Recent food trends, such as the "New Nordic Kitchen" have focussed on the sensory qualities of regionally available raw materials. Current knowledge of variations in wheat sensory attributes is lacking. Especially lacking is knowledge of how wheat sensory attributes and volatile compounds may be connected. This Ph. D. project aims to study sensory attributes and volatile compounds of wheat and its products in order to investigate variations between wheat species, landraces and modern varieties and to reveal their impact on bread odours and flavours. Furthermore to examine whether bread could be replaced by, easier to produce, wheat porridge or cooked grain in future sensory analyses.
Sixty four field- and 81 greenhouse cultivated wheat varieties were used. Wheat grain volatiles were extracted by Dynamic Headspace Extraction (DHE) and trapped volatiles were analysed by Gas Chromatography-Mass Spectrometry (GC-MS). Sensory studies were conducted of cooked grain, wheat porridge and bread using a generic version of Quantitative Descriptive Analysis (QDA). Multivariate methods: Partial Least Squares (PLS), Partial Least Squares-Discrimination Analysis (PLS-DA) and Principal Component Analysis (PCA) were used to analyse sensory and GC-MS data.
Differences in odours and flavours were found in all wheat products. Variation also occurred between volatile peak-areas of wheat grain- and bread samples. Twenty four selected wheat species, landraces and varieties were evaluated by trained sensory panels as cooked wheat grain, flour porridge and for eight of these varieties, baked bread. Descriptors for odours of cocoa, oat porridge and øllebrød1) and flavours of sweet, bitter, oat porridge and øllebrød1) were common to all three wheat products. Wheat porridge shared 6 odour and 10 flavour descriptors with bread.
Furthermore variations in descriptors for “cocoa” and “oat porridge” correlate between wheat porridge and bread samples. Cooked grain shares 5 odour and 6 flavour descriptors with bread (Paper III). This indicates that wheat flour porridge would best substitute bread, although cooked grain could also be used, in future sensory evaluations of new wheat varieties. Whole-meal bread is mostly described by stronger descriptors of cocoa, grain, hazelnut, øllebrød, bitter, sour and salt indicating that the causes for these descriptors are mostly in the bran. Low-extraction bread made from 4 wheat varieties was mostly described by descriptors for odour and flavour of maize, bulgur, oat porridge and sweet. Low-extraction bread is also influenced by variations between wheat varieties. Five odour and 6 flavour descriptors differentiated the 4 low extraction bread samples (Paper III). Eighty eight volatile compounds were identified in whole-meal bread crumb and 30 of these were common to wheat grain (Paper II). Of these 30, 16 are reported in literature as having low odour thresholds that could potentially have an impact on bread sensory attributes. Twenty two volatile compounds were tentatively associated with bread sensory descriptors including: 2-pentylfuran, benzaldehyde, 6-methyl-5-hepten-2-one, 2-methyl-1-butanol, 3-methyl-1-butanol, 2-methylbutanal, 2-nonenal, hexanal, and hexanol. These are reported in literature as being important in bread and are also found in wheat grain (Paper II). Seventy two volatile compounds were identified in the grain of 81 wheat varieties (Paper II). Out of these, 7 selected wheat volatile compounds were significantly varied among 14 wheat varieties, indicating huge variation in volatile compound profiles among wheat varieties. Multivariate analysis showed that several wheat samples retained their configuration of distribution throughout the sensory tests. The same varieties also retained the same distribution configuration when analysed for volatile compounds which could link volatile profiles to sensory evaluation results (Papers II and III). Landraces were distinguishable from modern varieties and varieties from Austria could be distinguished from Danish, French and British varieties based on volatile profiles. This suggests that wheat volatile composition has genetic causes.
The results in this study provide a strong case that there is wide variation among wheat varieties for odour and flavour and that the odour and flavour of bread crumb is influenced by the odour and flavour of the wheat variety used. They infer a huge potential for extensive wheat odour and flavour variation globally. These results could benefit consumers and producers of wheat based products and could be of use to plant breeders. Further studies of odour and flavour variations in wheat should be made of wheat grown in other wheat growing regions or the world in order to reveal the extent of wheat odour and flavour variation.