基于多元素分析的贵州绿茶产地鉴别研究

Study on the origin identification of Guizhou Province green tea based on multielement analysis

中文关键词:  绿茶  微量元素  产地判别  差异性元素  正交偏最小二乘法判别分析

英文关键词:green tea  trace elements  origin discrimination  differentiating element  orthogonal partial least squares discriminant analysis

中文摘要:

      目的 探究以多元素统计对贵州不同产地绿茶判别分析的有效性和可行性, 筛选产地间差异性元素。方法 采用电感耦合等离子体发射光谱法(inductively coupled plasma emission spectrometry, ICP-OES)和电感耦合等离子体质谱法(inductively coupled plasma mass spectrometry, ICP-MS)对贵州4个产地63个绿茶样品中47种元素进行定量分析, 结合正交偏最小二乘法判别分析(orthogonal partial least squares discriminant analysis, OPLS-DA)建立贵州绿茶的产地判别模型。结果 4个产地的绿茶中元素含量有明显的差异; K、P、Ca、Mg、Mn、Al和Fe元素含量规律相同, 说明4个产地的茶叶对土壤中部分高含量元素的富集能力具有一致性; 4个产地的污染物Pb、Cu、Cd、As和Cr的含量均低于茶叶相关标准的限量要求。基于元素含量建立的6组OPLS-DA分析模型可以有效区分产地, 其中黔西南州与铜仁市模型(QXN-TR)参数最优, 该模型用50.3%的变量可解释93.8%的组间差异, 模型预测能力也能达到89.7%; 6组模型中共筛选出20种差异性元素, 在6组产地判别模型中未找到共有元素; 稀土元素在4个产地的判别上贡献有限。结论 综合结果表明, 以多元素分析和统计学模型针对贵州不同产地的绿茶样本进行判别分析是可行的, 整体区分效果良好; 不同产地的绿茶间元素含量的差异不同; 为贵州绿茶产地区分提供了思路和研究基础。

英文摘要:

      Objective To explore the validity and feasibility of multi-element statistics for discriminant analysis of green tea from different origins in Guizhou Province, and screen the differentiating elements among origins. Methods The quantitative analysis of 47 kinds of elements in 63 green tea samples collected from 4 tea-producing counties in Guizhou Province were determined by inductively coupled plasma emission spectrometry (ICP-OES) and inductively coupled plasma mass spectrometry (ICP-MS). The orthogonal partial least squares discriminant analysis (OPLS-DA) were used to develop classification models for the tea samples from different geographic origins. Results The results indicated that there were obvious differences in the elemental content of green tea from 4 origins. The elemental content of K, P, Ca, Mg, Mn, Al and Fe were in the same pattern, which indicated that the tea from 4 origins had the consistency in the enrichment ability of some macronutrients in the soil, and the content of the pollutants Pb, Cu, Cd, As and Cr in 4 origins were lower than the limit requirements of the relevant standards for tea. The 6 groups of OPLS-DA analytical models based on elemental content could effectively distinguish the origins, among which the QXN-TR model had the best parameters, which could explain 93.8% of the inter-group differences with 50.3% of the variables, and the model predictive ability could also reach 89.7%. A total of 20 differential elements were screened in 6 groups of models, and no common elements were found in the 6 groups of origin-discriminating models. The contribution of rare earth elements in the discrimination of 4 origins was limited. Conclusion The overall results suggest that the combination of multielement analysis and statistic model can trace the geographical origins of green tea samples from different regions in Guizhou Province is feasible, and the overall differentiation effect is good. The differences in elemental content among green teas of different origins are different. It provides ideas and research basis for the differentiation of green tea origins in Guizhou Province.