Application of IR and UV-VIS spectroscopies and multivariate analysis for the classification of waste vegetable oils

Due to the ever-increasing worldwide interest in the exploitation of waste vegetable oils, the development of analytical tools able to detect their adulteration with edible oils, is considered a priority for the scientific and industrial community. In this work, edible and waste vegetable oils have been analysed by Fourier TransformInfraRed (FT-IR) and Ultraviolet-Visible (UV-VIS) spectroscopies and the corresponding spectral data subjected to statistical multivariate analysis for classification purposes. In particular, Principal Component Analysis (PCA) and Partial Least Square Discriminant Analysis (PLS-DA) were performed in order to develop an analytical tool which is able to distinguish between edible and waste vegetable oil. Qualitative analysis of the spectra suggested FT-IR and UV-VIS as the more suitable techniques to distinguish between wastes and edible samples. Also, statistical multivariate analysis revealed that FT-IR-based methodology is more adequate for the target, even if the elevated sensibility of the method produces an undesired distinction between edible oils of the same type. Finally, further attempts on UV-VIS data obtained in reflection mode allowed to produce a good dataset which after statistical treatment gave a clear differentiation between edible and waste oil samples.

Automated summary

In this study, FT-IR and UV-VIS spectroscopy were used to develop an analytical tool for distinguishing between edible and waste vegetable oils. Qualitative analysis indicated that FT-IR and UV-VIS were more suitable techniques for distinguishing between waste and edible samples. Statistical multivariate analysis revealed that the FT-IR-based method was more appropriate, despite the undesired distinction between edible oils of the same type. Furthermore, attempts on UV-VIS data obtained in reflection mode successfully differentiated between edible and waste oil samples.