A simple and sensitive spectrophotometric method to the simultaneous determination of Mn2 and Fe3 in
foods, vegetable and water sample with the aid of artificial neural networks (ANNs) is described. It relies
on the complexation of analytes with recently synthesised bis pyrazol base ligand as 4,40[(4-cholorophenyl)
methylene] bis(3-methyl-1-phenyl-1H-pyrazol-5-ol)(CMBPP). The analytical data show that the ratio
of ligand to metal in metal complexes is 1:1 and 1:2 for Fe3 and Mn2 , respectively. It was found that the
complexation reactions are completed at pH 6.7 and 5 min after mixing. The results showed that Mn2 and
Fe3 could be determined simultaneously in the range of 0.20–7.5 and 0.30–9.0 mg l1, respectively. The
analytical characteristics of the method such as the detection limit and the relative standard error predictions
were calculated. The data obtained from synthetic mixtures of the metal ions were processed by
radial basis function networks (RBFNs) and feed forward neural networks (FFNNs). The optimal conditions
of the neural networks were obtained by adjusting various parameters by trial-and-error. Under the working
conditions, the proposed methods were successfully applied to the simultaneous determination of elements
in different water, tablet, rice, tea leaves, tomato, cabbage and lettuce samples.