Multi-response optimization of extraction yield, total phenols-flavonoids contents, and antioxidant activity of extracts from moroccan Lavandula stoechas leaves: Predictive modeling using simplex-centroid design

Abstract

This study aimed to improve the extractive effect of four solvents and their combinations on phenolic compounds from Moroccan Lavandula stoechas leaves. A simplex-centroid design was used in a solvent system containing ethyl acetate, methanol, ethanol, acetone, and their mixtures to optimize the five responses: extraction yield, total phenol content (TPC), total flavonoids content (TFC), and antioxidant activity by DPPH and FRAP assays. HPLC-DAD then analyzed the obtained extracts from each pure solvent. Sinapic acid, epicatechin, and rosmarinic acid were identified as major compounds. After optimizing each response separately according to the validated special cubic models, the desirability function has allowed the optimization of all responses simultaneously. The optimal solvent system was the ternary mixture (50:36:14 v/v/v) methanol: ethanol: acetone, while the optimal values were 14.63%, 292.23 mg GAE/g DE, 95.73 mg QE/g DE, 27.55 μg/mL and 3.13 μg/mL for yield, TPC, TFC, DPPHIC50, and FRAPEC50, respectively. This study shows that the combination of the solvent mixture may be applicable to giving better antioxidant activities than those given by some commercial antioxidants and may subsequently be useful as natural food preservatives.