Multivariate Assessment of Ground Coffee Samples from Bahia, Brazil on the Basis of Mineral Composition

This study determined metals in ground coffee samples from different producing regions in Bahia and used Principal Component Analysis (PCA) and Kohonen neural networks as multivariate techniques to assess sample similarity based on their mineral composition. The samples were prepared using a decomposition method with dilute acid and hydrogen peroxide in a closed system under high pressure. Microwave Induced Plasma Optical Emission Spectrometry (MIP OES) was used to quantify the elements Al, Co, Cr, Cu, Fe, Mn, Ni, V and Zn. For all samples, the elements with the highest concentrations were Mn, Fe, Al, Zn and Cu; and the lowest concentrations were obtained for the elements Cr, Ni, Co and V. The results of this study provide and aggregate relevant nutritional information about coffee produced in the Southwest and Chapada Diamantina regions of Bahia.