基于无机元素指纹分析的桔梗产地溯源研究

目的:比较四川梓潼、云南玉龙、安徽太和、湖北英山、河南桐柏无机元素含量差异,结合计量学方法,探讨桔梗产地溯源应用的可行性,为桔梗产地溯源提供新方法。方法:利用ICP-MS和ICP-AES测定四川、云南、安徽、湖北、河南5个主产区共75份桔梗样品中K、P、Fe、Mn、B等37种元素的含量,通过主成分分析对5大产区桔梗进行聚类,运用KNN算法、Bayes算法、SVM-ga算法、BP-NN算法、LS-SVM算法、SVM-grid算法,建立桔梗产地溯源模型。结果:五大产区桔梗中无机元素含量差异显著(P＜0.05),其中Sb、Sn、Ge、Nb、Ag 5种元素,在不同产区间差异最为明显,由主成分分析发现不同产地桔梗样品分布集中且相对独立,SVM-grid桔梗溯源模型能够100%将五大产区桔梗进行区分。结论:Sb、Sn、Ge、Nb、Ag 5种元素含量分布具有明显的地域特征,以无机元素结合计量学方法可以对桔梗进行有效的产区判别,且准确率高,是产地溯源的有效方法。

Traceability study on the origin of Platycodon grandiflorum(Jacq.)A.DC. based on fingerprint analysis of inorganic elements

Abstract: Objective: To compare the differences in inorganic element contents of Zitong in Sichuan (SCZT), Yulong in Yunnan (YNYL), Taihe in Anhui (AHTH), Yingshan in Hubei (HBYS) and Tongbai in Henan (HNTB), and to explore the feasibility of traceability application of the origin of Platycodon grandiflorum（Jacq.）A.DC. in combination with metrological methods, and to provide a new method for traceability of the origin of Platycodon grandiflorum（Jacq.）A.DC. Methods: The contents of 37 elements including K, P, Fe, Mn and B in a total of 75 samples of Platycodon grandiflorum（Jacq.）A.DC. from five main producing areas in Sichuan, Yunnan, Anhui, Hubei and Henan were determined by ICP-MS and ICP-AES. The Platycodon grandiflorum（Jacq.）A.DC. in the five major production areas was performed by principal component analysis. Using KNN algorithm, Bayes algorithm, SVM-ga algorithm, BP-NN algorithm, LS-SVM algorithm, SVM-grid algorithm. Establishing a traceability model for the Platycodon grandiflorum（Jacq.）A.DC. Results Significant differences (P < 0.05) in the content of inorganic elements in Platycodon grandiflorum（Jacq.）A.DC. from the five major production areas, Among them, Sb, Sn, Ge, Nb, Ag 5 elements, the most obvious differences between different production areas, From the principal component analysis, it was found that the samples of different origins of Platycodon grandiflorum（Jacq.）A.DC. were concentrated and relatively independent in distribution, The SVM-grid oranges traceability model was able to distinguish 100% of the five major production areas of Platycodon grandiflorum（Jacq.）A.DC. Conclusion The distribution of the content of the five elements Sb, Sn, Ge, Nb and Ag has obvious geographical characteristics, The inorganic elements combined with the metrological method can be used for effective appellation discrimination of Platycodon grandiflorum（Jacq.）A.DC. And the accuracy rate is high, which is an effective method for origin traceability.Keywords: Platycodon grandiflorum（Jacq.）A.DC.; inorganic elements; Pattern recognition; Origin traceability