化学
浸渍(污水)
萃取(化学)
生化工程
响应面法
过程(计算)
绿茶
制浆造纸工业
工艺工程
生物炼制
能源消耗
生产(经济)
色谱法
人工神经网络
机制(生物学)
绿茶提取物
计算机科学
生物系统
葡萄酒
高效能源利用
工艺优化
香兰素
作者
Hassan A. Saeed,Rosa H. Andriamamonjy,Yue Zhao,Mingzhe Wang,Tingting Chen,Yong Deng,Balarabe Bilyaminu Ismail,Wenjun Wang,Donghong Liu
摘要
ABSTRACT Huangshan‐Chaoqing (HC) green tea holds significant promise as a sustainable source of bioactive compounds. This study holistically evaluates its valorization through ultrasound‐assisted extraction (UAE), combining process optimization with mechanistic and industrial‐viability assessments. Comparing conventional optimization using response surface methodology (RSM) with machine learning algorithms suggests that an artificial neural network (ANN) model was better at predicting optimal UAE conditions than RSM, yielding 28.7 min, 37.6°C, 36.5% amplitude, and a solvent‐to‐solid ratio of 25.5 mL/g. Under these conditions, the predicted and experimentally verified total phenolic content (TPC) were 385.2 mg/100 g dm and 379.7 mg GAE/100 g dm, respectively, which were 34% and 52% higher than those under heat‐assisted extraction (HAE) and maceration (MAC), respectively. Mechanistic analysis revealed that extraction kinetics followed a two‐stage (phenomenological) model of rapid washing and slow diffusion, while mass‐transfer modeling confirmed internal diffusion as the rate‐limiting step (Bi > 100). An energy efficiency analysis was performed to assess industrial scalability. It revealed that while the lab‐scale UAE system's energy consumption (approximately 1348 kJ/mg GAE) was influenced by ancillary equipment, the core ultrasonic process is highly efficient. These findings provide a comprehensive blueprint spanning optimization, mechanisms, and energy considerations for the sustainable, industrial‐scale production of high‐value extracts from HC green tea.
科研通智能强力驱动
Strongly Powered by AbleSci AI