(6d) Solubility of Light Gases in Water and NaCl Brines at High Pressures

A CO₂ stream destined for geological carbon storage (GCS) usually contains impurities such as H₂, CO and N₂. Besides being an impurity in GCS, H₂ is also the stock gas for underground hydrogen storage (UHS). In order to understand the long-term fate of both H₂ and impure CO₂ after injection into a subsurface formation, it is important to understand the solubility of these gases in water and formation brine at elevated temperature and pressure conditions. The solubility of CO₂ in water and brine has been extensively studied over wide temperature and pressure ranges. On the other hand, the solubility of H₂, CO and N₂ has not been fully explored, especially the solubility of these gases in concentrated brines at high temperatures and high pressures. Therefore, in this work, we have conducted experimental measurements of the solubility of H₂, CO and N₂ in water and NaCl(aq) solutions with molality up to 4 mol/kg at pressures up to about 40 MPa and at temperatures from 323.15 to 423.15 K.

The measurements were conducted by a static synthetic method [1] using a high-pressure view cell. The view cell was closed at one end and had a sapphire window at the other end which allowed visual observation of the transition from gas-liquid two-phase to single liquid-phase. Fig. 1 [1] is a schematic diagram of the measurement system. The camera opposite to the window was used to capture live video inside the cell. The stirrer bar was located near the closed end of the view cell. The water or brine that was stored in the bottle was injected and pressurized by the syringe pump.

Initially, the view cell was filled with a known amount of gas, after which water or brine was injected under stirring until all the gas dissolved. At this point, the equilibrium temperature and pressure were recorded. The amount of gas was obtained from the gas-filling pressure, and the volume and temperature of the cell at the gas filling condition, while the amount of water or brine was obtained by the displacement of the injection pump.

A crucial feature of this design was that it prevented gas from being trapped in any dead volume in the cell without contact with liquid. Although there was some unavoidable dead volume around the sapphire window, the view cell was vertically positioned during the experiment, which guaranteed that the pre-filled gas was compressed towards the closed end of the cell as the liquid advanced. In this way, the gas that was compressed between the closed end and the liquid had nowhere to go other than dissolution into the liquid. This design also minimized the leakage possibility of gas. Another key feature of the apparatus was the use of materials that are resistant to corrosion in concentrated brines: titanium alloy, sapphire and perfluoro-elastomer.

The bubble pressures were determined by a combination of graphical analysis of pressure-volume data, a physical model and visual observation. The relative uncertainty of the bubble pressure of dissolved gas at a given temperature and salt molality was estimated to be 4%. The new results from this study were used to develop a simple model to predict H₂, CO and N₂ solubility in water and NaCl brines as a function of temperature, pressure and salt molality up to 4 mol/kg. This model is an extended Krichevsky-Kasarnovsky (KK) equation [2] given by:

ln(f₂ / b₂) = ln(k₁₂) + k_sb_s + (v₂ / RT)(p-p_1,sat). (1)

Here, f₂ is the fugacity of gaseous solute in the gas phase, b₂ is the molality of the solute in the aqueous phase, k₁₂ is Henry’s constant, k_s is the Sechenov coefficient, b_s is the molality of salt, v₂ is the partial molar volume of the solute in aqueous solution, p is pressure and p_1,sat is the solvent vapour pressure. Three parameters were regressed for each isotherm: k₁₂, v₂ and k_s.

Fig. 2 (a) shows the experimental results for CO solubility in water and NaCl(aq) solutions, in comparison with both the KK model and literature data, while Fig. 2 (b) shows the Henry’s constant k₁₂, and Sechenov coefficient k_s as a function of temperature for CO solubility in water and NaCl(aq) solutions.

The new gas solubility data in concentrated brines from this study extend the available data to a higher pressure up to about 40 MPa, which is of relevance to the geological carbon storage and underground hydrogen storage conditions. To the best of our knowledge, this study provides the first data for CO solubility in brine with high molality of salt. The extended KK model developed by this work can be used to predict H₂, CO and N₂ solubility in water and NaCl brines at elevated temperatures and pressures and at salt molalities of up to 4 mol/kg. The solubility data of H₂, CO and N₂ are important for the design of a GCS project and UHS project.

Reference

[1] G.A. Torín-Ollarves, J.P.M. Trusler, Fluid Phase Equilibria, 539 (2021) 113025.

[2] I.R. Krichevsky, J.S. Kasarnovsky, Journal of the American Chemical Society, 57 (1935) 2168-2171.

[3] C. Gillespie, G.M. Wilson, Research report, United States, (1980) pages: 38.

[4] K. Fischer, M. Petri, J. Chen, O. Noll, J. Gmehling, Unpublished data (1995).

[5] J. Jung, O. Knacke, D. Neuschütz, Chemie Ingenieur Technik - CIT, 43 (1971) 112-116.

[6] R. Rettich, R. Battino, E. Wilhelm, Berichte der Bunsengesellschaft für physikalische Chemie, 86 (1982) 1128-1132.

[7] W. Winkler, Berichte der Deutschen Chemischen Gesellschaft, 24 (1891) 89-101.

[8] G.G. Power, H. Stegall, Journal of Applied Physiology, 29 (1970) 145-149.

[9] E. Douglas, Journal of Physical Chemistry, 71 (1967) 1931-1933.