Process optimization of microwave-assisted aqueous extraction of tannins and saponins from Malabar Spinach (Basella alba) leaves using ANN-GA and RSM methodology

Abstract

The effect of microwave-assisted aqueous extraction (MAE) conditions on the recovery and quantification of tannins and saponins from Basella alba leaves has been explored. The experiment was carried out according to the Box-Behnken experimental design of response surface methodology (RSM) with 17 runs. The independent variables of the study were microwave power (100 to 300 W), treatment time (5 to 15 min), and feed/solvent ratio (0.025 to 0.05). The effect of extraction conditions on the response variables, total condensed tannin, total hydrolyzed tannin, and total saponin content, were predicted using RSM and artificial neural network-genetic algorithm (ANN-GA) methodology. The extracted values of total condensed tannin, total hydrolyzed tannin, and total saponin content, using RSM and ANN-GA were compared and evaluated for goodness of fit. The predicted values using RSM and ANN-GA models for total condensed tannin were found to be 0.18 and 0.26 mg TAE g−1, total hydrolyzed tannin were 0.14 and 0.15 mg TAE g−1, and total saponin content were 1.66 and 1.49 mg DE g−1, respectively. When compared to RSM, ANN-GA demonstrated higher value of R2 and lower values of other statistical parameters. Additionally, the predicted value of ANN-GA was more closely aligned with the experimental value.