Green is no longer the only color used to describe tea leaves. As tea plants with different leaf colors-white, yellow, and purple-yield significant economic benefits, scholars are growing increasingly curious about whether these differently colored leaves possess unique aromatic characteristics. Headspace solid-phase microextraction (HS-SPME) combined with GC-MS was used to analyze the volatile metabolites of buds and leaves from 7 white-leaf tea plants, 9 yellow-leaf tea plants, 4 purple-leaf tea plants, and 7 normal (green) tea plants. A total of 125 aroma metabolites were identified. The aroma compounds of heterochromatic tea leaves and green-leaf tea were compared separately. It was found that white-leaf tea had the most upregulated compounds (63 up), mainly floral and fruity aromas, including nerol, Z-isogeraniol, and E-3-hexen-1-yl acetate. Purple-leaf tea had the most downregulated compounds (31 down), including β-myrcene, benzyl alcohol, and methyl salicylate, which are related to fresh and fruity aromas. According to variable importance in projection (VIP > 1) and a p-value < 0.05, a total of 40 differential compounds were detected, among which Z-3-hexenol, 1-nonanol, 2,4-di-tert-butylphenol, and 2,6,10,15-tetramethyl-heptadecane were common in all heterochromatic tea. The random forest model constructed using differential metabolites screened out five aroma metabolites, including Z-3-hexenyl isobutyrate, E-3-hexen-1-yl acetate, 2,4-di-tert-butylphenol, Z-jasmone, and Z-isogeraniol. These metabolites demonstrated high accuracy in the model (AUC = 1) and have the potential to serve as characteristic aroma compounds for distinguishing tea leaf colors.

Keywords: Aroma metabolomics; Biomarker; GC–MS; Tea plant.

Copyright © 2025 Elsevier Ltd. All rights reserved.