A New QSPR Model for Predicting the Densities of Ionic Liquids

A new mathematical model for the prediction of ionic liquids (ILs) densities using quantitative structure-property relationship approach is presented. The model was developed based on a data set containing 918 experimentally measured density data for 747 pure ILs which were quoted from Shen et al. (Chem Eng Sci 66:2690-2698, 2011). MATLAB� software was used in the development of the algorithm, and the code was written based on a combination of multiple linear regression (MLR) method and polynomial equation. The results showed that the model is capable of predicting IL densities with very high accuracy at a correlation of determination value R2 = 99.5 . The proposed model can be considered comparable to (if not more accurate than) other established models which have been developed using the traditional MLR, polynomial, and artificial neural network methods. In addition, this model is practical, cost-effective and can be used as alternative to experimental measurement of density of ILs. © 2014 King Fahd University of Petroleum and Minerals.