Odor approximation of fruit flavors using a QCM odor sensing system

Odor approximations of orange and melon flavors were performed using an array of quartz crystal microbalance (QCM) odor sensors coated with different materials, such as lipids and stationary phase materials of gas chromatography (GC). The steady state sensor responses to single flavor components were measured and the data were used to estimate the important odor components by means of a computational two-level quantization method. From these results, the experimental approximations using the steady state sensor responses and principal component analysis (PCA) were performed. In both cases, orange and melon flavors, it was possible to approximate the recipes with only three odor components. Moreover, the similarities between the approximated odors and the original ones were evaluated by a human sensory test. It was found that these similarities were high, especially in the case of orange flavor.