Abstract:

A discrimination method for 6 kinds of Chinese famous tea by combination of organic components and
multielement analysis was established. Totally 67 different tea samples from their places of origin were collected, including
green tea (Huangshan Maofeng, West Lake Longjing, Biluochun and Lushan Fragrant) and oolong tea (Wuyi Rock Tea,
Ti Kuan Yin). The concentrations of 9 elements (K, Ca, Mg, Al, Mn, Fe, Zn, Rb and Ti) were determined by ICP-AES, 12
elements (Pb, As, Cd, Cr, V, Ni, Co, Cu, Sr, Cs, Ba and Se) by ICP-MS, and 10 organic compounds by UPLC. Partial least
squares variable selection method was applied in the establishment of the discrimination models for elements and organic
components, and a tracing procedure was set up for the confirmation of the known and unknown tea samples. The analysis
of real samples showed that the 6 kinds of Chinese famous tea were well discriminated and misjudgment was reduced
effectively by cross validation of the two models.

Key words: tea, inductively coupled plasma-atomic emission spectrometry (ICP-AES), inductively coupled plasma-mass spectrometry (ICP-MS), ultra high performance liquid chromatography (UPLC), discrimination, cross-validation