
	Coefficient of Expansion
2	1.39	1.36	1.32	1.29
3	1.44	1.38	1.35	1.31
4	1.47	1.43	1.38	1.36
5	1.61	1.56	1.51	1.46

. The research results indicate that, when the ratio of injected fluid CO₂ to methane is 9:1, 7:3, and 1:1 and pure CO₂ is used, the expansion coefficient of CO₂ is positively correlated with the system pressure. When the pressure is 4 MPa and 5 MPa, the expansion coefficient of CO₂ shows a significant increase. As the system pressure rises from 2 MPa to 5 MPa, the expansion coefficient of CO₂ increases from 1.39 to 1.61 after being treated with pure CO₂. After being injected with a fluid (CO₂-to-methane ratio of 9:1), the expansion coefficient of CO₂ increases from 1.36 to 1.56. After being injected with a fluid (CO₂-to-methane ratio of 7:3), the expansion coefficient of CO₂ rises from 1.32 to 1.51. After being injected with a fluid (CO₂-to-methane ratio of 1:1), the expansion coefficient of CO₂ rises from 1.29 to 1.46. From this, it can be seen that CO₂ can improve the flowability of CO₂, thereby enhancing the oil recovery rate, which rises with the continuous rise in CO₂ content in the injected fluid.

$$\mathrm{Coefficient} \mathrm{of} \mathrm{expansion}={\displaystyle \frac{{\mathrm{V}}_{\mathrm{n}}}{{\mathrm{V}}_{\mathrm{i}}}}$$

(1)

In the formula: V_n is the volume of crude oil after the simulation is completed; V_i is the initial volume of crude oil.

3.2. Interaction Energy

In molecular simulations, the interaction energy refers to the energy changes that occur between different molecules in a system due to their interactions. The interaction energy can describe the strength of intermolecular interactions, and an accurate, efficient analysis of the interaction energy is the key to clarifying the microscopic transport and distribution laws of molecules. The variation characteristics of the interaction energy between CO₂ and different injected fluids under varied pressures are displayed in Figure 3 and

Table 3. Data table of interaction energy between CO₂ and injected fluid under different pressures.

P/MPa	CO2	CO2-Methane (9:1)	CO2-Methane (7:3)	CO2-Methane (1:1)
	Interaction Energy/kcal·moL−1
2	−0.291	−0.275	−0.269	−0.265
3	−0.365	−0.361	−0.321	−0.292
4	−0.402	−0.372	−0.325	−0.302
5	−0.536	−0.518	−0.509	−0.492

. The research findings suggest that, as the pressure rises from 2 MPa to 5 MPa, the interaction between the injected fluid and CO₂ gradually increases. Among them, after pure CO₂ interaction, the interaction energy increases from −0.291 to −0.536 kcal/mol, and under the same pressure the interaction energy decreases as the CO₂ ratio gradually increases. From this, it can be seen that CO₂ can promote the degree of contact between the injected fluid and CO₂ and increase the solubility of the injected fluid in CO, thereby improving the flowability of CO₂ and ultimately enhancing CO₂ recovery.

3.3. Interfacial Tension

Oil–gas interfacial tension is a key physical parameter for evaluating the strength of intermolecular forces at the interface between CO₂ and injected fluid, which directly affects the distribution, migration, and recovery rate of fluids in oil and gas reservoirs. We calculate the oil–gas interfacial tension via the Gibbs interfacial tension calculation formula in LAMMPS, as shown in Equation (2). For the oil and gas system in bulk, the oil and gas interface is perpendicular to the z-axis and parallel to the x and y planes. The minimum miscible pressure of oil and gas is calculated through the expression of pressure tension. The interfacial tension between oil and gas phases at different pressures for different injected fluids is shown in

Table 4. Characteristics of interfacial tension changes between CO₂ and injected fluid under different pressures.

P/MPa	CO2	CO2-Methane (9:1)	CO2-Methane (7:3)	CO2-Methane (1:1)
	Interfacial Tension/mN·m−1
2	24.53	24.85	25.16	25.29
3	13.49	13.89	14.58	14.68
4	13.64	14.08	14.12	14.33
5	10.21	10.85	11.21	11.36

. The research results indicate that, as the pressure increases, the intermolecular distance shortens, enhancing the solubility and mixing ability of the injected fluid in CO₂, and the interfacial tension of the system gradually decreases. The order of interfacial tension from highest to lowest is CO₂, CO₂-CH₄ (9:1), CO₂-CH₄ (7:3), and CO₂-CH₄ (1:1).

$$\gamma ={\displaystyle \frac{1}{2}}\left[L_Z\left(p_{zz}-{\displaystyle \frac{p_{xx}+p_{yy}}{2}}\right)\right]$$

(2)

In the formula: L_Z is the z-axis length of the box; P_α (α = x, y, z) is the diagonal element of the pressure tensor.

3.4. Diffusion Coefficient

This article uses Einstein’s formula to calculate the diffusion coefficient of injected fluid, as shown in Equation (3). After the simulation is completed, the mean square displacement at different time points is exported and fitted with time. The diffusion coefficient is computed according to the slope and converted into units. The diffusion coefficients of injected fluids under varied pressures in the simulation system are displayed in Figure 4 and

Table 5. Data table of diffusion coefficient of injected fluid in bulk state.

P/MPa	CO2	CO2-Methane (9:1)	CO2-Methane (7:3)	CO2-Methane (1:1)
	Diffusion Coefficient/10−9 m·s−1
2	7.8	7.5	7.3	7.1
3	7.2	7.1	6.8	6.6
4	6.9	6.2	5.8	5.5
5	6.2	5.8	5.5	5.2

. The simulation findings show that, as the pressure rises, the density of fluid molecules increases, the interaction energy between molecules increases, and the diffusion ability of gas molecules weakens. When the pressure is 5 MPa, the diffusion coefficient of pure CO₂ is 6.2 × 10⁻⁹ m²/s, which is higher than that of other injected fluids. From this, it can be seen that CO₂ can enhance the migration and diffusion ability of injected fluids in CO, elevating the oil recovery rate.

$$D_{\infty }={\displaystyle \frac{1}{N_a}}\underset{t\to \infty }{\lim }{\displaystyle \frac{d}{dt}}\sum _{i=1}^{N_a}\left\{{\left[r_i\left(t\right)-r_i\left(0\right)\right]}^2\right\}$$

(3)

In the formula, N_a represents the number of diffusing molecules in the system; t is the simulation time, fs; r_i (t) is the displacement of the molecule at time t, Å; r_i(0) is the displacement of the molecule at the initial time, Å; [r_i(t) − r_i(0)]² is the displacement of the molecule at the initial time, Ų.

To further demonstrate the characteristics of combined gas recovery in TOR, the research conducted experiments on the characteristics of CO₂ flooding under different injection volumes using a core flooding apparatus. The variation characteristics of CO₂ recovery efficiency under varied injection volumes are displayed in Figure 5. The results of the research suggest that core flooding with CO₂, CO₂-methane (9:1), CO₂-methane (7:3), and CO₂-methane (1:1) can increase the core oil recovery by 39.25%, 38.57%, 37.89%, and 36.88%, respectively. It can be thus concluded that the higher the CO₂ content in the injected gas, the better the enhancement effect on oil recovery.

During the gas injection recovery process in TOR, it is hard to obtain an adequate supply of CO₂, and the feasibility of displacing CO₂ with pure CO₂ is relatively low. However, reservoir-associated gas is abundant at oilfield sites. Therefore, investigating the microscopic migration and swelling mechanisms of CO₂-methane systems with different mixing ratios helps to further clarify the enhanced oil recovery (EOR) technology for tight oil, thereby providing guidance for the efficient development of oilfields. Subsequently, this study plans to undertake the following two parts of research: ① By combining X-ray diffraction experiments on core minerals, the mineral composition of tight reservoirs is clarified. Based on the main minerals in the reservoir, an MDS method is employed to construct a molecular dynamics model for both oil–gas two-phase and oil–gas–water three-phase models within the nanopores, thereby revealing the characteristics of the occurrence state of oil and gas molecules in the nanopores. ② Building on nuclear magnetic resonance and microscopic visualization experimental methods, we clarify the occurrence state of oil, gas, and water in micro–nanopores, as well as the recovery limit of the remaining oil.

4. Conclusions

(1)

Under the same pressure conditions, CO₂ can enhance the expansion coefficient of CO₂ in the simulation system. As the CO₂ content increases, the expansion capacity of CO₂ gradually rises. Compared with the expansion coefficient of CO₂ when the CO₂-CH₄ ratio is 1:1, pure CO₂ can increase the expansion coefficient by 0.15.

(2)

The interaction energy between the injected fluid and CO₂ is positively related to system pressure. CO₂ can enhance gas–oil molecular interactions between injected fluid molecules and CO₂ molecules and increase the interaction energy between oil and gas molecules, thereby enhancing the expansion coefficient of CO₂ and improving its flow capacity.

(3)

The simultaneous co-injection of gases reduces the interfacial tension, thereby elevating the flow capacity of CO. The interfacial tensions (from highest to lowest) were pure CO₂ > CO₂–CH₄ (9:1) > CO₂–CH₄ (7:3) > CO₂–CH₄ (1:1).

(4)

CO₂ can accelerate the solubility of mixed fluids in CO₂. Compared with the CO₂-CH₄ system, the pure CO₂ crude oil system has a larger gas diffusion coefficient, which can more effectively improve CO₂ utilization.