Authentication of edible oils using Fourier transform infrared spectroscopy and pattern recognition methods

Abstract

A potential method to determine whether two edible oils can be differentiated by infrared spectroscopy is proposed. IR spectra of the pure edible oils and mixtures of these edible oils in known amounts are compared using pattern recognition techniques to solve a ternary classification problem. The edible oil mixtures span a large concentration range. If the IR spectra of the two edible oils and their binary mixtures are differentiable then differences between the IR spectra of the two edible oils are of sufficient magnitude to ensure that a reliable classification of these two edible oils can be obtained by infrared spectroscopy. The mixtures were prepared gravimetrically or from digitally blended data of the pure edible oils using an edible oil spectral library. The feasibility of authenticating edible oils such as extra virgin olive oil was demonstrated using this approach. For these studies, both digital and experimental data were combined to generate training and validation data sets to assess detection limits for adulterants.