Research Progress on CO₂ Capture, Utilization, and Storage (CCUS) Based on Micro-Nano Fluidics Technology

Abstract

The research and application of CO₂ storage and enhanced oil recovery (EOR) have gradually emerged in China. However, the vast unconventional oil and gas resources are stored in reservoir pores ranging from several nanometers to several hundred micrometers in size. Additionally, CO₂ geological sequestration involves the migration of fluids in tight caprock and target layers, which directly alters the transport and phase behavior of reservoir fluids at different scales. Micro- and nanoscale fluidics technology, with their advantages of in situ visualization, high temperature and pressure resistance, and rapid response, have become a new technical approach to investigate gas–liquid interactions in confined domains and an effective supplement to traditional core displacement experiments. The research progress of micro–nano fluidics visualization technology in various aspects, such as CO₂ capture, utilization, and storage, is summarized in this paper, and the future development trends and research directions of micro–nano fluidics technology in the field of carbon capture, utilization, and storage (CCUS) are predicted.

1. Introduction

Current global conventional oil and gas resources are tending to depletion, and unconventional oil and gas is gradually becoming a strategic replacement energy source. Its contribution and position in global oil and gas production are increasingly prominent, and in the future, it will shoulder the three major missions of global energy security, stable oil and gas production, and green and low-carbon development [1]. Developing unconventional oil and gas resources can not only accelerate the achievement of the “dual carbon” goal but also to some extent solve the energy supply and demand contradiction in our country. Compared to other enhanced oil recovery (EOR) methods that increase energy supply and promote carbon neutrality by sequestering greenhouse gases, CO₂ flooding for improving oil recovery in unconventional reservoirs has both economic and social benefits. However, although CO₂ flooding for unconventional reservoirs has been ongoing for several years, the understanding of the multi-phase flow and behavior mechanism of multiscale reservoirs under displacement is still lacking. The reason for this is that unconventional reservoirs exhibit the characteristics of coexisting micro–nano pores across scales [2], with pore size distributions ranging from several nanometers to several micrometers, and even millimeter-scale fractures.

The average free path of nanoscale confined fluids is comparable to the pore size, and this confinement in a small space restricts molecular thermal motion (as shown in Figure 1), leading to phenomena different from the bulk phase, such as surface adsorption, density differences, increased capillary forces, and the contraction of phase envelopment lines. These phenomena not only affect oil production and gas injection process selection [3] but also impact the efficiency and cost of oil and gas extraction [4,5]. Therefore, gaining a comprehensive understanding of the transition of wall wetting in micro–nanochannels, the mechanical characteristics of fluids, and their impact on phase behavior is of significant engineering importance for applications such as CO₂ geological storage and petroleum extraction [6,7]. CO₂ flooding has become one of the most important extraction methods since the early 1980s due to its excellent extraction and dissolution capabilities. However, the development of CO₂-enhanced oil and gas recovery in China still lags behind foreign countries, as shown in Figure 2, and currently, CO₂-enhanced oil and gas recovery and geological sequestration face the challenge of the aforementioned multiscale pore size distribution. As a complex multiphase dissolution and migration method, CO₂ flooding and geological storage are influenced by various factors such as reservoir and fluid physical properties, flow parameters, etc. The key to its technical application lies in understanding the mechanisms of gas–liquid interactions across micro- and nanoscales. In cracks and large pores (>50 nm [8]), fluid flow exhibits bulk phase characteristics, while in nanopores (<50 nm), the flow behavior deviates significantly from the bulk phase [9,10]. Existing theoretical models for studying the multiscale mechanisms of multiphase flow mainly rely on molecular simulations, digital rock models, and equations of state (EOSs). These methods require certain assumptions and lack sufficient experimental data support. In addition, the results of traditional theoretical simulations are inconsistent with existing microfluidic experimental results [11,12] and are depicted in Figure 3a. Most experimental methods simulate the limiting effects of pure component fluid static and dynamic behaviors based on a single nanopore size. Therefore, further understanding the multiscale pore-scale oil and gas seepage and migration mechanisms is the basis for developing effective enhanced oil recovery techniques and large-scale CO₂ sequestration, as well as a prerequisite for multiscale reservoir flow simulation and parameter optimization [13].

Thanks to the advances in chip processing technology, micro–nanofluidics technology has the advantages of the fine description of nanoscale pores and in situ visualization detection, providing a new experimental perspective for the study of unconventional oil and gas microscale seepage and phase characteristics [14]. It has become one of the most promising tools for visual analysis of oil and gas. Currently, it has been successfully applied in the petroleum and natural gas industry for physicochemical analysis [15], heavy oil recovery [16,17], hydraulic fracturing [18], chemical flooding [19,20,21], gas flooding [22], and other studies on multiphase flow mechanisms. It serves as an effective complement to traditional oil displacement mechanism evaluation techniques such as theoretical simulation and experimental analysis.

This paper summarizes the current research status in the field of micro–nano fluidics technology applied to CO₂ capture, storage, and utilization, as shown in Figure 3b, and identifies the shortcomings in current research.

2. Micro–Nanofluidic Devices and Chip Manufacturing

Extensive theoretical research has indicated that when the pore size reaches the sub-nanometer scale [12,23,24,25,26,27], the interactions and wettability characteristics between the liquid and solid substrates intensify the flow–solid interaction in nanoscale pore throats [28], and the phase characteristics and seepage behavior of the fluid are significantly altered. The properties of reservoir fluids and wettability [29,30] further complicate the confinement effects, necessitating experimental tools to address this highly coupled phenomenon. In recent years, with the development of unconventional oil and gas resources, micro–nanofluidic experimental platforms [31,32,33,34], as shown in Figure 4, have been introduced into the petroleum and natural gas industry. Different imaging modes of optical microscopy (bright field, dark field, and fluorescence) enable the in situ visualization monitoring of micro–nanochannel samples. In addition, spectroscopic methods such as infrared Raman can be used for the fine characterization of micro–nanofluidic systems [35]. The materials used for the chips include silicon, glass, ceramics, and other polymers, such as polydimethylsiloxane (PDMS) [36]. However, the high-temperature and high-pressure experimental conditions in the oil and gas industry and the research requirements for organic nonpolar fluids make silicon the most ideal manufacturing material. As depicted in Figure 5, the micro–nanofluidic chip transfers the characteristics of reservoir porous media micro–nanochannels onto a silicon substrate using techniques from the semiconductor industry [37], and the silicon substrate is bonded to a glass plate to form enclosed channels [38]. The chip channels can be modified on the wall surface [39,40] or filled with modified nanoparticles [41] to simulate the real conditions of reservoir porous media and address the fundamental issues of reservoir dynamics and fluid flow [42]. The so-called nanochannels have at least one dimension (width and depth) within the range of 1 to 100 nm. During processing, care must be taken to avoid channel collapse caused by capillary or van der Waals forces. Due to the high aspect ratio and challenging fabrication of high-throughput nanochannels, current research focuses mainly on low-aspect-ratio channels with nanoscale depth and micrometer-scale width [38].

3. Research on CO₂ Capture

Carbon capture is the process of collecting greenhouse gases, such as those emitted from fossil fuels, through fixed emission channels, and then compressing and transporting them for utilization or storage. As the first step in the CCUS process, carbon capture technology is crucial for the long-term feasibility of CCUS technology. Existing research has demonstrated the potential of micro–nanofluidic technology in CO₂ capture. Currently, CO₂ capture technologies mainly include absorption, adsorption [43,44,45], and membrane separation. Absorption methods involve physical and chemical absorption, which utilize organic solvents or alkaline solutions with high solubility, selectivity, and stability to separate CO₂. Adsorption methods mainly rely on carbon-based materials, zeolites, nanomaterials, and other chemical adsorbents to capture CO₂. The mechanisms of solvent absorption, porous medium absorption, and separation dissolution related to CO₂ capture using organic solvents and alkaline solutions, as well as nanomaterials, can be further studied using micro–nanofluidic control technology, as demonstrated in previous research [46]. In particular, the carbon capture applications of deep eutectic solvents (DESs) and ionic liquid analogues [47] hold significant engineering significance.

4. Research on CO₂ Enhanced Oil and Gas Recovery

Phase behavior is crucial in gas displacement processes as it determines gas–liquid equilibrium, swelling effects, gas solubility, and minimum miscibility pressure [48,49]. The complexity of phase behavior in unconventional reservoir fluids is attributed to the diversity of oil and gas components and the influence of porous media and reservoir rock properties [50]. In unconventional reservoirs, the ratio of pore surface area to volume increases significantly, leading to phenomena such as bubble point depression, dew point trajectory deviation, and variations in critical parameters, which directly result in a rapid decline in initial production rates (approximately ten times faster than conventional reservoirs) [14]. Although theoretical simulations [27,51,52] and instrumental analyses [53,54] have provided reliable means for studying restricted fluid phase behavior, as shown in

Table 1. The research methods of the phase behavior.

Research Methods			Principles	Advantages	Disadvantages
Experimental methods	Rock core displacement experiment [55]		Analyze recovery rate and other data based on experimental parameters	Realistic rock core displacement with high fidelity	Long cycle, not visually observable, and poor repeatability
	Adsorption–desorption method [56,57]		Isotherm adsorption line	Direct approach to observe the phase transition point and critical behavior of hydrocarbons in nanoporous carbon	Most are non-realistic rock cores
	Differential scanning calorimetry (DSC) [58]		Determining material thermal properties by measuring the rate of heat release or absorption during temperature increase	Simulating the bubble point temperature of confined hydrocarbons	Difficulty in controlling and measuring phase transition rates
Theoretical simulation	Molecular simulation	Molecular dynamics (MD) [59]	Numerically solving the classical equations of motion allows for the determination of the phase trajectory of a molecular system and the characterization of its macroscopic thermodynamic properties	High accuracy for simple components	High computational cost, difficulty in calculating near the critical point
		Monte Carlo (MC) [12]	Repeatedly sample different configurations of a molecular system and calculate the total energy of each configuration
	The equation of state [27,51]		Phase equilibrium and force calculation	Consider capillary forces and critical displacement	Depend on the equation of state
	Density functional theory (DFT) [52,60]		Fluid molecular density as the fundamental variable to describe the thermodynamics of a system	Exhibits minimal discrepancies with molecular simulation results	Limited to simple molecular density statistics
Instrumental analysis	Nuclear magnetic resonance online scanning (NMR) [61]		1H and 13C nuclei resonate in the magnetic field	Full-scale pore observation in the range of nanometers to millimeters	In situ imaging of oil or water requires phase shielding, resulting in low imaging accuracy and high costs
	CT scanning		Three-dimensional image scanning and reconstruction	Digital rock	High sample requirements and high costs

, experimental studies to validate the effectiveness of correction calculations are still in their early stages. Micro–nanofluidic technology allows for high-resolution imaging of fluids in nanochannels under high-temperature and high-pressure conditions, providing visual support for understanding the phase transition characteristics and flow behavior of fluids in confined spaces. It serves as a new experimental approach to investigate and validate predictions of nanoscale properties and the phase behavior of fluids in multiscale media.

4.1. Phase Behavior

In oil and gas extraction, bubble nucleation and vaporization are more favorable in larger pores, while they are constrained in nanoscale porous matrices, significantly affecting the fluid critical parameters and occurrence states. For instance, with the enhancement of confinement effects, the bubble point pressure is significantly reduced, while the upper dew point pressure increases and the lower dew point pressure decreases. Such phase behavior differences may have a significant impact on oil and gas recovery or data fitting.

Currently, methods for predicting phase behavior in nanoscale confined spaces include density functional theory [62] (DFT), molecular simulations [63,64,65], modified equation of state considering various parameter influences (such as the Peng–Robinson equation) [66,67], or a combination of the aforementioned methods [24]. However, the predicted results sometimes remain controversial [68], necessitating experimental verification of prediction accuracy. For instance, in the case of the offset phenomenon in confined space bubble points, the most commonly used method is to modify critical properties in vapor–liquid equilibrium calculations and consider the capillary pressure in the modified equation of state. The bubble point pressure lines generated using these two different methods exhibit slight differences, particularly with significant discrepancies near the critical point. Therefore, more experimental data are required for the accurate comparison and calibration of these methods. Experimental methods include differential scanning calorimetry (DSC) [58], the constant volume method [69], or traditional PVT experiments to determine the bubble point through volume–pressure curves. The testing apparatus is generally a plunger or piston-type PVT instrument, which consumes a large amount of oil sample and requires long equilibration time. Micro–nanofluidic chip devices have small thermal mass and rapid response, enabling the continuous measurement of multiple saturation pressures [70]. By confining the fluid sample within a closed cavity at the end using CO₂ and oil properties, the precise isolation of the fluid sample within the cavity is achieved, compensating for the existing experimental limitations. Currently, research on the evaporation and condensation of single-component fluids at ambient temperature and pressure based on micro–nanofluidic technology is relatively well-developed (as shown in

Table 2. The investigation of phase behavior in single-component systems.

Experimental Objective	Component	Size	Phenomenon
Evaporation	n-pentane [71]	50 μm 145 nm	Nanoporous capillary confinement enhances liquid capture and significantly impedes liquid evaporation, reducing the evaporation rate by approximately 16 fold.
	n-pentane [72]	100 nm 5 μm	The confinement effect leads to the preferential evaporation of the fluid in microchannels over nanochannels.
Condensation	propane [11]	30 nm 50 nm 500 nm	The condensation pressure for the 50 nm chip is close to the prediction of the Kelvin equation, while for the 30 nm chip, the condensation pressure is significantly lower than the predicted value.
	propane [73,74]	70 nm 100 nm	There exist disparities between the two length scales.
	n-butane [12]	2 nm	The deviation of the condensation pressure from the bulk phase reaches as high as 22.9%.
Bubble point temperature	n-hexane, n-octane, n-heptane [50]	4 nm 20 nm 50 nm 100 nm	The bubble point temperatures for 4 nm and 10 nm confinement exhibit significant deviations compared to those for 100 nm and 50 nm confinement. The confinement effect is more pronounced in the 4 nm channel, leading to a noticeable increase in the bubble point temperature.
Dew point pressure	n-butane [75]	4 nm 10 nm 50 nm	At 4 nm, the dew point pressure exhibits a deviation of up to 14% compared to the bulk phase.

).

Factors influencing the phase behavior of multi-component systems are more complex compared to single-component systems [72], and research in this area is relatively scarce. Zhang et al. [76] conducted experimental studies on the static phase behavior and fluid flow behavior of pure CO₂ and CO₂-C₁₀ systems in shale dual-scale porous reservoirs. The results showed significant changes in the static behavior of the fluid from the bulk phase to the nanoscale. Jatukaran et al. [77,78] used chip simulations to investigate the evaporation rates of methane, propane, and butane ternary hydrocarbon mixtures in nanoscale pores, revealing a 3000-fold reduction in evaporation rate, which further decreased at low temperatures. Alfi et al. [79] studied the bubble point temperatures of butane, hexane, and heptane mixtures using etched chips with depths of 10 nm, 50 nm, and 100 nm, and observed significant deviations in the bubble point temperatures at the 10 nm scale. Despite the progress made in multi-component phase-behavior studies across different length scales, these studies have not considered the mass transfer diffusion and the impact of flow on evaporation. Therefore, further improvement is needed in this area of research.

4.2. Seepage

Capillary forces, as the main driving force of seepage, make the seepage effect in nanopores more significant. Multiple studies have shown that the pore size of reservoir rocks reaches the nanoscale, and capillary pressure increases to several megapascals. The seepage of fluids in microchannels can be described as:

$$q=\frac{P_{ca}}{f}=wh\frac{dl}{dt}$$

(1)

where $q$ is the mass flow rate, $P_{ca}$ represents the capillary force, $f$ is the flow resistance, $w$ represents the width of the nanochannel, $h$ represents the depth of the nanochannel, and $l$ represents the seepage distance.

The capillary pressure in a circular channel can be represented as:

$$P_{ca}=\frac{2\gamma \cos \theta }{r}$$

(2)

However, microfluidic chip designs often have a high aspect ratio with channel widths in the micrometer range and depths in the nanometer range ($w$ » $h$). Therefore, the fluid flow in such channels resembles that of a flat plate flow. Consequently, combining Equations (1) and (2), the capillary pressure $P_{ca}$ in these channels can be expressed as:

$$P_{ca}=\frac{2\gamma \cos \theta }{h}$$

(3)

where $\gamma$ represents the surface tension, and $\theta$ is the contact angle.

Flow resistance:

$$f=\frac{12\mu l}{w{h}^{{}^3}}$$

(4)

Therefore, the seepage distance can be obtained as shown in Equation (5):

$$l=\sqrt{\frac{\gamma h\cos \theta t}{3\mu }}$$

(5)

where $\mu$ is the fluid viscosity, t represents the time.

According to the Washburn equation [80], seepage is closely related to surface tension and viscosity, and as the pore size decreases, the surface tension and viscosity of fluids deviate significantly from the bulk phase [81]. However, there are differences in the results of seepage experiments in confined spaces. For example, Li et al. [82] studied the effect of surface tension, viscosity, and contact angle on seepage in polar and non-polar mixtures at the 8 nm scale using microfluidic chips, and the experimental results showed that the Washburn equation was generally applicable in confined spaces. Another study showed that the oil uptake rate of shale in a 34 nm channel was only 40% of the theoretical calculation [83]. The reason for this is that the strong interaction between the nanochannel wall and the fluid increases the resistance, thereby slowing down the self-seepage process.

4.3. Diffusion Mass Transfer Study

The effectiveness of CO₂ for enhanced oil recovery depends on its potential to alter the properties of the reservoir oil, known as the phase-mixing ability, which in turn depends on the mass transfer capability between CO₂ and oil [84,85]. Studying the mass transfer and reaction kinetics in gas–liquid two-phase flow is of great significance for improving oil recovery.

The key parameter in the diffusion process is the molecular diffusion rate. At the nanoscale, the size of substance molecules approaches the limiting size, leading to a deviation between nanoscale diffusion rate and bulk diffusion rate, even though some convection [86] is restricted, resulting in relatively high mass flow rates [87], which affects mass transfer. However, under supercritical conditions, the interfacial effects are significantly weakened, and both convection and diffusion transport are enhanced. Previous studies on oil and gas mass transfer and diffusion laws have mostly relied on traditional PVT experiments or bubble towers to determine the kinetic data in the liquid phase. Under certain conditions, after gas makes contacts with oil, it diffuses into oil under the influence of concentration difference, and a diffusion model is used to fit the pressure drop curve to obtain the diffusion coefficient [88]. However, such methods not only have low heat and mass transfer efficiency but also have difficulties in separating diffusion from convection in large-scale measurement systems. For example, convection caused by changes in liquid-phase density due to gas dissolution is often ignored, leading to uncertainties and inaccuracies in measurements of phase, composition, velocity, and concentration.

Microfluidic technology, as a small-scale and non-invasive method for precise measurement of gas–liquid diffusion coefficients, is necessary for revealing quantitative information about gas–liquid contact and mass transfer diffusion. This method can calibrate the relationship between different concentrations and fluorescence emission intensity and convert fluorescence intensity into pH values using a standard curve [89]. The Fick diffusion equation is used to fit the change in concentration with diffusion distance to obtain the diffusion coefficient. Qiu et al. [90] used a fluorescence-based microfluidic chip method to measure the diffusion coefficient of NO₂ in H₂O₂ solution. Hu et al. [91] studied liquid flow in microscale pores at various flow rates and heat fluxes, analyzing the relationship between liquid supply capacity in the two-phase region and enhanced heat transfer in porous structures. The above studies indicate that at the nanoscale, the classical Fick diffusion law still has some applicability, but the diffusion coefficient of molecules will be significantly smaller than the macroscopic theoretical prediction.

4.4. Minimum Miscibility Pressure

The minimum miscibility pressure (MMP) is a key parameter for evaluating and optimizing CO₂ flooding in oil reservoirs, representing the lowest pressure at which gas and oil form a miscible phase. When the pressure (P) is lower than MMP, CO₂ and oil do not mix, and the displacement process is dominated by mobility and capillary forces, leading to fingering and premature breakthrough. On the other hand, when P is higher than MMP, CO₂ and oil reach a near-miscible or miscible state, the gas–oil interface disappears, the crude oil expands, and hydrocarbons evaporate. In the miscible flooding process, the capillary forces caused by pore size differences are significantly reduced, resulting in a significant increase in oil recovery efficiency [92] (as shown in Figure 6). Conventional reservoirs are associated with convection-dominated transport, while unconventional reservoirs are associated with diffusion-dominated transport [93], indicating that pore size to some extent affects MMP. The determination of MMP is generally based on the solubility theory, which can be expressed as:

$$\delta =\sqrt{T{\left(\frac{\partial P}{\partial T}\right)}_{\nu }-P}$$

(6)

$$P=\frac{RT}{\nu -b}-\frac{a}{\nu \left(\nu +b\right)+b\left(\nu -b\right)}$$

(7)

where $\nu$ is the molar volume, $\delta$ is the solubility, $R$ is the universal gas constant, and $a$ and $b$ represent the pressure and volume terms, respectively, in the Peng-Robinson equation of state.

$$a=\frac{0.45724R^2{T}_{cp}{}^2}{P_{cp}}{\left[1+k\left(1-\sqrt{\frac{T}{T_{cp}}}\right)\right]}^2$$

(8)

$$b=\frac{0.0778R{T}_{cp}}{P_{cp}}$$

(9)

where $k$ is a function of the eccentric factor, $T_{cp}$ and $P_{cp}$ are the supercritical temperature and pressure in the confined space, and their relationship with the critical temperature and pressure can be expressed as:

$$T_{cp}=T_c-T_c\left(0.22958\frac{{\left(\raisebox{1ex}{$T_c$}\!\left/ \!\raisebox{-1ex}{$P_c$}\right.\right)}^{\frac{1}{3}}}{r}-0.014378\frac{{\left(\raisebox{1ex}{$T_c$}\!\left/ \!\raisebox{-1ex}{$P_c$}\right.\right)}^{\frac{2}{3}}}{r^2}\right)$$

(10)

$$P_{cp}=P_c-P_c\left(0.22958\frac{{\left(\raisebox{1ex}{$T_c$}\!\left/ \!\raisebox{-1ex}{$P_c$}\right.\right)}^{\frac{1}{3}}}{r}-0.014378\frac{{\left(\raisebox{1ex}{$T_c$}\!\left/ \!\raisebox{-1ex}{$P_c$}\right.\right)}^{\frac{2}{3}}}{r^2}\right)$$

(11)

where $r$ is the pore radius, and $T_c$ and $P_c$ are the critical temperature and pressure.

In conclusion, the solubility $\delta$ is a function of the critical temperature and pressure parameters $T_c$, $P_c$, the molar volume $\nu$, and the pore radius $r$. This indicates that the transition from bulk phase to nanochannel critical properties dominates the miscibility of fluids in confined spaces [94].

Traditional experimental methods for determining MMP mainly include the slim tube experiment (ST) [95], rising bubble method (RBA) [96], and the disappearance of interfacial tension method (VIT) [97]. However, these experimental methods are time-consuming, subjective, lack quantitative information, and are not suitable for complex mixtures [98]. Theoretical methods simulate the situations of immiscible and miscible phases by coupling the Navier–Stokes equation, interface tracking equation, and convection–diffusion equation, or calculating critical parameters using state equations. For example, Zhang et al. [99] calculated that the critical temperature and pressure decrease as the pore radius decreases based on the van der Waals state equation and the semi-analytical state equation. The Peng–Robinsion EOS (PR EOS) proposes different fitting formulas based on the magnitude of the eccentricity factor, thereby exhibiting better performance near the fluid’s critical point. As a result, the PR EOS is currently the most popular equation of state in the petroleum industry [100,101]. However, these calculations are computationally expensive and depend on boundary conditions. In recent years, methods such as nuclear magnetic resonance and CT scanning have emerged for determining MMP, which can measure the minimum miscibility pressure of oil and gas without invasive procedures. However, the high cost of equipment limits their large-scale application [102]. Therefore, the existing methods are unable to meet the requirements for the in situ detection of oil–gas miscibility pressure at the micro- and nanoscale. Microfluidic experimental platforms estimate MMP by observing the appearance of the interface between CO₂ and oil [103], or quantitatively observe the mixing process of CO₂ and crude oil in microchannels using fluorescence imaging. The design of quasi-static (dead-end) microfluidic chips can also avoid complex experimental operations and potential measurement errors caused by pressure drop in dynamic flow, enabling a large number of tests to be performed in a short period of time. Compared to standard ST tests, this method provides a sufficient number of data points, and existing experiments have demonstrated its effectiveness [104].

4.5. Displacement Efficiency

The spreading law refers to the behavior of fluid displacement in different scales, including the flow and transport mechanisms in nanoscale pores and fractures. Gas injection for enhanced oil recovery (EOR) includes CO₂ [105,106,107], natural gas [108], and nitrogen [109], among which CO₂ has been widely studied due to its high displacement efficiency and applicability in low-permeability reservoirs [110]. The improvement of oil recovery by CO₂ involves the flow of oil and gas in nanoscale pores, gas dissolution, and diffusion [111] in low-permeability matrix, oil swelling [112], wettability alteration [113], the reduction in interfacial tension [114], and interaction between fractures and matrix. Therefore, a comprehensive understanding of the flow and transport mechanisms in porous media and fractures at different scales is crucial for the development of effective techniques to enhance oil recovery. In China, there is an urgent need to scale up the application of CO₂ flooding technology for stable and increased oil production. So far, various CO₂ injection schemes have been developed by the industry and academia for carbon sequestration and oil and gas extraction. Fluids in nanochannels enter the supercritical state in a sequence of smaller pores before entering larger ones, which exhibits a positive influence on the microscale spreading efficiency and flow path of CO₂, while the curvature of porous media may have a negative impact. Therefore, studying the miscible and immiscible displacement patterns across scales is of great engineering significance for understanding the displacement processes in real reservoirs. However, few experimental studies have considered the impact of cross-scale effects on unconventional reservoir displacement spreading.

Conventional core displacement experiments rely on analyzing changes in core mass before and after displacement or using characterization techniques such as nuclear magnetic resonance, which are time-consuming and have low repeatability [115]. Additionally, the spatial distribution and surface morphology of reservoir fractures are complex, and the unique flow patterns are the focus of displacement studies. For example, in actual production, as the reservoir pressure decreases, CO₂ in the oil grows and condenses into large gas bubbles that escape from the matrix into the fractures. The oil in the fractures is drawn into the matrix and reoccupies the pore space, significantly affecting the oil recovery in the matrix phase and hindering the geological storage of CO₂ in the reservoir. However, existing technologies are unable to directly observe and fully simulate the complex flow characteristics of reservoirs.

Microfluidic experiments based on micro- and nanofluidic chips with different matrices (as shown in Figure 7) can not only investigate the spreading laws of oil displacement in micro- and nanoscales but also directly observe the coupled mechanisms of oil flow and phase behavior under different displacement modes. For example, Lu et al. [116] studied the multiphase transport mechanisms during CO₂ imbibition under different wettability conditions, and the experimental results showed that the presence of water phase can slow down the rapid transport and gas channeling of CO₂ from the matrix to the fractures, thereby improving oil recovery. Huang et al. [117] used a high-pressure visualization system at the pore scale to study the influence of the presence of aqueous phase on CO₂ exsolution under different initial states, pressure drop rates, and wettability conditions [118]. Guo et al. [119] found that the residual saturation of the displacing fluid increases with the increase in the number of capillary tubes. Zhong et al. [120] studied the pressure threshold effect of N₂ immiscible displacement and CO₂ miscible displacement in confined spaces of 60 nm.

5. Research on CO₂ Geological Sequestration

CO₂ geological sequestration primarily involves CO₂-enhanced oil (natural gas) recovery and storage, CO₂ displacement for coalbed methane storage, and CO₂ storage in saline aquifers [121]. Among them, reservoir storage and saline aquifer storage utilize mechanisms such as structural trapping, capillary trapping, dissolution trapping, and mineral trapping (as shown in Figure 8). The sequestration process is controlled by various parameters such as fluid properties, interfacial tension, wetting [122,123], solubility, and pore throat characteristics. Due to the coupling phenomena of fluid flow, geochemistry, and biogeology in porous media during CO₂ sequestration, traditional core displacement experiments are insufficient to understand the pore-scale mechanisms of CO₂ storage, and there are significant changes in thermophysical properties near the critical point. Therefore, accurate data on the interaction between CO₂-saline water-porous media are key to improving storage capacity and the reliability of numerical simulations for geological sequestration [124]. Conventional methods are time-consuming, and there is a greater need for rapid, accurate, and reproducible methods to precisely describe key data. Micro–nanofluidic experiments can be used to optimize numerical and modeling methods from the pore scale to the reservoir scale, making it an ideal experimental approach for studying CO₂ sequestration.

5.1. CO₂-Enhanced Oil and Gas Recovery and Sequestration Technology

During the process of CO₂-enhanced oil and gas recovery and storage, the majority of CO₂ is stored in underground structural and capillary spaces, a portion dissolves in residual underground fluids, and a small fraction is mineralized in underground rocks. Some CO₂ is also produced to the surface along with oil and gas flow. The dissolution of supercritical CO₂ into the saline water of the target reservoir, the physicochemical reactions between acidic saline water and solid matrix, the parameters of the aquifer, and the presence of fractures [125], as well as the information on the geometry of the pores introduce uncertainties to large-scale CO₂ storage applications. Additionally, capillary trapping, deposition structures, and the permeability of the target layer significantly affect the migration pathways of CO₂ [126]. Micro–nano fluidics technology provides a new approach for simulating the migration of CO₂ in complex geological structures and porous media.

5.2. Saline Aquifer CO₂ Sequestration Technology

Saline aquifers have great potential for CO₂ storage, but the density difference between CO₂ and the surrounding fluids increases the risk of upward migration and leakage. Salt precipitation in porous media can hinder carbon capture and storage in saline aquifers. Therefore, when CO₂ is dissolved into the formation brine, saline aquifer storage has a higher level of safety. Injected CO₂ diffuses in the porous media, displaces formation water, and is eventually stored underground after undergoing a series of physical and chemical reactions [127,128]. The dissolution of CO₂ in water depends on temperature, pressure, and salinity conditions [129]. Therefore, developing more effective methods to represent the interaction between CO₂ and water and the behavior of water and gas in porous media is the basis for calculating the amount of carbon storage in saline aquifers [130]. CT scanning is commonly used to observe sediment distribution and changes in pore size and permeability to determine the impact of salt precipitation on geological carbon storage (GCS) [131]. Additionally, scanning electron microscopy (SEM) and X-ray diffraction (XRD) can analyze the products of the interaction between nanoscale porous media and CO₂ [132]. Microfluidic technology is an emerging method for studying salt precipitation and has been used to investigate the effects of two configurations of precipitates and pore structure on salt crystal growth rate [133], as well as co-confocal Raman spectroscopy for solubility studies [134].

6. Limitations

(1) The dimensions of microfluidic chips currently mainly focus on longitudinal depth differences, and the manufacturing of high aspect ratio chips with higher throughput still faces technological and cost issues. In addition, there are discrepancies between existing microfluidic experimental results and theoretical simulations (such as mixed-phase pressures and phase envelopes), and the reasons for these discrepancies still need further exploration. Furthermore, there is no unified definition for the size boundary of the confinement effect, and extensive experimental research and simulation predictions are still needed.

(2) Most existing studies mainly focus on the phase behavior of small molecule hydrocarbons or binary mixtures at a single-pore-size scale. However, the multi-component nature of reservoir fluids further complicates the confinement effect, and the mass transfer diffusion between components cannot be ignored. There are still certain deficiencies in the study of multi-component mixture phase behavior across scales based on micro-fluidic technology. In addition, idealized pore geometries and pure fluid systems are not sufficient to represent the inherent complexity of multi-component mixtures in unconventional reservoirs.

(3) Target formations for CO₂ geological sequestration often contain brine, and the interaction between CO₂ and pore media and formation water under high temperature and pressure conditions still needs further clarification. The rate of salt growth and in situ distribution of precipitation at the pore scale require extensive experimental research.

7. Prospects

(1) Pores below 10 nm approach the scale of CO₂ and hydrocarbon molecules more closely, and the effects of fluid continuity, solid–liquid interface properties, and critical parameters on phase behavior may have important implications for recovery and storage rate analysis.

(2) Micro–nano fluidics experiments can provide support for future studies on CO₂ migration in water and tight caprock, and serve as a reference for the design of CO₂ geological storage and simulation of supercritical CO₂ dissolution processes.

(3) The development of micro–nano fluidics technology should be combined with innovation from multiple disciplines to achieve more precise fluid control and response. For example, the simulation of reservoir wettability in CO₂-enhanced oil recovery and geological storage can refer to surface modification of chips from disciplines such as biomedicine. The study of droplet condensation, adsorption, and separation in CO₂ capture processes can also draw inspiration from micro–nano fluidic technologies. Additionally, insights can be gained from filtration and separation applications in other disciplines.

(4) In the future, micro–nano flow control experiments can investigate how the efficiency of CO₂ flooding and storage can be improved by studying the effects of additives such as surfactants, foams, and nanoparticles, as well as changes in physical and chemical conditions. They can also explore emerging fields such as microbial enhanced oil recovery or geothermal energy to enhance the understanding of fine-scale mechanisms in pore structures and establish connections with reservoir-scale phenomena.

(5) Supercritical CO₂ is in a state that exists between gas and liquid at critical conditions. It has wide-ranging industrial applications. In the future, research on supercritical CO₂ based on micro–nanofluidic experimental platforms can refer to low-carbon extraction technology using supercritical CO₂ as an organic solvent, the facile control of morphological structure in the synthesis of nanoparticles, supercritical drying technology, carriers for nanoscale drug synthesis, and reaction media for organic synthesis under special conditions.

8. Conclusions

The capture, utilization, and storage of CO₂ have broad application prospects, and the application of micro–nanofluidics technology in CO₂ capture is just beginning to emerge and requires further exploration. Although the application of CO₂-enhanced oil and gas recovery has been ongoing for many years, it is currently limited to the study of simple mixtures at a single scale due to manufacturing processes and fluid control limitations. There is a significant lack of research on more realistic scenarios such as crude oil degassing, CO₂ extraction, and stripping. Additionally, there is a scarcity of research on caprock leakage in CO₂ geological storage, and further clarification is needed on the mechanisms of CO₂ interaction with saltwater and rocks. In conclusion, there is a tremendous demand for research on CO₂ capture, utilization, and storage based on micro–nanofluidics technology, making it an indispensable and important tool in future experimental developments.