Effect of roast degree, preparation method, and variety in the sensory and chemical characteristics of coffee (coffea arabica): a mid-infrared spectrum analysis

Abstract

In coffee beverages, there are several factors that affect the final compounds and generate sensory variations. This study evaluated the effect of five preparation methods, three roast degrees, and three different varieties (coffea arabica) on the physicochemical compounds of coffee (coffea arabica) be fore and after preparation by using information obtained from the mid-infrared spectrum. The effect on some sensory attributes was assessed by means of a panel of 54 habitual coffee consumers. Spectrum data were processed using hierarchical clustering and principal component analysis (PCA), while a mixed general linear model was applied for sensory analysis. The results showed that each factor behaves independently, showing a significant effect (p < 0.05) on a greater number of attributes. The preparation method and the roast degree are attributed to the changes generated in the chemical character istics of coffee during these processes. Through the analysis of the infrared spectrum (IR) by hierarchical cluster, it was found that, before the preparation of the coffee drinks, the samples are grouped by roast degree, regardless of the type of variety. Spectrum analysis by PCA after brewing indicated that there is a greater effect of the roast degree and variety of coffee (coffea arabica) on the chemical markers of the IR spectra. Finally, wavelengths 1,800, 1,740, 1,650, 1,550, 1,480, 1,150, and 710 cm-1, which had a greater contribution in the PCA and allowed the evaluated samples to be grouped differently, are those related to non-volatile compounds such as caffeine, chlorogenic acids, and trigonelline, which provide sensory characteristics to the final drink.