Resolution of HPLC-DAD highly overlapping analytical signals for quantitation of pesticide mixtures in groundwater and soil using multicomponent analysis and neural networks

Along with the development of hyphenated chromatography techniques, such as high performance liquid chromatography (HPLC) with diode array detection ( DAD), three-dimensional data matrix for each sample can be easily obtained. In this paper, a comparative study of three methods using this type of dat a is presented. The capabilities of inverse calibration through ordinary le ast squares (OLS), partial least squares (PLS), and an artificial neural ne twork (ANN) model, have been investigated for the simultaneous multicompone nt analysis of synthetic mixtures of iprodione, procymidone, chlorothalonil , folpet, and triazophos when highly overlapping analytical signals are pre sent. Also, the methods have been satisfactorily applied to the determination of the pesticides in groundwater and soil samples, although the ANN gave the b est results.