Label-Free Detection of Adulteration in Extra Virgin Olive Oils from the Mut (Mersin) Region Using ATR–MIR Spectroscopy Integrated with Chemometric Approaches

Abstract This research investigates the use of mid-infrared (MIR) spectroscopy combined with chemometrics for the label-free analysis of adulteration in extra-virgin olive oils (VOOs) from the Mut (Mersin) region. Overall, 301 unadulterated oil samples and adulterated VOO samples with pomace oil (PO) and Riviera olive oil (ROO) samples were assessed. The MIR data were processed using both supervised and unsupervised multivariate analyses, including linear discriminant analysis (LDA), soft independent modeling of class analogies (SIMCA), hierarchical cluster analysis (HCA), principal component analysis (PCA), and partial least square-regression (PLS-R), by utilizing the selected spectral regions. The SIMCA models predicted exceptional grouping for different oil types. The PLS-R graphs showed perfect predictions ( R 2 > 0.9992), and the predicted calibration and validation parameters were observed to be 0.4894%–2.7660% and 0.8906%–3.3628%, correspondingly. Therefore, the MIR methods, along with chemometrics worked out in this research, can be used to quantify adulterant at levels < 0.49% in unknown oil samples.