Experimental Design and Partial Least Squares for Optimization of Reversed-Phase Ion-Interaction Liquid Chromatographi Separation of Nitrite, Nitrate, and Phenylenediamine Isomers

Ion-Interaction chromatographic data were treated by chemometric methods in order to optimize resolution and analysis time. A mixture of nitrite, nitrate, and 1,4-1, 3-, and 1,2-phenyienedlamine at average 0.5 ppm levels was considered. The effect on retention of alkyl chain length of the ammonium salt used as the Interaction reagent, Its concentration, and flow rate were studied. Since a nonlinear dependence is expected, the starting experimental design selected was a central composite design. The regression model was calculated using the partial least squares (PLS) method, and a cross-validated stepwise technique of selection of the independent variables was employed to obtain the best regression model. The fitted models were used for the optimization of the chromatographic conditions with respect to the resolution of the components of the mixture. The differences between experimental retention times obtained in the predicted optimal condition and the experimental times were comparable with the experimental errors (correlation coefficient R²= 0.9854).