Vapor-liquid equilibrium data evaluation for systems of interest in the biodiesel production
Thermodynamic modeling, Soybean oil, Ethanol, Cottonseed oil, Hexane.
In an attempt to mitigate the environmental impacts arising from the burning of fossil fuels, many studies have sought to make biodiesel production even more feasible, since it is a renewable, biodegradable fuel that has a lower content of polluting compounds. In this sense, this work focuses on the description the vapor–liquid equilibrium (VLE) data of three systems that are of interest to the biodiesel industry. The following systems were studied: ethanol (1) and oleic acid (2) at isobaric data at 760 mmHg, isothermal at 318.15 K and activity coefficient at infinite dilution; n-hexane (1) and cottonseed oil (2) with isobaric data at 310 mmHg; ethanol (1) and soybean oil (2) with isobaric data at 760 and 600 mmHg. A thermodynamic modeling was carried out with the ideal (Raoult's Law), UNIQUAC correlation and UNIFAC prediction approaches. The vapor pressure data of the pure substances were extracted from the literature. The relative mean deviations of the fitted Antoine's constants for ethanol, oleic acid, n-hexane, cottonseed oil, and soybean oil were 0.34%, 3.30%, 0.55%, 5.00% and 2.53%, respectively. The activity coefficient at infinite dilution values and the VLE binary data for the ethanol and oleic acid system were extracted from the literature. For the other systems, binary data were obtained in a modified Othmer ebulliometer from a previous study in our laboratory (FOTEQ). The correlation study with the UNIQUAC model described with quality the behavior of all systems under the conditions of interest. This indicates that the obtained binary parameters can be used in process simulation studies, as well as in equipment sizing projects, such as distillation columns. The predictive study carried out with the UNIFAC model showed accurate descriptions as UNIQUAC model. Furthermore, for all systems, a deviation from ideality was observed, which reiterates the indispensability of using an activity coefficient model, such as UNIQUAC.