Enhancing the fluorescence spectrum of frying oil using a nanoscale probe.

This study attempted to develop a probe-based fluorescence technology as a rapid method to discriminate the oxidation degree of oil. The fluorescence probe was made by dissolving the selected probe in a chloroform solvent. A fluorescence landscape profile for each oil sample was obtained to discriminate fluorescence changes during the oxidation process. Parallel factor analysis (PARAFAC) was performed to extract characteristic fluorescence peaks. Then, these characteristic fluorescence peaks and chemical indicators were used to build regression models based on support vector regression (SVR). The correlation coefficients (R) and mean squared error (MSE) in the prediction sets were applied as the assessment parameters for the SVR models as follows: R = 0.9724, MSE = 0.0088 for an acid value. This research suggests that probe-based fluorescence spectroscopy is useful for oxidation evaluation of oil and perhaps other food containing high amounts of oil.