Submitted to: Experimental Biology
Publication Type: Abstract Only
Publication Acceptance Date: 11/6/2007
Publication Date: 4/5/2008
Citation: Harnly, J.M., Luthria, D.L., Chen, P., Lin, L. 2008. Spectral Fingerprints for Authentication and Identification of Variation in Foods and Botanical Materials. Experimental Biololgy Meeting, April 5-9, 2008, San Diego, CA.
Technical Abstract: Spectral fingerprints were acquired using ultraviolet/visible (UV/Vis), infra-red (IR), near infra-red (NIR), and mass spectrometry (MS) for broccoli, dry beans, and Ginkgo biloba with no prior chromatographic separation of the samples. Patterns were elucidated from the complex overlapping spectra using analysis of variance-principal component analysis (ANOVA-PCA). Using this approach, it was possible to detect differences in the chemical composition of the samples attributable to experimental factors such as the cultivar, growing conditions, growing site, plant-to-plant variability, and analytical variability. In addition, it was possible to determine the fraction of variance arising from each of the factors. All four analytical methods, including UV/Vis, were found to be equally effective for these analyses, although each focuses on different chemical components. An aqueous methanol extract was used for the UV/Vis and MS analyses, while finely powdered solid samples were used for IR and NIR. Only MS provided sufficient resolution to allow tentative identification of specific compounds. This research was funded by USDA and the Office of Dietary Supplements at the National Institutes of Health.