Partial Least Squares and Partial Least Squares-Discriminant Analysis to Detect and Quantify Adulterations in Olive Oil Using Optical Methods of Analysis. A Comparative Study between NIR and UV-Vis Spectroscopy

Due to its high nutritional value, olive oil is more expensive than most of other vegetable oils, making it the most common edible oil to be adulterated by oil suppliers in small percentages to obtain economic benefits. These fraudulent practices become more sophisticated every day, what gives rise to the necessity of the implementation of new more innovative and efficient methodologies. For that reason, in this work, the great advantages of some optical methods of analysis combined with chemometric techniques are used to study the possibilities they present in the determination of adulterants in olive oil.The main objective of the study is to evaluate the presence of five different adulterants (vegetable oils) in olive oil using NIR and UV-Vis-NIR spectroscopy combined with chemometric techniques. On the one hand, a qualitative analysis was performed using Partial Least Squares-Discriminant Analysis, and on the other hand, Partial Least Squares models were built in order to quantify the adulterants. Using these techniques, it is possible to classify between pure olive oil samples and those that have been adulterated. In addition, good estimation of the adulterant percentage in olive oil is achieved, obtaining the best results when UV-Vis-NIR spectroscopy is used.