[关键词]
[摘要]
本研究为探讨稀土元素指纹分析对于葡萄酒原产地判别的可行性,采用电感耦合等离子体质谱法(inductively coupled plasma-mass spectrometry,ICP-MS)测定了三个原产地228个葡萄酒中的15种稀土元素含量,并对数据进行相关性分析、方差分析和判别分析。葡萄酒原产地和稀土元素含量间显著相关(p<0.01),不同原产地间存在显著性差异(p<0.001),沙城葡萄酒中稀土元素含量最低,通化最高,贺兰山东麓居中。Fisher线性判别分析(fisher linear discriminant analysis,FLD)模型对沙城、贺兰山东麓、通化三产地的交叉验证判别率分别为92.98%、98.25%、100.00%,外部验证判别率分别为84.21%、89.47%、100.00%;偏最小二乘判别分析(partial least squares-discriminant analysis,PLS-DA)模型判别能力略差,三个产地的交叉验证判别率分别为82.46%、98.25%、91.23%,外部验证判别率仅为73.68%、84.21%、100.00%。说明稀土元素含量结合多元统计分析可以应用于葡萄酒原产地的鉴别。
[Key word]
[Abstract]
In this study, in order to explore the feasibility of using rare earth element fingerprinting to identify wine of origins, inductively coupled plasma-mass spectrometry (ICP-MS) was employed to measure the contents of 15 rare earth elements in 228 wine samples from three origins. The data were analyzed by correlation analysis, analysis of variance (ANOVA), and discriminant analysis. The wine of origin was significantly correlated with the rare earth element contents (p < 0.01), while the rare earth element contents were significantly different for the three origins (p < 0.001). The highest rare earth element content was found in the wines from Tonghua, followed by wines from the eastern foot of the Helan Mountains, and the lowest was found in the wines from Shacheng. The cross-validated accuracy rates of the wines from Shacheng, the eastern foot of the Helan Mountains, and Tonghua, using Fisher linear discriminant analysis (FLD), were 92.98%, 98.25%, and 100.00%, respectively, while the accuracy rates by external validation were 84.21%, 89.47%, and 100.00%, respectively. The discrimination capacity of partial least squares discriminant analysis (PLS-DA) was slightly lower; the cross-validated accuracy rates of the three wine origins were 82.46%, 98.25%, and 91.23%, respectively, while the accuracy rates by external validation were 73.68%, 84.21%, and 100.00%, respectively. Therefore, it is possible to identify wine of origins by rare earth element fingerprinting and multivariate statistical analysis.
[中图分类号]
[基金项目]
国家公益性行业科研专项(2012104019-3);国家十二五科技支撑计划(2012BAD31B07)