Prediction of Liquid Phase Heat Capacity of Ionic Liquids: Comparison of Existing Methods and Development of New Hybrid Group Contribution Models

Estimation of the liquid phase heat capacity of ionic liquids (ILs) is necessary in order to select and design the optimum ILs for specific industrial applications, particularly those involving thermal storage and heat transfer. Most attempts at estimating the heat capacity (at the constant pressure) of ILs have followed the group contribution model (GCM) approach, but these have involved the use of limited databases and cannot be applied to a wide range of ILs. In this study, an extensive database of over 11500 data points with 273 unique ILs, made up of 155 unique cations and 74 unique anions, was compiled and used to assess three commonly used GCMs. GCM-1 is less complicated than the other two in terms of its mapping parameters but is more restricted in terms of its applicability. GCM-2 and GCM-3 are more complex and have been further improved by incorporating additional functional groups and indirect parameters. The results have shown that GCM-2 and GCM-3 have performed better than GCM-1 in terms of their mean absolute percentage error (MAPE) (2.37% for GCM-1, 2.27% for GCM-2, and 2.17% for GCM-3) and enable heat capacity prediction for a wider range of ILs.