1H NMR spectroscopy combined with multivariate data analysis for differentiation of Brazilian lager beer according to brewery

Abstract

In this study, 1H NMR spectroscopy was used to classify samples of beer, considering three categories (Ambev, Heineken, and Grupo Petrópolis), employing chemometric methods: principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), and soft independent modeling of class analogies (SIMCA). The full NMR spectra were evaluated, although only the aliphatic region (0–3 ppm) was used for multivariate analysis, since it provided superior results, compared to the use of other regions or the full spectrum. It was necessary to use an alignment procedure to eliminate small deviations in the chemical shifts caused by variations of pH and intermolecular interactions. Organic acids (lactic, acetic, and succinic acids) were the chemical compounds most susceptible to these variations. In the PCA, the first two components explained 82.1% of the variability of the dataset, while PLS-DA and SIMCA both provided accuracy higher than 92% in the prediction sets.