Near Infrared Spectroscopy and Pattern Recognition Methods Applied to the Classification of Vinegar According to Raw Material and Elaboration Process

Abstract

Near infrared (NIR) spectroscopy is used to classify vinegar samples according to the raw material of origin (white, red, aged and balsamic wine, malt, cider and molasses) and elaboration processes such as ageing in wood or must addition. Balsamic vinegar from Modena and Sherry wine vinegar are also included in the study; these vinegars are sometimes aged using a battery of barrels or “ soleras and criaderas” system that give the final product a special character. The differences between these samples and the other vinegar samples are also described in this study. Multivariate analysis techniques, cluster analysis, minimum-spanning tree, potential functions and discrimination via regression methods, apart from soft independent modelling of class analogy (SIMCA) and unequal class models (UNEQ) class-modelling techniques, were used to extract the information from the NIR spectral variables. These methods enable the development of models to classify the different groups of vinegar samples and provide accurate information on the specific characteristics of the vinegars related to the elaboration process of the several raw materials. A high percentage of correct classifications were obtained for wine, cider, malt, Balsamic and Sherry vinegars, ranging from 85.7 to 100%. Molasses vinegar was grouped with non-aged wine or cider samples depending on its acetic grade.