A crossover‐UNIQUAC model for critical and noncritical LLE calculations

Abstract

A new model, named the crossover‐UNIQUAC model, has been proposed based on the crossover procedure for predicting constant‐pressure liquid–liquid equilibria (LLE). In this manner, critical fluctuations were incorporated into the classical UNIQUAC equation. Coexistence curves were estimated for systems having a diverse range of asymmetries. These systems included the LLE of five different mixtures, composed of nitrobenzene with one of the members of the alkane homologous family (either pentane, octane, decane, dodecane, or tetradecane), as well as an extra system having a different chemical nature, namely the mixture of n‐perfluorohexane and hexane, to further check the validity of the proposed approach. Using these nonideal mixtures, the validity of the new model was investigated within wide ranges, covering near‐critical to regions falling far away from the critical point. The graphical trends, as well as the quantitative comparison with experimental data indicated the good agreement of the proposed model results with the experimental data. A maximum AARD% value of 3.97% was obtained in calculating molar compositions by the proposed model for such challenging systems covering noncritical, as well as critical regions. In addition, to show the strength of the proposed crossover approach to describe properties other than LLE, molar heat capacities were investigated for the system of nitrobenzene + dodecane. © 2015 American Institute of Chemical Engineers AIChE J, 61: 3094–3103, 2015

Thermodynamic Modeling of Aqueous Electrolyte Systems: Current Status

“…Correlations often permit empirical functions to follow trends in the data too easily, for example as shown by Maribo-Mogensen et al in their graphic exposure of the mean spherical approximation (MSA) , and by the remarkably free-wheeling explanation of specific interactions based on a “volume exclusion cluster” where species can have negative equilibrium constants! These concerns apply to all the empirical activity coefficient models including Pitzer, ,, eNRTL, − MSA, , MSA-NRTL, UNIQUAC ,, and MSE. − As the ever-growing multiplicity of “extensions” to these approaches makes perturbingly clear, the “temptation to improve the fitting of thermodynamic property models by repeatedly introducing new variants with system-specific basis functions and additional adjustable parameters is not confined to the Pitzer equations” . The recent proliferation of SAFT-related modifications − is a prime example: although nearly 700 papers have appeared in the literature referring to Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT), some dozens of which deal with electrolyte solutions (ePC-SAFT), one has to query the merits of a “theory” that must so often be attached to various other equations of state and/or requires some specific alteration to suit different contexts.…”

“…Correlations often permit empirical functions to follow trends in the data too easily, for example as shown by Maribo-Mogensen et al 193 in their graphic exposure of the mean spherical approximation (MSA) 194,195 and by the remarkably free-wheeling explanation of specific interactions 196 based on a "volume exclusion cluster" where species can have negative equilibrium constants! These concerns apply to all the empirical activity coefficient models including Pitzer, 140,141,147 eNRTL, 1 7 2 − 1 7 5 MSA, 1 9 4 , 1 9 5 MSA-NRTL, 1 9 7 UNI-QUAC 31,188,198 and MSE. 167−171 As the ever-growing multiplicity of "extensions" to these approaches makes perturbingly clear, the "temptation to improve the fitting of thermodynamic property models by repeatedly introducing new variants with system-specific basis functions and additional adjustable parameters is not confined to the Pitzer equations".…”

Development of a Model for Electrostatic Contribution to the Osmotic Coefficient of Aqueous Electrolytes

“…As a result, when the parameters having a similar definition, obtained from one model, are used in another model, inaccurate results are obtained . Therefore, these models have the impression of requiring continuous upgrading. − …”