Combinando modelos quimiométricos e portabilidade para o controle de qualidade de cerveja e outras bebidas fermentadas

• Ano de defesa: 2023
• Autor(a) principal: Ana Carolina da Costa Fulgêncio
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; ICX - DEPARTAMENTO DE QUÍMICA; Programa de Pós-Graduação em Química; UFMG
• Palavras-chave em Português: Cerveja; Quimiometria; Controle de qualidade cervejeiro; Portabilidade; Miniaturização; Beer; Chemometrics; Brewery; Portability; Miniaturization; Química analítica; Cerveja – Indústria – Controle de qualidade; Mínimos quadrados; Calibração; Espectroscopia de infravermelho; Bebidas fermentadas – Indústria; Espectrofotômetro; Análise de componentes principais; Processamento de imagens – Técnicas digitais; Análise multivariada; Preparação de amostra (Química)
• Link de acesso: http://hdl.handle.net/1843/51383

This thesis combined portability and chemometrics in the development of new analytical methods for the quality control of beer. The beer market moves an extensive network that aggregates from the research, cultivation, processing, and marketing of inputs and raw materials to the delivery of the finished product. Given this fact, it is necessary to consider quality control as a fundamental step in the brewing industry. In this study, new simple, rapid, robust, and reliable analytical were developed. Color is the first sensory characteristic of beer that draws the consumer's attention, and thus was chosen as the initial parameter to be determined. The developed method combined digital images recorded by a smartphone and multivariate calibration (PLS, partial least squares) for the color determination of beers. The following two applications aimed at determining the alcohol content using different strategies and matrices. For this aim, a portable near-infrared (NIR) spectrophotometer was applied in combination with multivariate calibration (PLS). The second application focused on the brewing industry, thus developing a method for determining the alcohol content in beers of different styles. On the other hand, the third application focused on regulatory issues, thus developing a multi-product quantitative model for four fermented beverages (beer, wine, cider and mead). The last application combined portable NIR spectroscopy and PLS discriminant analysis (PLS-DA) to develop a fast screening method for the direct detection of diethylene glycol in beer. This is a forensic application based on a recent tragic occurrence of poisoning of dozens of people, including some deaths, by consuming beer contaminated with this substance. Multivariate analytical validation was carried out for all developed methods. The dissemination of portability opens up new perspectives. It can change the way users communicate with science, opening new markets and being able to be incorporated into consumer devices and smart appliances, becoming part of the Internet of Things in the coming years, and contributing to the democratization of analytical chemistry.