Spectral Fingerprints for Chemical Differentiation of Botanical Materials

Spectral fingerprinting of solids or un‐fractionated extracts in combination with chemometric analysis was used to detect differences in chemical composition arising from genetic and environmental factors. We examined 15 botanical materials using near infrared, ultraviolet, and mass spectrometry and determined that discrimination between genera was easily accomplished. In addition, we demonstrated that it was possible to discriminate on the basis of species, growing location, and processing of materials as illustrated by two specific examples. For Ginkgo biloba, the raw leaves were easily distinguished from the processed, commercial product. Spiking of common flavonol glycosides was very apparent in some of the products. Three Panax species (P. ginseng, P. notoginseng, and P. quinquefolius) were easily separated and samples of P. quinquefolius (American ginseng) grown in Wisconsin, Canada, and China were identifiable. Processing differences for “red” and “white” Chinese ginseng (P. ginseng) were easily distinguished. Data processing using principal component analysis (PCA), soft independent modeling of class analogy (SIMCA), and analysis of variance‐principal component analysis (ANOVA‐PCA) provided score plots that were easily interpreted visually and statistically.