Characterization of the key odorants of fennel essential oils of different regions using GC-MS and GC-O combined with partial least squares regression.

The key odorants and the aroma profile of six fennel essential oils from different regions were investigated by using gas chromatography-olfactometry (GC-O), gas chromatography-mass spectrometry (GC-MS) and sensory evaluation. A total of 30 volatile compounds were determined by GC-O with aroma extract dilution analysis (AEDA) and the odor activity values (OAV) of them were calculated. Among these compounds, α-pinene, α-phellandrene, limonene, α-cubebene, β-caryophyllene, estragole, α-humulene, trans-anethole, δ-cadinene and p-anisaldehyde contributed greatly to the aroma of fennel essential oil due to their high flavor dilution (FD) factors and high OAVs. The aroma of fennel essential oils was described by 7 sensory terms as spicy, woody, grassy, floral, musty, sweet and green, and was correlated to the key odorants by partial least squares regression (PLSR). It was showed that spicy, woody, grassy, musty and green attributes were covaried well with aroma compounds. In addition, hierarchical cluster analysis (HCA) was used to find out the similarities among different samples and the result indicated that three main groups were identified.