文献类型： 外文期刊

作者： Liu, Hong-lin ¹ ; Meng, Qing ¹ ; Zhao, Xin ² ; Ye, Yu-long ³ ; Tong, Hua-rong ¹ ;

作者机构： 1.Southwest Univ, Coll Food Sci, Chongqing 400715, Peoples R China

2.Chongqing Univ Educ, Chongqing Engn Res Ctr Funct Food, Chongqing Collaborat Innovat Ctr Funct Food, Chongqing Engn Lab Res & Dev Funct Food, Chongqing 400067, Peoples R China

3.Sichuan Acad Agr Sci, Tea Res Inst, Chengdu 610066, Sichuan, Peoples R China

关键词： Tea; Discrimination; Chemometrics; Formula; ICP-MS; ICP-OES

期刊名称：FOOD CONTROL （ 影响因子：5.548； 五年影响因子：5.498 ）

ISSN： 0956-7135

年卷期： 2021 年 123 卷

页码：

收录情况： SCI

摘要： Multielement and stable isotopes analysis combined with chemometrics as a potential tool for first authenticity discrimination of teas. Eighty-four statistically significant elements and stable isotopes from Chongqing tuo tea with 4 formulations were determined using ICP-MS and ICP-OES. Four chemometrics were used to build discrimination models.The discrimination of PLS-DA and LDA were better than that of HCA and PCA. Cd-114, Mo-95, Rb-85, Co, Cs-133 and P were the most important markers for successful authenticity discrimination of teas. Under the validation by cross-validation, test sets and blind data sets, the precisions of prediction of the PLS-DA and LDA were not less than 90%. The LDA performs best, with precisions of prediction of 100%. This study helps to promote the application of fingerprint technology of multi-elements and stable isotopes in combination with chemometric analysis as a very promising tool for a worldwide inspection of the authenticity of teas.