Go to page

Bibliographic Metadata

 The document is publicly available on the WWW


Peru und Bolivia dispose of a unique variety of Capsicum accessions mainly from the species C. baccatum and C. chinense that have not been characterized by its biochemical and sensory traits. This research aims to study the relationship between sensory attributes like sweet, sour, and bitter as well as the content of sugars and organic acids obtained by instrumental analysis in Capsicum powders. Prior to the analysis, the capsicum fruits were harvested, solar or oven-dried, milled in their native countries, and then sent to the University of Wuppertal. After a basic and product-specific training, a sensory descriptive Panel of 14 judges was established. 31 native chili powders were analysed in six different product sets by their olfactory (pro- and retronasal), gustatory and trigeminal sensations. The sensory profiling method used was Free Choice Profiling (FCP), in which each assessor created and evaluated its own list of descriptive terms without agreeing in the meaning of the attributes with the rest of the panel. The sensory data was evaluated by Generalized Procrustes Analysis (GPA). GPA creates a consensus sample space, in which the highest statistical agreement among the assessors is presented. FCP was a quick and inexpensive method that could be performed with the small available amount of powder. It mainly reveals the interrelationships (similarities and differences) among the samples of each set, and provides a coarse characterization of the samples. Regarding the characterization, the most agreement was observed by the attributes: sweet, bitter, pungent and burning. Besides that, a few olfactory characteristics were able to be interpreted for some sample groups such as acidic, caramel-like, chocolate-like, toasted/smoky, musty and fruity. Due to the drying process, typical odour notes of fresh Capsicum fruits such as fresh, flowerly, and green bell pepper were not observed. Moreover, a rapid instrumental method was developed to simultaneously detect sugar and organic acids. Fructose, glucose, sucrose, citric acid and malic acid were first formed into trimethylsilyl derivates with an oximation and silylation step, and then analysed by gas chromatography (GC) with a flame ionization detector (FID). The total sugar content (sum of fructose, glucose and sucrose) and the total organic acid content (sum of citric acid and malic acid) of 191 native samples (involved in this project) were quantified. Following concentration ranges in Capsicum powders were observed: 0,6-17,8 g glucose/100 g dry weight, 0,9-33,6 g fructose/100 g, 0-7,0 g sucrose/100 g, 0,2-8,3 g citric acid/100 g and 0-5 g malic acid/100 g. It was also possible to show the variability of these analytes in dry material for the species C. baccatum and C. chinense, which have not been investigated yet. With the developed method, the 31 sensory analysed samples were also characterized by their sugar and organic acid contents in order to prove a correlation between the sensory and instrumental data. Two correlation methods were tested: GPA and linear regression. Both methods showed that there is a positive correlation between the attribute sweet and the total sugar content in Capsicum powders. However, there was no correlation between the sourness and the total organic acid content. The attributes pungent and burning also were positively correlated with the total capsaicinoid content. Only the linear regression model revealed that a considerable linear relationship was present in the three cases. As a result, a rough forecast of the sweetness, pungency, and burning sensation with the linear equations based on the instrumental data of new samples, can be generated.