CO₂ Utilization and Sequestration in Organic and Inorganic Nanopores During Depressurization and Huff-n-Puff Process

Abstract

CO₂ injection in shale reservoirs is more suitable than the conventional recovering methods due to its easier injectivity and higher sweep efficiency. In this work, Grand Canonical Monte Carlo (GCMC) simulation is employed to investigate the adsorption/desorption behavior of CH₄-C₄H₁₀ and CH₄-C₄H₁₀-CO₂ mixtures in organic and inorganic nanopores during pressure drawdown and CO₂ huff and puff processes. The huff and puff process involves injecting CO₂ into the micro- and mesopores, where the system pressure is increased during the huffing process and decreased during the puffing process. The fundamental mechanism of shale gas recovery using the CO₂ injection method is thereby revealed from the nanopore-scale perspective. During primary gas production, CH₄ is more likely to be produced as the reservoir pressure drops. On the contrary, C₄H₁₀ tends to be trapped in these organic nanopores and is hard to extract, especially from micropores and inorganic pores. During the CO₂ huffing period, the adsorbed CH₄ and C₄H₁₀ are recovered efficiently from the inorganic mesopores. On the contrary, the adsorbed C₄H₁₀ is slightly extracted from the inorganic micropores during the CO₂ puffing period. During the CO₂ puff process, the adsorbed CH₄ desorbs from the pore surface and is thus heavily recovered, while the adsorbed C₄H₁₀ cannot be readily produced. During CO₂ huff and puff, the recovery efficiency of CH₄ is higher in the organic pores than that in the inorganic pores. More importantly, the recovery efficiency of C₄H₁₀ reaches the highest levels in both the inorganic and organic pores during the CO₂ huff and puff process, suggesting that the CO₂ huff and puff method is more advanced for heavier hydrocarbon recovery compared to the pressure drawdown method. In addition to CO₂ storage, CO₂ sequestration in the adsorbed state is safer than that in the free state. In our work, it was found that the high content of organic matter, high pressure, and small pores are beneficial factors for CO₂ sequestration transforming into adsorbed state storage.

1. Introduction

The potential for shale oil resources in China’s saline lake sediments is huge [1], which makes shale oil an important energy type, as the output of traditional fuels is declining sharply [2,3,4,5]. Although the original reserves of shale oil are abundant all over the world, commercially recoverable shale is quite limited due to its distinctive characteristics, such as extremely low permeability, etc. [6,7]. In recent years, horizontal drilling and multi-staged hydraulic fracturing have been widely recognized as the main technologies for shale oil production [8]. Although shale oil can be recovered at the initial stage after employing these developing methods, large amounts of shale resources are still left in shale reservoirs, up to 80% [9,10]. More advanced technologies should be developed to enhance shale oil recovery. There is a strong motivation for developing advanced extraction technologies in that even a 1.0% increase in oil recovery results in the production of an extra 1.6–9.0 billion barrels of oil [11].

Gas injection is accepted as being more advanced than traditional technologies for shale recovery. Gas is superior at entering the nanopores and therein has a higher sweep efficiency in shale reservoirs. Commonly used gases in oilfields are N₂, CO₂, natural gas, etc., of which CO₂ possesses the lowest minimum miscibility pressure (MMP) compared to other gases [12,13]. The basic mechanism of CO₂ for oil recovery includes oil swelling and considerable oil viscosity reduction [14,15,16,17,18]. When injected, CO₂ can become miscible with the crude oil when the pressure is higher than the MMP. However, the miscible conditions are not easy to achieve in shale reservoirs. In fact, the light and intermediate hydrocarbons are readily recovered by CO₂, while the heavy components remain in nanopores and are hard to recover [19].

Recently, methane, propane, butane, etc., have been proposed as cosolvents to assist CO₂ for oil recovery, through which the performance of CO₂ is reinforced [20,21,22,23]. With the assistance of a cosolvent, the miscibility performance of CO₂ is greatly improved in the in situ oil; in addition, a cosolvent can also dissolve and reduce the viscosity of crude oil, which is beneficial for oil recovery. Methanol is used with CO₂ to remove the liquid blockage in unconventional condensate reservoirs [24,25]. Liu et al. (2022) proposed dimethyl ether to assist CO₂ flooding for oil recovery, as well as carbon sequestration, so-called storage-driven CO₂ enhanced oil recovery (EOR) [26,27]. In order to improve shale oil recovery, more advanced technologies need to be developed. Hydrochloric saline lake shale formations are dominated by inorganic pores and have a low incidence of organic pores [28], which is of great significance for the design of CO₂ injection wells. In order to improve the design of CO₂ injection, it is critical to have a basic understanding of the fundamental mechanisms of shale hydrocarbon recovery during CO₂ injection [29,30,31,32,33,34,35,36,37,38,39]. Advanced heating technologies like radio frequency and microwave heating can enhance carbon dioxide removal and recovery, improving energy efficiency and environmental sustainability. J.F. et al. demonstrated that radiofrequency(RF)-heated fixed-bed reactors utilizing CaCO₃ sorbent can enhance CO₂ capture from flue gas [40], while T.C. et al. found that Microwave Swing Desorption (MSD) accelerates CO₂ desorption from microporous activated carbon fourfold compared to Temperature Swing Desorption (TSD) [41], together, showcasing innovative approaches to improving CO₂ management and emissions reduction.

In this work, the Grand Canonical Monte Carlo (GCMC) simulation method is employed to investigate the adsorption/desorption behavior of CH₄-C₄H₁₀ and CH₄-C₄H₁₀-CO₂ mixtures in organic and inorganic nanopores during pressure drawdown and CO₂ huff and puff processes. Specifically, the influence of pore size, mineral types of nanopores, and pressure on adsorption and desorption behavior are thoroughly evaluated. The fundamental mechanism of shale gas recovery using the CO₂ injection method is thereby revealed from the nanopore-scale perspective. This study enhances the understanding of CO₂ injection in shale oil and gas recovery, offering a theoretical foundation for optimizing extraction technologies. By examining adsorption and desorption behaviors in nanopores, the research improves recovery efficiency and promotes sustainability. Additionally, the findings contribute to emission reduction, carbon sequestration, and the global energy crisis, supporting energy transition. Overall, the research holds significant academic value and potential applications, benefiting national energy security and environmental protection.

2. Molecular Models and Simulations

The mineral analysis indicates that Chang 7 shale is rich in organic kerogen and K-illite. The surface morphology of Chang 7 shale suggests that the shape morphology of the nanopores can be simplified as slit-shaped in the molecular models. The organic pore is usually represented by the carbon-based pore. In this molecular model, the carbon-slit pore is employed to represent the kerogen nanopore for simplicity, i.e., organic pores; the K-illite nanopore is used to represent the inorganic nanopore. In this study, the inorganic characteristics of the pores are predominantly influenced by the structural features of K-illite, known for its stable crystal framework and high thermal resistance. Conversely, the organic pores are derived from the organic kerogen-rich Chang 7 shale, which demonstrates excellent adsorption capacity and chemical reactivity. Consequently, both the inorganic and organic pores possess distinct advantages in terms of functionality and properties, collectively impacting the physicochemical performance of the shale. It should be noted that the K-illite has one Al-O octahedral sheet lying between each of the two Si-O tetrahedral sheets, i.e., one typical 2:1 clay mineral for Chang 7 shale [42]. The K-illite has the unit cell formula K_x[Al_xSi_(8-x)][Al_yMg_4-y]O₂₀(OH)₄ [43], in which the x and y represent 1 and 4, respectively. The clay sheet is negatively charged because one Si⁴⁺ cation is usually replaced by one Al³⁺ cation in every unit cell. Meanwhile, in the interlayers, the K⁺ cations are distributed randomly, counterbalancing the negative charges in the unit cell. It is noted that we investigate the recovery mechanism of shale gas in organic and inorganic nanopores with two sizes, i.e., 1 nm and 3 nm, representing micropores and mesopores, respectively.

The force field is used for the potential models of adsorbed gas [44]. The clay models are obtained from the CLAYFF force field. The C and H atoms in the alkanes are regarded as a united atom, while the partial charge of the C atom and O atoms in the CO₂ molecule are taken as +0.7e and −0.35e, respectively. The following potential model is used to describe the Lennard-Jones (LJ) potential and the electrostatic terms for the nonbonded interactions:

$$u\left(r_{ij}\right)=4{\epsilon }_{ij}\left[{\left({\displaystyle \frac{{\sigma }_{ij}}{r_{ij}}}\right)}^{12}-{\left({\displaystyle \frac{{\sigma }_{ij}}{r_{ij}}}\right)}^6\right]+{\displaystyle \frac{q_i{q}_j}{4\pi {\epsilon }_0{r}_{ij}}}$$

(1)

where ${\epsilon }_{ij}$ is the well depth of the LJ potential, and the ${\sigma }_{ij}$ is the LJ radius; r_ij is the distance between atoms i and j; and q represents the atrial atom charge, which is employed to calculate the Coulomb interactions.

The cross interactions between two different atoms are calculated by the Lorentz–Berthelot combining rules as follows [45]:

$${\sigma }_{ij}={\displaystyle \frac{1}{2}}\left({\sigma }_{ii}+{\sigma }_{jj}\right)$$

(2)

$${\epsilon }_{ij}=\sqrt{{\epsilon }_{ii}{\epsilon }_{jj}}$$

(3)

The Ewald summation method is used to estimate the long-range electrostatic interactions by placing a vacuum slab in the z direction in the simulation cell [46,47].

In this study, the adsorption and desorption behavior of alkane/CO₂ mixtures is investigated by the GCMC simulation method. The grand canonical μVT ensemble is employed in this work, where μ, V, and T represent the chemical potentials, volume, and temperature, respectively; within such an ensemble, μ, V, and T are regarded as constants [48]. Within the molecular simulation, gas molecules in nanopores can exchange with a fictitious bulk reservoir with a fixed chemical potential. At the beginning of constructing the GCMC simulation, we set the starting point of the simulation by selecting a certain number of C₄H₁₀ and CO₂ molecules and randomly distributing them in the nanopores to represent the initial state. This initial configuration undergoes multiple cycles of energy minimization to ensure the thermodynamic stability of the system. A configurational-biased GCMC algorithm is used to insert and remove C₄H₁₀ molecules from nanopores [48]. The chemical potentials of bulk CH₄ and C₄H₁₀ molecules in the canonical ensemble are calculated using Widom’s insertion method [49]. During GCMC simulations, the nanopore structure is fixed. The transitional move is implemented with CH₄ molecules, which removes them from or inserts them into the simulation box randomly, with equal probability. As for CO₂ and C₄H₁₀, in addition to the transitional move, a rotational move is implemented [48]. The Peng–Robinson equation of state (PR-EOS) is used to calculate the bulk densities [50]. The gas molecules are endowed with an equal probability; according to the chemical potential of the bulk reservoir, the gas molecules can be inserted or removed from the nanopores. The bulk reservoir has a given pressure and temperature. In order to initiate gas production, the bulk pressure is lowered from the initial pressure (P_i = 400 bar) to P₁ = 200 bar and P₃ = 100 bar. The equilibrium pressure (P₂) in the nanopores after gas injection is calculated using the PR-EOS [50]. The equilibrium fluid composition is calculated based on the chemical potentials in and out of the nanopores.

As for the depressurization and CO₂ huff-n-puff processes performed by GCMC, the primary pressure drawdown is first performed to initiate shale hydrocarbon production by lowering the bulk pressure from P_i = 400 bar to P₁ = 200 bar. Subsequently, CO₂ is introduced into the bulk volume. The PR-EOS is used to calculate the equilibrium pressure after CO₂ injection, i.e., P₂ [50]. After reaching equilibrium, the puffing process, i.e., a pressure drawdown process, is applied, starting from P₂ to P₃ = 100 bar. In order to achieve equilibrium, 0.3 million cycles are required for each molecule and 0.7 million cycles are required for sampling. This timescale ensures the stability and accuracy of the simulation process.

Through statistical analysis and comparison of the simulation results, we validated the effectiveness of the GCMC simulation method, which provides a solid theoretical foundation and an experimental basis for a deeper understanding of the adsorption and desorption behaviors of alkane/CO₂ mixtures.

3. Results and Discussion

3.1. Primary Production during Depressurization

3.1.1. Organic Nanopores

Figure 1, Figure 2, Figure 3 and Figure 4 present the density distribution of CH₄, C₄H₁₀, and CH₄-C₄H₁₀ mixtures in the micropores and mesopores at P_i and P₁, respectively. It is noted that carbon slit pores were employed to represent the organic pores in shale, considering that carbon slit has similar physical properties to that of the organic matter in shale. In the micropores and mesopores, the density of hydrocarbons near the surface was significantly higher than that in the pore center, indicating that hydrocarbons form adsorption layers on the pore surface, while hydrocarbon molecules tended to stay in a free-gas state due to the relatively weak surface attraction. In addition, the density of hydrocarbons in the adsorption layers was remarkably higher than that in the mesopores due to the coupling effect of surface attraction from both sides of the nanopores. The density at the pore center of mesopores approached the bulk density, indicating that the gas readily formed free gas in the mesopores. However, the density at the pore center of the micropores did not converge with the bulk density, suggesting that there was no free-gas region.

As shown in Figure 1 and Figure 2, the density of CH₄ decreases both in the adsorption layers and at the center of nanopores as the pressure declines. However, there is no obvious density change for C₄H₁₀ whether in the micropores or in the mesopores. In addition, for CH₄-C₄H₁₀ mixtures, the density of CH₄ drops in the first adsorption layer as pressure decreases, while C₄H₁₀ shows enrichment near the pore surface during pressure drawdown, indicated by increasing density at a lower pressure. During the primary gas production, the lighter hydrocarbons are more likely desorbed from the pore surface and more easily produced as the reservoir pressure drops, like CH₄. On the contrary, due to the stronger interaction with the pore surface, the heavier components tend to be trapped in these organic nanopores, especially in micropores. This behavior is similar to that found in other adsorbents, such as zeolites [51] and graphene [52,53].

3.1.2. Inorganic Nanopores

Figure 5, Figure 6, Figure 7 and Figure 8 present the density distribution of CH₄ and C₄H₁₀ in a CH₄-C₄H₁₀ mixture in the inorganic micropores and mesopores at P_i and P₁, respectively. In this work, the inorganic nanopores are represented by K-illite pores. Similarly to that in the organic pores, CH₄ and C₄H₁₀ molecules form an adsorption layer near the pore surface and free-gas region at the pore center in mesopores, while there is no free gas in the micropores. Comparatively, the density in the adsorption layers in the inorganic pores is much smaller than that in the organic pores, demonstrating relatively weaker molecule–surface interactions in the inorganic pores. As the pressure drops, the density of CH₄ in the adsorption layer decreases, while the adsorption layer of C₄H₁₀ cannot be recovered during pressure drawdown. Specifically, as with the CH₄-C₄H₁₀ mixture, the density of C₄H₁₀ increases in the first adsorption layer, and the increase is remarkably more significant in the inorganic pores than that of the organic nanopores. In other words, it is more difficult to recover the heavier hydrocarbons from the inorganic pores, such as C₄H₁₀.

3.2. Improved Gas Production during CO₂ Huff-n-Puff Process

3.2.1. Organic Nanopores

In this process, CO₂ is injected into the micro- and mesopores; the system pressure is increased from P₁ to P₂ during the huffing process and then it decreases to P₃ in the puffing process. Figure 9, Figure 10, Figure 11 and Figure 12 present the density profiles of CH₄ and C₄H₁₀ in the CH₄-C₄H₁₀ mixture in the organic micropores and mesopores at P₁, P₂, and P₃, respectively. Comparatively, CO₂ presents the strongest adsorption capacity on the pore surface compared to CH₄ and C₄H₁₀. In addition, it is found that the adsorption layer of CO₂ is much closer to the pore surface than those of the alkanes, i.e., CH₄ and C₄H₁₀. After introducing CO₂, i.e., in the CO₂ huffing period (P₁ → P₂), the density in the first adsorption layer of CH₄ and C₄H₁₀ decreases drastically, indicating that the adsorbed hydrocarbons are replaced by the injected CO₂. This observation agrees well with the previous work [19,54]. In the puffing process, i.e., P₂ → P₃, the adsorbed CH₄ desorbs from the pore surface continuously and thus can be heavily recovered. However, the density of C₄H₁₀ in the adsorption layer is slightly changed, suggesting that the adsorbed C₄H₁₀ cannot be readily produced. More interestingly, as puffing time continues, the adsorbed CH₄ in the CH₄-C₄H₁₀ mixture is produced as expected, while the density of C₄H₁₀ in the adsorption layer increases, indicating that C₄H₁₀ re-adsorbs on the pore surface in the huffing period.

3.2.2. Inorganic Nanopores

Figure 13, Figure 14, Figure 15 and Figure 16 present the density profiles of CH₄, C₄H₁₀, and the CH₄-C₄H₁₀ mixture in the inorganic micropores and mesopores at P₁, P₂, and P₃, respectively. Unlike in organic pores, the density of CO₂ in the adsorption layer within inorganic pores is significantly higher. It was reported that CO₂ molecules have a neutral charge and zero dipole moment, while CO₂ molecules exhibit a strong quadrupole moment [55,56]. These properties enable CO₂ to have improved adsorption by combining with the charged clay atoms [46]. Due to the presence of CO₂ adsorption layers, CH₄ and C₄H₁₀ form a second minor adsorption layer towards the pore center. In the inorganic mesopores, the density of CH₄ and C₄H₁₀ decreases significantly in the adsorption layer during the CO₂ huffing period. That is, the adsorbed CH₄ and C₄H₁₀ can be recovered efficiently from the inorganic mesopores during the CO₂ huffing period. On the contrary, in the inorganic micropores, the density of C₄H₁₀ in the adsorption layer slightly increases during the CO₂ puffing period.

3.3. Shale Hydrocarbon Recovery

The oil recovery efficiency (RE) is obtained for CH₄ and C₄H₁₀ from the organic and inorganic micro- and mesopores. RE is calculated by ${\displaystyle {\int }_{z_0}^{z_i}{\rho }_{P_i}}dv/{\displaystyle {\int }_{z_0}^{{z^{\prime }}_0}{\rho }_{P_0}}dv$ (where ${\rho }_{P_i}$ and ${\rho }_{P_0}$ represent the density profiles at P_i and P₀). Figure 17, Figure 18, Figure 19 and Figure 20 present the recovery efficiency during pressure drawdown and the CO₂ huffing and puffing period. We find that RE during pressure drawdown is smaller in micropores than in mesopores, and vice versa for the heavier hydrocarbons, i.e., C₄H₁₀. In inorganic pores, the recovery efficiency of C₄H₁₀ is much higher than that of organic pores. However, RF is comparable in both kinds of pores for the lighter hydrocarbons, i.e., CH₄. Interestingly, the recovery efficiency for C₄H₁₀ can be negative in the micropores during pressure drawdown; in other words, the heavy hydrocarbons tend to accumulate in the small pores as pressure drops, which is in line with our previous findings. During CO₂ huff and puff, the recovery efficiency of CH₄ is higher in the organic pores than that of the inorganic pores; the efficiency during the CO₂ huffing process is comparable in the micro- and mesopores. More importantly, we observe that the recovery efficiency of C₄H₁₀ reaches the highest in both the inorganic and organic pores during the CO₂ huff and puff process. It suggests that the CO₂ huff and puff method is more suitable for heavier hydrocarbon recovery compared to the pressure drawdown method.

3.4. CO₂ Sequestration in Nanopores

Figure 21 and Figure 22 present the molar fraction of the free-state and adsorbed-state CO₂ in the micro- and meso-organic and -inorganic pores, respectively. In the organic pores, CO₂ is mainly sequestrated in the adsorbed state in micropores, while, in the mesopores, free-state CO₂ dominates at low pressures. In large pores, most CO₂ molecules are adsorbed on the pore surface during high-pressure conditions; however, as pressure decreases, the adsorbed CO₂ desorbs from the pore surface and is sequestrated in the free state. Comparatively, in the small pores, CO₂ molecules are mainly in adsorbed state storage due to the strong fluid–pore surface interactions. In the inorganic pores, CO₂ is stored mainly in the free state due to the relatively weak fluid–pore surface interactions, except in the micropores at high-pressure conditions. Similarly, as pressure decreases, CO₂ sequestration transfers from the adsorbed state to the free-gas state in the inorganic pores. In summary, CO₂ sequestration in the adsorbed state is safer than that in the free state. A high content of organic matter, high pressure, and small pores are beneficial factors for transforming CO₂ sequestration into adsorbed state storage.

4. Conclusions

In this work, the GCMC simulation methods were applied to study the mechanisms of shale hydrocarbon extraction in organic and inorganic nanopores with sizes of 1 nm and 3 nm, analyzing the impact of pressure drawdown and the CO₂ huff and puff process on gas recovery rates. During the primary gas production, CH₄ readily desorbs from the pore surface and is produced as reservoir pressure drops. On the contrary, C₄H₁₀ tends to be trapped in these organic nanopores and is difficult to produce, especially in the micropores. Comparatively, it is more difficult to recover C₄H₁₀ from the inorganic pores. During the CO₂ puffing process, the adsorbed CH₄ desorbs from the pore surface continuously and can therein be heavily extracted. However, the adsorbed C₄H₁₀ cannot be readily produced. During the CO₂ huffing period, the adsorbed CH₄ and C₄H₁₀ can desorb and escape from the inorganic mesopores, enhancing shale gas production. However, during the CO₂ puffing period, C₄H₁₀ in the inorganic micropores is hard to recover. Research findings indicate that during pressure drawdown and the CO₂ huff and puff process, CH₄ is more easily desorbed and released from the pores, whereas C₄H₁₀ is more difficult to generate and recover, especially in inorganic micropores.

Recovery efficiency during pressure drawdown is smaller in micropores than that of the mesopores. In inorganic pores, the recovery efficiency of C₄H₁₀ is much higher than that in the organic pores. However, it is comparable in both pores for CH₄. In addition, the recovery efficiency of C₄H₁₀ can be negative in the micropores during pressure drawdown. During CO₂ huff and puff, the recovery efficiency of CH₄ is higher in the organic pores than that in the inorganic pores. More importantly, the recovery efficiency of C₄H₁₀ reaches the highest for both the inorganic and organic pores during the CO₂ huff and puff process. These findings indicate that the CO₂ huff and puff method is more effective for recovering heavier hydrocarbons than the pressure drawdown method.

Beyond traditional carbon sequestration methods, CO₂ mainly sequestrates in the adsorbed state in organic pores, while it mainly sequestrates in the free-gas state in the inorganic pores. According to the content of this study, CO₂ sequestration in the adsorbed state is safer than that in the free state. In our work, it is found that a high content of organic matter, high pressure, and small pores are beneficial factors for CO₂ sequestration transforming into adsorbed state storage.