Reactivity of Shale to Supercritical CO₂: Insights from SEM-EDS and Mineral Phase Evolution in Caney Shales for CCUS Applications

1. Introduction

Geologic carbon sequestration (CCS) is a leading strategy for reducing atmospheric CO₂ levels through long-term storage in deep subsurface formations [1]. Shale formations, commonly utilized as caprocks, are now also being explored as storage media due to their fine-grained texture, high clay content, and nanoporous structure [2]. These characteristics enhance surface reactivity and support mechanisms such as sorptive trapping and geochemical immobilization.

Exposure of shale to supercritical CO₂ (scCO₂) can initiate a range of mineral-fluid interactions. Carbonate phases may undergo dissolution; phyllosilicates, particularly illite, can experience elemental leaching and structural disruption; and sulfides like pyrite may oxidize. These transformations can alter porosity, permeability, and long-term seal integrity [3,4]. Clay minerals are especially reactive due to their high surface area and exchangeable interlayer cations, which contribute to buffering and ion transport under scCO₂ conditions [5,6]. Additionally, scCO₂ has a higher diffusivity than water, allowing it to access internal shale surfaces more efficiently and drive reactions within nanopores and microfractures [7].

Previous studies have frequently incorporated brine or gas impurities, making it difficult to isolate the effects of CO₂ alone. Mineral transformations occur in shales under scCO₂–SO₂ conditions [8], but the independent behavior of shale minerals in the presence of pure CO₂ remains poorly understood. Moreover, comparative data on reservoir and caprock lithofacies from the same shale formation under identical conditions are limited.

This study addresses that gap by examining mineralogical and geochemical changes in reservoir- and caprock-proxy samples from the Caney Shale after static exposure to pure scCO₂ at 60 °C and 2,000 psia for 30 days. Pre- and post-exposure characterization was performed using scanning electron microscopy with energy-dispersive X-ray spectroscopy (SEM-EDS), micro-computed tomography (microCT), and RAMAN spectroscopy to assess mineral phase changes, elemental redistribution, and structural modifications. The objective is to determine: (1) which mineral phases exhibit early-stage reactivity under pure scCO₂ conditions; (2) how reactivity differs between reservoir and caprock proxies; and (3) what microstructural features indicate the onset of mineral trapping or porosity evolution. By integrating SEM-EDS, microCT, and RAMAN spectroscopy with systematic phase classification across distinct lithofacies, this study establishes a baseline for understanding the intrinsic reactivity of shale formations under pure CO₂ conditions. The next section outlines the experimental setup, sample selection, and analytical procedures used to evaluate these transformations.

Figure 1. Summarized context of the research and the key questions our study answers.

Figure 1. Summarized context of the research and the key questions our study answers.

Insights for literature review accounts for various research aspects, which our paper will seek to answer. Table 1. A summarized comprehensive literature.

2. Materials and Methods

Shale samples were sourced from the Caney Formation in Southern Oklahoma, representing fine-grained, organic-rich lithofacies. Two distinct zones were selected to represent reservoir and caprock analogs: Reservoir-proxy samples (R1, R2, R3) originated from the brittle zone and Caprock-proxy samples (D1, D2) came from the ductile zone.

Samples were cut and polished into 1 × 1 × 0.5 cm cubical for exposure to scCO₂. For each sample, two orientations were obtained: one parallel and one perpendicular to the bedding plane. This approach captured anisotropic responses to scCO₂ exposure. Each analyzed surface was further segmented into four quadrants and a central point (Point 5) to account for heterogeneity during post-exposure analysis.

Figure 2. Comprehensive methodology. Accumulator B is filled with CO₂ and set to the supercritical conditions of 35^oC and 1200 psi, and increased progressive in steps of 100psi and 5^OC to reaction conditions (60 ^oC and 2500 psi). This is done with Acc A closed. Accumulator A contains the shelve on which the samples are stacked on trays, avoiding the any contaminations. Pressure is maintained using an Isco dual piston pump.

Figure 2. Comprehensive methodology. Accumulator B is filled with CO₂ and set to the supercritical conditions of 35^oC and 1200 psi, and increased progressive in steps of 100psi and 5^OC to reaction conditions (60 ^oC and 2500 psi). This is done with Acc A closed. Accumulator A contains the shelve on which the samples are stacked on trays, avoiding the any contaminations. Pressure is maintained using an Isco dual piston pump.

Pre and Post-exposure characterization involved a combination of imaging and spectroscopic methods to evaluate structural and chemical alterations (Figure 3). Due to the scale of interpretation, the CT scan wasn’t adequate for this experiment and next experiments we plan to employ nano-CT. Samples were analyzed using scanning electron microscopy with energy-dispersive X-ray spectroscopy (SEM-EDS). Quadrant-based mapping and a central point (Point 5) were examined at magnifications ranging from 500× to 5,000×. Phases were classified as silicates (quartz, illite, feldspar), carbonates (calcite, dolomite), sulfides (pyrite), and organic matter. Elemental maps were used to track Fe, Mg, Ca, Si, K, and Al redistribution. Micro-computed tomography was used to evaluate internal porosity and structural evolution at high resolution. Pre- and post-exposure scans were performed to compare structural changes. RAMAN analysis was conducted using 532 nm and 785 nm lasers at ~1 µm resolution over a 1x1 cm area. This method identified shifts in carbonate peak positions, disordering in clay mineral structures, and potential alterations in organic matter. Spectra were collected at 20x magnification to support correlation with SEM-EDS data.

3. Results

Post-exposure SEM-EDS analyses revealed distinct mineralogical changes across reservoir and caprock facies, driven by phase-specific reactivity to scCO₂. This section details the identified mineral phases, their compositional shifts, and morphological alterations after 30-day exposure.

3.1. Mineral Phase Identification

Qualitative and morphological analyses from SEM-EDS revealed clear mineralogical heterogeneity across the Caney Shale facies. Samples were mapped at five locations per face four quadrants and a central point to ensure spatial representativity. SEM-backscattered imaging and EDS revealed distinct phase compositions and early-stage alterations across the Caney Shale samples after 30 days of pure scCO₂ exposure. Reservoir and caprock proxies showed both compositional heterogeneity and phase-specific responses following scCO₂ exposure. All five samples exhibited increases in organic matter visibility, while silicate and sulfate phases developed locally under reactive conditions. Primary mineral phases identified include quartz, illite, calcite, dolomite, K-feldspar, pyrite, and organic matter as shown in Figure 4, with a cross mineralogical comparison in Figure 5.

• Quartz remained the principal framework silicate across all facies. In reservoir proxies, pre-exposure abundances were 43.77% (R1), 42.23% (R2), and 45.96% (R3), with post-exposure values of 43.94%, 42.22%, and 45.68%, respectively. In caprock proxies, quartz accounted for 33.79% (D1) and 33.69% (D2) prior to exposure, increasing marginally to 34.13% and 32.75% post-exposure. Across all facies, quartz grains retained angular, sharp morphologies with no significantly observable structural or chemical alteration
• K-Feldspar (KAlSi₃O₈). K-feldspar was consistently present in all samples. In reservoirs, values ranged from 5.06% to 5.84% pre-exposure and from 5.89% to 5.99% post-exposure. In D1 and D2, K-feldspar was measured at 7.06% and 6.54% pre-exposure, increasing slightly to 7.13% and 6.36% post-exposure. No dissolution or surface roughening was evident under SEM imaging.
• Illite [(K,H₃O)(Al,Mg,Fe)₂(Si,Al)₄O₁₀(OH)₂]. Illite occurred in all facies and was typically distributed along grain boundaries or within clay-rich matrices. Pre-exposure content ranged from 17.61% (R3) to 18.40% (R2), and post-exposure values decreased slightly to 17.69%–18.67%. In ductile samples, illite was recorded at 17.26% (D1) and 16.68% (D2) before exposure, declining to 17.47% and 16.31%, respectively. Platy textures remained intact, although localized thinning and roughening of particle edges were noted in R2 and R3.
• Kaolinite (Al₂Si₂O₅(OH)₄). Kaolinite was identified in D1 and D2. Pre-exposure values were 10.03% (D1) and 9.89% (D2), increasing slightly to 10.13% and 9.76%, respectively, post-exposure. Kaolinite maintained blocky morphology with no signs of chemical erosion or micro-pitting.
• Paragonite (NaAl₂(Si₃Al)O₁₀(OH)₂). Paragonite was not detected in any sample prior to exposure. Post-exposure, it appeared in R2 (0.38%), R3 (0.88%), D1 (0.72%), and D2 (0.90%). It was typically observed near altered illite flakes and within fine-grained matrix zones, forming as secondary Na-bearing phyllosilicate lamella.
• Calcite (CaCO₃). Calcite was present in all samples, particularly in the reservoir facies. In R1–R3, calcite decreased slightly from 9.92%, 9.97%, and 8.80% pre-exposure to 9.01%, 8.76%, and 9.05%, respectively. In caprocks, it declined from 9.41% (D1) and 9.62% (D2) to 9.75% and 8.80%, respectively. SEM images revealed surface pitting and edge retreat, especially in R1 and R2.
• Ankerite [Ca(Fe²⁺,Mg)(CO₃)₂]. Ankerite occurred in both reservoir and caprock proxies. In R1–R3, pre-exposure values ranged from 4.40% to 5.49%, decreasing post-exposure to 4.08% – 4.33%. In D1, it was not detected pre-exposure and remained absent post-exposure. In D2, it decreased from 4.02% to 3.55%. Morphologies were retained but with localized surface dulling near grain boundaries.
• Wollastonite (CaSiO₃). Wollastonite was absent prior to exposure and formed in all samples post-exposure. In R1–R3, abundances were 0.67%, 1.08%, and 1.02%, respectively. In D1 and D2, wollastonite was recorded at 0.56% and 0.93%. It appeared as fibrous or acicular precipitates localized around sites of prior carbonate dissolution.
• Albite (NaAlSi₃O₈). Albite was present in every sample. Pre-exposure values in R1–R3 ranged from 4.88% to 5.96%, decreasing slightly to 5.75%–5.94% post-exposure. In D1 and D2, albite changed from 4.71% and 4.82% to 4.93% and 4.65%, respectively. Grains retained sharp outlines and showed no dissolution features.
• Pyrite (FeS₂). Pyrite was present across all facies. In reservoirs, pre-exposure values ranged from 4.88% (R1) to 5.87% (R2), declining to 4.67%–5.41% post-exposure. In D1 and D2, pyrite decreased from 6.28% and 6.26% to 5.56% and 4.84%, respectively. SEM showed edge diffusion and oxidation halos near OM and clay interfaces in caprock samples.
• Jarosite [KFe₃(SO₄)₂(OH)₆]. No jarosite was detected prior to exposure. It was identified post-exposure in R3 (0.58%), D1 (1.24%), and D2 (1.49%), with trace detection in R1. No formation was observed in R2. It formed as fine-grained aggregates, frequently bordering pyrite and organic-rich regions.
• Calcium Sulfate (CaSO₄). Calcium sulfate was not detected pre-exposure and was present only post-exposure in caprock samples: 0.80% in D1 and 1.07% in D2. It appeared as thin, patchy coatings at mineral boundaries.

Organic Matter (CxHyOz). Organic matter was found in all samples, increasing post-exposure in every case. In R1–R3, OM increased from 5.17%, 6.57%, and 5.99% pre-exposure to 10.79%, 12.77%, and 11.89%, respectively. In D1 and D2, OM rose from 6.66% and 6.99% to 12.15% and 14.19%, respectively. Post-exposure OM showed increased surface roughness, irregularity, and porosity development.

3.2. Elemental Mobilization

Variations in K⁺, Ca²⁺, Mg²⁺, Na⁺, SO₂^4-, and Fe^2+/3+ are recorded across R1. K⁺ and Na⁺ signals decrease alongside a reduction in K-feldspar abundance (13.6% to 10.2%) and Albite (5.0% to 3.4%). Ca²⁺ and Mg²⁺ shifts correspond to marked declines in Calcite (18.0% to 11.2%) and Ankerite (5.0% to 2.1%). SO₄²⁻ release is supported by the disappearance of Anhydrite (2.5% pre-exposure). A drop in Pyrite content (3.2% to 1.4%) aligns with trace-level Fe²⁺ observations. Illite (27%) and Quartz (24.7%) show no significant quantitative change. Minor components such as Kaolinite (<1.5%) remain within detection limits without notable shift.

Ionic reductions are inferred for K⁺, Ca²⁺, Mg²⁺, SO₄²⁻, and Fe²⁺ in D1. These align with the measured loss in K-feldspar (14.3% to 9.8%) and Illite (26.7% to 22.9%), as well as Calcite (15.4% to 9.3%) and Dolomite (12.2% to 8.0%). Sulfur-bearing phases show change, with Anhydrite declining from 3.0% to <1.0%. Pyrite decreases to 2.1%, and minor Fe signals are present in adjacent altered zones. Na⁺ remains stable, with Albite retaining 3.2%. Quartz (24%) and minor phases such as Kaolinite and organic matter (1%) do not register significant variation.

In R2, changes in K⁺, Ca²⁺, Mg²⁺, SO₄²⁻, and Fe²⁺ are evident. K-feldspar content declines from 12.7% to 8.4%, and Illite from 21.9% to 19.2%. Calcite and Dolomite decrease to 10.1% and 6.3%, respectively, reflecting a reduction in Ca²⁺ and Mg²⁺ during carbonates reprecipitation. Anhydrite, originally present at 2.8%, is no longer detected post-exposure, consistent with increased SO₄²⁻ signal. Pyrite falls to 2.3%, with Fe^2+/3+ signal diffusion around previous grain boundaries. Albite (3%) and Na⁺ levels remain stable. Quartz holds at 24.6%. Kaolinite (1.1%) and organic matter (0.9%) are preserved without apparent shifts.

The reduction of Ca²⁺, Mg²⁺, K⁺, and SO₄²⁻ is confirmed across D2. This corresponds with decreased proportions of Calcite (17.5% to 11.6%), Dolomite (11.8% to 7.5%), and K-feldspar (13.2% to 8.9%). Illite shifts slightly (19.6% to 17.1%), while Anhydrite (2.6%) is fully absent post-exposure, with elevated sulfur signals noted in former anhydrite-bearing regions. Pyrite declines to 1.7%, with minor Fe presence. Albite ranges from 3.1% to 2.6%, with negligible Na⁺ signal deviation. Quartz (~23.9%) and minor minerals such as Kaolinite, organic matter, and trace phosphates show no measurable alteration.

SEM-EDS elemental mapping indicates selective post-exposure mobilization of K⁺, Ca²⁺, Mg²⁺, Fe²⁺, and SO₄²⁻ across facies. Carbonate-bearing zones in R1, R2, and R3 show marked Ca²⁺ and Mg²⁺ depletion, consistent with mineral phase volume loss from Calcite, Dolomite, and Ankerite. Pyrite-associated Fe^2+/3+ and S(SO₄^2-) signals decrease, with partial redistribution observed in caprock proxies such as D2. Illite grains exhibit localized leaching of K⁺, Mg²⁺, and Fe^2+/3+, particularly in the reservoir proxies. In contrast, Si and Al remain stable, reflecting Quartz and K-feldspar persistence. Organic matter becomes more spatially distributed post-exposure, especially along former carbonate-clay boundaries. No secondary precipitates were detected within the 30-day interval, though surface alterations indicate potential reactivity under extended conditions. Figure 6 and Figure 7 give a sample observation of a key feature observed post reaction for a caprock and reservoir proxy.

Table 2. From ionic mobilizations to sequestration, a summary of the potential ionic mobilizations.

Ionic Species	Primary Mineral Phase Sources	Facies Observed	Post-Exposure Observation	Possible Geochemical Path
K+	K-feldspar, Illite	R1, R2, D1, D2, R3	K-feldspar reduction (13.6% to 7-10%); slight Illite shift	Leaching from feldspars and clay edges
Na+	Albite	R1, R3, D2	Minor Albite decline (5.0% to 2.6-3.4%)	Limited Na+ exchange
Ca2+	Calcite, Dolomite, Ankerite	R1, R2, R3, D1, D2	Redistribution among carbonate phases; net Ca2+ preserved	Partial dissolution and re-precipitation
Mg2+	Dolomite, Illite, Ankerite	R2, R3, D2, D1	Mg-bearing carbonates reduced; Dolomite often retained	Phase transition and reallocation
Fe2+ / Fe³+	Pyrite, Illite, Ankerite	D1, D2, R2, R3	Pyrite decreased (up to 50%); Fe detected near former grains	Oxidation and surface destabilization
SO42⁻	Anhydrite, Pyrite	R1, D1, D2, R2, R3	Anhydrite loss, S redistributed	Sulfate release from dissolution/oxidation
Al³+	K-feldspar, Albite, Illite	All facies	No significant compositional change	Structurally retained in aluminosilicates
Si 4+	Quartz, K-feldspar, Illite, Albite	All facies	Quartz (~24–25%) stable throughout	Framework remains chemically inert
C (elemental) CO32⁻	Calcite, Dolomite, Ankerite	All facies	Carbon and carbonates retained via phase shifts, not net loss	Re-precipitation or phase conversion
S (Elemental)	Pyrite, Anhydrite	D2, R2, R3	Sulfur detected post-Anhydrite; diffused spatially	Sulfate migration from sulfates/sulfides
P / PO4³⁻	Apatite, trace organics	D2, R3 (trace levels)	Stable in isolated inclusions	Largely inert under dry CO2

Table 2. From ionic mobilizations to sequestration, a summary of the potential ionic mobilizations.

Ionic Species	Primary Mineral Phase Sources	Facies Observed	Post-Exposure Observation	Possible Geochemical Path
K+	K-feldspar, Illite	R1, R2, D1, D2, R3	K-feldspar reduction (13.6% to 7-10%); slight Illite shift	Leaching from feldspars and clay edges
Na+	Albite	R1, R3, D2	Minor Albite decline (5.0% to 2.6-3.4%)	Limited Na+ exchange
Ca2+	Calcite, Dolomite, Ankerite	R1, R2, R3, D1, D2	Redistribution among carbonate phases; net Ca2+ preserved	Partial dissolution and re-precipitation
Mg2+	Dolomite, Illite, Ankerite	R2, R3, D2, D1	Mg-bearing carbonates reduced; Dolomite often retained	Phase transition and reallocation
Fe2+ / Fe³+	Pyrite, Illite, Ankerite	D1, D2, R2, R3	Pyrite decreased (up to 50%); Fe detected near former grains	Oxidation and surface destabilization
SO42⁻	Anhydrite, Pyrite	R1, D1, D2, R2, R3	Anhydrite loss, S redistributed	Sulfate release from dissolution/oxidation
Al³+	K-feldspar, Albite, Illite	All facies	No significant compositional change	Structurally retained in aluminosilicates
Si 4+	Quartz, K-feldspar, Illite, Albite	All facies	Quartz (~24–25%) stable throughout	Framework remains chemically inert
C (elemental) CO32⁻	Calcite, Dolomite, Ankerite	All facies	Carbon and carbonates retained via phase shifts, not net loss	Re-precipitation or phase conversion
S (Elemental)	Pyrite, Anhydrite	D2, R2, R3	Sulfur detected post-Anhydrite; diffused spatially	Sulfate migration from sulfates/sulfides
P / PO4³⁻	Apatite, trace organics	D2, R3 (trace levels)	Stable in isolated inclusions	Largely inert under dry CO2

4. Discussion

4.1. Mineral Stability and Reactivity

The early-stage mineralogical and microstructural changes observed in the Caney Shale samples have important impact on the long-term safety and effectiveness of geologic carbon sequestration. The Caney Shale demonstrates that supercritical CO₂ (scCO₂) exposure initiates diverse mineralogical transformations, largely governed by facies-level differences in composition, porosity, and reactivity. While the broad categories of carbonate dissolution and clay interaction have been widely reported in other shales, this study reveals the importance of sulfate-related transformations and the catalytic role of organic matter. These findings support a nuanced understanding of geochemical reactivity across mechanically distinct lithologies [35] and provide some insights for carbon storage safety and long-term relevance.

4.1.1. Carbonate Phases

Reservoir facies such as R2A and R3A displayed extensive dissolution of calcite and ankerite. This pattern confirms earlier observations of carbonate reactivity under CO₂-rich conditions [36], where the release of Ca²⁺, Mg²⁺, and Fe²⁺ ions initiates buffering and supports the early stages of mineral trapping [37]. In contrast to previous findings where reprecipitation was considered minimal without brine [34], this study provides evidence of secondary precipitation even under nominally dry conditions. Carbonate reactivity is a major gateway to geochemical sequestration of CO₂. No discrete Ca-silicate phases such as wollastonite were identified. Ca–Si signals observed in Raman and EDS maps are likely attributable to mixed silicate-carbonate domains or poorly crystalline components associated with clay-carbonate interfaces [38]. Figure 8 is an illustration of these. The formation of calcium silicate (wollastonite) indicates that these carbonates did not simply dissolve but instead facilitated new stable mineral assemblages. Such transformations likely arise through coupled reactions involving silicate-structural and carbonate-derived cations [32]. Their spatial localization near feldspar-carbonate-OM boundaries suggests that geochemical trapping in Caney Shale is interface-driven rather than homogeneous. This behavior underscores the importance of interfacial mineralogy in shaping the spatial pattern of storage efficiency.

4.1.2. Clays and Feldspars

Clay minerals exhibited distinct responses across facies. In the ductile caprock settings of D1 and D2, illite displayed edge thinning alongside localized transformation into paragonite, accompanied by an overall increase in illite presence. This trend is consistent with illitization, potentially driven by minor Na⁺ availability from albite or residual interstitial fluids [30,39]. Alterations observed near clay boundaries under low-water conditions suggest that clay phases can undergo mineralogical adjustment, contrary to prior assumptions of their inert behavior in scCO₂ environments, particularly where thin water films persist [31,40]. The increased clay content, along with progressive illitization, is associated with enhanced ductility and greater sealing capacity in the caprock proxies [28,32]. Feldspars, while structurally preserved, show signs of active chemical exchange. Slight K⁺ depletion and surface etching, particularly in R1A and R3A, point to early-stage hydrolysis [33]. These reactions may facilitate the formation of secondary clays or contribute mobile cations relevant to CO₂ trapping processes. The involvement of feldspars in such geochemical interactions reflects long-term alteration patterns observed in sedimentary basins [34], although their direct contribution to mineral trapping remains limited. Clays and feldspar serve as alteration controllers and ion exchange sites. Figure 9 illustrates the clay mobilization in the caprock proxy D1.

4.1.2. Sulfide Oxidation and Sulfate Reaction Pathways

Formation of jarosite and additional anhydrite (CaSO₄) was observed in both reservoir and caprock facies following scCO₂ exposure. These phases are associated with pyrite oxidation and redistribution of sulfate from pre-existing anhydrite. Their presence in a closed system, containing only scCO₂ and no added SO₂, indicates that sulfate-bearing phases can form under localized oxidative conditions. Co-occurrence with organic matter and likely trace water suggests that redox activity can occur in the absence of externally supplied oxidants, supporting Fe²⁺ and SO₄²⁻ mobility in the various facies [35]. Sulfate-bearing phases, although less frequently emphasized in shale carbon storage contexts, are observed to persist under the experimental conditions. Jarosite was identified in multiple points, and both jarosite and anhydrite are known to exhibit stability under low pH and moderate temperature [36]. Jarosite spikelets as seen in Figure 6, show vast porosity created, giving more room for sequestration. Figure 10 shows the pre and postexposure EDS of the jarosite linked spot. Here, the reaction is hypothesized to initiate through pyrite oxidation, likely accelerated by the presence of trace porewater and redox-active OM [37]. This process releases Fe²⁺ and S²⁻, which, under oxidizing conditions, transition into Fe³⁺ and SO₄²⁻ [38], ultimately forming jarosite through a precipitation reaction:

$${\mathrm{K}}^{+}\left(\mathrm{a}\mathrm{q}\right)+3{\mathrm{F}\mathrm{e}}^{3+}\left(\mathrm{a}\mathrm{q}\right)+2{\mathrm{S}\mathrm{O}}_4^{2-}\left(\mathrm{a}\mathrm{q}\right)+6{\mathrm{H}}_2\mathrm{O}\to \mathrm{K}{\mathrm{F}\mathrm{e}}_3{\left({\mathrm{S}\mathrm{O}}_4\right)}_2{\left(\mathrm{O}\mathrm{H}\right)}_6\left(\mathrm{s}\right)+6{\mathrm{H}}^{+}\left(\mathrm{a}\mathrm{q}\right)$$

(1)

Equation 1 shows the jarosite formation reaction which is thermodynamically favored under acidic, low ionic strength conditions [39]. Its appearance in a brine-free, scCO₂ setting underscores the capability of shale-hosted systems to support complex redox mineral transformations without exogenous oxidants. Our results indicate that sulfate mineral formation can occur alongside carbonate reactivity, particularly in organic-rich zones where sulfide-CO₂ interactions are spatially linked.

4.1.3. Organic Matter: A Chemically Active Interface

Organic matter played a dual role across all facies, two proxies are used in Figure 11. It acted both as a high-affinity sorbent for CO₂ and as a catalytic interface for redox-mediated mineral reactions [41]. In ductile caprock facies, organic surfaces underwent fragmentation and porosity development, which enhanced their ability to store CO₂ in micropores [18]. These changes also supported electron transfer reactions, especially near pyrite and clay edges, fostering the nucleation of new mineral phases [42]. Organic matter can be defined in our context as a chemically active interface. The formation of illite-OM aggregates and secondary mineral patches at OM-clay boundaries confirm that organic matter supports not only CO₂ uptake but also chemical transformation. Previous studies have shown the sorptive value of OM [24], but the current results demonstrate its capacity to induce mineral precipitation and stabilize reaction fronts. This behavior extends the functional relevance of organic matter from passive retention to active geochemical regulation within shale-hosted sequestration systems.

4.2. Relevance for Geochemical Sequestration

The mineral transformations and reactivity patterns observed in this study suggest that Caney Shale facies contribute to CO₂ storage in distinct yet complementary ways. Reservoir units serve as the primary trapping domains through carbonate dissolution and reprecipitation. Ductile caprock facies, on the other hand, operate as chemical seals, limiting vertical migration and stabilizing CO₂ through redox and sulfate-based reactions.

4.2.1. Reservoir Proxies

Proxies R1 through R3 are characterized by high quartz content combined with moderate amounts of reactive carbonates, clays, and organic matter. Spatially localized dissolution and precipitation were observed, primarily concentrated along mineral boundaries and microstructural interfaces. R2 exhibited the most extensive carbonate loss, particularly of ankerite, followed by irregular precipitation of both carbonate and silicate phases. R3, with lower carbonate content, displayed greater expression of organic matter and sulfate reactivity, indicating multiple coexisting trapping mechanisms [29]. Such mineral–fluid interactions align with findings in other mixed-lithology reservoirs, where uneven carbonate distribution contributes significantly to chemical trapping [34]. The detection of Ca–Si-rich areas at silicate–carbonate-OM interfaces suggests the formation of calcium-bearing silicate assemblies [42]. While wollastonite (CaSiO₃) was not conclusively identified, Raman and EDS data reveal spatially correlated Ca²⁺ and SiO₂ signals consistent with possible low-crystallinity or transitional Ca–silicate compounds [43]. These mineral associations, though unresolved in structure, may represent additional trapping phases not typically emphasized in standard CCS models [7,43]. Such variability in mineral phases supports localized reactivity that could potentially enhances CO₂ retention in texturally heterogeneous zones. The reservoir proxies are thus accounted for reactive porosity and multiphase trapping.

4.2.2. Caprock Proxies

Caprock proxies D1 and D2 retained structural integrity and low porosity following scCO₂ exposure. Their composition rich in clays and organic matter, and the emergence of sealing-associated phases suggest a dominant geochemical sealing role rather than bulk-phase CO₂ mineralization [44]. Compact microtextures were preserved post-exposure, and the presence of sulfate and redox-active phases was confirmed by both SEM-EDS and Raman analyses [43]. These features are consistent with current caprock performance models that emphasize ion retention and interfacial buffering over mechanical deformation. Sulfate mineralization and redox processes involving organic matter occur in the absence of external oxidants or brine immersion. The sealing behavior is reinforced by microscale dissolution–precipitation reactions that operate across reactive boundaries and enhance caprock resilience without compromising structural integrity [45]. These processes are not just passive but actively reinforce sealing through coupled dissolution and precipitation at the microscale. Caprock proxies are thus accounted for ensuring sealing (geochemically proven), hence confirming the dual nature of shales as for CO₂ containment and leak prevention.

4.2.3. Integrated Storage Performance and Relevance for CCS Design

Facies-dependent responses in Caney Shale illustrate the function of an integrated trapping system in which porosity access, geochemical transformation, and redox buffering act concurrently. Reservoir zones enable CO₂ migration and support mineral conversion along carbonate and sulfate pathways. Caprock layers contribute by limiting leakage through chemical immobilization and formation of clay barrier phases at reactive interfaces. Formation of sulfate-bearing minerals, calcium–silicate domains, and clay–organic composites under scCO₂ exposure broadens the range of sequestration pathways beyond traditional carbonate-based considerations.

SEM-EDS analyses of Caney Shale samples exposed to scCO₂ reveal dual effects: local mineral transformations support both self-sealing and mechanical weakening processes. Carbonate dissolution was frequently accompanied by precipitation of secondary phases at grain boundaries and within pore spaces, especially in reservoir facies such as R2 and R3. These fine-scale precipitates suggest activation of self-sealing pathways, even without brine, aligning with prior findings in clay–carbonate systems [38].

Similarly, microstructural weakening was evident in caprock facies. Illite exhibited sheet-edge thinning and minor delamination; pyrite showed early oxidation; and organic matter displayed fragmentation with increased surface porosity. Such alterations, most prominent in D1 and D2, raise concerns about clay integrity under prolonged exposure, consistent with reported CO₂-induced softening of sealing units under low-water conditions [3]. The interplay between sealing and weakening is spatially variable and phase-dependent. In carbonate-rich zones, reprecipitation may enhance capillary sealing. In phyllosilicate-dominated domains, structural degradation may reduce long-term integrity, particularly under stress cycles associated with injection operations [2,35]. Observed phase transformations support a wider trapping framework that incorporates both primary reactions and secondary mineral development without reliance on brine or supplemental oxidants. Caney Shale demonstrates geochemical and structural reliability for long-term CO₂ retention, with distinct yet complementary contributions from reservoir and caprock settings. Table 3 provides an overview of literature with the common mineral phases associated with CCUS and how they fit the purpose of our research.

4.3. Geochemical Insights

Mineral reactivity in the Caney Shale under scCO₂ exposure is highly localized, driven by the spatial arrangement of reactive phases and interfacial microenvironments. Transformations occurred not uniformly, but along grain boundaries where carbonates, clays, sulfides, and organic matter converge. These interfaces enabled coupled processes such as ion exchange, redox cycling, and secondary phase nucleation. Consequently, this extends trapping pathways beyond carbonate dissolution. In Figure 12, illustrations of the observed mineral evolution reflect not just composition, but the connectivity and proximity of reactive constituents in both facieses. Organic matter enhanced reactivity by supporting electron transfer and acting as a nucleation site, while stable phases like quartz and feldspar constrained reactions spatially. Overall, trapping efficiency in the Caney Shale appears controlled less by bulk mineralogy than by the distribution and interaction of components at the microscale.

5. Conclusions

Exposure of Caney Shale to pure scCO₂ at 60 °C and 2500 psia induces rapid, phase-specific mineral reactions relevant to long-term carbon storage. Quartz remains chemically inert, while carbonates dissolve and locally reprecipitate as calcium- and iron-rich secondary phases. Illite exhibits edge alteration and cation leaching; pyrite undergoes early-stage oxidation; and organic matter becomes more porous, supporting redox activity at mineral interfaces. SEM-EDS analysis confirms that mineral trapping initiates under dry conditions, with nanometer-scale precipitates forming at grain boundaries and pore surfaces. Reactions are not uniformly distributed but are instead governed by spatial variability in mineral associations, surface properties, and interface geochemistry. Insights gained from this experiment includes:

• Localized porosity development enhances CO₂ injectivity, while secondary mineral precipitation at grain contacts and pore throats contributes to self-sealing behavior, supporting containment stability.
• Demonstrated mineral trapping in dry scCO₂ (no added brines) systems confirms that water is not a prerequisite for initiating geochemical containment, with in situ precipitation providing a viable mechanism for immobilizing (sequestering) injected CO₂.
• Facies-dependent reactivity, mineral phase and ionic species distribution support a naturally evolving balance between fluid migration pathways and geochemical seals. This allows reactive zones (reservoirs) to co-exist with stable, low-permeability zones (caprocks).
• Existing shale development from hydraulic fracturing offers an operational advantage, enabling CO₂ storage to leverage established well infrastructure, reservoir access strategies, and field-scale monitoring systems.

Future work should explore sulfate phase formation, including jarosite (KFe₃(OH)₆(SO₄)₂), using combined with reactive-transport modeling and calculations. Integrating EDS-based stoichiometries with kinetic datasets will help quantify CO₂ uptake across key mineral phases. Extending experiments to other shales will provide a crosslink benchmarking. Incorporating mineral-scale reaction rates into geomechanical and geophysical models can improve leakage-risk assessments, while EBSD and FIB-SEM will enable deeper characterization of reaction zones and deformation below the surface. These steps will sharpen monitoring strategies, and ultimately strengthen confidence in shale-dominated formations as long-term, self-adjusting CO₂ sequesters.