Abstract
The aim of this study was to develop the optimal conditions for supercritical fluid extraction (SFE) of bioactive trans-resveratrol from peanut kernels using an experimental design. The variables taken into account were extraction pressure, extraction temperature, extraction time and amount of modifier. The model was first set for significant factor screening by full factorial design, then, optimized by central composite designs. The optimal extraction parameters were a pressure of 7000 psi, temperature of 70 °C and time of 50 min while amount of modifier did not show significant effect. The quantity of trans-resveratrol was predictable by a full quadratic regression equation with R2(predict) = 95.56%. The predicted trans-resveratrol concentration in peanut samples was 0.7998 µg/g while the experimental concentration was 0.7884 ± 0.1553 µg/g. Conventional solvent extraction demonstrated less selectivity and needed more clean-up process prior to HPLC analysis. Our optimized SFE condition was effective to maximize trans-resveratrol extraction with less contaminants and gave the comparable amount of trans-resveratrol between actual and predicted values.