Effect of Cation Chloride Concentration on the Dissolution Rates of Basaltic Glass and Labradorite: Application to Subsurface Carbon Storage

Abstract

The steady-state dissolution rates of basaltic glass and labradorite were measured in the presence of 10 to 700 × 10⁻³ mol·kg⁻¹ aqueous NaCl, KCl, CaCl₂, and MgCl₂ at 25 °C. All rates were measured in mixed flow reactors, and at pH~3.6 by the addition of HCl to the reactive fluids. The steady-state basaltic glass dissolution rates, based on Si release, increased by ~0.3 log units in the presence of 10⁻³ mol·kg⁻¹ of either CaCl₂ or MgCl₂ compared to their rates in 10⁻³ mol·kg⁻¹ of NaCl or KCl. In contrast, the steady-state dissolution rates of labradorite decreased by ~0.4 log units in the presence of 10⁻³ mol·kg⁻¹ of either CaCl₂ or MgCl₂ compared to their rates in 10⁻³ mol·kg⁻¹ of NaCl or KCl. These contrasting behaviours likely reflect the varying effects of these cations on the stability of rate controlling Si-rich activated complexes on the surface of the dissolving solids. On average, the Si release rates of these solids are similar to each other and increase slightly with increasing ionic strength. As the pH of water charged with 10 to 30 bars CO₂ is ~3.6, the results of this study indicate that both basaltic glass and labradorite dissolution will likely be effective at increasing the pH and adding Ca to the aqueous phase in saline fluids. This observation supports potential efforts to store carbon through its mineralization in saline aquifers containing Ca-bearing feldspar and in submarine basalts.

1. Introduction

This study is focused on quantifying the effect of the ionic strength and the concentration of some common aqueous chloride salts on the dissolution rates of basaltic glass and Ca-rich plagioclase. The dissolution rates of these solids have received significant recent attention due to their potential use as feedstock for carbon mineral storage in the subsurface [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27]. Notably, there have been a large number of studies reporting the measured dissolution rates of basaltic glass (e.g., [17,28,29,30,31,32,33,34,35,36,37], as well as of Ca-rich feldspar labradorite [38,39,40,41,42,43,44,45,46,47,48,49]. These past studies have explored, in detail, the effects of temperature, the presence of aqueous organic species, pH, the distance from equilibrium, and crystallographic orientation on measured dissolution rates. The present study builds on these past efforts by measuring the dissolution rates of basaltic glass and labradorite as a function of reactive aqueous concentrations of NaCl, KCl, CaCl₂, and MgCl₂.

The motivation for this study is to more accurately quantify the effect of major cations on the dissolution kinetics of basaltic glass and Ca-rich plagioclase to understand the potential applicability of saline fluids in carbon mineralization efforts. Notably, carbon dioxide storage in the subsurface is increasingly being targeted for injection into saline reservoirs to preserve freshwater resources (e.g., [50,51,52]). Some of these efforts are aimed at injecting CO₂ into reactive basaltic or ultramafic rocks containing saline fluids or seawater to enhance the long-term security of carbon storage (e.g., [53,54,55]. Towards the improved quantification of these efforts, the dissolution rates of basaltic glass and labradorite have been determined at steady-state, far-from-equilibrium conditions in flow through reactors at a pH of 3.6 and 25 °C and in the presence of aqueous metal chloride solutions with ionic strength values of up to 700 × 10⁻³ mol·kg⁻¹. The purpose of this study is to report on the results of these measurements and to use them to assess the effect of water salinity on subsurface mineral carbonation.

2. Theoretical Background

The standard state adopted in this study is that of the unit activity of pure minerals and H₂O at any temperature and pressure. For aqueous species other than H₂O, the standard state is the unit activity of the species in a hypothetical one molal solution referenced to infinite dilution at any temperature and pressure. All thermodynamic calculations reported in this study were performed using the geochemical modelling code, PHREEQC3 [56], together with the carbfix.dat database [57].

The dissolution rates of basaltic glass and labradorite are controlled via the detachment of metals from the surface of these solids [58]. Within the context of the Transition State Theory, surface reaction-controlled dissolution rates, r, can be considered to be the difference between the forward rate (r₊) and the reverse rate (r₋), such that

$$\mathit{r}=r_{+}-r_{-}=r_{+}\left(1-\frac{r_{-}}{r_{+}}\right)$$

(1)

Taking the law of detailed balancing into account, it can be shown that Equation (1) is equivalent to [59,60,61,62,63]

$$\mathit{r}=r_{+}(1-\mathit{exp}(-A/\sigma RT\left)\right)$$

(2)

where A refers to the chemical affinity of the reaction, σ stands for Temkin’s average stoichiometric number equal to the ratio of the rate of destruction of the activated or precursor complex relative to the overall rate, R designates the gas constant, and T denotes the absolute temperature. The experimental evidence suggests that the value of σ in Equation (2) is three for both basaltic glass and labradorite [63,64]. The form of Equation (1) is such that overall rates (r) are equal to forward rates (r₊) when A >> σRT. All dissolution rates in the present study were measured under far-from-equilibrium conditions, such that A >> σRT. At these conditions $r_{-}<<r_{+}$, and thus, r ≈ r₊.

According to the Transition State Theory applied to surface-controlled dissolution reactions, the forward rate, r₊, is proportional to the concentration of a rate-controlling surface complex, such that [65,66]:

$$r_{+}=k_{+}\left[\mathsf{\Theta }\right]$$

(3)

where k₊ refers to a rate constant, and [$\mathsf{\Theta }$] denotes the concentration of the rate-controlling surface complex. Previous work on basaltic glass and plagioclases with less than ~75% Ca in their cation site (<An₇₅) suggests that their far-from-equilibrium dissolution rates are proportional to the concentration of a Si-rich surface complex. This surface complex is formed by an Al-proton exchange reaction [32,45,63,64,66,67]. This observation leads to an equation describing basaltic glass and labradorite forward dissolution rates of the following form:

$$r_{+}=k_{+}^{\prime }s{\left(\frac{a_{{\mathrm{H}}^{+}}^3}{a_{{\mathrm{Al}}^{3+}}}\right)}^n$$

(4)

where $k_{+}^{\prime }$ represents a constant rate, s refers to the specific surface area of the mineral, and n denotes an exponential factor equal to ~0.33. Monovalent and divalent cations, including Na, K, Mg, and Ca, also need to be removed from the structure of basaltic glass and labradorite to create a dissolution rate-controlling Si-rich surface complex. The degree to which these cations influence the dissolution rate of these solids, however, has yet to be quantified. The effect of these cations on the dissolution rate of these solids is explored in detail in the present study.

The rates measured in the present study are normalized to the geometric surface areas of the basaltic glass and labradorite samples. These surface areas, A_geo, were calculated using [68]:

$${\mathrm{A}}_{geo}=\frac{6}{\rho ·d_{eff}}$$

(5)

where $\rho$ refers to the density of the solid, and d_eff refers to the effective particle diameter. This latter term, d_eff, was calculated using:

$$d_{eff}=\frac{d_{max}-d_{min}}{\ln \left(\frac{d_{max}}{d_{min}}\right)}$$

(6)

where d_max and d_min correspond to the minimum and maximum grain sizes of the prepared powders, respectively.

3. Materials and Methods

3.1. Solids

Basaltic glass used in this study was originally collected from Stapafell Mountain in SW Iceland. This basaltic glass has been previously used in a large number of different experimental studies (e.g., [17,20,32,36,69,70,71,72,73,74,75]). The chemical composition of this glass, as determined via X-ray fluorescence spectrometry, is shown in

Table 1. Chemical composition and surface areas of the basaltic glass and labradorite used in this study.

Solid	Chemical Composition	Ageo (m2/g)
	Normalized to 1 silicon atom
Basaltic glass	Si1.000Al0.365Fe0.191Mn0.003Mg0.294Ca0.263Na0.081K0.008Ti0.025P0.004O3.405	0.0251
Labradorite	Si1.000Al0.684Ca0.286Na0.139K0.002Fe0.008O3.107	0.0284
	Normalized to 8 oxygen atoms
Basaltic glass	Si2.281Al0.833Fe0.436Mn0.007Mg0.669Ca0.6 Na0.184K0.019Ti0.058 P0.008O8	0.0251
Labradorite	Si2.359Al1.612Ca0.674Na0.327K0.006Fe0.018O8	0.0284

. Backscattering images obtained using a scanning electron microscope confirm that while some minor microcrystalline phases are present in this glass, they are confined within glass shards [34].

The labradorite sample used in this study was collected from an anorthosite intrusion located on the Hrappsey Islands in Breidafjördur, western Iceland. This intrusion has been described in detail by Kirstmannsdottir [76]. Its chemical composition, provided in Table 1, was determined using standard wavelength dispersive techniques using the JEOL Superprobe JSL 8200 electron microprobe located at the GET/CNRS in Toulouse, France. Analyses were performed using an acceleration voltage of 15 kV, a beam current of 15 nA, and a beam diameter of 2 µm.

Solids were first dried at room temperature for several days before being crushed in plastic bags using a plastic hammer. Solids were then further ground with an agate mortar and dry sieved to obtain the 45–125 µm size fraction. This size fraction was gravitationally settled to remove fine particles, and subsequently ultrasonically cleaned five times in acetone. The resulting powder was oven dried overnight at 60 °C. The geometric surface areas of these powders were calculated using Equations (5) and (6), taking account of the densities of basaltic glass and labradorite of 3.05 and 2.70 g cm⁻³, which were equal to 251 and 284 cm² g⁻¹, respectively.

3.2. Reactive Fluids

Reactive inlet fluids were created by dissolving Merck/Sigma-Aldrich analytical-grade NaCl, KCl, CaCl₂·2H₂O, or MgCl₂·6H₂O in deionized Millipore™ water with ionic strengths ranging from 0.01 to 2.1 mol·kg⁻¹. The concentration of these chloride-bearing inlet solutions was selected to cover a broad range of ionic strengths. A sufficient amount of reagent grade HCl was added to each fluid to adjust the pH to 3.6. This reactive fluid pH was selected so that it would have a similar pH to that of the likely CO₂-charged injection waters and to avoid the precipitation of most secondary phases. The pH of pure water in equilibrium with 25 bars of CO₂ pressure at 25 °C is 3.2 [12].

3.3. Experimental Design

Basaltic glass and labradorite dissolution experiments were performed in mixed-flow reactors. The reactor system was similar to those used in past dissolution rate studies as illustrated in Figure 1 (c.f. [36,46]). This system consisted of 300 mL acid-washed high-density polyethylene reactors maintained at a constant temperature in a thermostat-controlled water bath. The temperature was kept constant during the experiment at 25 °C. All the reactor components were made of polyethylene to avoid corrosion. All reactors, connectors, and tubing were cleaned with 0.1 mol·kg⁻¹ HCl solution for 24 h and rinsed with Millipore™ water prior to each experiment. All outlet fluid sample bottles went through the same cleaning procedure prior to sampling to prevent contamination. These reactors were stirred using floating magnetic stir bars located on the bottom of the reactors. Stirring bars were rotated with a magnetic stirring motor located underneath the water bath. Basaltic glass and labradorite dissolution experiments were initiated by placing 4 g of cleaned powder and the selected initial reactive fluid into the reactor. The reactors were then sealed, and a Masterflex™ cartridge pump delivered the inlet fluid into the polyethylene mixed-flow reactors at a rate of 0.85–1.0 g·min⁻¹. The reactive inlet fluids were stored in 12 L compressible plastic bags during the experiments. Inlet solutions of selected concentrations of NaCl, KCl, CaCl₂, and MgCl₂ were sequentially pumped into the reactor. The reactor fluid passed through a 10 µm filter when it exited the reactor and was further filtered through a 0.2 µm cellulose acetate filter prior to chemical analysis. The outlet fluid pH was measured at 25 °C using a Eutech Instruments© pH meter coupled with a Eutech Instruments© electrode with 3 mol·kg⁻¹ KCl outer filling solution. The electrode was calibrated with NBS standards at pH values of 4.01 and 7, with an average error less than 0.03 pH units. Fluid samples were collected and acidified with 0.5% concentrated supra-pure HNO₃ prior to analysis. Si concentrations of all inlet and outlet fluids were measured with inductively coupled argon plasma using a Spectro Vision optical emission spectrometer (ICP-OES,). Analytical uncertainties on ICP-OES analyses were estimated to be 3%–5% based on repeated analyses. As was the case for a number of previous mineral dissolution rate studies (e.g., [70,77,78], the experiments in this study were performed in several distinct series on individual solid powders. At the start of each experimental series, the initial reactive fluid was injected into the rector at a constant flow rate. The reactors were then operated for at least 48 h before the first outlet fluid sample was collected. Subsequent fluid sampling was timed to allow at least 3 residence periods to pass between each sampling. The residence time is defined as the reactor volume divided by the fluid flow rate, and it had a range of 5–6 h. A steady state was assumed when four consecutive rate determinations of Si outlet concentrations were constant, with analytical uncertainty. A steady state was usually attained after 7–8 days. After reaching a steady state with the initial NaCl-bearing fluid, the inlet fluid was replaced with KCl of the same cation concentration, and the procedure was repeated. For the attainment of this second steady state, inlet fluid was then replaced with CaCl₂-bearing inlet fluid of a similar concentration, resulting in a third steady state over time. Finally, CaCl₂ inlet fluid was replaced with MgCl₂-bearing inlet solution of a similar concentration until a fourth, and final, steady state was attained. In total, each solid was dissolved in the presence of four different cation-chloride salts during each series in the order of NaCl, KCl, CaCl₂, and MgCl₂. Using several distinct salts in each series facilitated the identification of distinct effects of each of them on the measured rates.

The basaltic glass and labradorite dissolution rates reported in this present study are based on Si release. Silicon release rates are commonly used as a measure of the dissolution rate of silicate minerals, as this metal is critical for maintaining their structure (cf. [58,63,79,80]). These rates (r_Si,_i) were calculated from steady-state Si concentrations in the outlet fluids of the reactors using:

r_Si,i = (F C_Si)/(ν_i Ageo,_i m_i)

(7)

where F represents the fluid flow rate, C_Si stands for the concentration of Si in the outlet fluid at a steady state, ν_i signifies the stoichiometric number of Si in one mole of the solid (1 for basaltic glass and 2.36 for labradorite), Ageo,_i denotes the specific geometric surface area of ith solid, and m_i represents the initial mass of the ith solid in the reactor.

4. Results

A representative variation in the measured outlet fluid Si concentration as a function of time is shown in Figure 2. The measured Si concentrations of all samples collected during this study is provided in Table S1 of the electronic supplement. Si release rates rapidly attain a near constant value in all experiments during each reactive series, although some scatter is evident. Steady-state Si concentrations for each experiment are reported in

Table 2. Summary of measured steady-state dissolution rates generated in the present study. All experiments were performed under atmospheric P_CO2. All inlet fluid also contained sufficient HCl to attain a pH of 3.6.

Experiment ID	Inlet Fluid Composition 1	Ionic Strength (mol·kg−1)	Exp. Duration (h)	pH In	pH Out	(Si) Out (mol × 105)	Log r+si Geo 1
G-NaCl-10	10 × 10−3 mol·kg−1 NaCl	0.01	210	3.59	3.67	0.95	−8.85
G-KCl-10	10 × 10−3 mol·kg−1 KCl	0.01	176	3.57	3.60	0.81	−8.92
G-CaCl2-10	10 × 10−3 mol·kg−1 CaCl2·2H2O	0.03	193	3.55	3.61	1.77	−8.58
G-MgCl2-10	10 × 10−3 mol·kg−1 MgCl2·6H2O	0.03	209	3.64	3.67	2.00	−8.52
G-NaCl-100	100 × 10−3 mol·kg−1 NaCl	0.1	210	3.86	3.80	1.06	−8.80
G-KCl-100	100 × 10−3 mol·kg−1 KCl	0.1	176	3.65	3.66	0.71	−8.97
G-CaCl2-100	100 × 10−3 mol·kg−1 CaCl2·2H2O	0.3	193	3.53	3.66	1.75	−8.59
G-MgCl2-100	100 × 10−3 mol·kg−1 MgCl2·6H2O	0.3	209	3.59	3.64	1.37	−8.69
G-NaCl-700	700 × 10−3 mol·kg−1 NaCl	0.7	210	3.57	3.71	1.92	−8.55
G-KCl-700	700 × 10−3 mol·kg−1 KCl	0.7	176	3.64	3.58	1.94	−8.54
G-CaCl2-700	700 × 10−3 mol·kg−1 CaCl2·2H2O	2.1	193	3.56	3.74	3.77	−8.25
G-MgCl2-700	700 × 10−3 mol·kg−1 MgCl2·6H2O	2.1	209	3.63	3.67	1.62	−8.62
L-NaCl-10	10 × 10−3 mol·kg−1 NaCl	0.01	208	3.68	3.76	1.80	−9.02
L-KCl-10	10 × 10−3 mol·kg−1 KCl	0.01	223	3.62	3.73	1.38	−9.09
L-CaCl2–10	10 × 10−3 mol·kg−1 CaCl2·2H2O	0.03	236	3.64	3.72	0.67	−9.41
L-MgCl2–10	10 × 10−3 mol·kg−1 MgCl2·6H2O	0.03	237	3.64	3.70	0.53	−9.51
L-NaCl-50	50 × 10−3 mol·kg−1 NaCl	0.05	208	3.64	3.74	1.64	−9.03
L-KCl-50	50 × 10−3 mol·kg−1 KCl	0.05	223	3.66	3.73	1.26	−9.14
L-CaCl2-50	50 × 10−3 mol·kg−1 CaCl2·2H2O	0.15	236	3.61	3.65	0.70	−9.40
L-MgCl2-50	50 × 10−3 mol·kg−1 MgCl2·6H2O	0.15	237	3.65	3.65	0.84	−9.13
L-NaCl-200	200 × 10−3 mol·kg−1 NaCl	0.2	208	3.63	3.71	3.12	−8.75
L-KCl-200	200 × 10−3 mol·kg−1 KCl	0.2	223	3.66	3.65	3.57	−8.69
L-CaCl2–200	200 × 10−3 mol·kg−1 CaCl2·2H2O	0.6	236	3.66	3.69	2.37	−8.86
L-MgCl2–200	200 × 10−3 mol·kg−1 MgCl2·6H2O	0.6	237	3.68	3.70	1.73	−9.00

¹ Rates for basaltic glass are based on 1 Si per formula unit, whereas rates for labradorite are based on 2.36 Si per formula unit.

. The integration of temporal Si release was used to estimate the total mass of solid dissolved during each experimental series. In each case, no more than 1% of the total mass of the solid was dissolved during any experiment.

Measured Si concentrations were used to determine the steady-state dissolution rate of the solids. These rates are reported in Table 2 and illustrated as a function of the concentration of each added chloride salt in Figure 3. The rates were relatively unaffected by the addition of the selected chloride salts to the reactive fluids. The geometrically measured dissolution rates vary from 10^−9.0 to 10^−8.3 mol·m⁻²·s⁻¹ for basaltic glass and from 10^−9.5 to 10^−8.5 mol·m⁻²·s⁻¹ for labradorite. Nevertheless, some trends are apparent. The addition of divalent metal chlorides tended to increase the steady-state dissolution rate of basaltic glass compared to that of the monovalent chloride salts. The steady-state basaltic glass dissolution rates measured in the presence of 0.01 mol·kg⁻¹ NaCl and 0.01 mol·kg⁻¹ KCl are approximately 0.3 orders of magnitude slower than rates measured in either 0.01 mol·kg⁻¹ CaCl₂ or MgCl_2. These differences appear to decrease somewhat with increasing salt concentrations. In contrast, the addition of divalent metal chlorides tended to decrease the steady-state dissolution rate of labradorite more compared to that of the monovalent chloride salts. The steady-state labradorite dissolution rates measured in the presence of 0.01 mol·kg⁻¹ NaCl and 0.01 mol·kg⁻¹ KCl are approximately 0.3 orders of magnitude slower than the rates measured in either 0.01 mol·kg⁻¹ CaCl₂ or MgCl₂. Similar to the dissolution behavior of basaltic glass, the differences in these rates appeared to decrease with an increasing salt concentration.

The variation in measured steady-state rates with the ionic strength of the reactive fluid is depicted in Figure 4. The steady-state dissolution rates of both basaltic glass and labradorite increased slightly in response to the ionic strength of the reactive fluid, as indicated by the trend line in this figure. Nevertheless, a systematic scatter is evident by the variation in labradorite rates with increasing ionic strength, suggesting a small inhibitory effect of the presence of divalent metal cations on the rates.

5. Discussion

5.1. Comparison of Basaltic Glass and Labradorite Si Release Rates

The results of this study allow the direct comparison of the dissolution rates of basaltic glass and an intermediate feldspar. These solids are both commonly present in basaltic rocks and have similar atomic Ca/Si ratios. The atomic Ca/Si ratios of Stapafell basaltic glass and of Hrappsey Islands labradorite are 0.263 and 0.286, respectively. The availability of Ca is essential for the mineralization of CO₂ as solid calcium carbonate during carbon storage efforts. The measured Si release rates of these two solids are compared directly in Figure 5. As can be seen in this figure, the Si release rates of these solids are identical within uncertainty. This concurrence suggests that glassy and crystalline basalts might be equally effective at capturing and storing CO₂ via calcium carbonate precipitation.

5.2. Comparison to Previously Measured Rates

A comparison of the measured steady-state Stapafell basaltic glass dissolution rates with those of previous studies is presented in Figure 6a. This comparison is confounded by the variety of conditions and fluid compositions considered in the various experiments. Oelkers and Gislason [32] measured the dissolution rates at pH 3 and pH 11 at 25 °C in the presence of added aqueous Al, Si, and organic acids and found that the rates depended on the aqueous Al activity of the reactive fluids. Gislason and Oelkers (2003) reported the dissolution rates of Stapafell basaltic glass as a function of pH and temperature [68]. Stockmann et al. [36] measured the dissolution rates of this glass in the presence of carbonate precipitates on its surface. They found a small effect of precipitated carbonate minerals on the dissolution rates of the basaltic glass surfaces. Stockmann et al. [71] measured the dissolution rates in the presence of dead and alive bacteria and concluded that the nutrients and the presence of bacteria substantially slowed dissolution. Wolff-Boenisch et al. [17,33] reported the dissolution rates of this basaltic glass in the presence of inorganic cations, aqueous fluoride, and organic acids. They found that fluoride and aqueous organic acids substantially accelerated basaltic glass dissolution.

The rates summarized in Figure 6a suggest that the presence of increased concentrations of NaCl, KCl, CaCl₂, and MgCl₂ do not strongly influence the rates of basaltic glass dissolution. This behavior contrasts that of aqueous Al, F, and organic acid anions. The effect of these aqueous species on basaltic glass dissolution rates were attributed to a dissolution mechanism that required the removal of aqueous Al to form the rate limiting Si-rich activated complex on the basaltic glass surface (e.g., [32,63,66].

Gudbrandsson et al. [46] reported on the dissolution rates of labradorite samples identical to the ones considered in this study. These previous dissolution rates were measured in a mixed flow reactor system in the presence of an aqueous solution with 0.01 mol·kg⁻¹ ionic strength using NH₄Cl as a background electrolyte at 22 °C. These previously reported rates are compared with those measured in the present study as a function of pH in Figure 6b. It can be seen that the rates measured in the present study are consistent with those of previous efforts. The rates of Gudbrandsson et al. [46], which were normalized to BET surface area, were recalculated to geometrically normalized dissolution rates using Equations (5) and (6) and the grain sizes reported in the original publication. The previously reported rate measured at pH = 3.76 is directly comparable to that of the present study and equal to 10^−9.23 mol·m⁻²·s⁻¹. The rates measured at this pH and in the presence of 0.01 mol·kg⁻¹ NaCl and 0.01 mol·kg⁻¹ KCl in the present study are 10^−9.02 and 10^−9.09 mol·m⁻²·s⁻¹, respectively. Although these rates are similar, they suggest that the presence of Na and K in the reactive solution may have a slight accelerating effect on the rate of labradorite dissolution at this pH.

5.3. Implications for the Dissolution Mechanism of Basaltic Glass and Feldspars

Past studies have reported that the far-from-equilibrium dissolution rates of basaltic glass and feldspars having an anorthite content less than 75% slow significantly and systematically as a function of increasing aqueous aluminum activity. This observation was attributed to the dissolution mechanism of these solids. In both cases, the dissolution rate variation in aqueous Al activity was attributed to the control of these rates by a Si-rich activated complex formed by the removal of Al from the aluminosilicate structure.

The results of the present study suggest that the presence of CaCl₂ and MgCl₂ decreases the labradorite dissolution rates by as much as 0.3 log units compared to the corresponding rates in the presence of aqueous NaCl and KCl solutions. This observation, which contrasts the results of Oelkers and Schott [63], suggests that these divalent metals may limit the formation of the Si-rich activated complex at the surface of Ca-rich feldspars. A potential pathway for this effect is that the presence of these divalent metals in solution may not favor the removal of divalent metals from the near surface of dissolving felspar. This possibility could be further tested through a detailed study of the surface chemistry of Ca-rich feldspars after their interaction with compositionally distinct divalent metal-bearing fluids.

In contrast to the inhibitory effect of solute CaCl₂ and MgCl₂ on labradorite dissolution, these salts are observed to increase the basaltic glass dissolution rates. It seems likely that this observation is due to the role of these salts in increasing the ionic strength of aqueous solutions. This interpretation is supported by the distribution of rates illustrated in Figure 4. Basaltic glass rates appear to plot as a linear function of ionic strength. This is in contrast to the behavior of labradorite dissolution rates, which appear to plot as two distinct linear functions of ionic strength, one for monovalent and one for divalent metal chlorides. An increase in silica glass dissolution rates with increasing ionic strength was reported by Icenhower and Dove [82]; this effect was attributed to the increased charging of the mineral surface by Brady and Walther [83].

5.4. Consequences for Carbon Mineral Storage in Saline Aquifers

This study considered the dissolution rates of basaltic glass and labradorite. The dissolution of these solids favor the mineralisation of CO₂ via the combination of adding Ca to the aqueous phase and increasing the pH of mildly acidic waters. The results presented in this study indicate that the dissolution rates of neither basaltic glass nor labradorite are inhibited by the presence of elevated concentrations of NaCl, KCl, CaCl₂, or MgCl₂ in the aqueous phase at pH~3.5. In fact, the rates appeared to increase to some extent as the ionic strength of the aqueous phase increased. The pH of the reactive fluids considered in this study is similar to that of water in equilibrium with 10 to 50 bars of CO₂. It therefore seems likely that both basaltic glass and labradorite dissolution are, at least, equally able to promote the mineral storage of CO₂ in saline and fresh water aquifers.

6. Conclusions

The rates generated in the present study indicate that the dissolution of basaltic glass and labradorite are equally efficient at releasing Ca into aqueous solution and adding alkalinity to the aqueous phase at pH 3.6. This pH is close to that expected in waters charged with CO₂ at pressures from 10 to 30 bars. Notably, the dissolution rates of these minerals are found to increase mildly with increasing fluid ionic strength to as much as 2.1 mol·kg⁻¹. Basaltic glass and labradorite dominate the compositions of fresh basaltic rock. Consequently, the results of this study support the likely success of mineral carbonation of fresh basaltic rocks located in saline aquifers.