Avaliação comparativa da acidez de cafés espresso em cápsulas por espectroscopia no infravermelho próximo, cromatografia líquida de alta eficiência e análise sensorial

• Ano de defesa: 2018
• Autor(a) principal: Mariana Helena Cardoso Costa
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Ciência de Alimentos; Análise sensorial; Café Análise; Cromatografia líquida de alta eficiência; Espectroscopia de infravermelho
• Link de acesso: http://hdl.handle.net/1843/BUOS-B59NCG

The objective of this work was to use Near Infrared Spectroscopy (NIRS), acid analysis by High Performance Liquid Chromatography (HPLC) and sensorial analysis to differentiate coffee marketed in capsules of different brands, according to their acidity characteristics. The HPLC analysis was performed for the quantification of chlorogenic acid and citric acids, the sensorial analysis was carried out by the evaluation of the acidity, aroma and flavor attributes, and a titratable acidity analysis was also performed. Chemometric analyzes such as Principal Component Analysis (PCA), Partial Least Squares Regression (PLS) and Discriminant Analysis by Partial Least Squares Regression (PLS-DA) were used. The PCA analysis of the NIR spectra allowed the separation of different brands, as well as the separation by roasting intensity, showing the ability of the spectroscopic technique to differentiate the samples analyzed. The titratable acidity analysis presented values from 63.30 ± 6.99 to 112.85 ± 2.83 mL of NaOH 0,1M 100 g-1 of sample. The average values of chlorogenic acid and citric acid quantified by HPLC were 0.454 ± 0.176 g100g-1 and 0.853 ± 0.348 g100g-1. The quantitative models obtained by PLS for the sensorial attributes of acidity, aroma and flavor presented RMSEC values of 0.141, 0.128 and 0.226 and RMSEP values of 0.194, 0.199 and 0.362, respectively. For chlorogenic acid and citric acid the RMSEC values were 0.035 and 0.145 and RMSEP of 0.048 and 0.139, respectively. The comparison between the PLS models for acidity and chlorogenic acid showed the potential to use the PLS model of chlorogenic acid to classify the coffee samples for the acidity attribute, showing that this compound is directly related to acidity perceived by consumers, and that NIR spectroscopy is able to differentiate the samples according to the chlorogenic acid content