Wine discrimination based on chemometric analysis of untargeted markers using FT-Raman spectroscopy

In this study, wine discrimination possibilities given by FT-Raman spectroscopy associated with chemometric methods, with respect to variety, vintage and geographical origin was discussed. As far as we know, the potential of Raman spectroscopy to include anti-Stokes range in the wine discrimination was explored here for the first time. The classification of 30 wines, from two sorts, known as Feteasca Regala and Sauvignon Blanc, produced in three Romanian viticulture regions, during five consecutive vintages was performed; an initial and cross-validation percent of 100% being obtained for all three classifications. A control sample set formed by four French samples (2 Sauvignon Blanc and 2 Chardonnay) was used for testing the efficiency of the proposed approach. By testing the control set, a discrimination of wine variety and geographical origin of 100% in both initial and cross-validation was obtained, while for vintage differentiation a percentage of 100% in initial and 94.1% cross-validation was reached.