The optimization of extraction process on Quzhou Fructus Aurantii flavonoids with natural deep eutectic solvents: A comparative study between orthogonal analysis and genetic neural network

Quzhou Fructus Aurantii (QFA) is the dried immature fruit of the Rutaceae plant Citrus changshanhuyou Y. B. Chang (CSHY). In this study, six flavonoids including narirutin, naringin, hesperidin, neohesperidin, nobiletin, tangeretin were obtained by the ultrasonic-assisted extraction. Subsequently, nine types of natural deep eutectic solvents (NaDESs) were systematically screened and evaluated by the extraction yields determined by high-performance liquid chromatography (HPLC). Based on the single factor experiment, the influences of critical parameters on the extraction rate were investigated. Finally, the extraction process was optimized by orthogonal analysis and genetic neural network (GNN), respectively. Besides, the yields and antioxidant capacities of the extracts derived from the two different optimized processes were measured and compared. Taken together, the results indicated that betaine-lactic acid used as an extractant showed significant advantages among the tested NaDESs. Compared with the optimization results of orthogonal experiments, better extraction yields and antioxidant abilities were produced by GNN. The optimal factors provided by GNN were as follows: ultrasonic power 100 %, ultrasonic temperature 76 °C, ultrasonic time 69 min, water content 36%, liquid-solid ratio 15:1 (mL: g), particle size 70 mesh. The optimum extraction process was simple, feasible, stable and efficient.