The Influence of Supercritical CO₂ Displacement at Different Temperatures on Porosity and Permeability Evolution in Marine Unconsolidated Strata

Abstract

CO₂ geological sequestration in marine sediment is one of the important ways to lower carbon emissions. To study the influence of CO₂ sequestration on the permeability and porosity of unconsolidated strata, this paper conducted overpressure permeability, isothermal adsorption and CO₂ displacement experiments. Through nuclear magnetic resonance (NMR) technology, the effects of supercritical CO₂ (SCO₂) at different temperatures on the permeability evolution and pore size variation of unstratified strata in marine hydrate reservoirs were studied. The experimental results show that: (1) When the pressure changed from 0 to 17.5 MPa, the permeability and porosity of the soil samples decreased sharply. The porosity dropped from 36.83% to 16.07%, and the permeability also decreased from 48.53 mD to 1.18 mD. (2) During the adsorption tests, the fitted absolute adsorption capacity of CO₂ and CH₄ gradually increased with pressure growth. The maximum fitted absolute adsorption capacity of CO₂ was 2.45 times that of CH₄. (3) Through displacement experiments, the porosity and permeability increments during SCO₂ displacement were much greater than those during non-SCO₂ displacement. From 30 °C to 70 °C, the increments of porosity and permeability all increased. After SCO₂ displacement, the pores’ proportions (>0.1 μm) increased for all samples, with the largest growth rate reaching 34.37%. Above all, these results indicate that environmental pressure significantly affects the permeability of soil samples, and that SCO₂ displacement can effectively enhance the proportion of large-sized pores, thereby further improving the permeability of unconsolidated strata.

1. Introduction

In recent years, the excessive use of fossil fuels has led to a gradual increase in atmospheric CO₂ levels. In 2023, global emissions reached 37 billion tons, with the annual average CO₂ concentration rising to 419 ppm, nearly 50% higher than pre-industrial levels [1,2,3]. Against the backdrop of global warming and increasingly frequent extreme weather events, reducing carbon emissions and enhancing carbon sequestration have become urgent priorities [4,5]. Currently, the geological storage of CO₂ in underground reservoirs stands as a crucial measure to reverse the worsening trend of global climate change and as one of the most effective methods for lowering atmospheric carbon dioxide concentrations [6].

Among these, CO₂-Enhanced Oil Recovery (CO₂-EOR) technology can enhance recovery rates while simultaneously achieving CO₂ sequestration, thereby fulfilling carbon reduction objectives [7,8,9]. This technology has undergone large-scale trials and industrial applications in oil and gas fields across multiple countries. CO₂-EOR technology was first field-tested in North America. Particularly in Texas’ Permian Basin, dozens of large-scale CO₂-EOR projects have been implemented since the 1970s. By 2020, over 140 CO₂-EOR projects were operational or under trial in the United States, with cumulative injection exceeding 1 × 10⁸ tons and annual crude oil production increases reaching millions of barrels [10]. In addition, the Weyburn Oilfield Project is a typical joint demonstration initiative in Canada for CO₂-EOR and carbon sequestration. Since 2000, it has injected industrial CO₂ tail gas from North Dakota, USA, achieving the dual objectives of enhanced oil recovery and stable carbon sequestration [11]. In Europe, Norway’s Sleipner gas field became the first project to inject CO₂ on a large scale into deep saline aquifers for storage. Although its primary objective was not to enhance oil and gas recovery, it demonstrated the technical feasibility and storage safety for subsequent CO₂-enhanced oil recovery projects [12,13,14]. Given the favorable economic and environmental benefits achieved by CO₂-EOR, a method of CO₂ displacement for extracting natural gas hydrates has been proposed [15,16], and numerous scholars have conducted many related studies. In 1993, Ebinuma et al. [17] first proposed the theoretical concept of utilizing CO₂ displacement technology for natural gas hydrate extraction. Subsequently, Ohgaki et al. [18] experimentally demonstrated the technical feasibility of the CH₄/CO₂ displacement method for gas hydrate extraction in 1996. Pandey et al. [19] conducted comparative analyses of CO₂ displacement characteristics for CH₄ hydrates in pure water systems, confirming that increasing the contact area between hydrates and CO₂ could significantly enhance displacement rates and efficiency. Furthermore, Ota et al. [20] observed via in-situ Raman spectroscopy that even after 307 h of displacement reaction, 65% of CH₄ molecules remained trapped within the hydrate cage, yielding a displacement efficiency of only 35%. Zhang’s research group [21] innovatively proposed a combined extraction method integrating CO₂ displacement with thermal stimulation, successfully increasing the CH₄/CO₂ displacement efficiency to 64.63%. However, current research indicates that the CO₂ displacement of natural gas hydrates proceeds at a relatively low speed and with low efficiency, failing to meet industrial application requirements [22,23].

To address the challenges of poor permeability and low efficiency in CO₂–hydrate displacement in marine silty clay reservoirs, CO₂ displacement experiments were conducted using remolded soil samples under confining pressure to simulate seafloor condition. In this research, SCO₂ at different temperatures (from 30 °C to 80 °C) was sequentially injected into saturated water-saturated soil samples. Using NMR technology, the evolution of soil sample permeability and porosity at different displacement temperatures was investigated. This result will provide an optimized approach for further research on the efficiency of SCO₂–hydrate displacement in marine unconsolidated strata.

2. Experiments

2.1. Experimental Materials

The study area was located in the Pearl River Mouth Basin, which is situated on the continental shelf in the Northern part of South China Sea (in Figure 1) [24]. It is a Middle and Cenozoic basin developed on the basis of a Mesozoic active continental margin basement structure. Marine gas hydrate reservoirs are mainly located in unconsolidated sedimentary strata beneath the seafloor at the continental margin [25,26]. The seabed temperature is mostly between 2 °C and 4 °C, with a geothermal gradient of 44 to 67 °C/km [27]. The pressure on the seabed is generally greater than 10 MPa. This stratum is essentially a low-permeability, mechanically weak sedimentary body with extremely low mechanical strength. Its formation resulted from the combined effects of specific temperature–pressure conditions and sedimentary dynamics, exhibiting characteristics of unconsolidated rock, low permeability, and the coexistence of gas hydrates.

To study its characteristics, several core samples from unconsolidated strata were taken, and detailed tests on these samples were conducted. The results are shown in

Table 1. Basic physical properties of the soil samples.

Density (ρ)	1.61 g/cm3
Water content (w)	28.56%
Porosity (n)	44.27%
Permeability (mD)	17.19

. The density of soil samples was found to be 1.61 g/cm³. Moreover, it exhibited extremely low permeability and was highly friable, presenting a non-consolidated state. The oxide content of the sampled soil was tested using XRF, and the particle size of soil was measured using a laser particle size analyzer. There was a large amount of SiO₂, Al₂O₃ and CaO in the target stratum (as shown in

Table 2. XRF analysis of the component content of soil samples.

Component	SiO2	Al2O3	CaO	Fe2O3	MgO	K2O	Na2O
Content (wt.%)	52.65	17.10	11.74	7.02	3.37	3.64	1.85

), with the particle size mainly ranging from 0.4 to 400 μm. Among them, D₁₀ was 3.673 μm, D₃₀ was 14.98 μm, D₅₀ was 48.30 μm, and D₆₀ was 66.11 μm. This soil sample was determined to be silty clay based on the particle size gradation distribution. In addition, the gas permeability of the soil sample was 17.19 mD, with a porosity of 44.27%, which fell within the data range for the porosity and permeability of hydrate reservoirs in the South China Sea [28]. Although the pore structure of remolded soil samples may differ from that of the in-situ samples, it still could, to some extent, reflect the characteristics of original soil samples.

Due to the difficulty in preserving in-situ hydrate soil samples, remade soil samples were prepared using soil from hydrate reservoirs to study the CO₂ displacement effects. The diameter of reconstructed soil samples was 2.5 cm and the height was 6.0 cm. Energy-dispersive X-ray spectroscopy (EDS) was performed on the remolded soils to determine their elemental composition and precise mineralogical information. The specific results were presented in

Table 3. EDS analysis of element content in the remolded soil.

Components	O2	Si	Al	Fe	Na	Mg	K	Ca	S	Ti
Content (wt.%)	47.17	29.02	7.55	4.75	3.46	2.80	2.45	1.20	0.63	0.32

and Figure 2, where oxygen constituted the largest proportion at 47.17%, followed by sodic feldspar at 41.70%.

2.2. Experimental Setup

The experimental samples were scanned using an NMR scanner provided by Suzhou Tainiu Testing Service Co., Ltd. (Suzhou, China), with an effective maximum core size measurement capacity of Φ 25.4 mm × H 60 mm.

The CO₂ displacement test system consisted primarily of a carbon dioxide cylinder, a CO₂ pressurization device, a pressure sensor, a confining pressure control device (with a maximum confining pressure of 22 MPa), a temperature control cabinet, soil samples, a clamper and a water collector. The clamper constitutes the central component of this test apparatus, serving to mechanically secure the core within a sealed chamber. It employs a hydraulic system to simulate the confining pressure of underground reservoirs, thereby preventing fluid leakage from the sides of the core. In this system, the intermediate container pressurized CO₂ to 8 MPa. Furthermore, the temperature control cabinet heated CO₂ to reach the supercritical state. The confining pressure control device produced and maintained confining pressure using water. A diagram of the experimental system is shown in Figure 3.

2.3. Experimental Processes

(1) Overburden permeability tests were conducted on the soil samples obtained from marine unconsolidated sediments at ambient temperature (25 °C). The test pressures ranged from 0.0 to 17.5 MPa. The porosity and permeability of the reconstructed soil samples was measured under different pressures. Additionally, isothermal adsorption tests of CO₂ and CH₄ were carried out at ambient temperature to evaluate their adsorption capacities, with pressures ranging from 0 to 25 MPa.

(2) CO₂ displacement experiments were conducted at different temperatures, with injection temperatures at 30 °C, 40 °C, 50 °C, 60 °C, 70 °C, and 80 °C. To simulate the temperature and pressure conditions of subsea sediment samples, the confining pressure was set at 10 MPa, with a CO₂ injection pressure of 8 MPa. CO₂ at temperatures exceeding 31.1 °C and pressures exceeding 7.38 MPa is classified as supercritical CO₂ [29]. For comparison, non-supercritical CO₂ was injected at 6 MPa and 25 °C. The test parameters are detailed in

Table 4. Parameters of the CO₂ displacement experiments.

Number	Confining Pressure	CO2 Injection Pressure	Injection Temperature	CO2 Phase State
No. 1	10 MPa	6 MPa	25 °C	non-supercritical phase
No. 2	10 MPa	8 MPa	25 °C	non-supercritical phase
No. 3	10 MPa	8 MPa	30 °C	supercritical phase
No. 4	10 MPa	8 MPa	40 °C	supercritical phase
No. 5	10 MPa	8 MPa	50 °C	supercritical phase
No. 6	10 MPa	8 MPa	60 °C	supercritical phase
No. 7	10 MPa	8 MPa	70 °C	supercritical phase
No. 8	10 MPa	8 MPa	80 °C	supercritical phase

Note: Due to variations in the initial permeability and porosity of each soil sample, duplicate tests could not be performed on any single sample. Therefore, each CO₂ displacement experiment was conducted only once.

. The CO₂ used in all experiments was of commercial grade with a purity of 99.95% to ensure the accuracy and reproducibility of the results. The detailed procedures were as follows:

① Sample Preparation: Reconstituted soil specimens were prepared according to displacement test specifications, based on measured in-situ soil density and moisture content. The specimens were sealed using a combination of heat-shrink tubing and spacers to prevent deformation or fragmentation during subsequent displacement procedures. Their initial weights (M₁) were determined using an electronic balance with a precision of 0.1 mg.

② Vacuum water retention and T₂ measurement: The sample was immersed in deionised water and placed in a vacuum water saturation apparatus at −0.1 MPa for 10 h. After saturation, the sample (M₂) was reweighed. The water-saturated sample was wrapped with PTFE tape to prevent water loss and contamination of the nuclear magnetic resonance (NMR) probe. Subsequently, the sample was placed in the NMR system for scanning. The porosity and permeability was measured using Equation (1). The primary CPMG sequences sampled by the NMR system were as follows: Spectral Width (SW) = 200 KHz, Wait Time (TW) = 3000 ms, and Number of Scans (NS) = 16.

$${\displaystyle \frac{1}{T_2}}\approx {\displaystyle \frac{1}{T_{2S}}}={\rho }_2·{\displaystyle \frac{S}{V}}$$

(1)

where T₂ denotes transverse relaxation time, ms; T_2S is the surface relaxation time, ms; ${\rho }_2$ is the surface relaxation rate, μm/ms; and S/V is the ratio of surface area to the volume of the pores, μm⁻¹.

③ Clamper preheating: First, a dummy core matching the sample dimensions was inserted into the clamper to prevent damage to the inner sleeve from high temperatures during preheating. Subsequently, the clamper containing the dummy core was placed within the temperature control system and heated for 20 min to ensure the interior reached and maintained the set test temperature, thereby establishing constant thermal conditions for subsequent sample testing.

④ SCO₂ displacement: The preheated clamper was removed, the dummy core was replaced with the actual saturated soil sample, and it was reinserted into the temperature control system. The CO₂ pressurization device was connected. As the section of the SCO₂ pipeline passing through the temperature control system had already reached the preset temperature prior to entering the sample, no additional heating of the SCO₂ was required. Displacement experiments were conducted on the saturated soil sample at various temperatures using the displacement system, maintaining a confining pressure of 10 MPa throughout. A continuous SCO₂ injection was maintained for a displacement duration of 4 h. At the outset of the experiment, continuous water discharge was observed at the outlet end of the apparatus. As displacement progressed, bubbles gradually appeared in the outlet fluid. The injection process of SCO₂ is illustrated in Figure 4.

⑤ Vacuum water retention and T₂ measurement: Following displacement, the sample was placed back into the vacuum water saturation apparatus and subjected to saturation treatment at −0.1 MPa for 10 h to ensure complete saturation of the internal pore space. Subsequently, NMR testing was conducted to obtain porosity and permeability data. The sample was weighed again (M₃).

⑥ Weight porosity was calculated using Equation (2). Then, the weight porosity was compared with NMR porosity to verify the accuracy of NMR porosity.

The weighting porosity of the soil sample after displacement was calculated according to Equation (2), as follows:

$$\left\{\begin{array}{c}{n}_1={\displaystyle \frac{M_2-M_1}{V}}\\ n_2={\displaystyle \frac{M_3-M_1}{V}}\end{array}\right.$$

(2)

where n₁ denotes the porosity of soil samples before displacement, %; n₂ denotes the porosity of saturated soil samples after displacement, %; M₁ is the mass of the original reconstructed soil sample, g; M₂ is the mass of the saturated reconstructed soil sample before displacement, g; M₃ is the mass of the saturated soil sample after displacement, g; and V is the volume of the soil sample, cm³.

3. Results and Discussion

3.1. Overpressure Permeability Tests

The overpressure permeability tests were conducted to investigate the effects of soil porosity and permeability changes on high-pressure marine environments. Figure 4 shows the variation curves of the porosity and permeability of the soil samples under ambient temperature as the confining pressure increased from 0 to 17.5 MPa. As shown in Figure 5, the pore structure and fluid flow capacity of soil underwent significant changes with increasing pressure. The porosity dropped markedly from 36.83% to 16.07%, owing to pore volume shrinkage from compaction under high-pressure conditions. At the same time, the permeability declined from 48.53 mD to 1.18 mD, suggesting that the pores available for fluid flow were markedly reduced, thereby greatly suppressing the seepage capacity of soil. During the pressure range of 0–4 MPa, permeability dropped sharply. Beyond this range, although the soil continued to compact and porosity steadily decreased, permeability exhibited little change. This phenomenon may be associated with the pore water within the soil; although the interparticle pores were compressed under external pressure, the incompressibility of water prevented water-filled pores from being compacted, and, in some cases, previously unconnected pores may have become interconnected under confining stress, thereby enhancing fluid mobility [30,31,32]. Therefore, pressure has an important and significant influence on the porosity and permeability characteristics of marine unconsolidated strata.

3.2. Isothermal Adsorption Experiments

The adsorption capacities of soil samples for CO₂ and CH₄ under different pressure conditions were tested to evaluate their gas storage capacity and sequestration potential, providing fundamental data for CO₂ geological storage in unconsolidated marine sediments. Isothermal adsorption experiments of CO₂ and CH₄ were carried out separately to investigate the adsorption behavior of the soil samples and the variation in adsorption capacity for both gases. To quantitatively describe the adsorption behavior of gases in reservoir media, the Langmuir isothermal adsorption model (Equation (3)) was employed to fit the experimental data [33]. This model assumes that adsorption occurs as a monolayer process with a finite number of adsorption sites that are independent of one another. It effectively represents the fundamental adsorption characteristics of gases in porous media such as coal and shale. From the fitting results, the Langmuir volume VL and Langmuir pressure PL indicate the maximum adsorption capacity and the adsorption strength of the gas, respectively, serving as key parameters for characterizing reservoir adsorption properties. Owing to its simple structure, clear physical meaning, and high fitting accuracy, the Langmuir model has been widely applied in studies of adsorption characteristics in unconventional gas reservoirs, analyses of gas displacement mechanisms, and evaluations of geological CO₂ sequestration.

$$n=n_0\times {\displaystyle \frac{P}{P_L+P}}\times (1-{\displaystyle \frac{{\rho }_g}{{\rho }_{ad}}})$$

(3)

In Equation (3), n is the adsorption capacity at the current pressure adsorption equilibrium, cm³/g; n₀ represents the Langmuir maximum adsorption capacity, cm³/g; P_L is the Langmuir pressure, MPa; P is the environmental pressure, MPa; ρ_ad is the adsorption phase density, g/cm³; and ρ_g is the free gas density, g/cm³.

Figure 6 and Figure 7 show the adsorption curves of CO₂ and CH₄ under different pressure conditions. As illustrated, due to differences in the molecular properties and thermodynamic states, the two gases exhibit distinctly different adsorption behaviors. The excess adsorption capacity of CO₂ increases rapidly with pressure, then decreases gradually, reaching a peak at approximately 7.5 MPa. This phenomenon is associated with the phase transition of CO₂ into the supercritical state. Fitting results indicate that the absolute adsorption capacity of CO₂ continued to rise, reaching a maximum of 4.36 cm³/g, with an adsorbed phase density of 0.938 g/cm³. In contrast, CH₄ adsorption increased monotonically with pressure and stabilized at approximately 1.78 cm³/g, with an adsorbed phase density of 0.537 g/cm³, consistent with the typical Langmuir monolayer adsorption characteristics. At 25 MPa, the adsorption capacity of CO₂ in the soil samples was 2.45 times that of CH₄.

The experimental results demonstrate that CO₂ has a stronger adsorption capacity than CH₄. As CO₂ transitions into the supercritical state, its adsorbed phase density increases significantly. Under identical pressure conditions, both the Langmuir volume and Langmuir pressure of CO₂ indicate stronger adsorption performance. These differences primarily arise from the distinct molecular properties of the two gases. Firstly, CO₂ is a linear polar molecule with stronger intermolecular forces, particularly enhanced electrostatic attraction and van der Waals interactions with organic surfaces or oxygen-containing functional groups. In contrast, methane, as a non-polar molecule, exhibits relatively weak adsorption affinity [34,35]. Secondly, CO₂ molecules have slightly smaller diameters, higher diffusion rates, and greater accessibility to micropore systems, which strengthen their competitive advantage in occupying nanoscale pores. Furthermore, CO₂ demonstrates stronger competitiveness at adsorption sites, effectively displacing pre-adsorbed methane molecules and thereby achieving displacement effects [36,37].

3.3. Effect of SCO₂ Displacement on Soil Sample Porosity at Different Temperatures

The porosity of the soil samples before and after CO₂ displacement at different injection temperatures was characterized using both NMR and weighting methods, as summarized in

Table 5. NMR porosity and weight porosity before/after displacement of the soil samples.

Number	NMR Porosity Before Displacement	NMR Porosity After Displacement	Increment of NMR Porosity	Weight Porosity Before Displacement	Weight Porosity After Displacement	Increment of Weight Porosity
No. 1	25.24%	26.76%	1.52%	25.08%	26.47%	1.39%
No. 2	26.08%	27.55%	1.47%	25.27%	27.28%	2.01%
No. 3	22.97%	24.97%	2.00%	22.54%	24.64%	1.90%
No. 4	22.26%	24.51%	2.25%	22.62%	24.19%	1.57%
No. 5	21.96%	27.82%	5.86%	21.79%	27.37%	5.58%
No. 6	21.91%	29.14%	7.23%	21.65%	28.94%	7.29%
No. 7	24.51%	34.64%	10.13%	24.35%	33.99%	9.64%
No. 8	23.05%	30.47%	7.42%	22.91%	29.91%	7.00%

. Among the 8 test samples, the displacement processes of No. 1 and No. 2 occurred under non-supercritical conditions, whereas the displacement tests for the remaining samples were conducted using SCO₂.

Overall, as shown in Table 5, the increase in soil sample porosity following SCO₂ displacement was greater than that observed with non-SCO₂. From 30 °C (No. 3) to 70 °C (No. 7), the porosity increments of the soil samples increased in both NMR porosity and weight porosity. In most cases, the porosity increment measured using the weight method was slightly lower than that obtained from NMR analysis. This phenomenon also validates the effectiveness of the test results. At 70 °C, the NMR porosity of the soil samples after displacement reached a maximum of 34.64%. In addition, the increase in soil sample porosity also reached its maximum value. In 80 °C (No. 8), the porosity increment was slightly lower than No. 7. Therefore, the experimental results above demonstrate that SCO₂ exerted a significantly greater impact on the porosity increment of soil samples compared to non-SCO₂ displacement. Moreover, to a certain extent, as the temperature of injected SCO₂ increased, the increment in soil sample porosity also increased.

3.4. Effect of SCO₂ Displacement on Soil Sample Permeability at Different Temperatures

Figure 8 displays the permeability changes of saturated soil samples before and after CO₂ displacement under different temperature conditions. It was evident that the permeability of all samples increased significantly following displacement. Before displacement, the overall permeability of soil was low at 25 °C. The permeability ranged from 1.2 to 2.87 mD, indicating that the soil samples had relatively poor permeability. After displacement, the permeability generally increased substantially, particularly under high-temperature conditions. For example, at 70 °C, permeability rose to 7.38 mD, much higher than other supercritical conditions. From 30 °C to 70 °C, post-displacement permeability increased with temperature, whereas at 80 °C, a slight decrease was observed. This change pattern was consistent with the variation in soil sample porosity. As shown in Figure 9, the increment in soil permeability began to rise from 30 °C and reached a maximum of 5.98 mD at 70 °C, but it did not continue to increase at 80 °C. Under 25 °C, the permeability increment with non-SCO₂ was relatively lower, at 0.87 and 2.01, respectively. These results indicate that non-SCO₂ has a far weaker influence on soil permeability increment than SCO₂.

3.5. Characteristics of Pore Structure Changes in Soil Samples Before/After SCO₂ Displacement

Low-field NMR can induce resonance in hydrogen nuclei within pore fluids. When an external magnetic field is applied, hydrogen nuclei in the fluid align directionally under the field’s influence, generating a magnetic signal. Upon removal of the external field, the aligned hydrogen nuclei revert to a disordered state, causing the magnetic signal to gradually decay [38]. Based on this principle, mathematical inversion of the NMR relaxation signal from this process yields an NMR spectrum of transverse relaxation time T₂. A low T₂ value indicates smaller pore diameters, whereas a high T₂ value indicates larger pore diameters. Furthermore, the spectral area exhibits a quantitative relationship with the molar mass of hydrogen nuclei, enabling the determination of fluid mass within pores [39,40]. When applying this technique to measure sample porosity, it is essential to maximize fluid filling within the pore space to ensure the NMR-measured porosity closely approximates the sample’s true porosity.

Taking the soil samples displaced by SCO₂ for No. 5, No. 6, and No. 7 as examples, the NMR T₂ characteristics before and after displacement were analyzed in Figure 10, Figure 11 and Figure 12. It can be seen that three peaks appeared in the T₂ curves before and after the displacement. The amplitude of Peak 1 after displacement was obviously larger than that before displacement. The width of Peak 1 also significantly widened. In three tests, the greatest increase in Peak 1 was No. 7. Peak 1 before displacement was 2179.74, and Peak 1 after displacement was 2922.41, with an increase of 34.07%. In addition, the values of Peak 2 and Peak 3 were approximately the same before/after displacement. Compared to before displacement, Peak 2 after displacement was only slightly shifted to the right. This result proved that the pore structure of the soil sample changed before and after displacement.

The pores were categorized into five size ranges (≤0.01 μm, 0.01–0.1 μm, 0.1–1 μm, 1–10 μm, and ≥10 μm). The proportions of each pore type are presented in Figure 13, Figure 14 and Figure 15. Before and after displacement, the predominant pore size in the soil samples was 0.01–0.1 μm. The pore volume proportion for this size ranged from 76.47% to 90.05%. Compared to the pore size distribution in the soil samples before SCO₂ displacement, the pore proportions (>0.1 μm) after displacement increased for all samples. Their growth rates were 34.37%, 16.79%, and 21.87%, respectively. However, all the pore proportions less than ≤0.1 μm decreased. These test results demonstrate that SCO₂ displacement effectively increases the proportion of large pores, thereby significantly enhancing the permeability of soil samples (in

Table 6. Pore percentage change before/after displacement in No. 5, No. 6 and No. 7 soil samples.

Soil Sample Number	Pore Percentage Before Displacement (≤0.1 μm)	Pore Percentage After Displacement (≤0.1 μm)	Increment	Pore Percentage Before Displacement (>0.1 μm)	Pore Percentage After Displacement (>0.1 μm)	Increment
No. 5	90.05%	86.63%	−3.80%	9.95%	13.37%	34.37%
No. 6	81.78%	78.72%	−3.85%	18.22%	21.28%	16.79%
No. 7	87.29%	84.51%	−3.18%	12.71%	15.49%	21.87%

).

Regarding the above experimental results, there are three hypothesized mechanisms for pore changes: (1) During the displacement processes, the pores in the soil samples are under the combined effects of confining pressure and SCO₂. Changes in pore pressure caused by confining pressure and CO₂ injection lead to significant alterations in the internal stress of the soil samples. As confining pressure increases, pore spaces and throats undergo compression and shrinkage. Some throats close off, forming isolated “dead pores”. Once isolated, the water trapped within becomes immobile and cannot be displaced by injected CO₂, thereby reducing microporous connectivity [41]. This mechanism is particularly pronounced in unconsolidated formations and significantly impacts the assessment of pore structure under reservoir confining pressure conditions. However, when the SCO₂ was injected, the pressure and temperature generated by SCO₂ further increased pore sizes, particularly those larger than 0.1 μm. (2) Simultaneously, higher temperature can reduce the viscosity of the fluids within the pores, accelerating flow velocity and improving pore connectivity. (3) Due to a large amount of CaO in the soil samples, the injected SCO₂ may react with CaO, causing the dissolution of some soil particles near the pores and, consequently, altering pore size. Ultimately, the combined effects of these two processes resulted in significant alterations to the pore size distribution of soil samples after SCO₂ displacement, particularly for pores larger than 0.1 μm.

4. Conclusions and Prospects

This study involved CO₂ displacement experiments combined with NMR analysis to investigate the effect of CO₂ displacement on the evolution of porosity and permeability in marine unconsolidated strata. The main results are summarized below:

(1) Overpressure permeability tests and adsorption experiments were conducted on remolded soil samples. The results show that the porosity and permeability of the samples decreased sharply as the pressure increased. The fitted absolute adsorption capacities of CO₂ and CH₄ increased with pressure. Below 25 MPa, the fitted absolute adsorption of CO₂ was 2.45 times greater than that of CH₄. These experiments demonstrated that pressure significantly influences the permeability and adsorption properties of unconsolidated samples. Furthermore, CO₂ has a stronger adsorption capacity than CH₄, which facilitates the displacement of CH₄ from subsea strata by CO₂.

(2) Compared to non-SCO₂ displacement, SCO₂ displacement yielded greater increments in the porosity and permeability of soil samples. From 30 °C to 70 °C, these increments also increased. The predominant pore size in the soil samples was 0.01–0.1 μm. Following SCO₂ displacement, the proportion of pores (>0.1 μm) increased, with the largest growth rate reaching 34.37%. However, all pore proportions ≤0.1 μm decreased. These test results demonstrated that SCO₂ displacement could effectively increase the proportion of large pores to enhance the permeability of soil samples.

(3) Due to significant changes in pore characteristics before and after the CO₂ displacement experiments, no duplicate experiments were conducted. Examining trends in pore changes before and after displacement across multiple unconsolidated samples revealed that CO₂ displacement could moderately enhance the permeability of silty clay formations. This phenomenon may offer potential benefits to improve the seepage velocity of decomposed gas and water in hydrate production.