Comparativo de métodos sensoriais descritivos na avaliação de café torrado e moído

Abstract

Coffee has been considered one of the most valuable commodities, exported worldwide, increasing the demand for high-quality coffees. Among the methods used for coffee quality evaluation, the sensorial evaluation of the beverage stands out. Traditional descriptive methods require extensive training with assessors. In order to reduce the analysis time, this study sought for faster alternative descriptive methods such as Temporal Dominance of Sensation(TDS), Free Choice Profile (FCP) and Check-all-that-apply (CATA). With the evolution of sensory tests, it was also necessary to apply more sophisticated statistical methods for data processing. In this context, sensometric arose. Of these, the Common Dimensions Analysis (ComDim) and Correspondence Analysis (CA) are cited. ComDim was applied to the Free Choice Profile data analysis and then as a new approach to the Temporal Dominance of Sensation data analysis. Correspondence Analysis was applied to CATA data analysis.In addition to sensory analysis, the quantification of compounds such as caffeine, trigonelline,and chlorogenic acid was also performed by high-performance liquid chromatography (HPLC). Therefore, a comparative between the different descriptive sensory methods combined with sensometric and physicochemical analyzes was developed in the evaluation of 12 samples and roasted and ground coffee (A -K).In the FCP, samples A, C, H, I and L were described with the attributes translucent, tea and caramel color, sweet odor, mild taste and weightlessness to the taste. While samples B, D, E, F, G, J and K were characterized by the attributes: color, odor and taste characteristic of the coffee beverage, roasted bean odor and taste, woody odor, bitter taste, residual bitter and astringent.In the TDS test, samples B and L presented higher dominance for the acid attribute; sample D for the astringent attribute; samples B, D, E and F for the bitter attribute; samples A and L for the sweet attribute; C, E and K for the roasted attribute and, finally, the coffee flavor attribute was more dominant in beverages A, C and H.The CATA and Preference-ranking tests were performed with samples A, B,C and K. In CATA, after applying the Correspondence Analysis, four distinct groups were formed. Samples B and K, in the Preference-ranking test, were the samples most preferred by the assessors. Thus, by combining the data obtained through the different analytical methods applied, a sensory description of the 12 coffee beverages was obtained.