Waste Cooking Oil Classification through Raman Spectroscopy and Multivariate Analysis

Abstract Five benches of used vegetable oils were prepared by deep frying of commercial edible vegetable oil. Raman analyses of original and waste samples revealed few and small differences between samples. Nevertheless, statistical multivariate analysis, in particular unsupervised principal component analysis (PCA) followed by supervised partial least square discriminant analysis (PLS-DA), conducted on the raw Raman fid allowed to determine a chemical fingerprint charac-teristic of each sample. The technique in principle can be used for quality assessment of waste vegetable oils, and for detection of waste’s adulteration with edible oils.