Improvement of the reproducibility of retention indices in temperature-programmed gas-liquid chromatography. Part I. DB-1 column

Abstract

For calculation of retention indices, retention data for the n-alkane reference series has to be measured. In most cases when we analyse samples having many components, it is not possible to mix the n-alkanes into the analysed sample. When we measured the retention times of n-alkanes from separate injections the calculated retention indices showed unacceptably large random deviations under programmed temperature operations. We describe a simple correction for the retention times of the reference n-alkanes to improve the reproducibility of the retention indices. For this correction the analysed samples must contain one n-alkane identical with a member of the reference series. Using the differences or ratios of the retention times of this one common n-alkane measured in the sample and measured separately, we can calculate normalized sets of retention times of the n-alkane series. The retention indices of the analysed solutes obtained by using these normalized retention times of the separately injected and measured reference series showed a better reproducibility than without normalizing. © 1998 John Wiley & Sons, Ltd.