Zeta potential of CO₂-rich aqueous solutions in contact with intact sandstone sample at temperatures of 23 °C and 40 °C and pressures up to 10.0 MPa

Abstract

Despite the broad range of interest and applications, controls on the electric surface charge and the zeta potential of silica in contact with aqueous solutions saturated with dissolved CO₂ at conditions relevant to natural systems, remains unreported. There have been no published zeta potential measurements conducted in such systems at equilibrium, hence the effect of composition, pH, temperature and pressure remains unknown.

We describe a novel methodology developed for the streaming potential measurements under these conditions, and report zeta potential values for the first time obtained with Fontainebleau sandstone core sample saturated with carbonated NaCl, Na₂SO₄, CaCl₂ and MgCl₂ solutions under equilibrium conditions at pressures up to 10 MPa and temperatures up to 40 °C.

The results demonstrate that pH of solutions is the only control on the zeta potential, while temperature, CO₂ pressure and salt type affect pH values. We report three empirical relationships that describe the pH dependence of the zeta potential for: i) dead (partial CO₂ pressure of 10^-3.44 atm) NaCl/Na₂SO₄, ii) dead CaCl₂/MgCl₂ solutions, and iii) for all live (fully saturated with dissolved CO₂) solutions. The proposed new relationships provide essential insights into interfacial electrochemical properties of silica-water systems at conditions relevant to CO₂ geological storage.

Introduction

Quartz is a common mineral comprising 12% of the Earth’s crust [1]. Quartz is also the constituent mineral of sandstone formations, and it can be found in many subsurface settings including aquifers (e.g., [2], [3]), hydrocarbon reservoirs (e.g., [4], [5]) and geothermal sources (e.g., [6], [7]). To characterize the subsurface flows in such settings, a variety of electrical geophysical methods are available including electrical resistivity tomography (e.g., [8], [9]), electro-seismic (injecting electric current and measuring the resulting seismic energy; e.g., [10], [11]), seismo-electric (generating a seismic wave and measuring the resulting electric field; e.g., [11], [12]) and self-potential (SP) (voltage that arises in response to existing gradients in pressure, concentration or temperature; e.g., [13], [14]) measurement. The SP has been shown to be an efficient method to characterize single- and multi-phase flows in the subsurface, especially in sandstone reservoirs (e.g., [15], [16]). Moreover, the SP method can characterize permeability heterogeneities (e.g., fractures, faults, and variable permeability zones [17], [18]).

The SP method relies on electrochemical processes that arise in response to the establishment of an electrical double layer (EDL) at the rock-water interface; this can be characterized by the zeta potential (e.g., [19], [20]). The zeta potential also plays an important role in determining the wettability (e.g., [21]); while the wetting state controls the pore occupancy of aqueous solutions (hereafter referred to as water for simplicity) and non-aqueous phase fluids (NAPF) in multi-phase systems, and thus strongly influences fluid saturations and flow patterns, e.g. in CO₂ geological storage (CGS) [22], hydrocarbon recovery [23], or H₂ geo-storage [24].

There are three principal forces (namely van der Waals, structural and electrostatic forces [25]) that act between rock-water and NAPF-water interfaces; these forces determine the disjoining pressure, which in turn controls the wetting state. Structural forces are always repulsive, thus implying a positive (repulsive) contribution to the disjoining pressure [25], [26], while van der Waals forces depend on properties of all constituent phases (refractive index, dielectric constant and absorption frequency), and these forces can be characterized by the Hamaker theory, resulting in either positive or negative [25], [27], [28] to the disjoining pressure. Electrostatic forces can also be positive or negative [25], [29] depending on rock mineralogy, water pH, ionic strength and chemical composition. The magnitude and polarity of the electrostatic forces depend on the interfacial zeta potentials, which can vary substantially [26], therefore these forces play a key role in controlling the wettability.

In order to accurately characterize the wettability, the measured experimental data of zeta potential of rock-water and NAPF-water interfaces is essential. There are two common methods available for measuring zeta potential; namely the electrophoretic mobility and streaming potential. The electrophoretic mobility method (EPM) relies on the motion of the dispersed phase (either rock or NAPF) relative to the continuous stationary water phase under the influence of an applied electric field [30]. In contrast, the streaming potential method (SPM) is based on the flow of water through a stationary porous medium, which may also contain NAPF, under the influence of a pressure gradient [20], [30]. The benefits of using EPM include a relative ease of use commercially available instruments. However, the measurement conditions are far from representative of deep subsurface settings for several reasons. Firstly, EPM cannot currently be used under high pressure and elevated temperature conditions, or with high ionic strength electrolytes (>1M; M = mol∙L^-1), the conditions that are typical for deep rock formations [26]. Secondly, EPM requires either a powdered mineral sample or emulsified NAPF dispersed in water and therefore, it cannot capture the true complex pore space topology [26]. Finally, EPM cannot take into account a third phase, which is needed for multi-phase flow (e.g., water and gas) [31]. In contrast, SPM can be used on intact sandstone samples (e.g., [32], [33]), at elevated temperature (e.g., [16]), using low-to-high salinity electrolytes of simple and complex composition (e.g., [34]) and also on multi-phase systems containing water, NAPF and minerals at the same time (e.g., [35], [36]). However, conducting SPM experiments is challenging and time consuming. In addition, to the best of our knowledge, thus far there has been no published study that reported either EPM or SPM zeta potential measurements under high pore pressure and elevated temperature conditions, typical for deep subsurface settings. Acquisition of the high pore pressure and elevated temperature data is particularly important as gas under these conditions (e.g., CO₂ during carbon dioxide sequestration or CO₂ injection for improved oil recovery) dissolves in water to a higher degree and alters the ionic composition, reduces pH, so that the resulting aqueous solution becomes the so-called carbonated water (C_water). Such changes in water chemistry will have an impact on the C_water-rock and C_water-NAPF zeta potentials and will ultimately affect the wettability and dynamics of flow of each fluid. Note, that the term C_water used in this study corresponds to any aqueous solution with non-zero concentration of dissolved CO₂.

Several attempts have been made to measure the zeta potential in CO₂ containing systems. A recent study published by Kim and Kwak [37] reported the zeta potential of CO₂-water interfaces using EPM. The experiments were conducted by bubbling CO₂ gas through 0.01 M NaCl solution. The zeta potential was reported to be negative, but the experiments were conducted at atmospheric pressure and unreported temperature. Another study by Moore et al. [38] reported measurements of the zeta potential using SPM in Berea sandstone samples saturated with tap water and liquid CO₂. The experiments were conducted at a maximum pressure of 6.5 MPa and temperature of 20 °C, so that the latter value is not consistent with the expected temperature of 38 °C normally found at the depth that corresponds to 6.5 MPa [39]. The single-phase zeta potential was measured in rock sample fully saturated with water, which was not carbonated prior to the experiments, i.e., the amount of dissolved CO₂ corresponded to the atmospheric level. The experiment was repeated with water and immiscible liquid CO₂ and the effective (i.e., multi-phase) zeta potential was found to be negative and approximately ten-fold smaller in magnitude compared with the single-phase zeta potential. However, Moore et al. [38] did not report single-phase zeta potential measurements conducted with C_water under the same experimental conditions, hence the contribution of the zeta potential at the interface between water and immiscible liquid CO₂ could not be quantified. Moreover, Moore et al. [38] did not report the equilibrium pH of water during the experiments, to indicate whether chemical equilibrium between the mineral, water and liquid CO₂ was established. Since pH is known to have a strong effect on the silica-water zeta potential [16], [40], uncertainty exists in relation to Moore et al.’s [38] reported multi-phase zeta potential results. Furthermore, to the best of our knowledge, no experimental zeta potential data for C_water-rock or C_ water-immiscible CO₂ interfaces under high pressure and elevated temperature conditions has been reported (which are typical for deep subsurface formations). Note that in CGS, CO₂ is stored below a depth of 800 m, so that the CO₂ exists in a dense supercritical phase [41], [42] which correspond to the critical point of CO₂ is 7.38 MPa and 31.1 °C).

In the absence of such measured zeta potential data, several models have been proposed with which the wettability of sandstones can be predicted. For instance, Tokunaga [43] and Kim et al. [44] reported an analytical model of water film stability, based on DLVO (Derjaguin, Landau, Verwey, Overbeek) theory; the model was used to simulate CO₂ geological storage (CGS) conditions in sandstone reservoirs. The model required knowledge of the electrostatic interactions between silica-water and CO₂-water interfaces, to calculate the corresponding contribution to the disjoining pressure, and the model was implemented using compression approximation [45]. Tokunaga [43] and Kim et al. [44] assumed that the zeta potential of the silica-water interface was −25 mV for 0.01 M ionic strength solution, and −5 mV for 2 M ionic strength. Both, Tokunaga [43] and Kim et al. [44] assumed a zero zeta potential at the CO₂-water interface. However, neither of the assumed values was validated due to a lack of experimental data under true CGS conditions. Moreover, when the CO₂ dissolves in water at high pressure, and the pH of C_water becomes substantially lower [46], [47], the zeta potential of C_water-silica interfaces should become vanishingly small [48]. This, however, is inconsistent with the assumed values by Tokunaga [43] and Kim et al. [44], thus their wettability estimates are also doubtful.

Therefore, the main aim of this study is to develop an experimental methodology and for the first time measure the streaming potentials in intact sandstone samples under high pressure and elevated temperature, using C_water, to improve our understanding of the electrochemical processes that take place at silica-water interfaces. The outcomes of this study will, among other applications, better inform CGS, hydrocarbon recovery and geothermal projects. This work also provides fundamental petrophysical data essential for a broad range of Earth sciences.