A new group contribution method based on GCVOL model developed by Elbro et al. in 1991 [Elbro, H. S.; Fredenslund, A.; Rasmussen, P. Ind. Chem. Eng. Res. 1991, 30, 2576–2582] is proposed for the estimation of ionic liquids density over a wide range of temperature and pressure. A total of 102 new groups for ionic liquids were introduced to the already 60 existing groups revised and proposed in 2003 by Ihmels and Gmehling [Ihmels, E. C.; Gmehling, J. Ind. Chem. Eng. Res. 2003, 42, 408–412]. These groups were proposed based on a collection of density data from literature. The databank contains data of 864 different ionic liquids, including dicationic and tricationic species, and a total of 21 845 data points, covering a temperature range of 251.62–473.15 K and a pressure range of 0.1–300.0 MPa. An average absolute relative deviation (%AARD) of 0.83% was obtained, indicating that our model is able to predict densities of a great variety of ionic liquids accurately.

Biodiesel has many advantages because it is a biodegradable, nontoxic fuel and its production results in less particulate matter. For this reason it has been studied in many fields of science as a substitute for mineral fuels. Short chain alcohols such as ethanol have been used for extraction of soybean and sunflower biodiesel in the literature. The present paper reports liquid–liquid equilibrium data for systems containing soybean biodiesel + glycerol + ethanol at (293.15 and 323.15) K and sunflower biodiesel + glycerol + ethanol at (298.15 and 313.15) K. Binodal curves were obtained by the cloud-point method, while tie-line compositions were obtained by density measurements. The values of distribution coefficients and selectivities indicate that the ethanol is a good solvent for the extraction of soybean biodiesel and sunflower biodiesel from glycerol. The reliability of experimentally measured tie-line data can be ascertained by applying the Othmer–Tobias equation. The experimental data were correlated by the nonrandom two-liquid (NRTL) model, using the simplex minimization method with a composition-based objective function. The results found in this work were considered satisfactory, by analyzing statistical parameters using root-mean-square deviations.

Graphical abstract

No abstract

No abstract