Submitted to: Journal of the American Oil Chemists' Society
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: December 4, 2008
Publication Date: March 1, 2009
Citation: Eller, F.J., Taylor, S.L., Laszlo, J.A., Compton, D.L., Teel, J.A. 2009. Counter-current carbon dioxide purification of partially deacylated sunflower oil. Journal of the American Oil Chemists' Society. 86:277-282. Interpretive Summary: High oleic sunflower oil was treated with enzymes to produce a mixture of modified oils (i.e., structured lipids). After the reaction is finished, unwanted by-products (i.e., fatty acid propyl esters) must be removed from the mixture before the product can be used. Researchers at the National Center for Agricultural Utilization Research, Peoria, Illinois, studied the use of liquid carbon dioxide (L-CO2) in a continuous counter-current fractionation column to purify the mixture from this enzymatic reaction of sunflower oil. System parameters, such as solvent (i.e., CO2) to feed ratio and feed rate, were examined to optimize the separation. Product purity increased with solvent to feed ratio, though some product was lost at the highest ratios. Product purity peaked at approximately 99% with a feed rate of 2.5 mL/min. The optimized conditions were applied to large batches in both a semi-continuous feed mode, as well as a continuous feed mode, and gave product purities of 99.3 and 98.1%, respectively. This research benefits industry by providing an effective method to remove by-products from the enzymatic reaction mixture to give a very pure product free of contaminants.
Technical Abstract: High oleic sunflower oil was partially deacylated by propanolysis to produce a mixture of diglycerides and triglycerides. To remove by-product fatty acid propyl esters (FAPEs) from this reaction mixture, a liquid carbon dioxide (L-CO2) counter-current fractionation method was developed. The fractionation column was 1.2 m long and separations were done at 25 deg C and 11.0 MPa. Several solvent to feed ratios (S:FR) (i.e., 7.5, 15, 30, and 60 g/g) as well as feed rates (FR) (i.e., 1, 2, 2.5, 3, and 4 mL/min) at a constant S:FR of 15 were examined. Raffinate purity (i.e., percentage glycerides) as well as extract purity (i.e., percentage FAPEs) were both monitored. Percentage glycerides in both the raffinate and the extract increased with S:FR. The raffinate was ca. 83, 97, 100 and 100% glycerides at S:FRs of 7.5, 15, 30, and 60, respectively. The percentage glycerides in the extracts were ca. 3, 4, 8 and 17%, respectively. With a constant S:FR of 15, the raffinate purity peaked at ca. 99% glycerides with a FR of 2.5 mL/min and the extract at this FR contained ca. 96% FAPEs (i.e., ca. 4% glycerides). The optimized conditions were applied to large batches in both a semi-continuous feed mode, as well as a continuous feed mode, and gave raffinates of 99.3 and 98.1% glycerides, respectively, and extracts of 97.3 and 97.2% FAPEs, respectively. This work demonstrates that continuous L-CO2 fractionation effectively separates by-product FAPEs from product glycerides, providing both very pure raffinates (i.e., 98-99% glycerides) as well as very pure extracts (i.e., 97% FAPEs).