The development of the UNIFAC-CONDUCT model as a novel approach for the estimation of the conductivity of pure ionic liquids

The correlation and prediction of physical properties of ionic liquids (ILs) is important in the development and design of novel structures for a variety of applications. In particular, for the application as electrolytic media for electrochemical applications, the electrical conductivity of ILs is an important property. Herein, we define a novel model called the UNIFAC-CONDUCT for the estimation of IL conductivity which mimics the principles of the UNIFAC-VISCO model. 784 data points for 38 ILs were used to establish this group contribution model, covering a wide range of temperature (248.15-468.15 K) and a broad range of electrical conductivity (0.002-14.54 S center dot m(-1)). Three sets of UNIFAC-CONDUCT parameters were determined to calculate the conductivity of pure Its by three methodologies wherein the parameters of ion VFT equations, A, B and T-o, and binary interaction parameters, alpha(mn), are selectively either fixed (based on previously reported values) and/or optimized by Marquardt technique. Correlations by the three methods are compared and analyzed, and further correlated with the previously predicted viscosities using the Nernst-Stokes relationship to support the quality of both the UNIFAC-VISCO model, previously proposed by our group, and the UNIFAC-CONDUCT model developed in this paper. Finally, the quality of the proposed method was further assessed through the prediction of the conductivity of one binary system containing two different ILs as the function of the temperature and composition with an accuracy close to 4.1%. (C) 2017 Elsevier B.V. All rights reserved.