Green Ultrasound-Assisted Extraction of Flavonoids from Scutellariae Radix Using Deep Eutectic Solvents: A Hybrid BBD-ANN-GA Optimization Strategy and Mechanistic Insights

Deep eutectic solvents (DESs) have emerged as promising green alternatives to conventional organic solvents for the sustainable extraction of bioactive compounds from medicinal plants. In this study, a DES-based extraction system enhanced by ultrasound was employed to extract flavonoids from Scutellariae Radix (SR). A hybrid optimization strategy integrating Box-Behnken design (BBD), artificial neural network (ANN) modeling, and genetic algorithm (GA) was utilized to maximize the extraction efficiency. Under the optimized conditions, the predicted yield of flavonoids reached 194.58 mg/g, which was in close agreement the experimental value (194.03 mg/g). The purified DES extracts exhibited strong antioxidant activity. After five consecutive extraction cycles, the DES maintained its structure stability and retained a high extraction yield with only a 11.39% decrease. The characterization of the solid extraction residue indicated the microstructure of SR was disrupted by the DES, facilitating the release of flavonoids. In addition, mechanistic studies based on quantum chemical calculations revealed strong hydrogen bonding and orbital interactions between the DES and flavonoids as key contributors to the extraction process. These results demonstrate the potential of DESs as tunable and reusable media for the efficient extraction of natural products, supported by both experimental data and molecular-level insights.