Evaluation of feed-forward back propagation and radial basis function neural networks in simultaneous kinetic spectrophotometric determination of nitroaniline isomers.

Mixtures of 2-, 3-, and 4-nitoroanilines, are simultaneously analyzed with spectrophotometry, based on their different kinetic properties. These nitroanilines react differentially with 1,2-naphtoquinone-4-sulphonate (NQS) at pH 7 in micellar medium to produce colored product. The differential kinetic spectra were monitored and recorded at 500 nm, and the data obtained from the experiments were processed by chemometric approaches, such as back-propagation neural networks (BPNNs), radial basis function neural networks (RBFNNs), and partial least squares (PLS). Experimental conditions were optimized and training the network was performed using principal components (PCs) of the original data. A set of synthetic mixtures of nitroanilines was evaluated and the results obtained by the application of these chemometric approaches were discussed and compared. The analytical performance of the models was characterized by relative standard errors. It was found that the artificial neural networks model affords relatively better results than PLS. The proposed method was applied to the determination of considered nitroanilines in water samples.