Abstract
Oil was extracted from freeze-dried sardine (Sardinella lemur) fillets using supercritical carbon dioxide (SC-CO2) and a few milliliters of ethanol were optimized with response surface methodology (RSM). The impact of extraction pressure (200-400 bars) and temperature (40-70 degrees C) were studied on the total extraction yields, ratios of Eicosapentaenoic acid (EPA) and Docosahexaenoic acid (DHA). The results were compared with those of Soxhlet and modified Kinsella methods (MKM). The oils obtained using the SC-CO2 and MKM methods were significantly (P < 0.05) higher in oil yield (8.04% and 6.83%), EPA (5.43% and 5.45%) and DHA (18.76% and 18.54%), respectively, compared to the Soxhlet yield (5.10%), EPA (2.17%) and DHA (06.46%). Of the two independent variables, pressure had a critical effect on yield while EPA and DHA ratios were notably influenced by temperature. The combined optimal values were pressure at 328 bars and temperature at 40 degrees C, with corresponding responses of 7.20%, 5.68% and 20.09% for yield, EPA and DHA, respectively. The experimental values in this study were reasonably comparable to their predicted counterparts.