Application of response surface methodology for optimization of ultrasound‐assisted solid‐liquid extraction of phenolic compounds from Cenostigma macrophyllum

This study determines the optimal conditions for the ultrasound‐assisted solid‐liquid extraction (UA‐SLE) of the total phenolic compounds (TPC) of Cenostigma macrophyllum leaves. The experiments were carried out according to three‐level, four‐variable Box‐Behnken design combined with the response surface methodology (RSM). Extraction (sonication) time, liquid‐solid ratio, methanol concentration, and ultrasonic power were investigated as independent variables to obtain the optimal extraction conditions for TPC. Data were fitted to a second‐order polynomial model, and the model fitness was evaluated by the analysis of variance, which indicated the model (P < 0.0001), presenting precision and reliability with the experiments performed (R2 = 0.9947, coefficient of variance [CV] = 4.29, SD = 1.67). Individual and interactive effects of the independent variables on TPC extraction were interpreted using the proposed mathematical model. The results showed that for the maximum TPC extraction, the optimal conditions included 32 min extraction time, 29.73 mL·g−1 liquid‐solid ratio, 47.72% methanol concentration, and 145.71 W ultrasonic power. Under these optimal conditions, the predicted values were in agreement with the experimental values, thus validating the RSM model. When the optimal conditions were compared with those of conventional extraction techniques, the results showed that UA‐SLE presented significantly better results for TPC extraction (71.2% to 138.3% more efficient) and for antioxidant activity (115.0% to 179.2% more efficient). High‐performance liquid chromatography analysis of the extracts revealed that UA‐SLE did not interfere with the extract chemical composition, thus confirming that it was more effective than conventional techniques. Ultrasound‐assisted solid‐liquid extraction conditions were optimized for the maximum extraction of the total phenolic compounds of Cenostigma macrophyllum using a Box‐Behnken design combined with response surface methodology. Independent variables in the extraction processes were investigated to obtain the optimal conditions for the extraction. The results indicated optimal conditions for TPC extraction and validated the RSM model. These parameters also showed that UA‐SLE exhibited significantly higher TPC extraction efficiency than conventional methods and that it did not alter the chemical composition of the extracts.