Stepwise strategy based on 1H-NMR fingerprinting in combination with chemometrics to determine the content of vegetable oils in olive oil mixtures

H-1 NMR fingerprinting of edible oils and a set of multivariate classification and regression models organised in a decision tree is proposed as a stepwise strategy to assure the authenticity and traceability of olive oils and their declared blends with other vegetable oils (VOs). Oils of the 'virgin olive oil' and 'olive oil' categories and their mixtures with the most common VOs, i.e. sunflower, high oleic sunflower, hazelnut, avocado, soybean, corn, refined palm olein and desterolized high oleic sunflower oils, were studied. Partial least squares (PLS) discriminant analysis provided stable and robust binary classification models to identify the olive oil type and the VO in the blend. PLS regression afforded models with excellent precisions and acceptable accuracies to determine the percentage of VO in the mixture. The satisfactory performance of this approach, tested with blind samples, confirm its potential to support regulations and control bodies.

Automated summary

This study utilized H-1 NMR fingerprinting of edible oils and multivariate classification and regression models to accurately identify olive oils and their blends with other vegetable oils, providing support for regulations and control bodies. The models showed stable binary classification for olive oil type and precise determination of percentage of vegetable oil in the mixture. The approach demonstrated satisfactory performance with blind samples, confirming its potential for authenticity and traceability assurance.