Multivariate characterization of herbal drugs and rhizosphere soil samples according to their metallic content

Seven herbal drugs and corresponding rhizosphere soil samples from the plantation near Belgrade were analyzed for mineral content. Eleven metals (Cu, Zn, Mn, Fe, Cr, Ni, Pb, Cd, Ca, K and Mg) were selected as chemical features/descriptors and analyzed by FAAS/FAES and ETAAS. In a chemometrics evaluation of investigated plants–soil system, univariate, as well as multivariate statistics methods were applied. Principal component analysis (PCA) allowed considerable reduction in a number of variables and the detection of structure in the relationships between metals that give complementary information about the relations and elemental patterns within plants–soil system. The power of chemometrics was also used in exploring the potential natural and/or anthropogenic sources responsible for metal content in both medicinal plants and soil samples. Hierarchical cluster analysis (HCA) was used to explore herbal drugs grouping according to corresponding soil samples as additional information to the output obtained by PCA.