Excitation-emission matrix fluorescence spectroscopy combined with multi-way chemometric methods for rapid qualitative and quantitative analyses of the authenticity of sesame oil

• Published in: European food research & technology 2023-08, Vol.249 (8), p.2087-2099
• Main Authors: Song, Jia-Yu, Gu, Hui-Wen, Wang, Yan, Geng, Tao, Cui, Hui-Na, Pan, Yuan, Ding, Baomiao, Li, Zhenshun, Yin, Xiao-Li
• Format: Article
• Language: English
• Subjects: Agriculture; Analytical Chemistry; Biotechnology; Chemistry; Chemistry and Materials Science; Chemometrics; Discriminant analysis; Edible oils; Emission spectroscopy; Emissions; Excitation spectra; Fluorescence; Fluorescence spectroscopy; Food Science; Forestry; Least squares; Nutritive value; Original Paper; Predictions; Qualitative analysis; Quantitative analysis; Regression; Sesame oil; Soybean oil; Soybeans; Spectroscopy; Spectrum analysis; Vegetable oils; Vegetables
• ISSN: 1438-2377; 1438-2385
• DOI: 10.1007/s00217-023-04275-0

Sesame oil (SO) is a high-quality and very popular edible oil with pleasant odor and good taste, which has higher economic and nutritional value than other common vegetable oils. However, there is no effective and standardized method for detecting adulterated sesame oil, so there is an urgent need to develop a rapid and effective method for the detection of adulterated sesame oil. In this work, a strategy based on the excitation-emission matrix (EEM) fluorescence spectroscopy combined with multi-way chemometric methods was proposed to distinguish the plant origin of vegetable oils, identify adulterated sesame oil and quantify the level of adulteration of soybean oil (SBO) in SO. Three kinds of multi-way chemometric methods were used in this work. They are alternating trilinear decomposition coupled with partial least squares discriminant analysis (ATLD-PLS-DA), multilinear partial least squares discriminant analysis (N-PLS-DA) and unfolded partial least squares discriminant analysis (U-PLS-DA). The total recognition rates (TRRs) of plant origin of edible oils for the testing set were 79.17%, 100% and 100%, respectively, based on ATLD-PLS-DA, N-PLS-DA and U-PLS-DA model. For the identification of adulterated sesame oil, the TRRs for the testing set were 93.10%, 100% and 100%, respectively. What’s more, N-PLS regression and U-PLS regression were used for predicting the adulteration amount of SBO in SO. Satisfactory prediction accuracy with RMSEV = 2.74% was obtained by U-PLS regression. Therefore, it can be concluded that EEM fluorescent fingerprint combined with multi-way chemometric methods is an effective strategy to classify the plant origin of vegetable oils, identify the adulteration of SO and quantify the level of adulteration.