Comparison of an Ultrasound-Assisted Aqueous Two-Phase System Extraction of Anthocyanins from Pomegranate Pomaces by Utilizing the Artificial Neural Network–Genetic Algorithm and Response Surface Methodology Models
Qisheng Yue, Jun Tian, Ling Dong, Linyan Zhou
Abstract
As a by-product of pomegranate processing, the recycling and reuse of pomegranate pomaces (PPs) were crucial to environmentally sustainable development. Ultrasound-assisted aqueous two-phase extraction (UA-ATPE) was applied to extract the anthocyanins (ACNs) from PPs in this study, and the central composite design response surface methodology (CCD-RSM) and artificial neural network-genetic algorithm (ANN-GA) models were utilized to optimize the extraction parameters and achieve the best yield. The results indicated that the ANN-GA model built for the ACN yield had a greater degree of fit and accuracy than the RSM model. The ideal model process parameters were optimized to have a liquid-solid ratio of 49.0 mL/g, an ethanol concentration of 28 g/100 g, an ultrasonic time of 27 min, and an ultrasonic power of 330 W, with a maximum value of 86.98% for the anticipated ACN yield. The experimental maximum value was 87.82%, which was within the 95% confidence interval. A total of six ACNs from PPs were identified by utilizing UHPLC-ESI-HRMS/MS, with the maximum content of cyanidin-3-O-glucoside being 57.01 ± 1.36 mg/g DW. Therefore, this study has positive significance for exploring the potential value of more by-products and obtaining good ecological and economic benefits in the future.