Experimental Investigation of CO₂–Mineral Interactions in Tight Clastic Rock Reservoirs: Implications for Geological Carbon Sequestration

Abstract

Geological Carbon Sequestration (GCS) plays a crucial role in addressing climate change, particularly in oil and gas development. Understanding the reaction of supercritical CO₂ under in situ conditions and its effects on minerals is essential for advancing GCS technology. This study investigates the reaction mechanisms of feldspar (potassium and sodium feldspar) and clay minerals (chlorite, illite, montmorillonite, kaolinite) in CO₂ environments. The impacts on mineral crystal structures, morphologies, and elemental compositions were analyzed using X-ray diffraction (XRD), scanning electron microscopy (SEM), energy-dispersive spectroscopy (EDS), and ion concentration measurements (ICP-OES and ICP-MS). The results show that feldspar minerals exhibit lower reaction rates, with sodium feldspar dissolving faster than potassium feldspar, due to the higher solubility of sodium ions in acidic conditions. Chlorite showed significant crystal structure damage after 30 days, while montmorillonite underwent both dissolution and precipitation, influenced by interlayer cation dissociation. Kaolinite exhibited minimal reaction, primarily showing localized dissolution. Additionally, the formation of siderite (FeCO₃) was observed as Fe²⁺ substituted for Ca²⁺ in CaCO₃, highlighting the role of iron-bearing carbonates in CO₂ interactions. The study provides insights into the factors influencing mineral reactivity, including mineral structure, ion exchange capacity, and solubility, and suggests that chlorite, montmorillonite, and illite are more reactive under reservoir conditions, while kaolinite shows higher resistance to CO₂-induced reactions. These findings offer valuable data for optimizing GCS technologies and predicting long-term sequestration outcomes.

1. Introduction

Geological Carbon Sequestration (GCS) is a key carbon reduction technology with wide applications in oil and gas industry. The application of supercritical CO₂ has achieved significant progress across both conventional and unconventional oil and gas fields. In conventional reservoirs, its application is primarily based on miscible displacement theory for enhanced oil and gas recovery (CO₂-EOR/EGR). In unconventional reservoirs, it is utilized based on competitive adsorption theory for enhanced coalbed methane and shale gas recovery (CO₂-ECBM/ESGR), where CO₂ displaces methane adsorbed in the rock matrix. Furthermore, breakthroughs have been made in its use for fracturing and drilling technologies.

This study specifically focuses on tight sandstone and sandstone reservoirs, which are characterized by relatively low permeability. In these reservoirs, CO₂ injection interacts with the mineral phases present in the rock matrix, modifying the geochemical environment [1]. When coupled with Carbon Capture and Storage (CCS), CO₂-EOR is considered a promising method for reducing the carbon footprint of hydrocarbon production while enhancing recovery. This approach, often termed Carbon Capture, Utilization, and Storage (CCUS)-EOR, can store net CO₂ underground [2]. CO₂ not only carries and releases remaining oil from reservoirs but can also ‘stay’ in the reservoir after oil and gas extraction. Compared to water flooding, CO₂ has advantages such as reducing viscosity, lowering interfacial tension, and acidizing to clear blockages, making it especially suitable for low-permeability reservoirs. Compared to chemical flooding, CO₂ is more cost-effective, more easily obtained, less toxic, and less damaging to reservoirs. Furthermore, compared to other gas flooding methods (e.g., nitrogen or hydrocarbon gases), CO₂ has higher viscosity and lower volume coefficient, making it easier to inject and more efficient, while helping to alleviate ecological environmental pressure [3,4,5]. The effectiveness of CO₂-EOR varies with reservoir conditions—for example, miscible flooding at pressures above the minimum miscibility pressure (MMP) and in lighter oils typically yields higher gains, whereas sub-MMP pressures, heavier oils, and low-permeability/heterogeneous formations mainly see smaller improvements due to oil swelling, viscosity reduction, and gas channeling [4]. Previous studies have extensively investigated the mechanisms of dissolution and precipitation during CO₂ reactions and their dominance under different reservoir conditions, but experimental results for similar reservoirs still show differences, and no consistent understanding of the controlling factors for dissolution and precipitation has yet emerged [6,7,8].

Current research focuses on general evaluations of oil recovery and reaction effects, with little in-depth investigation of microscopic characteristics such as mineral composition and clay mineral microstructures. After CO₂ injection, CO₂–brine–rock interactions generate a suite of geochemical reactions that modify porosity and permeability through dissolution, precipitation, and secondary mineral formation [6]. The observed variability reflects mineral-dependent heterogeneity in reaction pathways and rates. For instance, dissolution of feldspars liberates K⁺/Na⁺ that couple with CO₂-derived dissolved inorganic carbon species to precipitate carbonates, thereby altering the reservoir geochemical milieu [9,10,11]. On the other hand, clay minerals such as illite, chlorite, and montmorillonite exhibit complex reaction behaviors during CO₂ injection, influencing the pH and major ion concentrations of fluids, thereby further affecting the stability of minerals and pore structures [12,13,14]. Due to the complex composition of reservoir cores, directly reacting cores with CO₂ makes it difficult to summarize reaction mechanisms, necessitating a detailed study of individual factors to reveal specific reaction pathways and interactions of different minerals during CO₂ injection.

2. Materials and Methods

This study was conducted at the National Key Laboratory of Oil and Gas Reservoir Geology and Development Engineering at Southwest Petroleum University. It aims to systematically investigate the reaction mechanisms of feldspar and clay minerals (illite, chlorite, montmorillonite) with CO₂, and their impact on the reservoir pore structure. The mineral samples used in the experiments include potassium feldspar, sodium feldspar, kaolinite, illite, chlorite, and montmorillonite. While quartz is a common component of most reservoirs, particularly those with feldspars and clay minerals, it was not included in the experimental samples due to its generally low reactivity under CO₂ injection conditions. The primary focus of this study is on the reactions of feldspars and clay minerals, as these minerals are more reactive and play a significant role in altering the geochemical environment of the reservoir. These mineral samples were obtained from laboratory specimens and from the Lixing Kaning and Tarim regions of Xinjiang, with a purity greater than 80%. Prior to the experiments, the mineral samples were mechanically crushed and sieved to a 200-mesh (75 µm) size. This step was performed to achieve a uniform particle size and ensure consistent exposure of the minerals to CO₂ during the reactions, allowing for more controlled and reproducible experimental conditions. Bulk powders for non-clay analyses were oven-dried at 105 °C for 12 h; however, clay XRD mounts were not oven-dried and were prepared as oriented mounts, analyzed in air-dried and ethylene-glycol-solvated conditions (≤40 °C for drying prior to glycolation).

Before the reaction experiments, the mineral samples were analyzed using a -Smartlab 9KW X-ray diffractometer (XRD, RIGAKU, Tokyo, Japan), under the conditions of Cu-Kα radiation (with an energy of 8.04 keV), a working voltage of 35 kV, and a current of 30 mA (Figure 1a). The scanning range was 2θ = 5–30°, with a scanning time of 5 min, a scanning speed of 2°/min, and a step size of 0.02°. The quartz d(101) peak was used as a reference for calibration. At the same time, a Magix PW-2424 X-ray fluorescence spectrometer (XRF, PANalytical, Almelo, The Netherlands) was used to determine the major elemental composition of the samples, using an Rh-target X-ray tube and a power setting of 4 kW (Figure 1b). The measurement time was 30 s, and the detection precision was up to 0.1%. The micro-morphology of the mineral surfaces was observed using an Quanta 650 FEG field emission scanning electron microscope (FE-SEM, FEI, Brno, the Czech Republic) at an accelerating voltage of 10–20 kV and a working distance of 10 mm, using secondary electron mode (SE) to record cleavage characteristics, fractures, and surface roughness of the samples (Figure 1c).

Static reaction experiments were conducted in a CWYF-1 high-temperature and high-pressure reactor- (Haian County Petroleum Scientific Research Instruments, Haian, China) (Figure 1e). The experimental conditions were set to simulate in situ reservoir conditions at a depth of approximately 2600 m, where the temperature is around 65 °C. This temperature was selected to match typical conditions at this depth in deep reservoirs. The pressure was set at 10 MPa, which is lower than the actual reservoir pressure of 31 MPa, to create more controlled conditions for studying reaction mechanisms without the complications of high pressure. The solid–liquid ratio was set at 1:3 (by mass) to ensure sufficient reactants while maintaining practical and reproducible conditions. The reaction times were 7 days and 30 days. Prior to the experiments, mineral powders were homogenized with deionized water and loaded into the reactor. To minimize sedimentation without disturbing system equilibrium, the reactor’s top-mounted sealed stirring rod (mechanical feedthrough) was used to manually agitate the suspension at 7-day intervals, without opening or venting the reactor. After the reaction, the samples were removed from the reactor, and the upper turbid liquid was allowed to stand for 24 h to separate the clear liquid from the rock powder. The upper clear liquid was analyzed for major elements (such as Si, Al, K, Na, etc.) using a PE Optima 5300 DV inductively coupled plasma optical emission spectrometer (ICP-OES, Perkin Elmer, Waltham, USA). The RF power was set to 1300 W, the cooling airflow to 10 L/min, the sample flow rate to 1 mL/min, and the detection limit was 0.01 mg/L (Figure 1d). Trace elements were analyzed using an Agilent-7900 inductively coupled plasma mass spectrometer (ICP-MS, Agilent, Tokyo, Japan) with the RF power set to 1550 W, the nebulizer flow rate to 0.1 L/min, and a detection limit of 0.1 µg/L. The remaining rock powder was rinsed and soaked in deionized water 2–3 times to remove soluble salts and then dried for further use.

After the reaction, the mineral samples were analyzed using X-ray diffraction (XRD) for mineral phase identification, with a scanning range of 5–60° and a scanning speed of 2°/min. The Rietveld method was used for quantitative phase analysis, using the quartz d(101) peak as the reference. Additionally, the surface morphology and elemental distribution changes of the rock powder after the reaction were observed using—Quanta 650 FEG field emission scanning electron microscope (FE-SEM, FEI, Brno, the Czech Republic) combined with energy dispersive spectroscopy (EDS). The magnification range was set between 500×–5000×. The X-ray fluorescence spectrometer (XRF) was used to measure the changes in the major elemental content of the solid-phase samples to monitor the migration and transformation of elements during mineral reactions.

3. Results

3.1. Mineral Identification

3.1.1. Feldspar Identification

Potassium feldspar (KAlSi₃O₈) belongs to the monoclinic crystal system, with the chemical formula KAlSi₃O₈. It typically appears pinkish-red, with two complete cleavage directions. Its density ranges from 2.54 to 2.57 g/cm³, and its specific gravity is between 2.56 and 2.59, with a hardness of 6. Energy-dispersive spectroscopy (EDS) analysis shows that the potassium feldspar used in this study contains small amounts of sodium, categorizing it as strip feldspar. The theoretical Si:Al ratio in potassium feldspar is 3:1, and the total content of potassium (K) and sodium (Na) is comparable to that of aluminum (Al). To compare the dissolution rates of alkali feldspar and plagioclase feldspar, we selected potassium feldspar as the K-feldspar host (matrix) and sodium feldspar as the exsolved albite lamellae for the CO₂ reaction experiments.

Potassium feldspar (KAlSi₃O₈) belongs to the monoclinic crystal system, with the chemical formula KAlSi₃O₈. It typically appears pinkish-red, with two complete cleavage directions. Its density ranges from 2.54 to 2.57 g/cm³, and its specific gravity is between 2.56 and 2.59, with a hardness of 6. Energy-dispersive spectroscopy (EDS) analysis shows that the potassium feldspar used in this study contains small amounts of sodium, consistent with a perthitic K-feldspar. The theoretical Si:Al ratio in potassium feldspar is 3:1, and the total content of potassium (K) and sodium (Na) is comparable to that of aluminum (Al). To compare reaction rates between alkali and plagioclase feldspars within a single specimen, we used a perthitic K-feldspar in which a K-feldspar host (matrix) contains exsolved albite lamellae; the CO₂ reaction experiments thus compared the host K-feldspar with the albite lamellae under identical conditions.

Sodium feldspar (NaAlSi₃O₈) belongs to the triclinic crystal system, with the chemical formula NaAlSi₃O₈. It typically appears white to grayish-white, with two complete cleavage directions and a hardness of 6–6.5. Its density ranges from 2.61 to 2.64 g/cm³ and is a common acidic plagioclase mineral. Sodium feldspar is rich in silicon and poor in aluminum, with a theoretical Si:Al ratio of 3:1. EDS analysis shows that the sodium feldspar used in this study is relatively pure (Figure 2,

Table 1. EDS analysis of feldspar.

Elements	Sodium Feldspar		Potassium Feldspar
	Weight (%)	Atomic (%)	Weight (%)	Atomic (%)
Na	5.01	7.06	11.80	13.94
Al	15.29	18.36	17.99	18.12
Si	51.00	58.84	70.22	67.94
K	16.59	13.75

).

3.1.2. Clay Mineral Identification

The clay minerals used in this study include chlorite, illite, kaolinite, and montmorillonite. Their specific identification is as follows (Figure 3,

Table 2. XRF analysis of different minerals.

Minerals	Al2O3/%	CaO/%	Fe2O3/%	K2O/%	MgO/%	Na2O/%	SiO2/%
Potassium feldspar	18.32	0.09	0.81	11.35	<0.01	3.47	65.38
Sodium feldspar	19.35	0.47	0.76	0.07	0.02	16.94	67.84
Kaolinite	13.46	0.20	0.89	1.39	0.30	0.10	49.70
Illite	17.04	1.35	0.64	2.40	0.73	4.70	70.36
Montmorillonite	34.26	1.36	1.13	3.36	1.24	2.29	70.87
Chlorite	12.03	5.52	3.40	4.95	4.57	1.4	58.06

):

Chlorite is a 2:1-type layered aluminum silicate mineral, with a structure similar to 2:1 montmorillonite but with an additional octahedral (alumina) sheet in the chlorite structural unit, which acts as an interlayer cation. Chlorite is typically monoclinic, with a specific gravity of 2.6–3.3 and a hardness of 2–3. Its molecular formula is (Mg, Fe)(AlSi₃O₁₀)(OH)₈ [15]. Based on the Fe²⁺/Fe³⁺ and Fe²⁺/Mg²⁺ ratios, chlorite can be classified as iron-rich, magnesium-rich, or intermediate types (Figure 3a). The chlorite used in this study is magnesium-rich.

Illite is a typical 2:1 (TOT)-type layered aluminosilicate mineral. Its fundamental structural unit consists of an aluminum-hydroxide octahedral sheet sandwiched between two silicon–oxygen tetrahedral sheets. The adjacent TOT layers are primarily held together by strong ionic bonds formed between the interlayer potassium ions (K⁺) and the tetrahedral sheets [16]. Additionally, hydrogen bonds, originating from the hydroxyl groups in the octahedral sheet, contribute to the stability of the layered structure. Illite is monoclinic, with a molecular formula of KAl₂(Si,Al)₄O₁₀₂·nH₂O, and it typically appears as a white, clayey material with a hardness of 1–2 and a specific gravity of 2.6–2.9 (Figure 3b).

Kaolinite is a 1:1-type (TO) dioctahedral layered silicate mineral. It is typically earthy or blocky, with a hardness of 2–2.5, a molecular formula of Al₄(Si₄O₁₀)(OH)₈, and belongs to the triclinic system, with a density ranging from 2.60 to 2.63 g/cm³ and a specific gravity of 2.16–2.68 (Figure 3c).

Montmorillonite belongs to the smectite group of minerals, typically occurring as blocky or earthy forms. Its molecular formula is (Na, Ca)₀.₃₃(Al, Mg)₂(Si₄O₁₀)(OH)₂·nH₂O. Its structure consists of a central aluminum–oxygen octahedral layer flanked by silicon–oxygen tetrahedral layers, forming a 3-layer structure (TOT) [17]. The layers contain water molecules and exchangeable cations, giving montmorillonite a high cation-exchange capacity and water absorption ability. It is monoclinic in structure (Figure 3d).

3.2. Feldspar Reactions with CO₂

In the first 7 and 30 days after CO₂ pressurization, both sodium feldspar and perthitic K-feldspar exhibited reaction, but the early-stage surface alteration of standalone albite (actual sodium feldspar) appeared modest and localized. According to SEM-EDS, the relative Na content in albite decreased from 21.73% to 20.93%, while Al and Si changed only slightly, indicating a mild early-stage response. Over the full reaction duration, however, solution chemistry (ICP-OES/ICP-MS) shows that sodium feldspar is overall more reactive than potassium feldspar [16]. In contrast, the perthitic K-feldspar shows more complex behavior. Electron-microscopy observations indicate preferential dissolution in Na-rich microdomains (albite exsolution lamellae), whereas K-rich regions (K-feldspar host) dissolve more slowly. Specifically, the relative Na content decreased from 7.35% to 6.90%, while K decreased from 14.36% to 13.81% [15]. The before-and-after differences—especially the contrast between Na and K—indicate preferential leaching of Na-rich domains within the perthite (Figure 4;

Table 3. EDS analysis results of different feldspars before and after reaction.

Reaction Time (d)	Mineral	Elements (%)
		K	Na	Al	Si
0	Perthite	14.36	7.35	20.44	57.85
	Albite	/	21.73	20.77	57.51
7	Perthite	13.75	7.06	20.36	58.84
	Albite	/	21.07	20.94	57.99
30	Perthite	13.81	6.90	20.75	58.94
	Albite	/	20.93	21.04	58.16

).

Regarding the changes in ion concentrations in the reaction solution, both sodium feldspar and potassium feldspar exhibited a significant increase in the concentrations of Na⁺, Al³⁺, and Si⁴⁺. Specifically, for sodium feldspar, over 0 to 30 days, the Na⁺ concentration increased from 9302 mg/L to 14,630 mg/L, the Al³⁺ concentration increased from 0 to 1.52 mg/L, and the Si⁴⁺ concentration increased from 0 to 11.13 mg/L. In comparison, the changes in the ion concentrations for potassium feldspar were also notable, With Na⁺ concentration increasing from 9302 to 11,720 mg/L, Al³⁺ from 0 to 0.20 mg/L, Si⁴⁺ from 0 to 11.9 mg/L, and K⁺ from 0 to 21.64 mg/L. While the ion concentrations in the reaction solution for both minerals increased significantly after 30 days of reaction, sodium feldspar showed a slightly higher reaction rate than potassium feldspar, indicating that the release of sodium ions was faster (Table 3).

Furthermore, the concentration changes of Ca²⁺ and Mg²⁺ ions in the reaction solutions of sodium feldspar and potassium feldspar were different. For sodium feldspar, the Ca²⁺ concentration decreased from 214.3 mg/L to 85.39 mg/L in the first 7 days, but then rebounded, reaching 118.69 mg/L by day 30. In contrast, the Ca²⁺ concentration in the potassium feldspar solution continuously decreased over the 30 days, suggesting that sodium feldspar may more readily undergo ion exchange with Ca²⁺. The concentration of Mg²⁺ in the reaction solutions for both minerals showed relatively small changes and remained fairly constant (Table 3).

In summary, sodium feldspar and potassium feldspar exhibit significant differences in their CO₂ reaction behavior. Sodium feldspar has a higher dissolution rate, and the release of sodium is more rapid, whereas potassium feldspar shows a slower dissolution rate, particularly in potassium-rich areas. Changes in the ion concentrations in the reaction solutions further confirm these differences, which may be related to the crystal structure, surface properties, and reaction kinetics of CO₂ with the minerals [18,19].

3.3. Clay Mineral Reactions with CO₂

3.3.1. Changes in Crystal Structure of Clay Minerals

In terms of crystal-structure evolution, all four clay minerals exhibited changes in interlayer reflections to varying extents (Figure 5).

Chlorite. After 7 days, the d(002) peak intensity decreased by 13.25%. After 30 days, the d(002) and d(004) peaks decreased by 31.93% and 33.06%, respectively, indicating substantial structural damage. Concurrently, new reflections at 4.33 Å and 3.032 Å emerged and are indexed to CaCO₃, consistent with mineral transformation during prolonged reaction (Figure 5a).

Illite. No obvious structural damage was detected after 7 days. By 30 days, the d(002), d(004), and d(006) peaks showed a gradual attenuation of 16.6–22.5%, indicating moderate damage to the illite structure after prolonged exposure, possibly due to the interaction of water within the structure with CO₂, resulting in the formation of carbonic acid (H₂CO₃) [16,20]. The presence of carbonic acid may have contributed to the dissolution of K-feldspar and chlorite, as evidenced by changes in their XRD peaks, without the formation of new crystalline phases (Figure 5b).

Kaolinite. The crystal structure of kaolinite remained relatively stable over the first 30 days. After 7 days, the d(001) and d(002) peaks decreased by 4.20% and 6.27%, respectively; by 30 days, the decreases were 2.79% and 6.70%. These changes are minor, indicating limited structural damage and no new crystalline phases [16,21]. The quartz peak remained essentially unchanged, suggesting no formation of crystalline SiO₂ (Figure 5c).

Montmorillonite. Montmorillonite underwent pronounced structural changes. After 7 days, the d(001) peak weakened by 30.61%, and the d(003) and d(004) peaks almost disappeared, indicating marked disintegration of the layered structure [17]. Additionally, a new reflection at d = 6.56 Å (2θ ≈ 13.4°, Cu Kα) appeared, consistent with a CaSO₄-type sulfate phase (gypsum/bassanite family) (Figure 5d).

Overall, chlorite and montmorillonite experienced the most significant structural changes, whereas illite and kaolinite showed weaker structural responses within the same period.

3.3.2. Mineral Morphology and Elemental Composition Changes

The SEM observations (Figure 6) corroborate the XRD trends and clarify the time dependence of morphological change.

Chlorite. At day 7, chlorite displays edge dissolution, yet the petal-/needle-like outline remains recognizable. By day 30, the surface becomes markedly fragmented, consistent with the strong XRD peak attenuation and carbonate formation reported in Section 3.3.1.

Kaolinite. Kaolinite retains its flaky aggregate at both day 7 and day 30, showing only slight edge corrosion and no significant disruption of aggregate integrity, corroborating its resistance to dissolution under the present conditions.

Illite. Illite exhibits minimal surface alteration at day 7 (fibrous habit largely preserved). By day 30, however, pronounced dissolution and fracturing render the fibrous structure nearly absent, indicating substantial consumption over time. This behavior is consistent with illite being comparatively more stable than smectite, yet not inert in CO₂–brine systems.

Montmorillonite. Montmorillonite shows morphology–composition coupling, with dissolution and precipitation occurring concurrently. Release of interlayer cations exerts first-order control on texture evolution from day 7 to day 30, consistent with the structural degradation seen in Section 3.3.1.

3.3.3. Element Release and Ion Concentration Changes

During the CO₂ reaction process, the ion concentration changes in chlorite showed a trend of increasing first and then decreasing, reflecting the dynamic balance between dissolution and precipitation (Figure 7). Compared with the pre-reaction solution, the concentrations of Ca²⁺, Fe³⁺, Mg²⁺, and Al³⁺ all increased at the early stage and then gradually decreased, while the Si⁴⁺ concentration increased initially and then stabilized. This indicates that, in the early stage, these ions were dissolved into the solution, and during the 7–30 day period, precipitation occurred for Al³⁺ and Fe³⁺. The decrease in ion concentrations after the initial increase is likely attributed to secondary reactions, such as precipitation of CaSO₄ or other mineral phases, which consume these ions and result in their gradual reduction in the solution (Figure 8a). For example, after 30 days of reaction, the Ca²⁺ concentration dropped to 33 mg/L, Mg²⁺ decreased to 87 mg/L, Fe³⁺ decreased to 0.2 mg/L, and Al³⁺ decreased to 0.3 mg/L. Meanwhile, the Si⁴⁺ concentration increased from 0 to 46 mg/L, indicating continuous release of Si⁴⁺ and a sustained dissolution process.

A similar trend was observed for illite, though the rates of ion release and the final equilibrium concentrations differed. During the reaction, the concentrations of Mg²⁺ and Al³⁺ increased initially and then decreased, whereas those of K⁺ and Si⁴⁺ rose in the early stage before stabilizing. For instance, after 7 days of reaction, the K⁺ concentration in the illite solution increased from 0 to 70 mg/L, Mg²⁺ from 39.13 mg/L to 170.91 mg/L, and Al³⁺ from 0 to 0.5 mg/L, while Si⁴⁺ increased from 0 to 37 mg/L; between 7 and 30 days, Mg²⁺ decreased to 114 mg/L and Al³⁺ dropped to 0 (Figure 8b).

Kaolinite reacted relatively slowly, with minor changes in the release of ions. After 30 days, the concentrations of Si⁴⁺, K⁺, and Al³⁺ in the kaolinite reaction solution increased at first and then leveled off; specifically, the Si⁴⁺ concentration increased from 0 to 15 mg/L and Al³⁺ from 0 to 0.35 mg/L (Figure 8c), with no obvious precipitation observed. This behavior is attributed to kaolinite’s higher chemical stability and its relatively closed structure.

Montmorillonite showed a distinct behavior compared to chlorite and kaolinite, particularly in terms of the rate of ion release and dissolution. In the montmorillonite system, the Ca²⁺ concentration increased sharply from 214.3 mg/L to 1935 mg/L within the first 7 days, followed by a decline to 459 mg/L between days 7 and 30. Similarly, the Mg²⁺ concentration rose from 39.1 mg/L to 473.0 mg/L in the first 7 days and then dropped to 228.0 mg/L by day 30. The Al³⁺ concentration increased from 0 to 2.8 mg/L in the first 7 days, then decreased to 0.8 mg/L by day 30, while the Si⁴⁺ concentration rose from 0 to 41 mg/L and then leveled off. This sharp initial increase followed by a decline in ion concentrations suggests a strong initial release of interlayer cations, which is more pronounced than in other clay minerals. The ion release behavior of montmorillonite is likely influenced by cation exchange processes, where the interlayer cations are exchanged with the surrounding aqueous phase, leading to the observed mobilization of ions. In contrast, while other clays also show a rise in ion concentrations initially, montmorillonite’s sharp early-stage release is more significant. The decreasing trend of K, Ca, and Mg concentrations in montmorillonite indicates that the interlayer cations were gradually released and mobilized during the reaction, highlighting the distinct ion release behavior of montmorillonite compared to other clays (Figure 8d).

In summary, the chemical stability and morphological changes of the minerals under CO₂ reaction conditions exhibit significant differences. Chlorite and montmorillonite show similar trends in elemental changes—dissolution followed by precipitation—while kaolinite remains relatively resistant to dissolution. In contrast, illite exhibits more pronounced changes, especially with a higher dissolution rate of Mg compared to K. These findings provide important insights into the dissolution and precipitation processes in CO₂ reaction systems and reveal the varied reactivity of different minerals in geochemical environments.

4. Discussion

4.1. Analysis of the Reaction Mechanism Between Feldspar and CO₂

Based on the SEM-EDS analysis and the observed ion concentration changes in the reaction solutions, both potassium and sodium feldspars undergo some dissolution, although the overall reaction is relatively slow. Sodium feldspar typically dissolves faster than potassium feldspar, as Na⁺ ions are more weakly bonded to the feldspar structure compared to K⁺ ions. This makes Na⁺ easier to release in acidic conditions, where ion exchange accelerates the dissolution rate, resulting in a faster reaction for sodium feldspar under reservoir conditions, i.e., sodium feldspar > potassium feldspar. The presence of carbonic acid (H₂CO₃), produced by the interaction of illite with CO₂ in the aqueous phase, may further promote the dissolution of K-feldspar and chlorite. The acidic conditions, coupled with the ion exchange processes facilitated by carbonic acid, lead to enhanced dissolution of these minerals [22,23,24].

During the feldspar dissolution process, the elements K, Na, and Al are preferentially released, while Si is released at a later stage. In the feldspar crystal lattice, the aluminum– oxygen tetrahedra (AlO₄) and silicon–oxygen tetrahedra (SiO₄) are connected by bridging oxygens, and the cations such as K⁺ and Na⁺ are coordinated by non-bridging oxygens. The dissolution reaction primarily occurs at the sites of both bridging and non-bridging oxygens. Notably, the (AlO₄)⁵⁻ tetrahedra in feldspar have a slightly larger volume than the (SiO₄)⁴⁻ tetrahedra, rendering the structure more unstable under acidic conditions. Consequently, aluminum is preferentially released compared to silicon [16]. This dissolution process is driven by the hydrogen ions (H⁺) that are generated when CO₂ dissolves in water, forming carbonic acid (H₂CO₃), which then dissociates into H⁺ and HCO₃⁻ ions. The reaction of CO₂ with water is as follows:

CO₂ + H₂O ⇌ H₂CO₃ ⇌ H⁺ + HCO₃⁻

(1)

These H⁺ ions actively participate in the chemical reactions on the mineral surfaces, promoting the release of K, Na, Al, and Si while accompanying the dissolution of silicate minerals [14,15]. The following reactions illustrate the dissolution under acidic conditions:

For potassium feldspar (KAlSi₃O₈) [25,26],

KAlSi₃O₈ + 4H⁺ → K⁺ + Al³⁺ + 3SiO₂ + 2H₂O

(2)

For sodium feldspar (NaAlSi₃O₈) [27],

NaAlSi₃O₈ + 4H⁺ → Na⁺ + Al³⁺ + 3SiO₂ + 2H₂O

(3)

These reactions indicate that the hydrogen ions (H⁺) in the acidic fluid actively participate in the chemical reactions on the mineral surfaces, promoting the release of K, Na, Al, and Si while accompanying the dissolution of silicate minerals [23,24].

Furthermore, the higher dissolution rate of aluminum from the feldspar crystals results in a significant increase in the Al³⁺ concentration in the solution, while the release of silicon is relatively slower. The changes in the ion concentrations in the reaction solution further verify this mechanism. In the CO₂ reaction system, the increase in Al³⁺ concentration is usually accompanied by the release of Na⁺ or K⁺, and the Si⁴⁺ concentration remains relatively lower, indicating that aluminum is preferentially released from the feldspar minerals [28].

The difference in solubility between potassium feldspar (K-feldspar) and sodium feldspar primarily stems from differences in chemical stability and ionic activities. Because potassium ions (K⁺) exhibit greater ionic stability compared to sodium ions (Na⁺), potassium feldspar generally has a lower solubility under normal conditions. Thus, when present individually, sodium feldspar tends to have a higher solubility than potassium feldspar [23]. However, when both feldspars coexist, the solubility behavior can change markedly. With increasing temperature, the solubility of sodium feldspar increases more rapidly than that of potassium feldspar [29]. This phenomenon is closely related to changes in ionic activities during the dissolution process in water and may vary under different temperature and pH conditions [22,29,30,31].

Further analysis indicates that the released sodium ions (Na⁺) can inhibit the dissolution of potassium feldspar. In environments with high Na⁺ activity, competition between Na⁺ and K⁺ may lead to reduced release of potassium ions, thereby suppressing the dissolution rate of potassium feldspar [23]. Therefore, when sodium feldspar and potassium feldspar coexist, the release of Na⁺ may significantly inhibit the dissolution of potassium feldspar through ion exchange.

In our experiments, the dissolution rates of sodium and potassium feldspar fragments were similar during the first 7 days, but between 7 and 30 days, sodium feldspar exhibited a significantly higher dissolution rate than potassium feldspar. No new minerals such as albite or kaolinite were detected post-reaction, and the overall dissolution rate remained low. Under the experimental condition of pH < 6, no formation of albite was observed, and the HCO₃⁻ concentration was high. Thus, both feldspars can react with acid as follows:

KAlSi₃O₈ + 4H₂O + 4H⁺ → K⁺ + Al³⁺ + 3H₄SiO₄

(4)

NaAlSi₃O₈ + 4H₂O + 4H⁺ → Na⁺ + Al³⁺ + 3H₄SiO₄

(5)

4.2. Analysis of the Reaction Mechanism Between Clay Minerals and CO₂

(1)

Reaction Mechanism of Chlorite with CO₂

As a typical iron-rich clay mineral, chlorite’s reaction in a CO₂ solution system is primarily controlled by changes in its crystal structure and ion exchange processes. Studies have shown that the appearance of Fe³⁺ does not necessarily imply the dissolution of chlorite; rather, it may be due to the extraction or leaching of iron (Fe³⁺) by H⁺ [32]. During the CO₂ reaction, the dissolution behaviors of Fe²⁺ and Fe³⁺ differ significantly. Due to its stronger solubility, Fe²⁺ is more easily replaced by H⁺ and enters the solution, potentially leading to the formation of siderite (FeCO₃). In this context, Fe²⁺, as a dissolved product, can react with CO₂ under low-pH conditions to form FeCO₃.

The dissolution reaction related to Fe²⁺ can be represented as:

Fe₂Si₄O₁₀(OH)₂ + 4H⁺ → 2Fe²⁺ + Si₄O₁₀²⁻ + 2H₂O

(6)

While the dissolution involving Fe³⁺ may lead to the formation of colloidal Fe(OH)₃, which can further settle as a solid under lower pressure:

Fe₂Si₄O₁₀(OH)₂ + 6H⁺ → 2Fe³⁺ + Si₄O₁₀²⁻ + 2H₂O

(7)

As the reaction progresses, Mg²⁺ and Al³⁺ in the chlorite structure are also gradually replaced by H⁺. According to the sequence of elemental stability (Mg²⁺ > Al³⁺ > Fe²⁺), the release sequence of Mg²⁺ and Al³⁺ follows this order. The dissolution of Mg²⁺ is particularly critical in this process and may be linked to the formation of carbonate minerals.

For example, Cai D. [33] found that after reaction, the reduction in Ca²⁺ and Mg²⁺ concentrations in dense sandstone was not only due to the precipitation of Mg²⁺ released from chlorite and illite (forming magnesite or sericite) but also related to the reaction of high concentrations of HCO₃⁻ in the solution. Kwak J.H. (2011) [12] suggested that the released Mg²⁺ might combine with HCO₃⁻ to produce magnesite (MgCO₃) and amorphous hydrated magnesium carbonate (MgCO₃·xH₂O); under CO₂ injection, magnesite predominates because the high concentration of HCO₃⁻ favors the formation of magnesite over other magnesium-based precipitates. This precipitation is particularly significant when CO₂ dissolves in the fluid and increases the concentration of HCO₃⁻, leading to magnesite formation. Once formed, magnesite is difficult to dissolve in acidic media, as its solubility is low in acidic conditions, making it more stable and less likely to re-dissolve. If this precipitation fills the pore spaces, it will affect the pore structure [8].

Rosenbauer R.J. (2005) proposed the following reaction in the presence of calcite [34]:

[Fe/Mg]₅Al₂Si₃O₁₀(OH)₈ + 5CaCO₃ + 5CO₂ → 5CaFe/Mg₂ + Al₂Si₂O₅(OH)₄ + SiO₂ + 2H₂O

(8)

Since no kaolinite was observed post-reaction and the calcite content in the sample was low—with Fe²⁺ and Mg²⁺ precipitates being detected—the reaction mechanism for chlorite in this study is proposed as:

2Mg₃Fe₂Al₂Si₃O₁₀(OH)₈ + 14H⁺ → 4Fe²⁺ + 3Mg²⁺ + 4Al(OH)₃↓ + 6H₄SiO₄

(9)

(2)

Reaction Mechanism of Illite with CO₂

The micro-morphological changes of illite are largely related to its occurrence and crystal structure. Illite often appears as hair-like coatings on particle surfaces or as flaky or fibrous fillings in pore spaces. Under high-pressure CO₂ solution conditions, the crystal structure of illite is disrupted, with hair-like crystals gradually fracturing into short columnar structures or crystal fragments that mix with the materials in the pore spaces [35]. This process is closely related to the solubility and acidity of CO₂, where CO₂ dissolved in the solution exchanges with interlayer cations such as K⁺ and Mg²⁺ in the illite lattice, leading to dissolution and disruption of the interlayer structure.

In experiments involving rock fragments rich in illite reacting with CO₂, the concentrations of Al and Si in the mineral increased slightly, whereas K⁺ and Mg²⁺ showed marked decreases, indicating the dissolution of these elements [14]. Morphological observations before and after the reaction reveal that the originally fibrous illite becomes dissolved and fragmented, with the residual illite showing significant structural damage. These changes further confirm that K⁺ and Mg²⁺ are leached out during dissolution and that the interlayer structure is severely disrupted.

Shi L. [14] reported that under conditions of 70 °C and 20 MPa, the d(001) peak of illite decreased by approximately 60%, indicating a more pronounced dissolution reaction under high temperature and pressure. This reaction depends not only on the solubility of CO₂ but also on temperature, pressure, and other conditions. Different reaction environments significantly affect the dissolution rate and product formation in illite, particularly under acidic conditions where the dissolution process is notably accelerated.

Under acidic conditions, the interlayer K⁺ in illite is rapidly replaced by H⁺, causing the disintegration and dissolution of the crystal lattice. In highly acidic environments the exchange reaction for K⁺ is fast due to its larger ionic radius and its relatively loosely held position within the hexagonal channels of layered minerals. The dissolution of K⁺ leads to changes in the interlayer spacing of illite, reducing its structural stability and further promoting dissolution [36,37].

In addition, new mineral phases may form during the reaction, especially under highly acidic or high-pressure CO₂ conditions. The increased CO₂ solubility under high pressure leads to a higher concentration of HCO₃⁻ ions, which in turn promotes the formation of carbonates such as siderite (FeCO₃) in the system. This behavior is similar to the mechanism observed in chlorite. High pressure enhances the carbonation process by increasing CO₂ dissolution and shifting the equilibrium toward carbonate precipitation. In illite, especially after long reaction times, some carbonate precipitation may occur, further altering the ion distribution and mineral composition in the system. Researchers have proposed that under reservoir conditions, illite can react with CO₂ according to the following equation:

Illite + H⁺ → 2.3Al³⁺ + 0.6K⁺ + 0.25Mg²⁺ + 3.5SiO₂ + H₂O

(10)

(3)

Reaction Mechanism of Kaolinite with CO₂

Kaolinite is a relatively stable aluminum silicate clay mineral. Its crystal structure lacks active interlayer cations, which gives it strong chemical stability under acidic and neutral conditions. In a CO₂ solution system, although kaolinite reacts only weakly with acids, under specific conditions—particularly at low pH—it may still undergo dissolution. Study by Guo H. (2016) [38] has shown that under pH conditions ranging from 4.6 to 5.7, the effect of acid on kaolinite was minimal, showing no significant dissolution. This indicates that the acidic environment does not directly compromise the structural stability of kaolinite. However, studies by Shi L. (2019) and Yuan B. (2018) have shown that when reacting with CO₂ solutions, the intensity of kaolinite’s characteristic peaks decreased by approximately 26%–36%, suggesting slight acid-induced erosion [14,39]. The reaction can be represented as:

Al₂Si₂O₅(OH)₄ + 6H⁺ ⇌ 2Al³⁺ + 2SiO₂ + 5H₂O

(11)

This reaction indicates that under strong acidic conditions, kaolinite releases Al³⁺ more readily than Si. Wu H. (2005) found that when pH < 4, a protective, Si-rich and Al-poor layer may form on the surface of kaolinite, slowing down the release of Al [40]. Moreover, numerical simulations by Liu N. (2018) [41] indicate that with increasing temperature, the dissolution rate of kaolinite increases, making the dissolution process at high temperatures more significant compared to other clay minerals. However, due to the presence of hydrogen bonding in its 1:1 layered structure, CO₂ cannot easily penetrate the interlayer, and the silico-aluminate framework remains largely intact. In our experiments, the reaction between kaolinite and CO₂ was considered to be weak.

(4)

Reaction Mechanism of Montmorillonite with CO₂

Montmorillonite, due to its high specific surface area and weak interlayer forces, exhibits a strong cation-exchange capacity and is highly reactive in CO₂ solution systems. The interlayer forces in montmorillonite are primarily van der Waals forces, which allow H⁺ to easily penetrate the (001) plane and initiate reactions. Some studies have indicated that under high-pressure CO₂ conditions, the adsorption of wet CO₂ can lead to deformation of the montmorillonite structure. Jeon P.R. (2018) conducted kinetic studies on the dissolution reaction and structural changes in a CO₂–brine–montmorillonite system at 45–65 °C and 15 MPa, finding that the (001) plane of montmorillonite underwent deformation following dissolution, as indicated by the reaction [42]:

5.92H⁺ + (Na₀.₁₄Ca₀.₀₆Mg₀.₁₅)(Al₁.₃₆Mg₀.₆₄)Si₄.₀₂O₁₀(OH)₂ + 4.08H₂O → 0.14Na⁺ + 0.06Ca²⁺ + 0.79Mg²⁺ + 1.36Al³⁺ + 4.02H₄SiO₄

(12)

Moreover, the reaction mechanism of montmorillonite is complex. Its strong affinity for water leads to hydration and dispersion, and upon hydration, it exhibits a basic character that promotes CO₂ dissolution. Various metal cations (such as Na⁺, K⁺, Ca²⁺, Mg²⁺) can be released from the interlayer via ion exchange, with H⁺ ions typically replacing them in acidic conditions. CO₂ affects the reactivity of silicon and aluminum in the montmorillonite structure differently; the aluminum octahedra are more easily destroyed under acidic conditions, while the silicon tetrahedra are relatively stable. Based on previous studies, the reaction mechanism of montmorillonite in our experiment can be divided into two steps [39,42]:

First step: The interlayer aluminum ions (Al³⁺) in montmorillonite are replaced by H⁺, leading to the disintegration of the aluminum octahedra and the subsequent release of silicon from the silicon tetrahedra:

KNaCa(Al,Mg)₂(Si,Al)₄O₁₀(OH)₂ + 14H⁺ → Na⁺ + K⁺ + Ca²⁺ + 2Mg²⁺ + 2Al³⁺ + 2SiO₂ + 8H₂O

(13)

Second step: Under the acidic conditions generated by CO₂ dissolution (via the formation of H₂CO₃ and its decomposition into H⁺ and HCO₃⁻), the released metal cations (Ca²⁺, Mg²⁺, Al³⁺) react with HCO₃⁻ to form precipitates:

Ca²⁺ + Mg²⁺ + Al³⁺ + HCO₃⁻ → (Ca, Mg)CO₃ + Al(OH)₃↓

(14)

The reaction of montmorillonite with CO₂ is of significant importance for CO₂-modified reservoir systems. On one hand, CO₂ reduces the montmorillonite content on rock surfaces and in pore spaces through chemical reactions, thereby increasing the pore volume and flow capacity. On the other hand, under weakly acidic conditions, the increase in H⁺ causes the montmorillonite surface to acquire a positive charge, reducing its cation exchange capacity (CEC). Additionally, the dehydration of H₄SiO₄ produces SiO₂ with a very low CEC, effectively inhibiting the swelling of montmorillonite, which in turn minimizes pore throat constriction and enhances reservoir permeability and stability [42,43,44].

(5)

Comparison of Reaction Intensities among Common Clay Minerals

Based on the changes in the characteristic peak intensities of the diffraction curves after 30 days of CO₂ injection, the decrease in peak intensities followed the order: chlorite > montmorillonite > illite > kaolinite. Combining the observations from electron microscopy, changes in the reaction solution ion compositions, comprehensive elemental analyses, and the inherent properties of the clay minerals (

Table 4. Comparison of clay minerals properties [5].

Properties	Clay Minerals
	Kaolinite	Illite	Montmorillonite	Chlorite
Cation exchange capacity mg/100 g (%)	3.00–15.00	10.00–40.00	80.00–150.00	10.00–40.00
Specific surface (m2/cm3)	8.80	39.60	34.90	14.00
Layer spacing (nm)	0.72	1.00	0.96–4.00	1.40
Main force	Hydrogen bond	Ionic bond	Van Der Waals force	hydrogen bond, Ionic bond, Van Der Waals force
Charge distribution	Edge	Si-O tetrahedron	Al-O octahedral	Magnesite octahedron

), it is evident that under reservoir conditions the clay minerals that undergo significant changes upon reaction with CO₂ are primarily montmorillonite, chlorite, and illite, whereas kaolinite exhibits only weak reactions. In reservoir conditions, the reactivity order of the clay minerals is: montmorillonite > chlorite > illite > kaolinite (Figure 9).

Chlorite, which is rich in Mg and Fe, tends to release these metal ions during the reaction, which can then react with CO₂ to form carbonate minerals (such as siderite, FeCO₃, or dolomite, CaMg(CO₃)₂) or form new precipitates (e.g., Al(OH)₃). These mineral transformations can significantly impact the reservoir pore structure and permeability, particularly the precipitation of siderite. The formation of siderite can block pore spaces, reduce permeability, and impair fluid flow, which may lead to failures in reservoirs, especially over prolonged periods of reaction with CO₂ [25]. Montmorillonite exhibits relatively strong reactivity due mainly to its readily exchangeable interlayer cations (Na⁺, Ca²⁺), leading to structural deformation and dissolution, and the formation of precipitates such as SiO₂ and Al(OH)₃, which can affect hydraulic conductivity [39,42]. In contrast, illite reacts more weakly; although its interlayer cation K⁺ is relatively stable, its crystal surfaces still suffer partial disintegration under supercritical CO₂ conditions. Kaolinite is the least reactive among them, largely due to its stable interlayer cations and compact crystal structure.

In actual CO₂–rock reactions, it is important to consider the effects of the active cations (such as Na⁺, Ca²⁺, K⁺) released from feldspar dissolution on the overall reaction process. Due to the salting-out effect, higher ion activities can inhibit the release of other ions [45]. Additionally, during the dissolution process, the deposition of Al³⁺, Fe³⁺, and SiO₂ on rock surfaces can reduce the effective reactive surface area, thereby inhibiting further reactions—especially when SiO₂ precipitates in its amorphous form, which can have a more pronounced inhibitory effect [12]. Moreover, colloidal Al(OH)₃ and Fe(OH)₃, though not easily precipitated under high-pressure production conditions, can adsorb onto rock surfaces and hinder further acid–rock reactions. When the outlet pressure drops, these colloids may precipitate, further impacting the reaction process.

5. Conclusions

This study systematically investigated the reaction behaviors and mechanisms of feldspar and clay minerals in a CO₂ environment and analyzed their potential impacts on reservoir pore structures. The main conclusions are as follows:

Feldspar Reaction Mechanism: Both sodium and potassium feldspars undergo dissolution reactions in a CO₂ environment, with sodium feldspar dissolving more rapidly due to the easier detachment of Na⁺ from its structure. The acidic conditions in the CO₂ environment accelerate this process. The dissolution of sodium feldspar releases Na⁺, Al, and Si, whereas the reaction in potassium feldspar is milder, indicating stronger structural stability.

Clay Mineral Reactivity: There are significant differences in the reactivity of different clay minerals. Montmorillonite and chlorite exhibit high reactivity, mainly through ion exchange and crystal structure disruption, leading to dissolution. In chlorite, the dissolution of Fe²⁺ and Fe³⁺ may lead to the formation of carbonate minerals (e.g., siderite), while montmorillonite shows pronounced interlayer cation dissolution and precipitation processes. In contrast, illite reacts more weakly, and kaolinite is the least reactive, displaying strong structural stability in a CO₂ environment.

Integrated Reaction Mechanisms: The reactivity of CO₂ with various minerals is closely related to the crystal structure, cation exchange capacity, and inherent solubility of the minerals. Feldspar dissolution is mainly influenced by the release of alkali cations (Na⁺, K⁺), while clay mineral reactions are governed by the interlayer cations and the stability of the crystal structure.

Implications for Geological Carbon Sequestration: This study provides essential data and theoretical support for understanding mineral reactions under CO₂ injection and storage conditions. However, the reactions here focus on individual minerals, while in a real reservoir, mineral interactions in the rock matrix could lead to different behavior. The complex multivariate system in a reservoir means that results from single mineral reactions may not directly apply to the entire reservoir. This limitation should be noted, as mineral interactions are crucial for understanding the overall stability and safety of CO₂ storage. Optimizing reactions between CO₂ and various minerals is important, but considering mineral interactions in the reservoir is key for long-term stability and safety.