西藏藏香中34种元素含量测定及产地判别分析

Determination of 34 Elements in Tibetan Incense and Geographical Origin Discrimination Analysis

通过分析不同区域燃香元素含量特征,结合化学计量学建立藏香产地判别模型,建立了微波消解-电感耦合等离子体质谱(ICP-MS)法测定藏香中34种元素含量的方法,对西藏藏香和区外燃香共67份样品进行测定,结合正交偏最小二乘判别分析(OPLS-DA)、线性判别分析(LDA)以及主成分分析(PCA)对数据进行了统计分析。结果表明,34种元素在相应浓度范围内,线性回归方程的相关系数r^(2)在0.999 5以上,方法的检出限为0.024~0.48μg/kg,方法的定量限为0.080~1.60μg/kg。在低、中、高三个不同添加水平下,方法的加标回收率为88.7%~121%,相对标准偏差RSD值为1.5%~5.6%,适用于藏香中34种元素含量的测定。两个区域样品中19种元素含量差异显著(p<0.05)。正交偏最小二乘法(OPLS-DA)判别模型预测指数(Q^(2))为0.576。线性(LDA)判别模型回代检验、交叉检验判别正确率分别为97.0%、95.5%。通过线性(LDA)判别模型筛选出了6种特征元素,用这6种特征元素进行主成分(PCA)分析,PCA分析提取了3个主成分,累计方差贡献率达97.8%,两个不同区域样品具有明显的分类趋势。为基于元素含量的西藏藏香产地判别提供了准确的检测方法和有效的判别模型。

Based on the analysis of characteristics of elements contents of incense from different regions,the geographical origin discrimination models were established combining with chemometrics.A method for determination of 34 elements in Tibetan incense by inductively coupled plasma mass spectrometry(ICP-MS)with microwave digestion was developed to determine the content of elements of a total of 67 samples from two regions,i.e.Tibet region and regions outside Tibet.The datas were analyzed by orthogonal partial least squares discriminant analysis(OPLS-DA),linear discriminant analysis(LDA)and principal component analysis(PCA).The results showed that the correlation coefficients of the linear regression equation were above 0.9995 in the corresponding concentration range,the limits of detection(LODs)and limits of quantitation(LOQs)were in the range of 0.024-0.48μg/kg and 0.080-1.60μg/kg.At three spiked levels of low,medium and high,the average recoveries of all elements were in the range of 88.7%-121%with relative standard deviations of 1.5%-5.6%.The method was suitable for the determination of 34 elements in Tibetan incense.There was a significant difference for 19 elements in incense from two regions(p<0.05).The prediction index(Q^(2))of the orthogonal partial least square(OPLS-DA)discriminant model was 0.576.The correct rate of regression test and cross test of linear(LDA)discriminant model was 97.0%and 95.5%.Six characteristic elements were used for principal component(PCA)analysis,and were screened out by linear(LDA)discriminant model.Three principal components were extracted by PCA analysis,and cumulative contribution ratio of the three components was 97.8%.The samples from different regions showed obvious classification trend.The study will provide an accurate detection method and an effective discriminant model for the origin discrimination of Tibetan incense based on element contents.