The usefulness of the deconvolution of chromatograms into orthogonal polynomials for characterizing the quality of separation

Simulated chromatograms are deconvoluted into series of orthogonal polynomials. The number of terms necessary to achieve a preset goodness of fit, the reproducibility of coefficients in the presence of noise and the effects of signal-to-noise ratio, peak-height ratio and peak asymmetry on the value of coefficients from the deconvolution are investigated. A convenient method for the deconvolution of real chromatograms is described in order to characterize the quality of separation in the chromatogram. A known quality criterion is modified by using coefficients from the deconvolution in order to enhance its performance. Performance is tested on a simulated separation of five sulfa drugs on a reversed- phase column.