44C-14 |
Mass transfer modelling of supercritical CO2 fluid extraction of carotenoids from tomatoes by neural networks |
J. SHI1, S. Yang2, H. Li2, and M. Le Maguer3. (1) Food Research & Development Centre, Agriculture & Agri-Food Canada, 93 Stone Rd. W., Guelph, ON N1G 5C9, Canada, (2) School of Engineering, Univ. of Guelph, Guelph, ON N1G 2W1, Canada, (3) Dept. of Food Science, Univ. of Guelph, GUelph, ON N1G 2W1, Canada Increasing clinical evidence suggest lycopene is being considered as an important nutrient, since it appears to provide protection against a broad range of epithelial cancers. Consumer demand for healthful food products provide an opportunity to develop a market for lycopene products. Industrial production of lycopene from tomatoes is in high demand by pharmaceutical companies and for functional food development. Lycopene and other carotenoids were extracted from tomatoes using supercritical CO2 fluid technology. To optimize the extraction process, experiments were conducted at temperature ranging from 35 to 75 °C, pressure ranging from 250 to 350 kg/cm2, fluid rate ranging from 3 to 10 mL/min. The solubility and extraction yield of carotenoids in the supercritical CO2 were determined by HPLC. The mass transfer flux was controlled by pressure, temperature, CO2 flow rate, and material matrix property.The proposed model assumes a three-layer structure with a fast back-propagation learning algorithm. Various temperatures, pressures, and solubility were used to train the proposed model. A novel neural network model is proposed for the mass transfer modelling of supercritical CO2 fluid extraction. A backpropagation network (BPN) was implemented for the modelling of supercritical CO2 extraction, and tested on data both from experiments and from literature. The prediction errors associated with the BPN were less than 3%. The results showed that the network was consistent in predicting the solubility of carotenoids in supercritical CO2, and is appropriate for application to the problem of supercritical CO2 fluid extraction modelling.
Session 44C, Fruit & Vegetable Products: Chemistry
|