RESPONSE SURFACE METHODOLOGY APPLIED TO THE EXTRACTION OF PHENOLIC COMPOUNDS FROM JATROPHA CURCAS LINN. LEAVES USING SUPERCRITICAL CO2 WITH A METHANOL CO-SOLVENT

摘要

Response surface methodology was used to analyze the results of experiments designed using the Box–Behnken method to extract three phenolic compounds, gallic acid (GA), corilagin (CG) and ellagic acid (EA), from Jatropha curcas Linn. leaves using supercritical CO2 and methanol as a cosolvent. Experiments were carried out from 10 to 30 MPa, 40 to 80C and 30 to 70% (v/v) aqueous methanol. A 3 × 3 Box–Behnken design was used to design the experiments to determine the effects of pressure, temperature and concentration of methanol (MeOH) as well as their interaction on the extraction yield. Three nonlinear equations and 3-D plots (one for each product) with 10 terms were developed. Analytical and numerical techniques were used to locate the optimal operating conditions. The highest experimental yields were obtained at 10 MPa, 60C and 30% (v/v) methanol modifier for GA; 20 MPa, 80C and 30% (v/v) methanol modifier for CG; and 30 MPa, 40C and 50% (v/v) methanol modifier for EA. The response surface models predicted that the maximum extraction yields of GA, CG and EA were 1,567.68 mg/kg of GA at 10 MPa, 80C and 30% (v/v) aqueous MeOH; 4,693.60 mg/kg of CG at 30 MPa, 80C and 30% (v/v) aqueous MeOH; and 1,089.02 mg/kg of EA at 10 MPa, 80C and 70% (v/v) aqueous MeOH, respectively. Because the theoretical optimum was on the limit of the range of the experiments, future work should focus on new experiments designed around the predicted optimum.