Document Type : Research Article
Department of Food Science and Technology, Science and Research Branch, Islamic Azad University, Tehran, I.R. IRAN
Department of Chemistry, Faculty of Chemistry, Tehran North Branch, Islamic Azad University, Tehran, I.R. IRAN
Department of Food Technology, College of Abouraihan, University of Tehran, Tehran, I.R. IRAN
Miscellas of canola were obtained by mixing its crude oil with hexane solvent at 80:20 and 70:30 ratios. Then, 16.1 M of phosphoric acid and 6.92 M of sodium hydroxide were mixed with the resulting micelles at 0.3% (w/w for degumming) and 13% (w/w for neutralizing), respectively, before two sequential Membrane Filtrations (MF). The MF unit had a crossflow mode equipped with three independent variables of transmembrane pressures or TMP (at 2, 3, and 4 bar), feed velocity (at 0.5 and 1 m/s), and temperatures (at 30, 40, and 50˚C) were used to determine the efficiencies of two MF processes and find out their optimum conditions. When the crude canola oil was mixed with 20-30% solvent and passed the two stages of MF (for degumming and refining) at TMP=2 bar, feed velocity=1 m/s, and temperature = 50˚C, the final polished canola oil had < 5% soap, < 5% phosphorus, < 5% fatty acids, and < 15% wax. Membrane refining, compared to chemical refining, significantly reduced the phosphorus content (50%), free fatty acids (29%), soap (99%), and wax (72%) of refined canola oil. While the permeate flux of canola miscella with 20% solvent increased with rising TMP, feed velocity, and temperature, the ones with 30% solvent did not increase with a similar trend. The highest permeate flux of refined canola oil reached 0.03 Kg/m2, s for miscella with 20% solvent when the feed velocity, TMP, and temperature of degumming or neutralization were 1m/s, 3-4bar, and 30-40oC, respectively. The dominant fouling changed from standard to cake blocking when the crude oil was mixed with 20 or 30% solvent, and the TMP of the MF process in each stage of degumming and neutralization was > 2 bar.