DEACIDIFICATION OF PALM OIL WITH ETHANOL: SIMULATION IN A MULTISTAGELIQUID-LIQUID COLUMN

Objective: The objective of this work was to apply the simulator Aspen Hysys 8.4 to describe the palm oil deacidification process by liquid-liquid extraction using anhydrous ethanol and ethanol+water mixtures as solvent. Methods: The experimental liquid-liquid equilibrium data from the literature of palm oil+fatty acids+ethanol and palm oil+fatty acids+ethanol+water were correlated to obtain the NRTL temperature-dependent binary interaction parameters using the Aspen Properties package. The palm oil deacidification using anhydrous ethanol, ethanol+water (6.10 wt.%) and ethanol+water (12.41 wt.%) as solvent was simulated in a multistage countercurrent liquid-liquid column. The NRTL model temperature-dependent binary interaction parameters adjusted previously were used. Results: The regressions of different systems show that the NRTL model was able to describe the liquid-liquid experimental equilibrium data for all the systems studied with RMSD between 0.15 to 1.72%. The highest palm oil recovery (99.05%), corresponding to the minimal loss of neutral oil to extract stream, was obtained using ethanol+water (12.41 wt.%) mixture as solvent with S/F=1 ratio and as S/F ratio increases the palm oil recovery decreases for each water concentration in ethanol solvent. These results are consistent with what is published in the literature. Conclusions: The simulations showed that, for all S/F ratios studied, only the deacidification of palm oil using ethanol+water (12.41 wt.%) mixture as solvent, the maximum limit accepted of 5 wt.% loss of neutral oil from physical refining industrial was reached. However it is necessary more than 15 theoretical stages and S/F ratio greater than 2.5 to palm oil reach the edible oil characteristics.