Study of FAME model systems: Database and evaluation of predicting models for biodiesel physical properties

The present paper reports a viscosity and density unpublished database of systems formed for fatty acid methyl esters (FAMEs), leading to 426 experimental data points of each property. Kay's mixing rule and Grunberg-Nissan equation were used to estimate data and the group contribution models GC-VOL and GC-UNIMOD were used to predict density and viscosity, respectively. For surface tension, parameters of a Wilson modified equation were adjusted and tested in systems with composition similar to biodiesel. Density estimations resulted in global average relative deviations (ARD) of 0.02%, 0.07% and 0.15% for Kay's mixing rule weighted in mass and molar fractions, and GC-VOL model, respectively. For viscosities, GC-UNIMOD was the most accurate model with global ARD of 5.17%. The surface tension prediction resulted in global ARD minor than 7.00%. These results are an important tool to improve the biodiesel production, its modeling and simulation. (C) 2019 Elsevier Ltd. All rights reserved.