A Feasibility Study of the Rapid Evaluation of Oil Oxidation Using Synchronous Fluorescence Spectroscopy

Synchronous fluorescence spectroscopy coupled with nonlinear regression analysis was developed and used to discriminate the degree of oil oxidation based on the acid and peroxide values. It is easy to discriminate the degree of oil oxidation after treatments with different heat treatment times and styles for both synchronous fluorescence spectroscopy and low-field nuclear magnetic resonance (NMR). The low-field NMR results proved that oil oxidation probably proceeds via more than one pathway for different treatment styles. To decompose the three-dimensional matrix into two-dimensional data, a parallel factor analysis (PARAFAC) algorithm was used to select an optimized of 120nm. Then, an artificial neural network (ANN) was used to build a regression model for both synchronous fluorescence and low-field NMR to evaluate the degree of oil oxidation. Compared with other regression results, synchronous fluorescence coupled with the ANN model achieved the best results, with 1.00 in the training set and 1.00 in the acid value test set at room temperature. The overall results suggest that synchronous fluorescence spectroscopy coupled with the ANN regression algorithm is useful for rapidly evaluating oil quality and perhaps foods with high oil contents.