Density and Electrical Conductivity for Aqueous Mixtures of Monoethylene Glycol and Sodium Chloride: Experimental Data and Data-Driven Modeling for Composition Determination

Monoethylene glycol (MEG) is a gas hydrate inhibitor widely applied for natural gas flow assurance. A series of density and electrical conductivity measurements of water + MEG + NaCl mixtures are reported, allowing the supervision of the MEG regeneration unit. Density (509 data points) and electrical conductivity (212 data points) measurements were performed in wide ranges of temperature, T = 278.15-363.15 K, and concentration of solvents and NaCl up to almost saturation. The theory of solutions was applied for density description using excess volume, which was correlated with the Redlich-Kister equation. The resulting absolute and relative mean deviations are 0.00127 g cm(-3) and 0.12%, indicating accurate representation. A semi-empirical correlation with 15 adjustable parameters was considered for electrical conductivity of water + MEG + NaCl mixtures. The obtained absolute and relative mean deviations are 1.49 mS.cm(-1) and 5.70%. The properties functions presented an approximately orthogonal behavior to each other, allowing the determination of mixture composition from experimental density and electrical conductivity data. The Matlab environment was found to be robust in solving the nonlinear system of two equations with constraints. The proposed methodology was extensively tested, and deviations less than 0.0060 and 0.0011 in solvents and NaCl mass fractions were obtained, respectively, demonstrating the required accuracy for industrial application.

Automated summary

Monoethylene glycol (MEG) is commonly used as a gas hydrate inhibitor for natural gas flow assurance. Density and electrical conductivity measurements of water + MEG + NaCl mixtures were conducted over a wide range of temperature and solvent concentrations, with accurate representation achieved through the use of theoretical solutions and semi-empirical correlations. The Matlab environment proved robust in solving the nonlinear system of equations, demonstrating the required accuracy for industrial application.