Determination of Pd(II) by application of an on-line microwave oven and artificial neural networks in flow injection analysis

A new methodology based on the coupling of an on-line microwave oven, employed to accelerate low-rate reactions, and artificial neural networks (ANNs), applied to the modeling and optimization of a new flow injection system, is proposed. In comparison with traditional heating, a microwave can accelerate low-rate reactions more remarkably and consume less energy. ANNs with a faster back-propagation (BP) algorithm are applied to model the system. Optimum experimental conditions are generated automatically by using jointly ANNs and optimization algorithms in terms of sensitivity, sampling rate and the energy consumed by a microwave oven. The methodology is tested on a new flow injection system for the spectrophotometric determination of Pd(II) with chlorophosphonazo-p-Cl (CPA-pC) in H2SO4 media, which has first been used as chromogenic reagent in the quantitative analysis of palladium. It is shown that the methodology can improve the ability of optimization, reduce analytical time, enhance sensitivity and consume less energy in comparison with traditional methods. (C) 2000 Elsevier Science B.V. All rights reserved.