Characterization of beers by liquid chromatography and chemometric data analysis

Abstract

Polyphenols are a group of substances commonly found in food and drinks which not only are responsible for organoleptic properties and have beneficial effects in health but also can be used to discriminate among samples from different regions and styles. In this project 63 beers from different styles and countries have been characterized based on their polyphenolic profile, which has been obtained by high performance liquid chromatography with UV detection at different wavelengths (280, 310, 370, 420 and 520 nm). A total of 13 polyphenols have been quantified. Preliminary studies have shown that neither the container nor the batch affects the polyphenolic profile and that degradation of polyphenols is not noticeable for a period of two weeks. Matrix effect was not observed. Hence, quantification has been carried out by external calibration curve. Three different approaches (chromatograms, areas and concentrations) have been used to build principal component analysis (PCA) models in order to discriminate between ale and lager beers but only the concentrations approach has been capable of successfully discriminate between both types of beer. Afterwards, the 13 polyphenols concentrations have been used to build a partial least squares discriminant analysis (PLS-DA) model which has allowed to discriminate beers according to their type and correctly assign new samples. Finally, a PLS-DA model has also been built in order to discriminate lager beers according to its country of origin (Spain and Czech Republic).