Background: Determination of the total phenolic content (TPC) in olive oils is of great interest, as phenolic compounds affect the health benefits, sensory attributes and oxidative stability of olive oils. The aim of this study was to explore the feasibility of direct front-face fluorescence measurements coupled with chemometrics for developing multivatiate models for discrimination between virgin olive oils with low and high TPC and determination of TPC concentration.
Results: Parallel factor analysis and principal component analysis of fluorescence excitation-emission matrices (EEMs) of virgin olive oils revealed different fluorescent properties for samples with low and high TPC. A perfect discrimination of oils with low and high TPC was achieved using partial least squares (PLS) discriminant analysis. The best regression model for the prediction of TPC was based on the PLS analysis of the unfolded entire EEMs (R2 = 0.951, RPD = 4.0).
Conclusions: The results show the potential of fluorescence spectroscopy for direct screening of virgin olive oils for TPC. This may contribute to the development of fast screening methods for TPC assessment, providing an alternative to conventional assays. The procedure is environmentally friendly and fulfils the requirements for green analytical chemistry. © 2018 Society of Chemical Industry.
Keywords: antioxidants; fluorescence; parallel factor analysis; partial least squares regression; virgin olive oil.
© 2018 Society of Chemical Industry.