The quality of green tea is influenced by many factors, geographic origin being one of high importance, as this parameter is typically associated with tea quality grade. However, in some cases, fraudulent practices such as mislabelling of geographic origin may take place. In this pilot study, aimed at green tea authentication, we investigated whether metabolic fingerprinting based on ultra-high-performance liquid chromatography coupled with high-resolution mass spectrometry (UHPLC–HRMS) enables sample classification to be achieved. In total, 54 authentic samples of green tea originating from China (n = 29), Japan (n = 17) and South Korea (n = 8) were available for experiments. To isolate even less polar metabolites, a dichloromethane/methanol (1:1. v/v) mixture was used for extraction. The data generated by the analysis of green tea extracts were processed by both unsupervised and supervised chemometric methods. The resulting predictive models document the applicability of this approach for green tea classification based on geographic origin.
