Improving the oil yield of Iranian Jatropha curcas seeds by optimising ultrasound-assisted ethanolic extraction process: A response surface method

Abstract

An ultrasound-assisted solvent extraction (UASE) procedure has been optimised to speed up oil extraction from Jatropha curcas L. seeds by response surface methodology-Box-Behnken design (RSM-BBD). The preliminary study to determine the best solvent for extracting oil from J. curcas containing more than 60% oil revealed that ethanol had the highest efficiency among chloroform, acetone, n-hexane and methanol. Three independent variables including UASE time (10-60 min), UASE temperature (30-60 °C) and ratio of ethanol to sample (15:45, v/w) were investigated to achieve the highest oil yield. RSM-BBD demonstrated that a second-order polynomial model can adequately be developed with a high R2 (0.9952) using multiple linear regression analysis. The linear effect of ethanol to sample ratio had the most significant (P<0.05) effect on yield value. The most suitable combination of variables for higher yield increase (61.94%) was UASE time of 51.69 min, UASE temperature of 44.57 °C and a ratio of ethanol to the sample of 44.78 (v/w). At this optimal condition, the yield increased to 62.12%, which was close to the amount predicted by the model which was nearly 2.5× more than the obtained yield in the Soxhlet extraction procedure (25.63%). Gas chromatography-mass spectrometry of the oils extracted under optimal conditions showed that the most dominant fatty acids were oleic (26.15%), linoelic (24.21%), palmitic (21.09%) and stearic (6.28%) acids, respectively.