Laboratory Experiments and Geochemical Modeling of Gas–Water–Rock Interactions for a CO₂ Storage Pilot Project in a Carbonate Reservoir in the Czech Republic

Abstract

The aim of this study was to characterize the influence of CO₂ in geological structures on mineralogical changes in rocks and assess the sequestration capacity in mineral form and solution of a potential pilot storage site in the Czech Republic. Rock samples from a dolomite reservoir and the overburden level, as well as the corresponding pore water, were used. The most important chemical process occurring in the reservoir rock is the dissolution of carbonate minerals and feldspars during the injection of CO₂ into the structure, which increases the porosity of the structure by approximately 0.25 percentage points and affects the sequestration capacity of the reservoir rock. According to the results of geochemical modeling, the secondary carbonate minerals (dolomite, siderite, and ephemeral dawsonite) were present only during the first 50 years of storage, and the porosity at this stage decreased by 1.20 pp. In the caprocks, the decomposition of K-feldspar and calcite resulted in an increase in porosity by 0.15 percentage points at the injection stage only, while no changes in porosity were noted during storage. This suggests that their insulation efficiency can be maintained during the injection and post-injection periods. However, further experimental research is needed to support this observation. The results of this study indicate that the analyzed formation has a low potential for CO₂ sequestration in mineral form and solution over 10,000 years of storage, amounting to 5.50 kg CO₂/m³ for reservoir rocks (4.37 kg CO₂/m³ in mineral form and 1.13 kg CO₂/m³ in dissolved form) and 3.22 kg CO₂/m³ for caprock rocks (3.01 kg CO₂/m³ in mineral form and 0.21 kg CO₂/m³ in dissolved form). These values are lower than in the case of the depleted Brodské oil field, which is a porous reservoir located in the Moravian part of the Vienna Basin.

1. Introduction

One of the current major environmental and climatic issues is the need to reduce atmospheric CO₂ concentrations. One method to achieve this aim is carbon capture and storage (CCS), which represents a technology based on CO₂ injection and its safe underground storage. In 2023, the global average atmospheric carbon dioxide was 419.3 ppm, which set a new record high. The increase between 2022 and 2023 was 2.8 ppm, which marked the 12th year in a row where the amount of carbon dioxide in the atmosphere increased by more than 2 ppm [1]. Deep geological storage in porous rock formations is considered the most appropriate strategy for CO₂ sequestration [2,3,4] and the injectivity of storage rocks is a key technical and economic issue for CCS projects [5]. The viability of CO₂ injection depends mainly on the porosity and permeability of the storage rocks. Further consequences of the CO₂ interactions with host rock such as dissolution–precipitation of minerals are also important [6,7].

Geological storage of CO₂ is possible through several physicochemical mechanisms, one of which is mineral trapping. These processes evolve over time after CO₂ injection and are dominated by structural, stratigraphic, or hydrodynamic capture early in the project. These mechanisms are primarily governed by the following physical processes: fluid flow in the liquid and gas phase under pressure and gravity, capillary pressure effects, and heat flow via conduction, convection, and diffusion. The transport of aqueous and gaseous substances via advection and molecular diffusion is considered in both the liquid and gaseous phases. After the CO₂ injection is complete, a number of capture mechanisms become important. CO₂ can be partially trapped through residual capture as the plume moves away from the wellbore. The gas also mixes with and dissolves in the formation water at the leading and trailing edges of the plume (solubility trapping). The dissociation of CO₂ dissolved in the formation water creates acidity that reacts with the minerals in the formation and can dissolve fast-reacting carbonate minerals (if present) in the acidified zone surrounding the injection well, leading to an increase in dissolved bicarbonate (called entrapment).

Theoretical, modeling, and experimental studies have been carried out to investigate CO₂–water interactions from rock formations under deep storage conditions. Most of these experimental studies were designed to simulate the injection of CO₂ mixed with formation water into the rocks at p–T conditions of a deep storage environment. The results of many of these experiments indicate an increase in the porosity/permeability of the storage rock caused by the partial dissolution of the carbonate minerals (mainly calcite) [4,8,9,10,11]. On the other hand, another set of experiments showed that porosity decreases due to initial carbonate dissolution followed by secondary precipitation/mineralization with porosity changes [10,12,13,14,15,16].

The essence of the research activity described in this article was to carry out model studies and laboratory experiments for the implementation of a pilot CO₂ storage project—the first in Central and Eastern Europe—in a mature, depleting, hydrocarbon reservoir in the southeastern part of the Czechia.

The Žarošice field was discovered in 2001. The field is currently considered “mature”, with a declining production trend, and consideration is therefore given to further use of the geological structure, which forms the current hydrocarbon field, to implement a pilot project of geological storage of carbon dioxide.

Research methods included data collection and quality control from core samples, water samples taken from the deposit, laboratory tests and experiments, computer modeling, and numerical synthesis of results. In the experimental stage, a high-pressure reactor was used to test the interactions of reservoir and caprocks from selected core samples with formation water and CO₂. The samples cover key types of fluid and rock chemical compositions. Geochemical modeling was performed using The Geochemist’s Workbench 11 software, in particular, the React module. The results of the reaction modeling were compared with experimental data from the tests in the reaction chamber. Comparing the results of both procedures is important to ensure the use of the most appropriate methods for predicting the evolution of the environment of a potential CO₂ storage site.

2. Methodologies and the Initial State Study

Models involving kinetic transport through porous media and thermodynamic issues of multiphase systems are very useful in forecasting the impact of CO₂ storage on the changes in the rock environment at the storage site. These models are based on information on the petrophysical, mineralogical, and hydrogeological properties of porous media, hydrochemical analysis of fluid composition, pressure and temperature values, and kinetic parameters.

2.1. Geological Setting

The area of interest, representing the surroundings of the structure referred to as “Zar-3” (Žarošice hydrocarbon field), covering an area of 94.5 km², is located in the southeastern part of the Czech Republic. The wider surroundings of the Žarošice oil and gas field include the contact zone between the Western European foreland plate (Western European Platform) and the Western Carpathian thrust belt in the territory of southern Moravia (south-eastern part of the Czech Republic) [17].

This work is focused on the study of rocks whose geological development in the studied locality began in the Mesozoic period.

The Mesozoic to Cenozoic Alpine cycle in southern Moravia began with continental rifting in the Early Jurassic and deposition of synrift detrital deposits, followed by marine transgression and deposition of mixed carbonate and siliciclastic rocks of the passive margins. The carbonate platforms occupied the shallow marginal zone and the uplifted rift blocks, whereas the organic-rich marly facies (Mikulov marls) formed in a deeper basinal environment. During the Late Cretaceous to early Paleogene, the European foreland in Moravia was uplifted and deeply eroded. Rivers cut the deep Nesvačilka and Vranovice valleys that turned into submarine canyons more than 1500 m deep, which were gradually filled with the Paleogene detrital organic-rich deposits [18,19,20].

In the early Late Cretaceous, the Carpathian passive margins were converted into a foreland-type synorogenic basin marked by deep-water turbiditic flysch sedimentation, followed by a late orogenic and post-orogenic shallow marine and continental molasse sedimentation. During the late phases of the Alpine orogeny, in the late Paleogene and early Miocene, the flysch sequences were deformed and thrust over the European foreland, including the inner zone of the Neogene foredeep [17]. The Carpathian thrust belt in Moravia is composed of Late Jurassic to early Miocene synorogenic sequences of the Flysch belt (Figure 1). The Neogene foredeep, except for the innermost zones adjacent to the front of the thrust belt, remained undeformed. Superimposed on the thrust belt is the Neogene Vienna Basin.

Žarosice field is a buried Jurassic patch reef, which was subsequently dolomitized and faulted and then partially eroded from the west. The seal is provided by two different caprocks, Jurassic Marl of Mikulov Fm. in the eastern part of the field and Paleogene claystone in the western and northern parts described as Nesvačilka Fm. The reservoir rock mainly comprises heavily fractured dolomite of the Vranovice Fm. and, in the southern part near the base, dolomitic sandstones of the Nikolčice Fm. Beds (Figure 1), both of Jurassic age. The primary porosity of the dolomite is around 1–3%. Much more important is the secondary porosity, which is based on log data reaching 2–9% (locally, even more). The permeability of the reservoirs is 193–630 mD, which was estimated based on hydrodynamic tests [22].

The most promising exploration targets proved to be the basal sandstones and the carbonates of the Jurassic, occurring as isolated erosional relicts on the northern slope of the Nesvačilka graben. These relicts (buried hills) are sealed mainly by impermeable shales and mudstones of the paleovalley fill. The isolated bodies of turbiditic and channelized sandstones enclosed within the Paleogene fill of the Nesvačilka paleovalley represent the secondary exploration play.

Geochemical studies [23,24,25] suggest that the hydrocarbons in the Nesvačilka graben area were predominantly sourced from Jurassic, organic-rich Mikulov marls. The Paleogene shales and mudstones may be considered as additional source rocks, especially for gas.

2.2. Hydrogeological, Mineralogical, and Reservoir Characteristics of the Storage Site

Data and outputs from laboratory experiments carried out in the laboratories of the Department of Geological Engineering at VSB—Technical University of Ostrava and data received from partners within the project were used in this study.

The reservoir rocks represented by dolomite from the Vranovice fm. (labeled ZA4A) were the focus of this research (

Table 1. Basic data for the samples of reservoir rocks.

Borehole	Core	Box	True Vertical Depth (m)	Stratigraphy	Lithostratigraphy	Lithology
ZA4A	3	3	1528.5–1537.5	Upper Jurassic	Vranovice fm.	Light gray dolomite
ZA4A	4	2	1571.0–1578.8	Upper Jurassic	Vranovice fm.	Light gray dolomite

). The Vranovice fm. lies at a depth of about 1500 m and is formed by an Upper Jurassic light gray dolomite about 100 million years old.

The caprock is represented by the Nesvačilka fm. rock sample (ZA4) (

Table 2. Basic data for the sample of caprock.

Borehole	Core	Box	True Vertical Depth (m)	Stratigraphy	Lithostratigraphy	Lithology
ZA4	1	5	1569.0–1577.0	Middle Paleogene	Nesvačilka fm.	Dark gray shaly siltstone

). The Nesvačilka fm. lies at a depth of approximately 1500 m and is formed by dark gray shale mudstone from the Middle Paleogene.

The composition of the pore water of the reservoir formation was determined based on the analysis of a sample from the Vranovice formation from the Žarošice 4A well (

Table 3. Physicochemical parameters of formation water from the Žarošice 4A well.

Formation Water Žarošice 4A
Ca2+	111.0	mg/L
Mg2+	107.0	mg/L
Na+	8570.0	mg/L
K+	252.0	mg/L
Cl−	12,300.0	mg/L
NH4+	32.6	mg/L
Br−	54.0	mg/L
I−	43.0	mg/L
HCO3−	2580.0	mg/L
SO42−	535.0	mg/L
Li+	3.2	mg/L
Sr2+	19.2	mg/L
Mn2+	0.0	mg/L
Fetotal	0.4	mg/L
B3+	43.0	mg/L
TDS	24,590.0	mg/L
Other parameters
T	53	°C
pH	7.1	-
Depth	1530	Meter

TDS—total dissolved substances; T—temperature.

). The water sample from the aquifer was taken by the project partner (MND a.s.) from a depth of 1530 m and chemical analysis of the formation water was carried out in the laboratories of the VSB—Technical University of Ostrava.

To model changes in the rock environment, it was also necessary to take into account the CO₂ phase, which becomes supercritical under the conditions of the bearing horizon: temperature 53 °C; pressure 150 bar (determined on site by the project partner MND a.s.). The mineralogical composition and porosity of the studied rock samples are presented in

Table 4. Mineralogical composition and porosity of the rock samples.

Borehole	Dol	Ank	Q	Calc	Mu	Ka	Mic	Alb	Chl	Pyr	Porosity
Reservoir ZA4A	(wt%)										(%)
	55.04	17.95	4.01	0.00	2.88	3.12	7.66	7.92	1.42	-	-
	(vol. %)										(%)
	50.60	16.56	3.68	0.00	2.76	3.23	7.36	7.39	1.12	-	8.00
Caprock ZA4	(wt%)										(%)
	-	18.10	57.10	0.30	4.00	-	16.68	1.18	1.14	1.50	-
	(vol. %)										(%)
	-	17.25	54.42	0.29	3.81	-	15.90	1.12	1.09	1.43	4.70

Dol—dolomite, Ank—ankerite, Q—quartz, Calc—calcite, Mu—muscovite, Ka—kaolinite, Mic—microcline, Alb—albite, Chl—chlorite, and Pyr—pyrite.

.

The hydrogeological and reservoir parameters of the Vranovice formation (

Table 5. Values of CO₂ density and viscosity and hydrogeological and reservoir parameters of light gray dolomite reservoir rock (measured by the project partner MND a.s).

Parameters	Value	Unit
Thickness	200	m
Porosity	8.00	%
Horizontal permeability, kR	500	mD
Vertical permeability, kz	1600	mD
Pressure P	116	bar
Temperature T	53	°C
Maxsalinity S	72	g/L
CO2 density	195	kg/m3
CO2 viscosity	1.5 × 10−4	Pa.s

) are important for understanding and predicting changes in the storage area. During the extraction of hydrocarbons from the studied deposit, it was found that the vertical permeability, kz, is greater than the horizontal permeability, kR, which results from geological processes occurring after the diagenesis of the dolomite body. The studied reservoir rock shows typical fractured permeability.

2.3. Methods

The methods selected to investigate the impact of CO₂ injection on changes in the rock environment at the studied site are based on information on petrophysical, mineralogical, and hydrogeological characteristics of porous media, hydrochemical analysis of the liquid composition, pressure and temperature values of the deposit, and kinetic parameters.

2.3.1. XRD Analysis and Porosity Determination

The mineralogical composition of rock samples was determined using XRD analysis by means of a Bruker AXS D8 θ-θ powder diffractometer. The experiments were conducted at room temperature in the Bragg–Brentan parafocusing geometry using CoK_α wavelength (λ = 1.7903 Å, U = 34 kV, I = 30 mA). The porosity of the reservoir and caprock formations was initially assessed by the project partner (MND a.s.), based on acoustic well logging. However, for the purposes of this work, it was precisely determined using mercury porosimetry (Autopore 9220 Micro-metrics Injection Porosimeter, Micromeritics Instrument Corporation, Norcross, GA, USA) on cylindrical samples cut from drill cores. The samples from the ZA4A and ZA4 wells were analyzed before the experiments in the reaction chamber and after the subsequent experimental intervals.

2.3.2. Experiments in the Reaction Chamber

The experiments in the reaction chamber were conducted in two phases at the VSB—Technical University of Ostrava. The first phase involved only reservoir rock experiments and the second phase involved only caprock experiments. The experiments were carried out in the reaction chamber in rock-formation water–gas systems at temperatures below 53 °C and a pressure of 150 bars. The experiments lasted 30, 90, 120, and 180 days. During reservoir rock and caprock testing, the tested rocks had their own separate cells for a defined time interval. The formation water from the Žarošice 4A well was used as the reaction solution (see Table 3).

The reaction chamber is a cylindrical vessel welded from a shell made of stainless steel and DN150 PN160 flanges with an outer diameter of 169 mm and a length of 955 mm. The lids (blind flanges DN150 PN160) are also made of stainless steel. The apparatus is divided into two parts with two reaction cells. The inner diameter of the reaction cells is 141 mm and the length of the cells is 368 mm. The insides of the cells have a protective coating and the cells are further lined. The lining is inert to CO₂. On the side of the casing, two pins are welded for attaching the pressure vessel in the bearings, which enables the positioning and rocking movement provided by the rocking mechanism.

The highest operating pressure is 200 bar. The range of operating temperatures is 5–80 °C. The heating for the reaction chamber is provided by an external heating coil. The operating medium can be reservoir water, hydrocarbons, CO₂, or nitrogen. The dynamic conditions of the geohydrodynamic system are simulated by changing the position of the reaction chamber with two swings per day (this simulation method was used when solving the issue of CO₂ geosequestration). Changes in position are secured using a swinging mechanism driven by an electric motor.

2.3.3. Geochemical Modeling

The geochemical modeling in this study focused on gas–water–rock interactions rather than on tracking the evolution of changes in CO₂ capture in individual zones of the storage site. For this purpose, kinetic modeling was performed in order to determine mineralogical transformations and identify chemical reactions occurring in the gas–water–rock systems, changes in porosity, and to estimate the mineral and solution sequestration capacity. This approach also allowed for a partial comparison of the results of batch experiments with the results of reaction kinetics modeling.

Simulations were performed using Geochemist Workbench 11 (GWB) [26] and the thermodynamic database “thermo.dat”. The software used did not provide dynamic modeling in which it would be possible to track the evolution of changes in CO₂ capture in individual zones of the storage site. Equilibrium models were aimed at assessing changes in the chemical composition of pore water in conditions of chemical equilibrium in the rock–water–CO₂ system.

Kinetic modeling enabled the evaluation of hydrogeochemical changes in the studied formation in two stages: the first stage reproduced the immediate changes in the aquifer and caprock impacted by CO₂ injection, and the second stage reproduced the long-term effects of sequestration.

This method further allowed for the assessment of the volume and amount of precipitated or dissolved mineral phases and their impact on the porosity of rocks in relation to the prediction of possible CO₂-related hazards.

The geochemical modeling was performed assuming the temperature and formation pressure relevant to the depth of the modeled environments, which are given in Table 5. The appropriate value of carbon dioxide fugacity f_CO2 75 bar was calculated based on the measured well pressure of 150 bar using the model in [27]. For the purpose of the simulation, the chemical compositions of the formation water in the reservoir and the caprock were obtained via equilibration of the formation water (Table 3) using the mineral assemblage typical for the modeled environments (Table 4).

The following kinetic dissolution/precipitation rate equation derived from the transition state theory [28] was used in simplified form in the calculations:

$${\mathrm{r}}_{\mathrm{k}}={\mathrm{A}}_{\mathrm{S}}{\mathrm{k}}_{\mathrm{T}}\left(1-\frac{\mathrm{Q}}{\mathrm{K}}\right)$$

(1)

where r_k is the reaction rate [mol·s⁻¹] (dissolution at r_k > 0, precipitation at r_k < 0), A_S is the mineral’s surface area [cm²], k_T is the rate constant [mol·cm⁻²·s⁻¹] at temperature T, Q is the activity product [-], and K is the equilibrium reaction for the dissolution reaction [-].

According to the above equation, a given mineral precipitates when it is supersaturated or dissolves when it is undersaturated at a rate proportional to its rate constant and the surface area. The Arrhenius law expresses the dependence of the rate constant k_T on temperature T:

$${\mathrm{k}}_{\mathrm{T}}={\mathrm{k}}_{25}\exp \left[\frac{-{\mathrm{E}}_{\mathrm{A}}}{\mathrm{R}}\left(\frac{1}{\mathrm{T}}-\frac{1}{\mathrm{298,15}}\right)\right]$$

(2)

where k₂₅ is the rate constant at 25 °C [mol·m⁻²·s⁻¹], E_A is the activation energy [J·mol⁻¹], R is the gas constant (8.3143 J·K⁻¹·mol⁻¹), and T is the absolute temperature (K).

The kinetic rate constants for the minerals involved in modeled reactions (

Table 6. Kinetic rate constants and specific surface areas applied in the models.

Mineral	Kinetic Rate k25 (mol/cm2·s−1)	Specific Surface (cm2/g)
Dolomite	7.760 × 10−10	161.1
Ankerite	7.760 × 10−10	146.6
Quartz	1.023 × 10−18	113.3
Calcite	3.311 ×10−8	170.3
Muscovite	1.413 × 10−16	212.0
Kaolinite	4.898 × 10−16	1156.0
K-feldspar	8.710 × 10−15	180.5
Albite	6.918 × 10−15	229.0
Clinochlore	7.762 × 10−16	1118.0
Pyrite	3.020 × 10−12	149.7

) were taken from the authors of [29]. The values of the specific surface areas calculated according to the spherical model in [30] are presented in Table 6.

3. Results and Discussion

Immediately after injection, CO₂ diffuses in its supercritical form and partially dissolves into the groundwater. This leads to acidification of the primary pore water, causes changes in the mineralogical composition of the reservoir and caprock, and may affect the well cement. All these reactions may pose potential risks in the CO₂ injection process in the reservoir structure and may lead to CO₂ leakage from the storage to the surface. Therefore, it is necessary to understand the mechanisms of these reactions and quantify and evaluate them to ensure the safety and sustainability of CO₂ storage.

3.1. Equilibrium Model

The reaction of CO₂ with groundwater and its effects on potential storage is of utmost importance for the process of mineral sequestration because only the dissolved form of CO₂ can react with the rock. CO₂ solubility is a function of the temperature, pressure, and salinity of the solution, with CO₂ solubility decreasing with increases in the temperature and molarity of the solution, while CO₂ solubility increases with increasing pressure. The dissolution simulation was carried out until the thermodynamic equilibrium was reached and the solution was saturated with CO₂. Simulations were performed using Geochemist Workbench 11 (GWB) [26] and the thermodynamic database “thermo.dat”.

Based on the input parameters used in the model (brackish water Žarošice 4A, Table 3), we report the theoretical acidification of brackish water Žarošice 4A as a result of CO₂ injection and calculated chemistry changes (

Table 7. Physicochemical parameters of original formation water from the Žarošice 4A well and model-calculated acidification of formation water Žarošice 4A.

	Original Formation Water Žarošice 4A	Acidified Formation Water Žarošice 4A
Ca2+	111.0	111.58	mg/L
Mg2+	107.0	108.03	mg/L
Na+	8570.0	8565.23	mg/L
K+	252.0	257.45	mg/L
Cl−	12,300.0	12,314.67	mg/L
NH4+	32.6	32.64	mg/L
Br−	54.0	54.15	mg/L
I−	43.0	42.76	mg/L
HCO3−/CO2(aq)	2580.0	11,584.62	mg/L
SO42−	535.0	528.89	mg/L
Li+	3.2	3.24	mg/L
Sr2+	19.2	18.91	mg/L
Mn2+	0.0	1.45	mg/L
Fe2+	0.4	0.28	mg/L
B3+	43.0	41.20	mg/L
TDS	24,590.0	24,590.29	mg/L
Other parameters
T	53	53	°C
pH	7.1	4.2	-
Ι	0.28	0.25	mol/L

Log P_CO2 = 2.176.

). The HCO₃⁻ ion is the dominant carbonate component in the solution at alkaline pH. When the pH drops to acidic values, the bicarbonate ions decrease in solution and carbonic acid, and CO_2(aq) becomes the dominant carbonate component [31].

The dissolution of CO₂ in the solution results in a decrease in pH and an enrichment of carbonate ions according to the following reactions:

CO_2(g) ↔ CO_2(aq)

(3)

CO_2(aq) + H₂O ↔ H₂CO₃

(4)

H₂CO₃ ↔ HCO₃⁻ + H⁺

(5)

HCO₃⁻ ↔ CO₃²⁻ + H⁺

(6)

After the reaction of brackish water Žarošice 4A with spCO₂, there was an increase in CO_2(aq) to a value of 11,584.62 mg/L as a result of the dissolution of CO_2(g) according to Equation (3) under the reservoir conditions. This process acidified the water, lowering the pH to a value of 4.2.

We observed a slight decrease in the concentration of sodium and sulfate ions; on the contrary, the concentration of chlorides increased slightly. During the reaction with spCO₂, Na⁺ cations reacted with sulfates and bicarbonates to form NaSO₄⁻ and NaHCO₃ complexes. The most important change during CO₂ injection into the pore environment of the collector saturated with formation water was an increase in the content of dissolved CO_2(aq) in the solution to a value of 11,584.62 mg/L and a decrease in pH to a value of 4.2. Therefore, the concentration of bicarbonates, which participated in the buffering of the system during CO₂ injection, dropped significantly.

3.2. Experiments in the Reaction Chamber

The mineralogical composition of all rock samples after the experimental intervals was determined using XRD analysis. Although data on mineral composition evolution derived from the XRD method have certain limitations, their presentation in this work shows additional possibilities for predicting mineral changes in the host rock as a result of CO₂ injection.

After the end of the experiments, the reaction solutions were analyzed in the laboratories of the Institute of Environmental Engineering at VSB—TUO in Ostrava (Czech Republic). The chemical composition of the formation water is shown in

Table 8. Chemical analysis of reaction solutions from experiments with reservoir and caprocks in the reaction chamber.

Parameter		pH	ρ at 15 °C	Cond.	SO4	Cl	HCO3	NH4	K	Mg	Na	Ca	Li	Sr	Mn	Fe	B
Units		(-)	(g/cm3)	(mS/cm)	(mg/L)
Input—MND analysis		7.1	1.0	36.5	535.0	12,300.0	2580.0	32.6	252.0	107.0	8570.0	111.0	3.2	19.2	0.0	0.4	-
Borehole	Days of experiment
Reservoir ZA4A	30	6.7	1.0	36.9	640.0	13,250.6	2702.3	9.1	257.0	99.6	4570	285.0	2.8	17.0	0.9	0.01	45.8
	90	6.9	1.0	35.3	660.0	12,727.6	3440.4	15.8	256.0	100.0	6260	315.0	2.9	13.6	1.1	94.9	43.0
	120	7.2	1.0	37.2	550.0	13,046.7	3225.7	19.4	463.0	94.8	8750	348.0	2.8	16.3	0.7	18.9	40.8
	180	6.8	1.0	38.0	540.0	12,444.0	3660.0	19.1	334.0	92.7	6770	702.0	2.7	13.4	0.4	60.8	41.6
Caprock ZA4	30	7.1	1.0	37.3	380.0	25,455.3	2840.2	10.8	279.0	103.0	8740	250.0	2.9	15.2	0.2	3.5	47.8
	90	7.7	1.0	37.9	500.0	13,188.5	2852.4	2.7	229.0	87.2	8100	253.0	2.8	18.5	0.7	25.0	47.4
	120	6.9	1.0	35.4	550.0	12,266.7	2976.8	14.8	289.0	95.2	5680	274.0	-	0.1	0.7	40.5	38.8
	180	6.8	1.0	35.6	600.0	11,664.0	3042.7	14.7	215.0	93.6	5380	266.0	-	0.1	0.4	23.5	39.6

.

The following graphs (Figure 2) show the mineralogical changes in the rock samples that occurred during the experiments in the reaction chamber.

Mineralogical changes in the reservoir rock over 30 days are shown in Figure 2a. As shown in the figure, the dolomite content decreased. The formation water after 30 days of experiments in the reaction chamber was depleted in magnesium but enriched with calcium and carbonate ions. A decrease in magnesium in the solution indicates the precipitation of ankerite. This is also confirmed by the decrease in iron concentration. Calcite precipitates because the dissolution of dolomite releases calcium and bicarbonate ions. Albite and muscovite remained stable throughout the experiment. Sodium concentration dropped significantly after 30 days of CO₂ injection, with no indication of chemical binding of the sodium ion in another mineral phase. The concentration of potassium in the solution increased slightly, although no precipitation or dissolution of muscovite was shown in the XRD analysis results. However, there was a slight decrease in the content of microcline, which also releases potassium. Quartz and kaolinite slightly increased, and minerals from the chlorite group were completely dissolved. Due to the dissolution of chlorite, the concentration of manganese increased in the solution.

Mineralogical changes in the reservoir rock from the injection of CO₂ over 90 days are shown in Figure 2b. Dolomite, kaolinite, and microcline dissolved only slightly, while chlorite was completely dissolved. This effect led to an increase in the concentrations of calcium, magnesium, iron, manganese, and carbonate ions. The potassium concentration increased only very slightly. Calcite precipitated due to the dissolution of dolomite and ankerite, and also due to the dissolution of minerals from the chlorite group. The content of albite increased after 90 days of the experiment and, simultaneously, the concentration of sodium in the solution decreased. Muscovite remained stable throughout the experiment.

Mineralogical changes in the reservoir rock over 120 days are shown in Figure 2c. An important phenomenon observed after 120 days of the experiment was the increase in dolomite content. This is related to the dissolution of ankerite during this experiment. Calcium, magnesium, iron, and carbonate ions are released into the solution. This also indicates the precipitation of calcite, although in a much lower content than in the previous experiments. Quartz, kaolinite, and albite dissolved (the latter is evidenced by an increased sodium concentration in the solution). Muscovite and microcline precipitated, but the potassium concentration increased significantly. This may indicate the existence of another mineral as a source of potassium, which was not revealed in the XRD analysis results, e.g., potassium feldspar or minerals from the potassium salts group (sylvite and carnallite). Chlorite was completely dissolved.

Mineralogical changes in the reservoir rock over 180 days are shown in Figure 2d. Only dolomite and muscovite were dissolved in this case. Calcium, magnesium, potassium, and carbonate ions were released into the solution. Dissolution of dolomite allows the precipitation of calcite and ankerite. In this experiment, chlorite also precipitated, so iron was consumed from the solution. Hence, another mineral (e.g., accessory pyrite) can be considered as the source of iron, which is needed for ankerite precipitation. Quartz, kaolinite, microcline, and albite slightly increased. The concentration of potassium in the solution after 180 days of the experiment increased due to muscovite dissolution, unlike sodium, which decreased due to the precipitation of albite. A slight increase in pH also contributed to the precipitation of carbonates.

Figure 3 presents the mineralogical changes in the ZA4 caprock samples that occurred during the experiments.

Mineralogical changes in the cap rock over 30 days are shown in Figure 3a. The content of CaFe–dolomite (ankerite) decreased. After 30 days of experiments, the formation water in the reaction chamber was depleted in magnesium but enriched with calcium and carbonate ions. The dissolution of ankerite is confirmed by the increase in the concentration of iron in solution. Some iron was consumed from the reservoir water solution during the precipitation of pyrite, whose concentration increased slightly after 30 days of the experiment. Albite remained stable throughout the experiment, but sodium concentration very slightly increased after 30 days of CO₂ injection, which indicates another source of sodium ions in some mineral phases. The concentration of potassium in the solution increased slightly, but the precipitation of muscovite was demonstrated by the XRD analysis results. There was a slight increase in the content of microcline, which also fixes potassium. The slightly higher potassium concentration after 30 days of experiment in the reaction chamber indicates an initial rapid dissolution of microcline and muscovite, enrichment of the solution with K⁺ ion, and subsequent secondary precipitation of muscovite. The concentration of manganese increased in the solution and minerals from the chlorite group were precipitated. The amount of manganese in solution indicates another source of this element in the caprock sample. Quartz increased slightly.

Mineralogical changes in the cap rock from the injection of CO₂ over 90 days are shown in Figure 3b. CaFe–dolomite and microcline dissolved. Quartz, muscovite, and pyrite were precipitated. These processes led to an increase in the calcium, iron, manganese, and carbonate ions concentrations in the solution. The potassium and magnesium concentrations decreased only very slightly. Calcite precipitated due to the dissolution of ankerite. Albite and minerals from the chlorite group remained stable throughout the experiment, but sodium concentration decreased. This indicates the fixation of sodium ions into other mineral phases, such as muscovite.

Mineralogical changes in the reservoir rock over 120 days are shown in Figure 3c. The decrease in CaFe–dolomite content was lower compared with experiments lasting 30 days. Calcium and carbonate ions were released into the solution, but secondary calcite was not detected in the case of this sample. The solution did not have suitable thermodynamic conditions for its precipitation. The concentration of iron ions was the biggest in this phase of the experiment due to the ankerite dissolution. Some iron was consumed from the solution during the precipitation of pyrite and minerals from the chlorite group, whose concentration increased slightly after 120 days of the experiment. Albite, muscovite, chlorite, and microcline precipitated very slightly. This was reflected by increased potassium concentrations in the solution, but this did not correlate with the very low sodium concentrations. Quartz was dissolved.

Mineralogical changes in the cap rock over 180 days are shown in Figure 3d. Only CaFe–dolomite was dissolved in this case. Calcium and carbonate ions were released into the solution again. The concentration of potassium, sodium, and magnesium in the solution after 180 days of the experiment decreased by precipitation of most of the defined mineral phases.

Secondary precipitation of most defined mineral phases is a favorable phenomenon in cap rock. It leads to a decrease in their permeability. However, precipitation of secondary phases is not a positive phenomenon in reservoir rock where decreasing permeability reduces the sequestration capacity of the reservoir rock.

3.3. Results of and Discussion on Kinetic Modeling

The kinetic modeling of mineralogical changes in caprock and reservoir rock was carried out by means of the GWB React package in two stages. The first stage, performed in sliding fugacity mode, represents short-term changes in the system under the influence of CO₂ injection over a period of 100 days, during which the gas fugacity rises in the system to the assumed value of f_CO2 75 bar. The second stage reproduces the long-term effects of sequestration over a period of 10,000 years.

3.3.1. Stage 1—Caprock Sample ZA4

At the first stage of the CO₂ injection, lasting for 100 days, gas fugacity (f_CO2) increased to 75 bar (corresponding to a reservoir pressure of 150 bar). In effect, the CO₂ concentration in solution, CO_2(aq), increased from 0.004 mol/kg to 1.384 mol/kg, and HCO₃⁻ increased from 0.034 mol/kg to 0.094 mol/kg. Due to the dissolution of primary minerals, the porosity increased slightly by 0.15 percentage points, from 4.79 to 4.94. At the same time, the pH dropped from 7.1 to 4.9 due to the hydrochemical balance between the species in solution (Figure 4).

The mechanisms of mineral sequestration of CO₂ are activated mainly as a result of Reaction (7), which involves the decomposition of primary K-feldspar, in which dawsonite is formed. The dissolution of primary calcite (Equation (8)) leads to an increase in bicarbonate and calcium concentrations in the pore solution (Figure 5).

Na⁺ + KAlSi₃O₈+ CO_2(g) + H₂O = NaAlCO₃(OH)₂ + K⁺ + 3SiO₂

(7)

CaCO₃ + CO_2(g) + H₂O = Ca⁺⁺ + 2HCO₃⁻

(8)

3.3.2. Stage 2—Caprock Sample ZA4

In this stage, gas fugacity (f_CO2) dropped to 0.2 bar, the CO₂ concentration in solution (CO_2(aq)) dropped to about 0.0003 mol/kg, and HCO₃⁻ decreased to 0.011 mol/kg. The porosity decreased slightly to 4.92. At the same time, the pH increased above the primary value to 7.7 (Figure 6).

Mineral sequestration of CO₂ is mainly associated with the formation of dolomite, which precipitates due to the decomposition of ankerite: Equation (9). The formation of dawsonite (Equation (10)) plays a minor role in only the first 20 years of the storage period (Figure 7 and Figure 8). It should be noted, however, that in reality, the processes of dissolution of ankerite and crystallization of dolomite will proceed at a similar, but not identical, rate. For modeling purposes, the rate constant for ankerite was assigned based on dolomite and for this reason, the formation of secondary minerals occurs effectively under conditions close to local equilibrium.

CaMg_0.3Fe_0.7(CO₃)₂ + 2.24CO_2(g) + 0.14Mg₅Al₂Si₃O₁₀(OH)₈ + 0.56H₂O = 2.24HCO₃⁻ + CaMg(CO₃)₂ + 0.7Fe²⁺ + 0.28Al³⁺ + 0.42SiO_2(aq)

(9)

CaCO₃ + CO_2(g) + 0.4HCO₃⁻ + 0.2Mg₅Al₂Si₃O₁₀(OH)₈ + 0.4Na⁺ = CaMg(CO₃)₂ + 0.6H₂O + 0.4NaAlCO₃(OH)₂ + 0.6SiO_2(aq)

(10)

3.3.3. Stage 1—Reservoir Rock Sample ZA4A

During stage 1, gas fugacity (f_CO2) increased to 75 bar, the CO₂ concentration in solution (CO_2(aq)) changed from 0.004 mol/kg to 1.396 mol/kg, and HCO₃⁻ increased from 0.033 mol/kg to 0.062 mol/kg. The porosity increased by 0.0023, from 8.15 to 8.38. At the same time, the pH dropped from 7.1 to 4.8 (Figure 9).

During the injection stage, changes in the mineral composition (Figure 10) follow Equation (7) of the K-feldspar dissolution, in which dawsonite is produced. Ankerite dissolution is also important, causing siderite and dolomite to precipitate (Equation (11)):

CaMg_0.3Fe_0.7(CO₃)₂ + 0.7CO_2(g) + 0.7H₂O = 0.7Ca²⁺ + 0.7FeCO₃ + 1.4HCO₃⁻ + 0.3CaMg(CO₃)₂

(11)

3.3.4. Stage 2—Reservoir Rock Sample ZA4A

During stage 2, gas fugacity (f_CO2) dropped to 0.06 bar, the CO₂ concentration in solution (CO_2(aq)) decreased to 0.0011 mol/kg, and HCO₃⁻ decreased to 0.013 mol/kg. The porosity decreased to 7.23. At the same time, the pH increased above the primary value to 7.2 (Figure 11).

During the storage stage (Figure 12), the mineral sequestration of CO₂ takes place in three steps in which the ankerite dissolution plays the main role:

(1)

In the first step, dawsonite formation occurs due to the dissolution of albite (Equation (12)) and the precipitation of dolomite and siderite (Equation (11)).

NaAlSi₃O₈ + CO_2(g) + H₂O = NaAlCO₃(OH)₂ + 3SiO₂

(12)

(2)

In the second step, dawsonite is dissolved and dolomite and siderite are precipitated (Equation (13)).

CaMg_0.3Fe_0.7(CO₃)₂ + 1.54CO_2(g) + 0.14Mg₅Al₂Si₃O₁₀(OH)₈ = 0.84HCO₃⁻ + CaMg(CO₃)₂ + 0.7FeCO₃ + 0.14H₂O + 0.28Al³⁺ + 0.42SiO_2(aq)

(13)

(3)

Dolomite precipitates due to ankerite dissolution in the third step (Equation (14)).

CaMg_0.3Fe_0.7(CO₃)₂ + 1.53CO_2(g) + 0.14Mg₅Al₂Si₃O₁₀(OH)₈ + 0.55H₂O + 0.29NaAlSi₃O₈ + 0.17O_2(aq) =
1.53HCO₃⁻ + CaMg(CO₃)₂ + 0.35Mg₀.₁₆₅Fe₂Al_0.33Si_3.67O₁₀(OH)₂ + 0.45Al³⁺ + 0.17Na⁺

(14)

3.3.5. Sequestration Capacity in Mineral Forms and Solution

The use of the reaction path model (GWB React) only allowed for the calculation of mineral and solution trapping; therefore, no assessment of the hydrodynamic trapping capacity was made. However, the mechanism of mineral trapping can be considered the most important because it ensures permanent trapping of CO₂.

The estimated mineral and solution sequestration capacities refer to closed systems in which, after CO₂ fugacity reaches the level assumed for the end of the injection stage, only the transformation of CO₂ aqueous phases into the mineral phases can occur. This type of approach allows for an approximate assessment of the size and mutual proportions of the effectiveness of the two trapping processes analyzed.

Based on kinetic modeling, changes in the volume of primary and secondary minerals during simulated reactions were calculated, which demonstrated an effect on the porosity of the analyzed formations. In addition, the amounts of dissolved or precipitated carbonates were determined and, based on their balance, the amount of sequestered CO₂ was calculated for each cubic meter of formation.

The chemical composition of pore water, especially the concentrations of aqueous species (e.g., HCO₃⁻, CO_2(aq), CO₃²⁻, and NaHCO₃) obtained based on simulations, enabled an assessment of the amount of carbon dioxide trapped in the solution. This assessment was calculated based on the volume of pore space in a 1 m³ rock formation multiplied by the amount of carbon dioxide contained in each component. During the injection stage lasting 100 days, it was determined that CO₂ trapping in solution was more effective than the mineral trapping mechanism (

Table 9. Calculated sequestration capacity.

Sample	Sequestration Capacity (kg/m3)
		t = 100 d	t = 10,000 y
ZA4 caprock	1. Mineral	0.18	3.01
	2. In solution	3.04	0.21
	sum (1 + 2)	3.22	3.22
ZA4A reservoir	1. Mineral	0.22	4.37
	2. In solution	5.28	1.13
	sum (1 + 2)	5.50	5.50

).

Both analyzed formations are characterized by a low total CO₂ sequestration potential in mineral form and solution for 10,000 years of storage, amounting to 3.22 and 5.50 kg CO₂/m³ for the ZA4 caprock and ZA4A reservoir rocks, respectively. These values are lower than the ones obtained in simulations for other considered storage formations [32,33,34,35].

The sequestration capacity values calculated in our research are also low compared with the results from the depleted Brodské oil field (Middle Badenian), which is also located in the Moravian part of the Vienna Basin. Capacities calculated for the latter location were 13.22 kg CO₂/m³ for the reservoir and 5.07 kg CO₂/m³ for the caprock, respectively [36].

During 100 days of injection, solution trapping predominated, while in the storage stage, mineral trapping showed a significantly greater potential (Table 9). The minerals responsible for mineral sequestration in each of the rocks are shown in

Table 10. Minerals responsible for mineral sequestration of CO₂.

Sample	Stage I		Stage II
	Dissolution	Precipitation	Dissolution	Precipitation
ZA4 caprock	Calcite	Dawsonite	Ankerite Calcite	Dawsonite Dolomite
ZA4A reservoir	Ankerite	Dawsonite Siderite dolomite	Ankerite	Siderite dolomite

.

4. Summary and Conclusions

Experiments and geochemical modeling of gas–water–rock interactions were carried out in order to characterize the impact of CO₂ on mineralogical changes in reservoir rocks and caprock and assess the sequestration capacity of a potential storage site in Czechia. The results of experimental research and geochemical modeling led to the following conclusions.

At the CO₂ injection stage, reservoir rocks with a significant share of dolomite and ankerite and a small percentage of feldspars showed a slight increase in porosity by 0.25 pp. due to the decomposition of ankerite and feldspar. The secondary carbonate minerals of the storage stage observed in this study were dolomite, siderite, and ephemeral dawsonite, which were present in only the first 50 years of storage; porosity decreased by 1.20 pp at this stage.

In the caprocks, with a high proportion of quartz, microcline, and ankerite, the decomposition of K-feldspar and calcite was responsible for the increase in porosity by 0.15 pp at the injection stage. Therefore, it can be concluded that the increase in porosity of the caprock is of little importance and corresponds to only 3% of the initial value, which has practically no effect on the sealing properties. During the storage stage, the decomposition of ankerite and calcite resulted in the precipitation of secondary dolomite; moreover, mineral changes had almost no effect on porosity.

The reservoir formation and caprocks have a low total CO₂ sequestration potential in mineral form and solution over 10,000 years of storage, amounting to 5.50 kg CO₂/m³ for reservoir formation (4.37 kg CO₂/m³ in mineral form and 1.13 kg CO₂/m³ in dissolved form) and 3.22 kg CO₂/m³ for caprock (3.01 kg CO₂/m³ in mineral form and 0.21 kg CO₂/m³ in dissolved form), respectively. These values may not be encouraging, especially when compared with the results of the depleted Brodské oil field, which is also located in the Moravian part of the Vienna Basin.

During the 100-day injection period, solution trapping dominates, while during the storage stage, mineral trapping has a greater potential.

The ability to store CO₂ in the aquifer depends on many parameters, such as the composition of the rock matrix, porosity, permeability, pressure, and salinity. These parameters have a significant impact on sequestration processes. Therefore, it should be noted that most of these parameters evolve over time. Moreover, in the model used in this study, it was not possible to take into account these changes, e.g., the specific surface of minerals or changes in reservoir temperature due to the gas injection.