Application of ionic liquid ultrasound-assisted extraction (IL-UAE) of lycopene from guava (Psidium guajava L.) by response surface methodology and artificial neural network-genetic algorithm

Junping Wang; Hongyi Zhao; Xuexue Xue; Yutong Han; Xin Wang; Zunlai Sheng

doi:10.1016/j.ultsonch.2024.106877

Application of ionic liquid ultrasound-assisted extraction (IL-UAE) of lycopene from guava (Psidium guajava L.) by response surface methodology and artificial neural network-genetic algorithm

Ultrason Sonochem. 2024 Apr 16:106:106877. doi: 10.1016/j.ultsonch.2024.106877. Online ahead of print.

Authors

Junping Wang¹, Hongyi Zhao¹, Xuexue Xue¹, Yutong Han¹, Xin Wang¹, Zunlai Sheng²

Affiliations

¹ College of Veterinary Medicine, Northeast Agricultural University, Harbin 150030, PR China; Heilongjiang Key Laboratory for Animal Disease Control and Pharmaceutical Development, Harbin 150030, PR China.
² College of Veterinary Medicine, Northeast Agricultural University, Harbin 150030, PR China; Heilongjiang Key Laboratory for Animal Disease Control and Pharmaceutical Development, Harbin 150030, PR China. Electronic address: shengzunlai@neau.edu.cn.

Abstract

Lycopene-rich guava (Psidium guajava L.) exhibits significant economic potential as a functional food ingredient, making it highly valuable for the pharmaceutical and agro-food industries. However, there is a need to enhance the extraction methods of lycopene to fully exploit its beneficial uses. In this study, we evaluated various ionic liquids to identify the most effective one for extracting lycopene from guava. Among thirteen ionic liquids with varying carbon chains or anions, 1-butyl-3-methylimidazolium chloride demonstrated the highest productivity. Subsequently, a single-factor experiment was employed to test the impact of several parameters on the efficiency of lycopene extraction using this selected ionic liquid. These parameters included extraction time, ultrasonic power, liquid-solid ratio, concentration of the ionic liquid, as well as material particle size. Moreover, models of artificial neural networks using genetic algorithms (ANN-GA) and response surface methodology (RSM) were employed to comprehensively assess the first four key parameters. The optimized conditions for ionic liquid ultrasound-assisted extraction (IL-UAE) were determined as follows: 33 min of extraction time, 225 W of ultrasonic power, 22 mL/g of liquid-solid ratio, 3.0 mol/L of IL concentration, and extraction cycles of three. Under these conditions, lycopene production reached an impressive yield of 9.35 ± 0.36 mg/g while offering advantages such as high efficiency, time savings, preservation benefits, and most importantly environmental friendliness.

Keywords: Artificial neural network-genetic algorithm (ANN-GA); Ionic liquids; Lycopene; Psidium guajava L.; Response surface methodology (RSM); Ultrasound-assisted extraction (UAE).