Development of quantitative structure gas chromatographic relative retention time models on seven stationary phases for 209 polybrominated diphenyl ether congeners
Yawei Wang,An Li,Hanxia Liu,Qinghua Zhang,Weiping Ma,Wenlu Song,Guibin Jiang
DOI: https://doi.org/10.1016/j.chroma.2005.11.034
2006-01-27
Abstract:Quantitative structure retention relationships (QSRRs) were developed to predict the gas chromatographic (GC) relative retention times (RRTs) for 209 polybrominated diphenyl ether (PBDE) congeners using the heuristic method included in the computer software Comprehensive Descriptors for Structural and Statistical Analysis (CODESSA). A total of 445 constitutional, topological, geometrical, electrostatic, and semi-empirical quantum chemical descriptors were derived for all PBDEs. Using experimental RRT data for 126 PBDE congeners from the literature, predictive regression models were built for seven individual GC capillary columns differing in stationary phases. Each model includes four descriptors which included Wiener index, Randic index, polarity parameter, etc., selected by CODESSA. High predictability was obtained. High multiple correlation coefficients R(2) indicated that >98.5% (except for stationary phase CP-Sil 19) of the total variation in the predicted RRTs is explained by the fitted models. The models were subsequently used to predict the RRTs of the remaining 83 PBDE congeners on seven different stationary phases. The statistical results show that, compared with others, DB-XLB column not only produces the least number of peak overlaps but also results in shorter retention times.