A Review of Coupled Geochemical–Geomechanical Impacts in Subsurface CO₂, H₂, and Air Storage Systems

Abstract

Increased demand for decarbonization and renewable energy has led to increasing interest in engineered subsurface storage systems for large-scale carbon reduction and energy storage. In these applications, a working fluid (CO₂, H₂, air, etc.) is injected into a deep formation for permanent sequestration or seasonal energy storage. The heterogeneous nature of the porous formation and the fluid–rock interactions introduce complexity and uncertainty in the fate of the injected component and host formations in these applications. Interactions between the working gas, native brine, and formation mineralogy must be adequately assessed to evaluate the efficiency, risk, and viability of a particular storage site and operational regime. This study reviews the current state of knowledge about coupled geochemical–geomechanical impacts in geologic carbon sequestration (GCS), underground hydrogen storage (UHS), and compressed air energy storage (CAES) systems involving the injection of CO₂, H₂, and air. Specific review topics include (1) existing injection induced geochemical reactions in these systems; (2) the impact of these reactions on the porosity and permeability of host formation; (3) the impact of these reactions on the mechanical properties of host formation; and (4) the investigation of geochemical-geomechanical process in pilot scale GCS. This study helps to facilitate an understanding of the potential geochemical–geomechanical risks involved in different subsurface energy storage systems and highlights future research needs.

1. Introduction

Interest in engineered subsurface storage systems as venues for retaining greenhouse gases or energy is growing. In such systems, a gas is injected into a deep geologic formation for temporary or permanent storage. Interactions between the injected gas, formation fluids, and formation mineralogy must be adequately assessed to evaluate the efficiency, risk, and viability of a particular storage site and operational regime. Here, an overview of the potential geochemical and geomechanical impacts on storage formations used for geologic carbon sequestration, underground hydrogen storage, and compressed energy storage will be considered.

Carbon capture and storage (CCS) has been proven to effectively reduce atmospheric greenhouse gas concentrations in the Earth’s atmosphere. The process in which CO₂ is captured and stored in a geological formation deep in the subsurface of the Earth is known as geological carbon sequestration (GCS) [1,2,3]. In this method, the CO₂ is mainly captured from a point industrial or energy production source and then injected into the subsurface; as such, interest and efforts associated with direct air capture are growing. The temperature and pressure conditions of the formation define the phase of the CO₂. When temperatures exceed 31 °C and the pressure exceeds 7.38 MPa, the CO₂ exists as a supercritical fluid, as shown in Figure 1 [4,5,6]. In this phase, CO₂ has characteristics similar to those of a liquid but is neither liquid nor gas. Storing CO₂ in these conditions has many advantages, such as requiring a smaller storage volume because of the higher density of CO₂ in this phase as compared to its density in the gas phase. As such, more CO₂ can be stored in a given geological formation if it is a supercritical fluid rather than a gas [4,6,7]. To ensure CO₂ exists in the supercritical state, the formation conditions must meet certain requirements. For example, for geological formations to be considered as storage reservoirs, they need to be at a depth of at least 800 m. This way, the pressures and temperatures will be sufficient to ensure CO₂ will exist as a supercritical phase.

Great candidates for CO₂ storage reservoirs are depleted oil and gas fields and saline aquifers, which have garnered the most considerable interest due to their storage potential volumes. Overall, sedimentary formations are largely targeted for CO₂ injection but igneous formations such as basalts are also promising [8,9]. Fundamental research has been carried out to understand the effects of carbon sequestration into formations with various lithologies [3,10,11,12]. Mineral precipitation is not the only way in which CO₂ is stored. Other types of trapping mechanisms include structural trapping, residual trapping, solubility trapping, and mineral trapping. A summary of these trapping mechanisms is shown in Table 1. These trapping mechanisms will depend on the rock formation’s composition and the reservoir conditions.

Figure 1. Phase diagram of CO₂ at different temperature and pressure conditions. Reprinted with permission from Beckingham and Winningham, Critical Knowledge Gaps for Understanding Water-Rock-Working Phase Interactions for Compressed Energy Storage in Porous Formations, ACS Sustainable Chemistry and Engineering [13]. Copyright 2020. American Chemical Society.

Figure 1. Phase diagram of CO₂ at different temperature and pressure conditions. Reprinted with permission from Beckingham and Winningham, Critical Knowledge Gaps for Understanding Water-Rock-Working Phase Interactions for Compressed Energy Storage in Porous Formations, ACS Sustainable Chemistry and Engineering [13]. Copyright 2020. American Chemical Society.

Table 1. Trapping mechanisms for CO₂ in geological formations.

Mechanism:		References:
Structural Trapping	CO2 is trapped physically. Faults and folds limit CO2 mobility within the formation.	(Saadatpoor et al., 2010) [14]
Residual Trapping	CO2 remains trapped in pore space. CO2 plume moves through the rock.	(Krevor et al., 2015) [15]
Solubility Trapping	CO2 is trapped by aqueous pore fluid. Injected CO2 dissolves in water, oil, and brine as aqueous species.	(Emami-Meybodi et al., 2015) [16]
Mineral Trapping	CO2 reacts with minerals found in the formation, creating an adequate environment for other minerals to precipitate such as calcite, magnesite, and siderite.	(Bachu, 2008; Zhang and DePaolo, 2017) [17,18]

Table 1. Trapping mechanisms for CO₂ in geological formations.

Mechanism:		References:
Structural Trapping	CO2 is trapped physically. Faults and folds limit CO2 mobility within the formation.	(Saadatpoor et al., 2010) [14]
Residual Trapping	CO2 remains trapped in pore space. CO2 plume moves through the rock.	(Krevor et al., 2015) [15]
Solubility Trapping	CO2 is trapped by aqueous pore fluid. Injected CO2 dissolves in water, oil, and brine as aqueous species.	(Emami-Meybodi et al., 2015) [16]
Mineral Trapping	CO2 reacts with minerals found in the formation, creating an adequate environment for other minerals to precipitate such as calcite, magnesite, and siderite.	(Bachu, 2008; Zhang and DePaolo, 2017) [17,18]

Underground Hydrogen Storage (UHS) is an efficient method for safe, long-term energy storage and recovery of the stored hydrogen [19,20,21]. In UHS, large amounts of H₂ can be effectively stored in geological formations such as empty salt caverns [21,22,23], deep hard rock aquifers [23,24], and depleted hydrocarbon reservoirs [25]. Understanding the geochemical interactions between H₂, rock, and brine is crucial for ensuring safe underground hydrogen storage. Injected H₂ can disrupt the existing equilibrium of the fluid–rock system, triggering H₂–rock–fluid reactions. These reactions can lead to mineral transformations within the reservoir rocks, potentially altering the geomechanical properties of both the reservoir rocks and the caprock, thereby affecting their structural integrity and storage capability [26].

Compressed air energy storage (CAES) is a physical energy storage technology with large-scale application potential [27]. It has the advantages of a large storage scale and site selection flexibility [28]. In CAES systems, a compressor is used to inject a storage gas for energy storage in a gas storage reservoir. The gas is then extracted and the energy recovered via an expansion machine [29]. The properties of the gas storage reservoir will directly affect the capacity of energy storage, efficiency, and operating costs. Common CAES geological gas storage reservoirs include natural salt caverns, saline aquifers, abandoned mines, etc. Among them, saline aquifers have the advantages of low construction costs, wide distribution, and controllable capacity [30,31]. The feasibility of such facilities have been fully proved in the Pittsfield CAES experiment and other practical cases [32,33].

Here, we review the current state of research regarding the coupled geochemical–gemechanical impacts of three subsurface energy storage systems: carbon capture and storage (CCS), hydrogen underground storage (UHS), and compressed air energy storage (CAES). First, we focus on the existing geochemical reactions in each system. Second, we address the impact of chemical reactions on the porosity and permeability of the formation rock. Third, we investigate the impact of chemical reactions on mechanical properties of formation rock. Finally, we review the pilot-scale studies.

2. Geochemical Reactions

2.1. CO₂ Geological Sequestration

Many geochemical interactions occur when CO₂ is injected into a geological formation such as a sedimentary formation. The geochemical reactions could impact the formation’s properties, such as its porosity and permeability. Some changes include water acidification, the buffering of water, dissolution of minerals within the formation, and mineralization of CO₂ [34]. When the CO₂ is injected into a formation, it first dissolves since CO₂ is very soluble in the aqueous phase. Next, the CO₂ combines with water, which results in carbonate (3) and bicarbonate ions (4). Such reactions can be represented as:

CO₂(sc/g) ⇌ CO₂(aq)

(1)

CO₂(aq) + H₂O ⇌ H₂CO₃

(2)

$${\mathrm{H}}_2{\mathrm{CO}}_3 \rightleftharpoons {\mathrm{H}}^{+}+{\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}$$

(3)

$${\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}\rightleftharpoons {\mathrm{H}}^{+}+{\mathrm{C}\mathrm{O}}_3^{2-}$$

(4)

Once carbonic acid forms and the pH conditions become more acidic, mineral dissolution and precipitation may occur. Carbonate minerals are typically the most reactive minerals in sedimentary formations. These minerals can also buffer the pH in the formation and reservoir system. Common minerals that may dissolve following CO₂ injection are calcite (5), dolomite (6), kaolinite (7), illite (8).

$${\mathrm{C}\mathrm{a}\mathrm{C}\mathrm{O}}_3+{\mathrm{H}}^{+}\to {\mathrm{C}\mathrm{a}}^{2+}+{\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}$$

(5)

$$\mathrm{C}\mathrm{a}\mathrm{M}\mathrm{g}{\left({\mathrm{C}\mathrm{O}}_3\right)}_2+2{\mathrm{H}}^{+}\to {\mathrm{C}\mathrm{a}}^{2+}+{\mathrm{M}\mathrm{g}}^{2+}+2{\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}$$

(6)

$${\mathrm{A}\mathrm{l}}_2{\mathrm{S}\mathrm{i}}_2{\mathrm{O}}_5(\mathrm{O}\mathrm{H}{)}_4+6{\mathrm{H}}^{+}\to 5{\mathrm{H}}_2\mathrm{O}+2{\mathrm{S}\mathrm{i}\mathrm{O}}_2+2{\mathrm{A}\mathrm{l}}^{3+}$$

(7)

$$\begin{gathered} {\mathrm{K}}_{0.6}{\mathrm{M}\mathrm{g}}_{0.25}{\mathrm{A}\mathrm{l}}_{2.3}{\mathrm{S}\mathrm{i}}_{3.5}{\mathrm{O}}_{10}{\left(\mathrm{O}\mathrm{H}\right)}_2 \left(\mathrm{s}\right)+8{\mathrm{H}}^{+}\to 5{\mathrm{H}}_2\mathrm{O}+0.6{\mathrm{K}}^{+}+ \\ 0.25{\mathrm{M}\mathrm{g}}^{2+}+3.5{\mathrm{S}\mathrm{i}\mathrm{O}}_2+2.3{\mathrm{A}\mathrm{l}}^{3+} \end{gathered}$$

(8)

When CO₂ is injected into a formation, mineral precipitation can occur and provide permanent, secure storage of the injected CO₂ [8,35,36]. Common carbonate minerals that could precipitate in sedimentary formations are calcite (9), magnesite (10), siderite (11), anhydrite (12), dolomite (13), Alunite (14).

$${\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}+{\mathrm{C}\mathrm{a}}^{2+}\to {\mathrm{C}\mathrm{a}\mathrm{C}\mathrm{O}}_3+{\mathrm{H}}^{+}$$

(9)

$${\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}+{\mathrm{M}\mathrm{g}}^{2+}\to {\mathrm{M}\mathrm{g}\mathrm{C}\mathrm{O}}_3+{\mathrm{H}}^{+}$$

(10)

$${\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}+{\mathrm{F}\mathrm{e}}^{2+}\to {\mathrm{F}\mathrm{e}\mathrm{C}\mathrm{O}}_3+{\mathrm{H}}^{+}$$

(11)

$${\mathrm{C}\mathrm{a}}^{2+}+{\mathrm{S}\mathrm{O}}_4^{2-}\to {\mathrm{C}\mathrm{a}\mathrm{S}\mathrm{O}}_4$$

(12)

$${\mathrm{C}\mathrm{a}}^{2+}+{\mathrm{M}\mathrm{g}}^{2+}+2{\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}\to \mathrm{C}\mathrm{a}\mathrm{M}\mathrm{g}{\left({\mathrm{C}\mathrm{O}}_3\right)}_2+2{\mathrm{H}}^{+}$$

(13)

$${\mathrm{K}}^{+}+3{\mathrm{A}\mathrm{l}}^{3+}+2{\mathrm{S}\mathrm{O}}_4^{2-}+6{\mathrm{H}}_2\mathrm{O}\to {\mathrm{K}\mathrm{A}\mathrm{l}}_3{\left({\mathrm{S}\mathrm{O}}_4\right)}_2(\mathrm{O}\mathrm{H}{)}_6+6{\mathrm{H}}^{+}$$

(14)

Although sedimentary formations are the most common choice for carbon storage, mineral trapping is not always feasible. When CO₂ is injected into a formation, many reactions can occur, and mineral precipitation will depend on numerous factors. In some cases, sedimentary formations will not contain carbonate minerals. Although CO₂ can interact with other minerals commonly found in porous sedimentary rocks, not all would dissolve. For example, quartz is one of the most common minerals in porous sedimentary formations. However, past studies have shown that this mineral has minimal reactions with CO₂. In the case of injecting CO₂ into a formation that contains minerals that have limited reactivity with CO₂, such as quartz and feldspars, the CO₂ will most likely remain mobile within the formation and trapping rely on structural, solubility, and residual trapping. All storage reservoirs need to have an impenetrable rock unit on top, or caprock, to prevent the upward migration of the CO₂ plume. Sedimentary formations targeted for CO₂ storage are similar to those that trap oil and gas, and in most cases, the CO₂ will remain mobile for many years, and the trapping will be highly dependent on the integrity of the caprock.

GCS is not only applicable in sedimentary formations. Recent studies have shown that mafic and ultramafic igneous formations are excellent for carbon sequestration. These igneous formations are rich in olivine (Mg₂SiO₄) as well as other minerals commonly found in mafic rocks such as serpentine (Mg₃Si₂O₅(OH)₄) and wollastonite (CaSiO₃). During carbon sequestration, the CO₂ interacts with these silicate minerals, resulting in the formation of magnesium or calcium carbonates as shown:

Mg₂SiO₄+ 2CO₂ → 2MgCO₃+ SiO₂

(15)

Mg₃Si₂O₅(OH)₄ +3CO₂ → 3MgCO₃ + 2SiO₂+ 2H₂O

(16)

CaSiO₃ + CO₂ → CaCO₃ + SiO₂

(17)

2.2. Underground H₂ Storage

When H₂ gas is injected into a dry reservoir rock or caprock, it is anticipated to remain inert and not react with minerals [37]. However, when H₂ is injected into a fluid–rock system, it disrupts the existing geochemical equilibrium and initiates various fluid–rock reactions that depend on the composition of the fluids and the rock. This causes the pH and redox state of the system to be re-equilibrated. It should be noted that only potential abiotic reactions are considered here but biotic reactions may also be important for H₂ storage systems.

The geochemical modeling studies conducted by [19,20,38,39,40,41] characterize the extent of mineral dissolution and hydrogen loss due to geochemical reactions during UHS, within a time frame of up to 30 years. The results indicate that carbonate dissolution is the primary reaction that occurs during UHS. All of these studies indicate that carbonate mineral dissolution can affect the porosity, permeability, and fluid flow behavior of the reservoir and influence the mechanical properties of the caprock and reservoir rocks [42,43,44]. It should be noted that geochemical interactions are anticipated to mostly occur under extreme underground conditions (extreme pressures, temperatures, and salinity) or over long time periods which may not be typical of hydrogen reservoirs made of sandstone, and further efforts are needed to assess the anticipated rate, extent, and impact of such reactions under pertinent storage conditions.

Reservoir and caprock formations typically contain a suite of potentially reactive minerals. Common minerals include carbonates (calcite [CaCO₃], dolomite [CaMg(CO₃)₂], magnesite [MgCO₃], and siderite [FeCO₃]), sulfates (anglesite [PbSO₄], anhydrite [CaSO₄], gypsum [CaSO₄·2H₂O], barite [BaSO₄], and celestite [SrSO₄]), sulfides (particularly pyrite [FeS₂]), and Fe³⁺ coupled minerals (hematite [Fe₂O₃], goethite [FeO(OH)]), where some clay minerals containing Fe³⁺ are particularly sensitive and may play critical roles in the system solution as they can react with stored hydrogen and lead to reductive dissolution [38,45,46]. When H₂ reacts with dissolved carbonate minerals such as (${\mathrm{C}\mathrm{O}}_3^{2-}$ and ${\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}$), CH₄ is generated through the redox reaction,

$${\mathrm{C}\mathrm{O}}_2+4{\mathrm{H}}_2={\mathrm{C}\mathrm{H}}_4+2{\mathrm{H}}_2\mathrm{O}$$

(18)

The redox reactions between ${\mathrm{C}\mathrm{O}}_3^{2-}$/${\mathrm{H}\mathrm{C}\mathrm{O}}_3^{-}$ and dissolved H₂ promote the dissolution of carbonate minerals. For sulfate minerals, the stored hydrogen can reduce ${\mathrm{S}\mathrm{O}}_4^{2-}$ from anhydrite (CaSO₄), anglesite (PbSO₄), barite (BaSO₄), gypsum (CaSO₄:2H₂O), and celestite (SrSO₄) to H₂S through the redox reaction:

$${\mathrm{S}\mathrm{O}}_4^{2-}+4{\mathrm{H}}_2+2{\mathrm{H}}^{+}={\mathrm{H}}_2\mathrm{S}+4{\mathrm{H}}_2\mathrm{O}$$

(19)

Geochemical simulations have indicated that redox reactions of H₂ with sulfate minerals can lead to the dissolution of sulfate minerals [39,47]. Similar results for sulfate mineral reductive dissolution have been observed during experimental studies [45,46,48], and have confirmed the dissolution of sandstones containing sulfates, such as anhydrite and barite, that were exposed to hydrogen and saline brine. Other studies [49,50,51] have also noted similar findings. Similar to carbonate minerals, the dissolution of sulfate minerals could partially affect the caprock’s integrity and rock’s mechanical behaviors [52]. In addition, the presence of redox-sensitive sulfate minerals in near-wellbore areas within the UHS system can trigger the dissolution of other pH-sensitive minerals, such as kaolinite and illite, that can be found near the wellbore. This dissolution can result in instability in these areas, potentially leading to downhole wellbore failures [53,54].

Pyrite is thermodynamically unstable in the presence of H₂, and therefore, their dissolution is marked as the main abiotic reaction that occurs during UHS under typical depleted hydrocarbon reservoir conditions [55]. The reduction of pyrite to pyrrhotite by hydrogen is also reported [20,56,57,58,59]:

$$\mathrm{F}\mathrm{e}{\mathrm{S}}_2+\left(1-\mathrm{x}\right){\mathrm{H}}_2=\mathrm{F}\mathrm{e}{\mathrm{S}}_{\left(1+\mathrm{x}\right)}+\left(1-\mathrm{x}\right){\mathrm{H}}_2\mathrm{S}(0<\mathrm{x}<0.125)$$

(20)

The dissolution of pyrite within the caprock presents a challenge to maintaining its structural integrity as pyrite, which typically exists in the form of framboidal cementation within host rock, may cause damage [41]. Conversely, carbonate minerals like calcite and dolomite are commonly present in both the reservoir and caprock, and their reductive dissolution can increase the in situ pH levels [20,41].

In addition to sulfate minerals, the ferric iron-bearing minerals such as goethite [FeO(OH)] and hematite (Fe₂O₃) can react with stored hydrogen through the redox reaction where Fe³⁺ could be reduced to Fe²⁺ [60]. However, such reactions minimally affect the reservoir’s porosity and permeability due to their slow reaction rates [37,38,61]. Dopffel [62] further highlighted that the efficacy of these reactions is often limited, unless accelerated by microbial activities. Furthermore, the reduction of Fe³⁺-bearing oxides predominantly occurs at high temperatures (300 to 700 °C) which far exceed typical geological storage conditions [63,64,65,66]. Apart from the reductive dissolution, clay minerals such as illite and kaolinite may dissolve, but not as a result of redox reactions. Instead, their dissolution primarily stems from geochemical shifts triggered by the reductive dissolution of other redox sensitive minerals such calcite that can raise the pH levels, could accelerate the dissolution of kaolinite and illite under increasingly alkaline conditions [20,41,67,68].

Salt caverns are one of the most commonly used geological formations for UHS [69]. Comparatively, H₂ storage in salt caverns is less challenging [70]. Salt caverns typically consist of layers of ~99% halite salt (NaCl) salt of and <1% anhydrite (CaSO₄) salt [71]. Halite is chemically inert under normal conditions and is not expected to react with H₂ during UHS in salt caverns. Therefore, salt caverns are less susceptible to being affected by hydrogen-induced geochemical reactions; hence they exhibit a significantly higher mechanical stability [22].

2.3. Compressed Air Energy Storage

Geochemical reactions between the injected air, formation fluid, and formation mineralogy are also anticipated in compressed air energy storage systems. The Pittsfield CAES experiment found a reduction in the O₂ concentration in the stored air due to the reactions between introduced air and native species in the reservoir [72]. Sulfide minerals are anticipated to be the main reactive species in the reservoir formation where the primary reactant in the Pittsfield case was pyrite (FeS₂). The complete oxidation of pyrite leads to the formation of hematite (Fe₂O₃):

$$4\mathrm{F}\mathrm{e}{\mathrm{S}}_2+11{\mathrm{O}}_2=2{\mathrm{F}\mathrm{e}}_2{\mathrm{O}}_3+8{\mathrm{S}\mathrm{O}}_2$$

(21)

If this reaction does not proceed to completion, partial oxidation might lead to the generation of ferrous sulfate (FeSO₄) or Fe(OH)SO₄, which can result in the production of colloidal ferric hydroxide and melanterite, respectively [30].

$$2\mathrm{F}\mathrm{e}{\mathrm{S}}_2+7{\mathrm{O}}_2+{2\mathrm{H}}_2\mathrm{O}=2\mathrm{F}\mathrm{e}{\mathrm{S}\mathrm{O}}_4+2{\mathrm{H}}_2{\mathrm{S}\mathrm{O}}_4$$

(22)

$$4\mathrm{F}\mathrm{e}{\mathrm{S}\mathrm{O}}_4+{\mathrm{O}}_2+{2\mathrm{H}}_2\mathrm{O}=4\mathrm{F}\mathrm{e}\left(\mathrm{O}\mathrm{H}\right){\mathrm{S}\mathrm{O}}_4$$

(23)

Gypsum (CaSO₄·2H₂O) is another potential oxidation product that may precipitate through the dissolution of carbonate minerals. Such reactions form scale deposits that might block pore space and impair the CAES system’s performance. Additional efforts are needed to assess the rate, extent, and impact of such reactions.

3. Geochemical Alteration of Porosity and Permeability

Chemical reactions, including primary mineral dissolution and secondary mineral precipitation, may result in changes in the reservoir’s mineralogy and pore structure, which in turn affect the reservoir’s porosity and permeability. In this section, we review recent efforts to quantify the impact of geochemical reactions on porosity and permeability for different rock samples and in different subsurface energy systems. It should be noted that experimental approaches are largely the focus here but there are additional extensive simulation efforts that will not be targeted here.

3.1. Experimental Approach

In the study of geochemical reactions in subsurface energy storage conditions, two methods are commonly used: batch experiments [73,74] and flow-through experiments [75,76]. In batch experiments, rock samples react with the working gas and brine in an autoclave over a wide range of temperature and pressure conditions to mimic the reservoir conditions. In a single autoclave, rock samples can react with different states of working fluids [77] (Figure 2). Some rock samples are immersed in the liquid phase and react with working gas-rich brine while other samples are located above the liquid phase and react with the water-rich gas.

To quantify the impact of chemical reactions, the rock samples are characterized before and after reaction. Their porosity can be determined using a helium porosimeter [78], Mercury Intrusion Capillary Pressure (MICP) [79], Scanning Electron Microscopy (SEM) [80], and X-ray Micro Computed Tomography (XMCT) [81]. Their permeability can be measured at steady state using Darcy’s Law (permeable sample) [76] and pulse decay methods (low permeable sample) [82]. The limitation of this approach is the lack of a confining environment. The ex situ characterization before and after the reaction cannot show the time-lapsed evolution of the sample.

In flow-through experiments, the rock sample is fixed in a core holder under confining pressure and reacted with the injected fluids. A flow-through system usually includes injection pumps, a core holder, confining pump, and a back pressure regulator (Figure 3). Various fluids can be injected in sequence into the rock samples under realistic reservoir conditions to reproduce the real applications [83]. During the flow-through experiments, the permeability of the tested sample can be measured in situ. With the help of X-ray computed tomography (X-ray CT), the porosity can also be determined in situ [84]. Once the downstream is closed, the flow-through set-up can work as a static system with a confining environment [85,86]. A summary of the research quantifying the impact of chemical reactions on the porosity and permeability of rock is shown in

Table 2. Summary of porosity and permeability alteration of reservoir rock due to geochemical reactions.

Author	Rock Type (Sample Amount)	Fluid	Method	Porosity	Permeability
Bemer and Lombard [85]	Carbonate (11)	Retarded acid	Flow-through	1.0–2.1% increase	Show both increase and decrease
Davila et al. [84]	Sandstone (1)	CO2 + brine	Flow-through	6.8% reduction	23% reduction
Han et al. [87]	Carbonate (1)	CO2 + water + decane	Flow-through	9% reduction	83% reduction
Harbert et al. [88]	Sandstone (2)	CO2 + brine	Batch	8% increase, 9% reduction	7–205% increase
Huq et al. [89]	Sandstone (2)	CO2 + water CO2 + brine	Flow-through	12% increase, 20% increase	170% increase
Mouzakis et al. [90]	Shale (2)	CO2 + brine	Batch	25% increase, 2% reduction	Not report
Rimmelé et al. [77]	Limestone and sandstone (40)	Wet Sc-CO2 CO2-saturated water	Batch	Limestone: 2–4% increase. Sandstone: 3–4% increase	Limestone: 56–180% increase Sandstone: 0–700% increase
Shi et al. [78]	Sandstone (9)	CO2 + brine	Batch	1.1–14.9% increase	250% increase
Tutolo et al. [91]	Sandstone (3)	CO2 + water CO2 + brine	Flow-through	Reduction in recycled test while increase in single-pass tests	26–37% reduction
Wang et al. [92]	Limestone (6)	CO2 + brine	Flow-through	1–7% increase	8–32% increase
Yu et al. [93]	Sandstone (3)	CO2 + brine	Flow-through	0–1.8% reduction	10–20% reduction
Yekta et al. [37]	Sandstone (3)	H2 + H2O	Flow-through	Insignificant	Insignificant
Flesch et al. [45]	Siltstone and sandstone (21)	H2 + brine	Batch	From 56% reduction to 107.8% increase	From 60.5% reduction to 38.5% increase
Shi et al. [42]	Sandstone, caprock, and cement (12)	H2 + brine	Batch	From −26.8% reduction to 4.2% increase	From −39.9% reduction to 10.2% increase
Bensing et al. [94]	Claystone (1)	H2 + brine	Batch	Increase	Not reported
Henkel et al. [46]	Caprock and sandstone (190)	H2 + CO2 + brine	Batch	Increase	Not reported
Pudlo et al. [48]	Not reported	H2 + brine	Batch	Increase in sulfate and calcite bearing sandstones.	Increase in sulfate and calcite bearing sandstones.
Wang and Bauer [95]	Sandstone	O2 + brine	Geochemical modeling	1% increase	5% increase

.

3.2. CO₂ Geological Sequestration

In CGS systems, carbonates are highly reactive to acidified brine (Equations (5) and (6)). The rapid dissolution of the carbonate phases may lead to greater porosity [96]. The induced alteration can also create preferential flow paths and enhance the permeability of the carbonate samples [97]. From Table 2, it can be seen that most of studies on carbonate samples show increases in porosity. For example, Bemer and Lombard used a retarded acid, which is activated under a specific temperature, to homogeneously alter Comblanchian limestone and Lavoux limestone samples for 5 days. The results showed an increase in porosity between 1.0 and 2.1%. However, there was no clear trend in the permeability variations [85]. A similar result was reported when Lavoux limestone samples interacted with wet scCO₂ and CO₂-saturated water for one month [77]. Both porosity and permeability increased due to microscopic dissolution. An opposite behavior was reported by Han [87]. A dramatic reduction in permeability was observed in their core flooding test. Their pore network model indicated the disintegration of rock grains, which developed abundant, non-connected pores due to dissolution and precipitation.

Sandstone and shale samples with more a complex mineralogy exhibit a greater uncertainty in the relationship between chemical reactions and porosity/permeability. For example, Shi [78] and Harbert [88] investigated the interaction between Mt. Simon Sandstone and CO₂/brine in reservoir conditions in batch experiments. They both observed mild increases in porosity and dramatic increases in permeability. Their results are consistent with Huq [89]. Shi [78] found a reduction in the mesopore volume and an increase in the macropore volume (Figure 4), which indicated a conversion from mesopores to macropores due to mineral dissolution. Harbert [88] observed the development of cracks inside the core (Figure 5), which were responsible for the high permeability. In contrast, Davila [84] found a reduction in the porosity and permeability of Mt. Simon Sandstone during flow-through experiments, which were explained by the clogged pore spaces observed in the CT images. Similarly, Yu [93] reported a reduction in both porosity and permeability during flow-through experiments. They claimed that the precipitation of new minerals, combined with the clay particles released by dissolution of the carbonate cement accumulated at pore throats, were responsible for the observed reduction in permeability (Figure 6).

3.3. Underground H₂ Storage

In UHS systems, the research into H₂-induced geochemical reactions is not as extensive as that into CO₂-indiced reactions in GCS systems, and the mechanisms of how geochemical reactions affect the porosity and permeability of the surrounding rock is still not clear. For example, Yekta [37] carried out batch experiments with sandstone, H₂, and water at 100–200 °C and a H₂ pressure up to 100 bar in a period ranging from 1.5 to 6 months. The results demonstrated that reactions between the minerals in sandstone and hydrogen are very limited, thus the influence on the rock porosity and permeability is minor. In contrast, a significant alteration in the porosity (ranging from −56% to +107.8%) and permeability (ranging from −60.5% to +38.5%) of Permian and Triassic sandstones after six weeks of interaction with H₂ and brine was reported by Flesh [45]. The changes in the samples’ petrophysical properties were due to the alteration/dissolution of pore-filling anhydrite and carbonate cements (Figure 7).

Shi [42] conducted an experimental study on sandstone, caprocks, and well cement extracted from an existing natural gas storage site to investigate the potential changes in porosity and permeability resulting from geochemical reactions. The findings highlighted mild changes in both the porosity and permeability of sandstones. However, the caprocks demonstrated a significant reduction in permeability while the cement showed an enhancement in permeability. The XRD analysis demonstrated minor change in sandstone and caprock’s mineralogy. A significant reduction in the amount of portlandite in cement sample may explain the observed increase in the cement’s permeability.

Pudlo [48] conducted a series of batch experiments to investigate the effect of hydrogen pressure, temperature, and salinity on the geochemical reactions in carbonate-containing sandstone and claystone samples. The samples were subjected to varying conditions between 4–20 MPa, 40–120 °C, and 16,000–350,000 mg/L. The study found no mineral reactions within the lower conditions, but significant calcite dissolution occurred at higher values. Similarly, Bensing [94,98] observed a significant dissolution of calcite fragments in claystone caprocks that were saturated with hydrogen and a 10 wt% NaCl solution.

3.4. Compressed Air Energy Storage

In CAES systems, there is a lack of experimental studies on the geochemical reactions and induced alterations in a formation’s porosity and permeability. As noted above, iron oxidation is noted to be a significant reaction in these systems. The major concern regarding iron oxidation (Equations (21)–(23)) is oxygen depletion during storage [27,72]. Wang [95] conducted geochemical modeling to quantify the changes that occur due to the induced geochemical reactions in both the stored air and the storage formation. The result indicated that the pH of the formation fluid after 20-year cyclic daily operation can drop significantly below 1 near the gas well, increasing the risk of well corrosion, but with smaller effects at larger distances. However, the mineral dissolution and precipitation found in the storage formation results in only minor increases in porosity and permeability with relative changes up to 1.0% and 5.0%, respectively. Although the air-induced reactions show insignificant impacts on the porosity and permeability of the host formation, CO₂, which is reactive to the formation rock, has been considered as s cushion gas for air storage [99]. Iloejesi [100] conducted reactive transport modeling to evaluate the geochemical reactions in a compressed energy storage site using CO₂ as a cushion gas. The result showed that porosity increased from 24.84% to 31.1% at the location closest to the injection well while it increased to 25.8% at the location furthest from the injection well.

4. Geochemical Alteration of Mechanical Properties

Geochemical reactions can result in changes in reservoir composition and morphology, which in turn modify the mechanical response of the formation. In this section, we review recent studies quantifying the impact of geochemical reactions on mechanical properties for different types of rock samples.

4.1. Experimental Approach

Strength and the elastic modulus are important mechanical properties used to describe the mechanical behavior of rock samples. These parameters can be measured by uniaxial compression tests [83,101,102], triaxial compression tests [103,104,105], and ultrasonic wave velocity measurements [78,88,106]. In the batch reaction method [78,88], the mechanical properties of reacted samples are measured under certain conditions. Several intact samples without fluid treatment are tested using the same protocol as that used for reacted samples. The results for the intact samples are used as a baseline to evaluate the change in mechanical properties due to reactions. In the in situ reaction method [86,107], the sample is loaded into the set-up (uniaxial or triaxial compression cell) before reacting with the fluids. Then, the reactive fluids are injected into the sample holder. The sample reacts with the fluid inside the set-up at reservoir conditions. The mechanical properties of the sample are then measured at relevant conditions. A summary of research quantifying the impact of chemical reactions on the mechanical properties of rock samples is shown in

Table 3. Summary of alterations to mechanical properties due to geochemical reactions *.

Author	Rock Type	Fluid	Method	Mechanical Test	Key Findings
AL-Ameri et al. [108]	Limestone	CO2 + brine	Batch	Indirect tensile strength, unconfined compression, acoustics testing	Reduction in YM, ITS, and UCS
Espinoza et al. [109]	Sandstone, siltstone, shale	CO2 + brine	Batch	Triaxial frame and deviatoric loading	Reduction in stiffness, strength, and brittleness of Sandstone and siltstone. Increase in stiffness in shale
Foroutan et al. [110]	Sandstone	CO2 + brine	Flow-through	creep, ultrasonic wave velocity, multi-stage failure tests	Reduction in strength and wave velocity
Foroutan et al. [111]	Sandstone	CO2 + brine	Flow-through	Isotropic compression, multi-stage elastic, and cyclic tests	Reduction in YM, BM, increase in PR
Fuchs et al. [112]	Sandstone	CO2 + brine	Batch	Scratch test	Reduction in fracture toughness
Grgic [86]	Carbonate	CO2 + water	Flow-through and no flow	Isotropic and deviatoric loading	Significant axial compaction in dynamic injection.
Grombacher et al. [113]	Carbonate	CO2 + water	Flow-through	Ultrasonic wave velocity	Reduction in wave velocity
Hangx et al. [114]	Sandstone	CO2 + brine	Flow-through	Triaxial creep experiments and ultrasonic wave velocity	Insignificant change in all mechanical properties
Huang et al. [83]	Sandstone	CO2 + brine	Flow-through	Uniaxial compression test, Brazilian splitting test, and fracture test	Reduction in UCS, BTS, and fracture toughness
Harbert et al. [88]	Sandstone	CO2 + brine	Batch	Ultrasonic wave velocity	Reduction/increase in YM
Kim and Makhnenko [115]	Sandstone and limestone	CO2 + water	Flow-through	Hydrostatic compression and triaxial compression	Increases in bulk compressibility, Skempton’s B coefficient decreases for sandstone but increases for limestones
Kim et al. [107]	Limestone	CO2 + water	Flow-through	Triaxial cell	Reduction in strength and YM
Lamy-Chappuis et al. [116]	Sandstone	CO2 + brine	Flow-through	Ultrasonic wave velocity and multiple failure test	Reduction in wave velocity and strength
Le Guen et al. [117]	Limestone and sandstone	CO2 + water	Flow-through	Triaxial cell	Increase in strain rates of the limestones, and weaker response of the sandstone.
Liteanu et al. [118]	Calcite aggregates	CO2 + water	Batch	Uniaxial compaction creep	Acceleration of strain rate
Lyu et al. [101,102]	Shale	CO2 + water CO2 + brine	Batch	Uniaxial compression, acoustic emissions	Reduction in UCS, YM, and brittleness index
Marbler et al. [103]	Sandstone	CO2 + brine	Batch	Triaxial compression	Reduction in UCS and modulus of deformation for carbonate sandstone
Oikawa et al. [119]	Sandstone	CO2 + water	No flow	Triaxial strength and triaxial creep	Reduction in YM. Creep life time increase with CO2 exposure time.
Rathnaweera et al. [120]	Sandstone	CO2 + brine	Batch	Unconfined compressive,	Reduction in UCS and YM
Rathnaweera et al. [121]	Sandstone	CO2 + brine	Batch	Triaxial strength	Reduction in strength
Rimmele et al. [77]	Limestone and sandstone	Wet Sc-CO2 CO2-saturated water	Batch	Unconfined compressive	Insignificant change in UCS and YM
Pimienta et al. [122]	Limestone and sandstone	CO2 + brine	No flow	P-wave velocity	Reduction in P-wave velocity
Rinehart et al. [123]	Sandstone	CO2 + brine	Flow-through	Hydrostatic compression and triaxial tests	Creep strain rates are accelerated and strength weakened
Samuelson et al. [124]	Sandstone and caprock	CO2 + water	No flow	Direct shear experiments	No significant effect on the coefficient of friction
Shi et al. [78]	Sandstone	CO2 + brine	Batch	Ultrasonic wave velocity	Reduction in YM
Simmons et al. [125]	Sandstone	CO2 + water	Flow-through	Indirect tensile strength tests, ultrasonic wave velocity	Tensile strength maintain, reduction is wave velocity
Vanorio et al. [126]	Sandstone	CO2 + water	Flow-through	Ultrasonic wave velocity	Increase wave velocity due to salt precipitation
Vialle and Vanorio [106]	Carbonate	CO2 + water	Flow-through	Ultrasonic wave velocity	Reduction in wave velocity
Yin et al. [127]	Shale	Sub-critical CO2 (SubCO2) Super-critical CO2 (ScCO2)	Batch	Uniaxial compressive strength, acoustic emissions	Reduction in UCS and elastic modulus
Zhang et al. [104,105]	Sandstone	CO2 + water	No flow	Triaxial compression	Reduction in strength and elastic modulus, enhancing the bulk compaction
Zhang et al. [128]	Shale	CO2 + brine	Batch	Uniaxial compression strength, acoustic emissions	Reduction in UCS and elastic modulus
Zheng et al. [129]	Sandstone	CO2 + brine	Flow-through	Triaxial compression and seepage creep	Reduction in compressive strength and threshold stress of shear dilatancy
Zou et al. [130]	Shale	CO2 + brine	Batch	Comprehensive test instrument and friction wear testing machine	decline in tensile strength and surface friction coefficient
Dabbaghi et al. [131]	Sandstone	H2 + brine	Batch	Uniaxial compression and triaxial compression	Reduction in peak strength

* YM = Young’s modulus; ITS = indirect tensile strength; UCS = uniaxial compressive strength; PR = Poisson ratio; BM = bulk modulus; BTS = Brazilian tensile strength.

.

4.2. CO₂ Geological Sequestration

The effect of CO₂-induced geochemical reactions on the mechanical properties of carbonates have been investigated via uniaxial compression tests [77,108], triaxial compression tests [107,115], and ultrasonic wave velocity measurements [106,113,122,126]. The general trend is a reduction in strength and the elastic modulus. However, Rimmele [77] reported no significant change in the mechanical properties of Lavoux limestone samples after a one-month interaction with scCO₂-saturated water and water-saturated scCO₂ at 28 MPa and 90 °C, although increases in porosity and permeability were observed. The authors pointed out that further tests performed at longer durations are necessary to track the evolution of the mechanical properties in CO₂ environments under reservoir conditions. AL-Ameri [108] reported reductions in indirect tensile strength, unconfined compressive strength, and the dynamic Young’s modulus of Khuff limestone and Indiana limestone samples after 14–90 day reactions with CO₂ and brine at 13.8 MPa and 100 °C. The extent of change in the mechanical properties increased with the duration of CO₂–brine contact time. Similarly, reductions in the Young’s modulus and Bulk modulus of Apulian limestone samples were reported after 3 days of CO₂ treatment (7 MPa, 22 °C) by Kim [107,115]. Based on the analysis of surface electron microscopy and mercury intrusion porosimetry results, the authors attributed these effects to carbonate dissolution and microstructural changes. Vialle and Vanorio [106] observed a continuous decrease in S and P wave velocity during the injection of CO₂-rich water into carbonate samples (Figure 8). The time-lapsed SEM images of rock sample (Figure 9) verified the change in microstructure due to carbonate dissolution and provided an explanation for the observed changes in wave velocity. Vanorio [126] further investigated the effect of salt precipitation on the elastic properties of Fontainebleau sandstone samples using the same set-up. An increase in the wave velocity due to salt precipitation was reported. Grombacher [113] studied the role of dissolution-induced compaction in the CO₂-rich water flow-through experiments. They concluded that compaction in chalky micritic facies favoring grain sliding suppressed the formation of flow channels and led to larger velocity reduction.

Some work has investigated the effect of CO₂ on the long-term creep behavior of carbonate samples. Le Guen [117] conducted flow-through triaxial testes on Estaillades limestone, Lavoux limestone, and sandstone. They observed a much more significant increase in strain rates of the limestone samples than that of sandstone samples. The different response between limestone and sandstone was attributed to higher reactivity of carbonate than quartz in CO₂ acidified fluids. Grgic [86] carried out triaxial tests under closed flow conditions and open flow conditions using oolitic limestone. In closed flow conditions, the buffer effect of water–CO₂–calcite re-equilibrium inhibits further calcite dissolution and leads to an insignificant amount of axial compaction. In open flow conditions, a significant amount of compaction was observed due to calcite dissolution.

Unlike carbonate samples, the effect of CO₂-induced geochemical reactions on the mechanical properties of sandstones does not show a clear trend due to the heterogeneous nature of sandstone samples. Conflicting results exist in the reported studies using the same method. For instance, in the studies using the uniaxial compression test, Rathnaweera [120] reported a significant reduction in uniaxial compressive strength and Young’s modulus of Hawkesbury sandstone after a four-month long saturation with CO₂ and water and CO₂ and brine at 8 MPa and 32 °C. The scanning electron microscope (SEM), X-ray diffraction (XRD), and X-ray Fluorescence (XRF) analyses demonstrated quartz mineral corrosion and the dissolution of calcite and siderite. However, no significant change in the uniaxial compressive strengths or the Young’s moduli of the Lavoux limestone samples after a one-month long interaction with scCO₂ saturated water at 28 MPa and 90 °C was reported in another study [77]. Huang [83] observed a reduction in the uniaxial compressive strength of Zunyi sandstone due to saturation with CO₂ and brine at 10 MPa and 32 °C. But in this study, an increase in the elastic modulus was observed. The author hypothesized that the increase in the modulus could be attributed to clay swelling [132].

In the studies using the triaxial compression test, a trend of CO₂ induced reduction in strength and elastic modulus of sandstone samples has been reported [104,105,111,121,129]. On the other hand, Hangx [114] carried out flow-through triaxial tests to investigate the effect of carbonate cement dissolution on the mechanical properties of the samples. The CO₂-rich brine injection did not affect the mechanical properties of Captain Sandstone even though the carbonate was completely dissolved. This is likely because grains in this material are cemented by quartz that is inert in CO₂-rich brine. Similar results retaining the strength of silicate sandstone after saturation with CO₂ and brine were reported in another study [103].

Reductions in wave velocity and the dynamic elastic moduli of CO₂-treated sandstone were reported in studies using ultrasonic wave velocity [78,88,125]. Similar results noting a reduction in the wave velocity of calcite-bearing sandstones after flooding with CO₂ saturated brine was reported by Lamy-Chappuis [116]. In this study, a series of failure tests were conducted, and a decrease in peak stress after calcite dissolution was observed.

Additional efforts to consider changes in creep behavior and fracture toughness have been carried out as well. Oikawa [119] and Foroutan [110] investigated the effect of CO₂ on the creep behavior of sandstone. Both studies reported significant increases in the creep response. Samuelson [124] conducted direct shear experiments with sandstone samples to understand CO₂ induced change in frictional strength. They found the presence of CO₂ and brine had no clear impact on frictional strength. Fuchs conducted a scratch test to investigate the alteration of fracture toughness of Mt. Simon Sandstone due to interaction with CO₂ and brine at 50 °C and 22.1 MPa for 4–8 weeks. Reductions in the fracture toughness of 32.1% after 4 weeks and of 69.5% after 8 weeks were observed.

For shale samples, the uniaxial compression tests showed a clear trend of reducing uniaxial compressive strengths and a decreasing elastic modulus regardless of the state of saturation fluid (CO₂, scCO₂, scCO₂+brine, scCO₂+water) [101,102,127,128]. However, the effect of CO₂ saturation on crack properties differs among studies. For instance, Yin [127] applied acoustic emission analysis to study the effect of sub- and super-critical CO₂ saturation on the crack properties of organic-rich shales. An increase in the crack initiation stress and decrease in the crack damage stress was observed. Zhang [128] reported the same trend in the case of sub- and super-critical CO₂ saturation in black shale samples and attribute this change to CO₂ adsorption-induced swelling. In the case of CO₂-brine saturation, lower values for both crack initiation stress and crack damage stress were observed due to mineral dissolution. Lyu [101,102] investigated the effect of saturation time on the crack properties of black shale samples when saturated with CO₂ and brine and CO₂ and water using acoustic emission analysis. They found a positive correlation between the total cumulative acoustic emission (AE) energy and saturation time. Zou investigated the impact of CO₂-brine saturation on the shear strength of Lujiaping and the Longmaxi shales. A reduction in the friction coefficient of up to 9.8% was observed at 120 °C.

4.3. Underground H₂ Storage and Compressed Air Energy Storage

Despite the multiple recent studies evaluating hydrogen–rock–brine interactions in UHS, there is limited knowledge on their potential impact on the geomechanical properties of both the reservoir and caprock. Dabbaghi [131] conducted uniaxial and triaxial compression tests on sandstone samples treated with brine and 50%H₂ and brine and 100%H₂. Reductions in peak strength, the elastic modulus, and the effective friction angle were observed due to the dissolution of dolomite and clay minerals. Simulation studies [38,133] have shown that H₂ dissolution in brine leads to a decrease in pH at relevant pressure and temperature conditions of UHS in sandstone reservoirs. This, in turn, results in a rapid short-term dissolution of carbonates (e.g., calcite and dolomite) and sulphate (e.g., anhydrite). This carbonate dissolution results in a rapid increase in the rock’s porosity and permeability, leading to a decrease in its compressive strength, Young’s modulus, and Bulk modulus [134,135].

In caprocks made of calcareous shale, which typically contain high levels of calcite, the dissolution of calcite compromises the structural integrity of the caprock. However, in caprocks composed of siliceous and argillaceous shale, where the carbonate content is lower, the hydrogen-induced carbonate dissolution can weaken the carbonate cementation, affecting the caprock’s stability [136]. When hydrogen reacts with iron-bearing clays, it can cause the reduction of Fe (III), causing hydrogen sorption in swelling clay-rich reservoir rocks and caprocks [58,137]. This reaction can cause local stress variations, potentially promoting mechanical fatigue, triggering slip and potentially enhancing hydrogen’s lubrication effect, which could lead to fault reactivation and leakage in geological seals during the lifespan of UHS operations [43].

Up to now, the effects of geochemical reactions on the mechanical properties of formation rock in CAES systems have not been explicitly studied. The production of colloidal ferric hydroxide and melanterite, as described in Equations (22) and (23), swell to as much as 500% of the original pyrite volume. Such a volume increase might lead to a deterioration of the expansive stresses acting on the caprock layers [30]. When CO₂ works as a cushion gas in compressed air storage, as described in the CGS section, the CO₂ induced mineral dissolution and precipitation can alter the mechanical properties of the formation. However, as pointed out by Iloejesi [100], the extent of dissolution is limited in the cyclic flow regime compared to the injection-only flow regime due to the lack of continuous undersaturated fluid.

5. Pilot-Scale Observations

Due to the lack of research on the geochemical–geomechanical processes in operational UHS and CAES systems, in this section, we will focus on pilot-scale studies about GCS sites. So far, there have been limited chances to directly investigate the geomechanical effects of massive CO₂ injection on the field scale in the real formation [138,139]. Based on the available data, we choose three locations here to highlight the influence of geochemical reactions on geomechanical properties of the host formation.

5.1. Decatur, Illinois, United States

The first phase of the Illinois Basin—Decatur Project (IBDP) was carried out over three years under the US Department of Energy, and was funded to show the safety of the injection of CO₂ by injecting approximately 1102 tons of CO₂ per day at depths beyond 2000 m into the Mount Simon formation [140,141,142,143]. Mt. Simon is made of a course-grained sandstone mostly composed of quartz, feldspar, and clay minerals.

Various monitoring methods, including geophysical monitoring wells, microseismic monitoring, surface deformation monitoring by satellite interferometry, and continuous GPS measurements, were used in the IBDP to measure the pressure and the movement of the storage formation [142,143,144,145]. A network of surface and borehole sensors (Geophones) was utilized to detect seismic activities [143,145]. Microseismic activities were measured in the pre-injection, injection, and post-injection periods [143,145]. The pre-injection data were gathered 1.5 years before the injection to provide a baseline. Microseismic activities were also measured for the injection period (3 years) and after shut-in [142,143]. For the baseline period, eight events were captured overall in the formation. In contrast, for the injection period, the overall average number of events per day in the formation was slightly over four, similar to other carbon-capturing projects [143]. To understand these observed microseismic activities due to CO₂ injection, several laboratory research using the Mt. Simon sandstone collected from the real site have been conducted [78,84,88,146,147]. The impact of CO₂-induced geochemical reactions on the transport and mechanical properties of Mt. Simon sandstone have been extensively investigated. The overall result showed reductions in mechanical properties and increase in porosity and permeability.

5.2. Heletz, Israel

The Heletz facility was established as a small-scale CO₂ injection experiment in the Heletz oil reserve located in Israel’s southern Mediterranean coastal plain [148,149]. This location has rich geological data due to prior petrological investigation (logs of several deep wells in a limited region). The CO₂ injection site was chosen to ensure that there was no oil in close proximity, an abandoned well with the potential for re-entry, and a geological data record [148]. The storage formation consists of three layers of poorly consolidated sand with an average thickness of 10.6 m separated by shales of varying thicknesses located at a depth of 1.65 km [142,148,149]. A limestone layer sits above the sand layers, and a thick layer of impermeable shale and marl serves as a caprock above that [148,149].

The injection well is located between two faults, 700 m apart. Because of this proximity, fault stability was analyzed for this site. The reactivation of faults is primarily affected by the stress ratio, original fault permeability, and confinement formation permeability [142,150]. The caprock and reservoir rock can be considered as confinement formations in this site. Analyses revealed that localized dissolution and precipitation process between CO₂/brine and reservoir rock, particularly the dissolution of dolomite and ankerite, can increase the permeability of the reservoir [151]. Higher permeability can lead to better fluid communication, followed by changes in pore pressure and effective stress, which raise the risk of failure (fault slip or fracture) [152].

5.3. Hontomín, Spain

The Hontomín CO₂ storage plant is a test facility located in Spain. The subsurface structure of the Hontomín site, placed at a depth of 1450 m, is a dome-like structure with a primary reservoir formation of Jurassic limestones and dolomites of the Sopeña Formation, characterized as a fractured carbonate reservoir, sealed by black shales and marls as the caprock [151,153]. While this reservoir formation has a matrix porosity of less than 1% and a matrix permeability of less than one mD, the fracture network contributes significantly to the fluid flow (fracture permeability of 0.5–15 mD) [142,151,154,155].

A passive seismic network of 30 surface stations and a borehole seismic array were used to record the microseismic events [156]. Microseismic events were observed to occur with permanent permeability increase, which shows that fractures opened as a result of the dilatancy that happens when rough fractures experience shear slip [157]. Additionally, the CO₂ injection and geochemical reactions increased the primary reservoir permeability from 0.5 mD to 15 mD [154] and the secondary reservoir from 0.022 mD to 0.085 mD [151]. The higher permeability of the sample after the reaction may allow the fluid to communicate with fractures or faults, leading to higher pore pressure and, as a result, lower effective stress, which can cause hydraulic fracture and shear slip of pre-existing fractures, leading to mechanical failure and microseismic occurrences [158].

6. Conclusions

Coupled geochemical–geomechanical processes in subsurface energy storage systems are complex and involve different reactions in different systems. For instance, CO₂-induced dissolution and precipitation in CGS systems, H₂-induced redox reactions in UHS systems, and air-induced oxidation in CAES systems. The effect of these reactions on the porosity, permeability, and mechanical properties of a rock formation are dependent on a larger set of parameters, including mineralogy, cementation state at grain-to-grain contact, reservoir conditions, initial pore structure, saturation fluids, composition of fluids, and so on. Another factor of complexity is the variety of experimental methods used in this study. It is hard for people to draw a general conclusion based on results obtained from different methods under different conditions, for example, batch or flow-through conditions with or without a confining pressure. Nevertheless, the major insights gained from this extensive review include the following:

• In GCS systems, mineral dissolution accounts for most of the reported increases in porosity. Generally, mineral dissolution generates a larger porosity and leads to an increase in permeability while mineral precipitation reduces the porosity and causes reductions in permeability. Due to opening or closure of pore throats and the creation of new flow paths, slight changes in porosity can significantly affect the permeability.
• In UHS systems, the H₂-induced reductive dissolution leads to complex changes in porosity and permeability. There is no clear trend in these changes.
• In CAES systems, the study on the impact of oxidation reactions on porosity and permeability is very limited. Geochemical modeling of a long-term operation reported insignificant changes in porosity and permeability.
• In GCS systems, CO₂-induced mineral dissolution is commonly reported, and leads to reductions in strength, the elastic modulus, and wave velocity. For sandstone samples with low carbonate content or quartz-cemented, even if carbonate dissolution occurs, their mechanical properties can be retained. Salt precipitation can lead to reduction in permeability and an increase in wave velocity.
• In UHS systems, the study of the geochemical alteration of mechanical properties of formation rock is limited. H₂-induced dissolution is anticipated to cause reductions in strength, the elastic modulus, and the effective friction angle. The complex H₂ induced reactions in the UHS systems raise the necessity of investigations into the H₂-induced alteration of mechanical properties.
• In CAES systems, the geochemical alteration of the mechanical properties of rock formations has not been studied yet due to their relatively low reactivity. However, potential reaction-induced local mineral swelling may have a negative impact on caprock integrity and is worth further investigation.
• The review of pilot scale observations in three GCS systems highlight the importance of studying the coupled geochemical–geomechanical impact of these subsurface energy storage systems.

7. Future Research

Although the coupled geochemical–geomechanical impact of GCS systems has been extensively studied, this review found that the relevant research in UHS and CAES systems is insufficient. The following research questions are suggested to fill in the knowledge gaps.

• More studies on the mechanical properties of formation samples under in situ conditions in GCS systems are suggested because ex situ measurements cannot ensure the tested samples are always under reservoir conditions.
• In GCS systems, the simultaneous collection of information about chemical reactions and mechanical properties during interaction between rock and fluid is valuable in studying coupled geochemical–geomechanical processes. Core-flooding experiments combined with wave velocity measurements are a promising way to reach this goal.
• In GCS system, more long-term investigations of coupled geochemical–geomechanical process are needed to simulate the real time scale of GCS.
• In UHS systems, more studies need to be carried out to investigate H₂-induced reductive dissolution with different types of rock because the reactivity of different minerals to H₂ is still not well understood.
• In UHS systems, an investigation into the effect of reactions on the porosity and permeability of rock samples should be conducted at wider temperature and pressure range, and longer periods.
• In UHS systems, the stress state of the reservoir changes frequently due to the injection–extraction operation mode. More studies about the effect of these reactions on mechanical properties of rock samples, which are insufficient now, need to be carried out.
• In CAES systems, more investigations into the reaction between O₂ and pyrite-bearing rock samples are suggested because this will also benefit other gas storage applications using air as cushion gas [159].
• In CAES systems, where CO₂ works as a cushion gas, studies on the impact of oxidation reaction and CO₂-induced reaction on mechanical properties of rock samples are suggested. Even though the CO₂-induced dissolution can be suppressed in this system, the porosity increase at the location close to the injection well was relatively significant [100].