Fe, Cu and Zn distribution in different components of commercial infant formulas

Fe, Cu and Zn determination by AAS was performed in 35 infant formula samples from different manufacturers. The trace element content was studied in whole, fat and whey milk, and different chemometric tools have been used in order to observe differences or correlations between infant formulas fortified with inorganic salts (iron, copper and zinc sulphates) or with organic and inorganic salts (iron and copper gluconates, or zinc and iron lactates and zinc oxides). Principal-components analysis (PCA) achieved a reduction from nine variables to three (accounting for 80.8% of the total variability), and some differences between infant formula from the two groups were observed.Cluster analysis gave similar results as PCA. Linear discriminant analysis (LDA) allowed the classification of infant formulas in two categories or classes: the first class formed by samples fortified with inorganic salts (category A) and the second one by samples fortified with organic and inorganic salts (category B). The percentages of samples correctly classified were 96.1 and 100.0 for the categories A and B, respectively. After the application of the soft independent modelling of class analogy (SIMCA) approach, percentages of cases correctly assigned of 87.5 and 12.5 were achieved for the categories A and B, respectively. However, it must be said that the application of SIMCA led to bad results because of the small number of sample used, mainly in category B.