LAUSR

Abstract The extraction of trans-anethole from fennel seeds is assessed by supercritical CO 2 together with ethanol as the modifier. The experiments and modeling are designed through response surface methodology (RSM). The fennel extract is analyzed by gas chromatography-flame ionization detector (GC-FID). RSM design and genetic algorithm (GA) are applied to optimize the SFE variables and trans-anethole recovery. The maximum trans-anethole recovery (56.38 w/w % and 55.50 w/w %) is obtained at optimal operating conditions at 21.3 and 21.55 MPa, 60.2 and 55.59 °C, 103 and 96.38 min and 1.74 and 1.49 ml/min of CO 2 through RSM and GA, respectively. According to response surface analysis, R 2 and modified R 2 of the model are 96.64% and 93.71%, respectively. The accuracy of modeling optimal trans-anethole recovery is validated through triplicate experiments providing the average extraction recovery of 55.3 ± 1.5% and 54.5 ± 1.2% for RSM and GA, respectively.