Classification of beer by thermogravimetric and chemometric techniques

• Autor(a) principal: Fernandes, Richard Perosa [UNESP]
• Data de Publicação: 2021
• Outros Autores: Ekawa, Bruno [UNESP], Ferreira, Laura Teófilo [UNESP], Carvalho, Ana Carina Sobral [UNESP], Freire, Rafael Teixeira, Caires, Flávio Junior [UNESP]
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.1007/s10973-021-10729-y; http://hdl.handle.net/11449/206119
• topic: Beer; Chemometrics; Thermal analysis

Thermogravimetric (TG) technique combined with partial least squares discriminant analysis (PLS-DA) was applied to distinguish five pilsner beers from different brands (Antarctica, Bohemia, Brahma, Budweiser and Imperio). Herein, we develop a TG methodology that use a small volume (40 µL), and an analysis time of 25 min. The major gases evolved during the thermal decomposition were water, ethanol and carbon dioxide. Energy-dispersive X-ray spectrometer of the ashes detected the major constituents: oxygen, sodium, magnesium, silicon, phosphorus, sulfur, chlorine, potassium and calcium. TG and PLS-DA technique together were able to classify the five brands with a classification rate for the model of 97% with a confidence interval of 92–99%, achieving high sensitivity and specificity between the calibration, cross-validation and predicted results.