关键词: 脱脂牛奶, 风味, GC×GC-TOF MS, PLS-DA

Abstract: Milk plays an important role in a healthy balanced diet. The demand for skimmed buttermilk is increasing as people are becoming more health conscious, however, flavour changes may occur during the skimming process. In this study, headspace solid phase microextraction in combination with comprehensive two-dimensional gas chromatography coupled to time-of-flight mass spectrometry (HS-SPME-GC×GC-TOFMS) technique to detect 49 volatile compounds in whole milk (WM), low fat milk (LFM) and no fat milk (NFM) samples, and it was determined that methyl ketones with odd carbon chains, such as 2-nonanone and 2-undecanone, constituted the main flavour compounds of WM. The results of Partial Least Squares Discriminant Analysis (PLS-DA) showed that the model could well discriminate among the three buttermilk samples and had good variance and cross-validation prediction ability. Next, nine compounds were identified as key aroma difference compounds, which might be the main factors contributing to the different flavours of the three buttermilks, screened by variable importance in prediction (VIP) > 1, p ≤ 0.01 and ≥1%. Finally, clustering heat map results indicated that skimmed buttermilk may have poor organoleptic performance due to the presence of off-flavour compounds (e.g., hexadecanal), whereas the presence of more aroma compounds in whole and low-fat buttermilk resulted in a full and rich aroma in terms of organoleptic properties. This study establishes a research methodology for the analysis of milk by HS-SPME-GC × GC-TOF MS, which provides theoretical guidance for the improvement of flavour and dairy flavouring of dairy products.

Key words: No-fat milk, Flavour, GC×GC-TOF MS, PLS-DA