Assessment of the chemical stability of clay materials with different reactive surface areas under varying pH environments: A numerical modeling approach

doi:10.21203/rs.3.rs-5172179/v1

Download PDF

Research Article

Assessment of the chemical stability of clay materials with different reactive surface areas under varying pH environments: A numerical modeling approach

https://doi.org/10.21203/rs.3.rs-5172179/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 06 Oct, 2024

Read the published version in Discover Geoscience →

Version 1

posted

You are reading this latest preprint version

This study examined the fluid-rock interactions to assess the stability of clay minerals under a subsurface heterogeneous reservoir. Three-dimensional reactive transport simulations were run up to 100 years using a grid representing the Morrow B Sandstone of Farnsworth Unit and parts of the adjacent Morrow Shale layers in the western Anadarko basin, Texas. Six model scenarios were tested where the minimum and maximum reactive surface areas of clay minerals were repeated with the systematically altered formation water pH level from neutral (pH = 7) to highly acidic states (pH = 3). A relative increase in total aqueous concentrations of Na⁺ and K⁺ and a lowering of SiO₂ (aq) were predicted under a higher acidic environment. The models predicted a continuous dissolution of illite and secondary precipitation of smectite minerals. In contrast the stability of kaolinite temporally varied depending on the size of the reactive surface areas considered for the simulations. This resulted in the largest increase in the net volumetric abundance of smectite compared to the kaolinite and illite minerals over the model domain. The sensitivity to the variations of the formation water pH level and the interaction effects of pH and surface area for the precipitation and dissolution of clay minerals was found to be relatively insignificant.

Geochemistry

Chemical stability

Clay mineral

Formation water

Reactive surface area

Previous literature extensively studied the impacts of clay mineral stability on deep well disposal, hydrocarbon production, carbon sequestration, construction of infrastructures, and many other Geo-environmental and Engineering applications [1–4]. Illites and kaolinites are pore-bridging and pore-filling clays, respectively. Hence, variations in the formation water pH environment and the mineral reactive surface areas might influence the precipitation and dissolution of those minerals, ultimately affecting the reservoir's hydraulic properties [5]. Qin and Beckingham [6] experimented with the influence of reactive surface areas of various carbonates and silicate minerals on mineral reactions and reaction rates. Koupai et al. [7] reported that the availability of relatively higher concentrations of soluble ions might increase the inter-particle attractive forces at low pH levels. Therefore, the formation water pH of the pore water significantly influences the mechanical properties, such as the shear strength of soils. This study particularly focused on the chemical stability of the clay minerals incorporating the site-specific model input parameters using an integral finite difference method. More specifically, this study sought to answer how the variations in minerals’ reactive surface areas and the formation water pH influence the formation water composition and the dissolution and precipitation of clay minerals in a subsurface heterogeneous media.

A grid representing the Morrow B Sandstone of Farnsworth Unit (FWU) and parts of the overlying Morrow Shale layers, as shown in Fig. 1, were considered for the experiments [8]. The FWU covers an approximately 73 km² area in the western Anadarko basin in northern Texas (Fig. 1a in supplementary material). The lithostratigraphy of the study area is provided in Fig. 1b in supplementary material.

The volume fraction of the reservoir materials consists of 84.26% quartz, 9% albite, and 1.25% carbonates; the rest are kaolinite, illite, chlorite, and smectite minerals [9, 10]. Gallagher [9] studied six samples collected from the coarse-grained sandstone facies, bioturbated mudstone facies, and fine-grained sandstone facies using X-ray Diffraction (XRD) analysis and reported that kaolinite is present as ghost grains or fills the pore spaces of most of the collected samples from the reservoir. These ghost grains are grain-sized accumulations of kaolinites originating from the weathering of feldspar minerals [9]. On the other hand, smectite and illite were present in small amounts in some of the measured samples. Since the later clay varieties could not be identified in optical petrography or backscattered electron (BSE) images, their distribution was unknown [9]. Photomicrographs of the minerals are provided in Fig. 2 in supplementary material. The relative clay abundance (weight percent) of the collected samples based on XRD Analysis from well Id: 13-10A is provided in Table 1 in the supplementary material.

The average thickness of Morrow B Sandstone is 9 m and lies at an average depth of about 2400 m in the subsurface [10, 11]. The hydrostatic pressure varied from 32 to 36.5 MPa, which caused the formation water to flow towards the north-western direction (Fig. 1b above). In the Morrow B formation water, the total dissolved solids (TDS) range from about 3500 to 4000 mg/L, pH of about 7, and Eh of about 140 mV [8]. The average reservoir temperature in Morrow B is about 75°C [12]. Earlier studies showed that the pore water in the Morrow B formation is supersaturated with respect to an assemblage of zeolite, clay, carbonate, mica, and aluminum hydroxide minerals and undersaturated for feldspars, ankerite, and chlorite [12]. All of these site specific datasets were considered to set up the model as mentioned in detail in the following section.

The model grid consisted of 110 cell columns, 80 cell rows, and seven cell layers, of which 15,357 cells in the western portion of the grid were considered for these model simulations (Fig. 1a above). Three-dimensional reactive transport simulations were run for up to 100 years using the TOUGHREACT software incorporating EQ3/ EQ6 thermodynamic database [13]. The numerical experiments were conducted by developing six different model scenarios where the minimum and the maximum reactive surface areas were considered, and the reservoir pH was systematically altered from a neutral to a highly acidic state (pH = 3, 5 and 7).

The average porosity of the Morrow B Sandstone is 14.5% (ranges 10–18%), whereas the average permeability is 48 millidarcies (mD) (ranges 1 to 250 mD) [8]. The spectrum of porosity-permeability covariance in Morrow B consisted of 36 discrete porosity-permeability combinations in the model considering the geologic model of Ampomah et al. [11], as shown in Table 1 below.

Table 1

Porosity (%) permeability (m²) covariance in the Morrow B modeled rock types
Porosity Permeability (m²)	0	> 0–0.1	> 0.1–0.125	> 0.125–0.15	> 0.15–0.2	> 0.2
7.99 x 10^− 17- 1 x 10^− 17	Rock 1	Rock 36
> 1 x 10^− 17 – 2.46 x 10^− 14		Rock 2	Rock 3	Rock 4	Rock 5	Rock 6
> 2.46 x 10^− 14– 4.93 x10^− 14		Rock 7	Rock 8	Rock 9	Rock 10	Rock 11
> 4.93 x10^− 14– 7.4 x 10^− 14		Rock 12	Rock 13	Rock 14	Rock 15	Rock 16
> 7.4 x 10^− 14– 9.87 x 10^− 14		Rock 17	Rock 18	Rock 19	Rock 20	Rock 21
> 9.87 x 10^− 14– 1.23 x 10^− 13		Rock 22	Rock 23	Rock 24	Rock 25	Rock 26
> 1.23 x 10^− 13– 1.48 x 10^− 13		Rock 27	Rock 28	Rock 29	Rock 30	Rock 31
> 1.48 x 10^− 13 – 1.72 x 10^− 13		Rock 32	Rock 33	Rock 34	Rock 35

Source: Ref. [11, 14]

For the model setup, the grid's topmost and bottommost cell layers were considered at a fixed state, and the lateral boundaries were open to fluid flow, heat, and solute transport. The mineral composition of the Morrow B Sandstone was calculated by averaging the values of the individual mineral types published by ref. [9, 10], as shown in Table 2 below.

Table 2

Mineral composition of the Morrow B Sandstone
Mineral	Volume (%)
Quartz	84.26
Albite	9.0
Calcite	0.75
Ankerite	0.25
Siderite	0.25
Smectite	0.1
Illite	0.88
Chlorite	1.79
Kaolinite	2.72

Source: Ref. [9, 10]

Information regarding the mineral radii, reactive surface areas of minerals (cm²/g of mineral), and the kinetic parameters for the acid and neutral mechanisms were compiled from ref. [9, 13, 15] (Table 3). The minimum reactive surface areas for the respective minerals were one magnitude smaller than the maximum values as shown in Table 3 below.

Table 3

Kinetic parameters of the Morrow B Sandstone formation materials used in the simulations
Minerals	Maximum Reactive Surface area (cm²/g of mineral)	Minimum Reactive Surface area (cm²/g of mineral)	Kinetic parameters
			Acid Mechanism			Neutral Mechanism
			H⁺>=	E _a	K ₂₅	E _a	K ₂₅
Quartz	9.8	0.98				87.7	1.023 x 10^− 14
Albite	11.45	1.15	0.457	65.0	1.35 x 10^− 10	69.8	9.12 x 10^− 13
Clinochlore	11.41	1.14	0.5	88.0	7.8 x 10^− 12	88.0	3.0 x 10^− 13
Illite	43.63	4.36	0.34	23.6	1.0 x 10^− 11	35.0	1.7 x 10^− 13
Kaolinite	46.15	4.62	0.777	65.9	4.9 x 10^− 12	62.76	1.01 x 10^− 13
Smectite	9.8	0.98	0.34	23.6	1.05 x 10^− 11	35.0	1.65 x 10^− 13
Calcite	11.07	1.11				23.5	1.55 x 10^− 7
Dolomite	10.49	1.05	0.5	36.1	6.46 x 10^− 4	52.2	2.95 x 10^− 8
Ankerite	9.84	0.98	0.5	36.1	6.46 x 10^− 4	62.76	1.26 x 10^− 9
Siderite	9.8	0.98	0.5	36.1	6.46 x 10^− 4	62.76	1.26 x 10^− 9
Magnesite	9.8	0.98	1.0	14.4	4.2 x 10^− 7	23.5	4.6 x 10^− 10

(Note: Activation Energy (E _a) in Kj/mol; Rate Constant (K ₂₅) in mol/m²/s)

Source: Ref. [9, 13, 15]

The initial formation water temperature was set to 75° C, and the salt mass fraction was 0.004, considering the homogeneous formation water salinity is close to 4000 mg/kg. The initial hydrostatic pressure in the model varied between 32.2 and 36.5 MPa based on the reservoir history matching datasets generated by Ampomah et al. [11]. The following regression equation was derived that explains 78% (R² = 0.78) of the variability of the initial pressure distribution within the Morrow B sandstone reservoir:

$$\:P=6.68\:\:x\:{10}^{8}\:+\:218.4\:\text{X}\:-\:179\:\text{Y}\:-\:8758\:\text{Z}$$

Here, P represents pressure (Pa), and X, Y, and Z represent the spatial coordinates (m).

The composition of Morrow B formation water was collected from Ahmmed et al. [12] (Table 4 below).

Table 4

Present composition of Morrow B Sandstone formation water
Basis species	Concentration (mg/kg)
Basis species	Well battery AWT3	Well battery AWT4
Al³⁺	1.05	0.00791
B(OH)₃(aq)	21.5	27.8
Ba²⁺	0.592	1.37
Br⁻	17.8	20.3
Ca²⁺	38.5	32.9
Cl⁻	1.47×10³	1.84×10³
F⁻	0.654	0.911
Fe²⁺	7.29×10^− 5	2.10×10^− 8
$\:\text{H}\text{C}{\text{O}}_{3}^{-}$	5.79×10⁴	795
K⁺	7.87	7.11
Li⁺	0.283	0.461
Mg²⁺	6.27	12.3
Na⁺	1.08×10³	1.41×10³
SO₄²⁻	21.4	12.9
SeO₃²⁻	0.0687	0.0827
SiO₂(aq)	24.5	42.1
Sr²⁺	2.68	1.85
Zn²⁺	1.51x10^− 5	0.108
pH	7	7
ORP (mV)	140	140

Source: Ahmmed et al. [12]

All of the numerical simulations were conducted using a preconditioned bi-conjugate gradient solver. Capillary pressure and relative permeability were computed according to the van Genuchten-Mualem model [16]. The changes in permeability due to the precipitation and dissolution of minerals as a function of porosity changes were estimated according to a “tube-in-series” model for the Morrow B formation [17, 18]. Aqueous activity coefficients were calculated using the extended Debye-Hückel equation [19]. Finally, analyses of variances (ANOVA) were conducted to assess the main effects of pH and mineral reactive surface areas and their interaction effects on the chemical stability of the kaolinite, illite, and smectite minerals.

Changes in Fluid composition

The models predicted an increase in dissolved Na⁺, K⁺, SiO₂, and AlO₂^- concentrations and a minor decrease of Mg²⁺ up to 10–20 years, and then it reached to an equilibrium state (Fig. 2a-c). Models that were run considering the larger reactive surface area of clay materials resulted in higher total aqueous concentrations of Na⁺, K⁺, and AlO₂^-, while the SiO₂ showed the opposite results. The concentration of Mg²⁺ was reduced in both cases. On the other hand, a relative increase of Na⁺ and K⁺ and lower concentrations of SiO₂ were predicted in the higher acidic environment (pH = 3). In contrast, the differences were minimal for the models considered a slightly acidic to neutral aqueous environment (i.e., pH = 5 to pH = 7). According to Ahmmed et al. [12], quartz is slightly supersaturated, and albite is undersaturated with respect to the initial Morrow B formation water composition. Due to the dissolution of albite and illite minerals, the aqueous concentration of Na⁺, K⁺, and SiO₂ initially increased. However, aqueous Na⁺ concentration gradually became depleted by the precipitation of smectite, dilution by ambient formation water, and reduced albite dissolution. At the lower pH environment, dissolved silica formed primary particles and precipitated out of the solution.

Changes in total volumetric abundances of minerals and reservoir hydraulic properties

Among the three clay varieties examined, kaolinite is the most abundant in all the six samples collected from the coarse and fine-grained sandstone and bioturbated mudstone facies. Illite was the second most abundant, except in one of the four samples collected from the coarse-grained sandstone facies [9].

The models predicted a continuous dissolution of illite and secondary precipitation of smectite up to 100 years of simulations (Fig. 3a & c). In both cases, a relatively higher dissolution and precipitation were noticed for the model scenarios considering the larger reactive surface areas because the coating layer prevents physical contact with the reactive fluid. This result is consistent with the findings by [20]. The secondary precipitations of kaolinite were noticed in the models that considered the least reactive surface areas during the simulation (Fig. 3b). Phukan et al. [21] also reported that the formation water initially supersatured with respect to kaolinite but became undersaturated over time. However, in the case of the largest surface area, minor kaolinite precipitated for up to 10 years and dissolved for the rest of the simulation period. This ultimately led to the highest increase in the volumetric abundance of smectite relative to kaolinite and illite minerals within the reservoir. In the entire model scenarios tested, the sensitivity to the variations of the formation water pH environment for the precipitation and dissolution of the clay minerals were found insignificant.

Among the carbonates, native calcite and ankerite initially dissolved, but later calcite, including Mg-rich calcite and siderite, precipitated in the reservoir. Among the non-carbonates, except the clay minerals, albite dissolved, and a minor amount of silica was predicted to precipitate. Overall, the dissolution of albite and illite increased aqueous silica concentrations, which allowed silica precipitation within the reservoir. Albite is slightly undersaturated with respect to formation water, promoting natural alteration to kaolinite, and the precipitation of smectite minerals at a higher acidic environment. Previous studies carried out by ref. [22–24] reported similar results.

The results from the ANOVA showed that the contributions of the differences in reactive surface areas to the dissolution/ precipitation of illite, kaolinite, and smectite are 46.4%, 63.8%, and 51.5% at 95% confidence intervals. Tukey's multiple comparisons showed neither the pH effect nor the interaction effects of pH and surface area showed any statistical differences on the variations of the clay minerals' stabilities in the reservoir.

Overall, in the model grid, the changes in the reservoir void space volume representing the total porosity were negligible (10^2.8 m³) during the first two decades of the model simulations. This was closely associated with the net decreases of the initial abundances of carbonates and noncarbonate minerals in the Morrow B Sandstone reservoir. The predicted net changes in porosity were not high enough to contribute to any significant changes in permeability and, thus, fluid flow within the reservoir.

Dissolutions of albite and illite resulted in an initial increase of Na⁺ and K⁺ ions, which gradually stabilized due to the dilution by ambient formation water and precipitation of smectite minerals. This ultimately led to a noticeable increase in smectite minerals in the reservoir. Overall, the chemical stability of illite, kaolinite, and smectite is not much influenced by the variations of the pH level due to buffering through the precipitation/ dissolution of the carbonates and non-carbonates and also subsequent changes in the reservoir's physical and chemical environments, include pH level and temperature. In contrast, the variations in the reactive surface areas cause a noticeable change in the magnitude of the reactions and, therefore, affect the stability of clay minerals. Regardless of the differences in the pH level, illite, and kaolinite dissolve more at higher reactive surface areas that tend to open pore spaces and reduce pore bridges. On the other hand, in the model with the minimum reactive surface area, more kaolinite precipitates and fills the pore spaces. Therefore, depending on the site characteristics, clay minerals' solubility may affect the reservoir's porosity and permeability, which is crucial information generally required for many Geo-Environmental applications.

Acknowledgments

The author is grateful to Professor Martin Appold at the University of Missouri-Columbia for providing the secondary datasets and sincerely thanks the anonymous reviewers for their valuable comments, which improved the quality of this manuscript.

Churchman GJ, Singh M, Schapel A, Sarkar B, Bolan N (2020) Clay minerals as the key to the sequestration of carbon in soils. Clays Clay Miner 68:135–143
Elsawy MB, Lakhouit A (2020) A review on the impact of salinity on foundation soil of coastal infrastructures and its implications to north of Red Sea coastal constructions. Arab J Geosci 13:1–12
Gaurina-Međimurec N, Pašić B, Mijić P, Medved I (2020) Deep underground injection of waste from drilling activities—An overview. Minerals 10:303
Radwan AE, Sen S (2021) Characterization of in-situ stresses and its implications for production and reservoir stability in the depleted El Morgan hydrocarbon field, Gulf of Suez Rift Basin, Egypt. J Struct Geol 148:104355
Hu C, Tan J, Lyu Q, Feng G, Shi J, Wang K, Wang Z (2022) Variations of shale's pore characteristics and hydraulic properties after long-term imbibition in hydraulic fracturing fluids, vol 8. Geomechanics and Geophysics for Geo-Energy and Geo-Resources, p 183
Qin F, Beckingham LE (2021) The impact of mineral reactive surface area variation on simulated mineral reactions and reaction rates. Appl Geochem 124:104852
Koupai AJ, Fatahizadeh M, Mosaddeghi MR (2020) Effect of pore water pH on mechanical properties of clay soil. Bull Eng Geol Environ 79:1461–1469
Khan R (2017) Evaluation of the Geologic CO2 Sequestration Potential of the Morrow B Sandstone in the Farnsworth. M.Sc. Thesis, University of Missouri-Columbia, Columbia, MO, USA
Gallagher SR (2014) Depositional and diagenetic controls on reservoir heterogeneity: Upper Morrow Sandstone, Farnsworth Unit, Ochiltree County, Texas. Master’s Thesis, New Mexico Institute of Mining and technology, Socorro, New Mexico. ; 215
Munson TW (1989) Depositional, diagenetic, and production history of the Upper MorrowanBuckhaults Sandstone, Farnsworth Field, Ochiltree County Texas. Oklahoma City Geological Society, Shale shaker. ; 40: 2–20
Ampomah W, Balch RS, Rose-Coss D, Hutton A, Cather M, Will R (2016) An integrated approach for characterizing a sandstone reservoir in the Anadarko basin. Offshore Technology Conference, Houston, TX, USA. ; OTC-26952-MS, 13
Ahmmed B, Appold MS, Fan T, McPherson BJOL, Grigg RB, White MD (2016) Chemical effects of carbon dioxide sequestration in the Upper Morrow Sandstone in the Farnsworth, Texas, hydrocarbon unit. J Environ Geosci 23:81–93
Xu T, Apps JA, Pruess K (2004) Numerical simulation of CO2 disposal by mineral trapping in deep aquifers. Appl Geochem 19:917–936
Ahmmed B (2015) Numerical modeling of CO2-Water-Rock interactions in the Farnsworth, Texas Hydrocarbon unit, USA. Master’s Thesis, Department of Geological Sciences, University of Missouri—Columbia. ; 63
Palandri JL, Kharaka YK (2004) A Compilation of Rate Parameters of Water-Mineral Interaction Kinetics for Application to Geochemical Modeling. U.S. Geological Survey, Reston, VA, USA
Pruess K, Oldenburg C, Moridis G (1999) TOUGH2 user’s guide, version 2.0. Lawrence Berkeley National Laboratory, Berkeley, CA. Rep LBNL – 43134
Dai Z, Middleton R, Viswanathan H, Rahn JF, Bauman J, Pawar R, Lee SY, McPherson B (2013) An Integrated Framework for Optimizing CO2 Sequestration and Enhanced Oil Recovery. Environ Sci Technol Lett 1:49–54
Pruess K (2005) ECO2N: A TOUGH2 fluid property module for mixtures of water, NaCl, and CO2: Lawrence Berkeley National Laboratory. Paper: LBNL – 57592
Xu T, Apps JA, Pruess K (2005) Mineral sequestration of carbon dioxide in a sandstone – shale system. Chem Geol 217:295–318
Deng H, Poonoosamy J, Molins S (2022) A reactive transport modeling perspective on the dynamics of interface-coupled dissolution-precipitation. Appl Geochem 137:105207
Phukan M, Vu HP, Haese RR (2021) Mineral dissolution and precipitation reactions and their net balance controlled by mineral surface area: an experimental study on the interactions between continental flood basalts and CO2-saturated water at 80 bars and 60° C. Chem Geol 559:119909
Li S, He H, Liang X, Tao Q, Ji S, Yang M, Zhu J (2022) Transformation of Ordered Albite into Kaolinite: Implication for the Booklet Morphology. ACS Earth Space Chem 6:1133–1142
Pearce JK, Khan C, Golding SD, Rudolph V, Underschultz JR (2022) Geological storage of CO2 and acid gases dissolved at surface in production water. J Petrol Sci Eng 210:110052
Xiong Z, Cao Y, Liang C, Liu K, Wang G, Zhu R, Wang Y (2022) Origin and significance of authigenic quartz and albite in lacustrine calcareous fine-grained sedimentary rocks. Mar Pet Geol 143:105799

The authors declare no competing interests.

Supplementarymaterials.docx

Download PDF

Journal Publication

published 06 Oct, 2024

Read the published version in Discover Geoscience →

Version 1

posted

You are reading this latest preprint version

Assessment of the chemical stability of clay materials with different reactive surface areas under varying pH environments: A numerical modeling approach

Status:

Journal Publication

Version 1

Abstract

Figures

I. Introduction

II. Description of the experiments

III. Results and discussions

IV. Conclusion

Declarations

Acknowledgments

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1