Protected Geographical Identification of Honey by Spark Discharge-assisted Laser-induced Breakdown Spectroscopy

Abstract

Honey is a natural food that is valued worldwide for its nutritional and therapeutic values. Therefore, authentication of honey according to the geographical origin is a guarantee of the genuine properties. In this article, an evaluation of spark discharge-assisted laser-induced breakdown spectroscopy (SD-LIBS) for certification of the geographical origin of honey is reported. Forty-nine samples of multifloral honey produced in four Argentine provinces were considered. The results showed the best classification performance was obtained using smoothing, generalized least squares weighting (GLSW) and mean centering for spectral preprocessing, added to the k-nearest neighbor (k-NN) or Support Vector Machine (SVM) classification algorithms, which provided 100% of correct classification. More importantly, the results of Partial Least Squares - Discriminant Analysis (PLS-DA) pointed to N, Ca, K, Cu, Fe and Mn as key elements for the certification of geographical origin. In addition, the greatest potential of N stands out for the discrimination of the origin of honey. These findings confirm SD-LIBS as a promising tool for authentication of honey quality, providing a simple, fast and environmentally friendly solution. The method can be useful for industry, the market and others related to food authenticity.