Independent components-discriminant analysis for discrimination of Brazilian Canephora coffees based on their inorganic fraction: A preliminary chemometric study

Essential element composition of new Brazilian Canephora coffees of specific origins, including both Robusta and Conilon varieties is reported for the first time. A sample set comprising 100 Canephora samples of different geographical origins (Conilon from Espirito Santo, Amazonian Robusta from indigenous and non-indigenous producers from Rondonia, and Conilon from Bahia) and Arabica coffee (25 samples) was analyzed in terms of their essential mineral content by flame atomic absorption spectrometry using diluted nitric acid to characterize their content in Fe, Ca, Zn, Mg, Mn, Cu, and K. A recent chemometric tool, independent components-discriminant analysis (IC-DA), was applied for the first time to analyze mineral composition data. IC-DA discriminated the samples according to the predetermined classes (geographical origins of Canephora and species - Arabica vs. Canephora), identifying the minerals responsible for the discriminations. IC-DA sensitivity and specificity were higher when compared to the partial least square-discriminant analysis (PLS-DA). The method is shown to be suitable to achieve supervised discrimination of samples based on their essential elemental composition, even when principal component analysis had failed. It can thus be used as a tool for origin and species discrimination based on the inorganic fraction of coffee.