Rigorous Analysis of Non-Ideal Solubility of Sodium and Copper Chlorides in Water Vapor Using Pitzer-Pabalan Model

TL;DR

This paper develops a rigorous fitting procedure for the Pitzer-Pabalan model to analyze the non-ideal solubility of sodium and copper chlorides in water vapor, validated against experimental data and quantum chemistry calculations.

Contribution

It introduces a systematic fitting method for the Pitzer-Pabalan model, improving reliability and applicability for salt-water vapor solubility analysis.

Findings

Successful parameterization of NaCl and CuCl solubility data

Benchmarking against quantum chemistry calculations confirms accuracy

Enhanced understanding of salt-water vapor interactions

Abstract

Gaseous mixtures of water vapor and neutral molecules of salt (e.g., NaCl, CuCl etc.) can be highly non-ideal due to the strong attractive interaction between salt and water molecules. In particular, this can result in high solubility of salts in water vapor and a strong dependence of solubility on vapor pressure. The analysis of salt solubility in water vapor can be done using the Pitzer-Pabalan model, which is based on the thermodynamic theory of imperfect gases. The original Pitzer-Pabalan work demonstrated that one can reproduce experimental data for NaCl solubility in vapor. No analysis was performed on the reliability of their original fits, which we believe has contributed to the lack of applications of the Pitzer-Pabalan model despite the apparent success of the original paper. In this work, we report recent progress in developing a rigorous fitting procedure to parameterize the Pitzer-Pabalan model using experimental data. Specifically, we performed fitting of the experimental results obtained elsewhere for NaCl and CuCl. We investigate the degree of underfitting/overfitting and the sensitivity of the fitting quality to variations in the resulting fitting parameters. The results, as represented by the thermodynamic parameters describing the energetics of formation of salt-bearing water clusters, were successfully benchmarked against Gaussian 16 ab initio quantum chemistry calculations. The resulting rigorous fitting procedure presented here can now be applied to other systems.