Densities of Aqueous MgCl₂(aq), CaCl₂(aq), KI(aq), NaCl(aq), KCl(aq), AlCl₃(aq), and (0.964 NaCl + 0.136 KCl)(aq) at Temperatures Between (283 and 472) K, Pressures up to 68.5 MPa, and Molalities up to 6 mol·kg^–1

Abstract

We report the densities of MgCl 2 (aq), CaCl 2 (aq), KI(aq), NaCl(aq), KCl(aq), AlCl 3 (aq), and the mixed salt system [(1 − x)NaCl + xKCl](aq), where x denotes the mole fraction of KCl, at temperatures between (283 and 472) K and pressures up to 68.5 MPa. The molalities at which the solutions were studied were (1.00, 3.00, and 5.00) mol·kg −1 for MgCl 2 (aq), (1.00, 3.00, and 6.00) mol·kg −1 for CaCl 2 (aq), (0.67, 0.90, and 1.06) mol·kg −1 for KI(aq), (1.06, 3.16, and 6.00) mol·kg −1 for NaCl(aq), (1.06, 3.15, and 4.49) mol·kg −1 for KCl(aq), (1.00 and 2.00) mol·kg −1 for AlCl 3 (aq), and (1.05, 1.98, 3.15, and 4.95) mol·kg −1 for [(1 − x)NaCl + xKCl](aq), with x = 0.136. The measurements were performed with a vibrating-tube densimeter calibrated under vacuum and with pure water over the full ranges of pressure and temperature investigated. An analysis of uncertainties shows that the relative uncertainty of density varies from 0.03 % to 0.05 % depending upon the salt and the molality of the solution. An empirical correlation is reported that represents the density for each brine system as a function of temperature, pressure, and molality with absolute average relative deviations of approximately 0.02 %. Comparing the model with a large database of results from the literature, we find absolute average relative deviations of 0.03 %, 0.06 %, 0.04 %, 0.02 %, and 0.02 % for the systems MgCl 2 (aq), CaCl 2 (aq), KI(aq), NaCl(aq), and KCl(aq), respectively. The model can be used to calculate density, apparent molar volume, and isothermal compressibility over the full ranges of temperature, pressure, and molality studied in this work. An ideal mixing rule for the density of a mixed electrolyte solution was tested against our mixed salt data and was found to offer good predictions at all conditions studied with an absolute average relative deviation of 0.05 %.

A Pitzer-based characterization of aqueous magnesium chloride, calcium chloride and potassium iodide solution densities to high temperature and pressure

“…The reference concentration was chosen as 5 mol.kg −1 since this is A c c e p t e d M a n u s c r i p t the highest concentration at which there is a significant number of data over a wide range of conditions (see Figure 1). The deviations between the model and the data of Al Ghafri et al [1] are minor (Figure 4), being mostly less than ±0.5 cm 3 .mol −1 . However, there is a small but noticeable cycling of the residuals with temperature at the highest solution concentration.…”

“…The parameters which need to be optimised are • usually exhibit a complex dependence on temperature and pressure [14,15]. As recommended by Archer [15], certain numerical difficulties associated with this parameter can be reduced [14] by re-writing equation (1) to make it relative to a non-zero reference concentration, m r .…”

“…On the other hand data for potassium iodide are comparatively scarce. While the recent values of Al Ghafri et al [1] extend the data to 473 K and high pressure, there are no measurements at concentrations above 1 mol.kg −1 at temperatures greater than 373 K. Since the data are lacking at non-ambient pressure, the β (1)V KI and C φV KI parameters were taken as pressure independent during model-fitting. As equations (2)- (6) are linear with respect to all of the unknown parameters, best-fitting values can be determined using singular value decomposition [6,19].…”

“…The recent publication of high quality density data for MgCl 2 (aq), CaCl 2 (aq) and KI(aq) under superambient conditions [1] provides an unusual opportunity to investigate the effectiveness of thermodynamic modelling capabilities at the current state-of-the-art. Taken together with previously described A c c e p t e d M a n u s c r i p t 6.17 [7] 0.1-2 273-473 1.0 [8] 0.1 273-373 7.0 CaCl 2…”

“…Mao and Duan [2] have criticised the better-known models [3][4][5] for aqueous chloride solutions identifying some frequent flaws such as non-thermodynamically conforming trends at high concentration and failure to extrapolate to the density of pure water. Al Ghafri et al's [1] modelling approach also re-optimized the density of pure water, producing differences with convincing correlations. Above all, the primary problem with the models in Table 1, as with any critical review of thermodynamic data, is that they become superseded whenever new, high-quality measurements are performed.…”