关键词: 浙江玫瑰醋, 固相微萃取技术, 主成分分析, 系统聚类分析, 偏最小二乘-判别分析

Abstract: In this study, response surface methodology was used to optimize the solid-phase micro-extraction conditions of volatile components from Zhejiang rosy vinegar. The optimal conditions were determined as follows: NaCl amount 2.4 g, extraction temperature 49.6 ℃, and extraction time 44.4 min. The volatile components during the fermentation process of Zhejiang rosy vinegar were determined under the optimized extraction conditions. The obtained data were analyzed by principal component analysis (PCA), systematic clustering analysis (SCA) and partial least square discriminant analysis (PLS-DA). PCA and SCA divided Zhejiang rosy vinegar samples into three categories: early, middle and late fermentation. PLS-DA analysis showed that the samples at the three fermentation stages could be well separated from each other based on the volatile components with R2Y = 0.994, and Q2 = 0.933, indicating that the developed model has good stability and extremely strong prediction ability. By PLS-DA analysis, the volatile components with variable importance in the projection (VIP) values larger than 1 were obtained. According to the PLS-DA factor loading plot, the characteristic aroma components at the early stage were ethanol, ethyl dodecanoate and 2-phenylethanolacetate; the characteristic aroma components at the middle stage were propionic acid-2-phenyl ethyl, cyclohexyl propionate, n-propanol, n-propyl acetate, isoamyl acetate, decanoic acid, ethyl acetate, ethyl phenylacetate, lauric acid, 2-ethylbutyrate, 3-methyl phenyl ester and ethyl oleate; and the characteristic aroma components at the late stage of fermentation were (E)-9-octadecanoic acid ethyl ester, ethyl octadecanoate, benzaldehyde, 4-ethyl phenol, diethyl succinate, ethyl hexadecate, furfural, acetate-2-ethylhexyl ester, phenyl ethanol, 2-methylpropionic acid, and 3-hydroxy-2-butanone.

Key words: Zhejiang rosy vinegar, solid phase microextraction, principal component analysis, hierarchical cluster analysis, partial least squares-discriminant analysis