A statistical thermodynamic model for prediction of vapor pressure of mixed liquid desiccants near saturated solubility

Abstract

Because the weight of interactions in liquid desiccants varies with solute type and concentration, it is difficult for existing typical models to accurately quantify the interactions in mixed liquid desiccants near saturated solubility. A statistical thermodynamic model is proposed to predict the vapor pressure of mixed liquid desiccant solutions near saturated solubility by re-quantifying the weight of interactions between all species in solution. The model considers that charge interaction and non-charge interaction are still dominant, and charge-dipole interaction between ions and solvent molecules increases with the increase of solute concentration. Rigorous expressions are given to calculate the interactions: ion-ion charge interaction is described by modifying the Fowler-Guggenheim theory, charge-dipole interaction is expressed by combining the McMillan-Mayer theory with the Debye-Hückel theory, and non-charge interaction is calculated based on the extended UNIQUAC equation. The prediction accuracy comparison between the new model and two typical models shows that the new model has better accuracy, and the superiority of the model increases with the increase of solute concentration and temperature. The model can be recommended for the liquid desiccant solutions consisting of Li+, Ca2+, Cl− or Br− with concentration ranging from infinite dilution to near saturation and temperature ranging from 283.15 to 353.15 K.