Quantitative Prediction of Normal Boiling Points for Organic Compounds Using Gas Chromatographic Retention Times on Two Columns of Different Polarity

Quantitative relationships between normal b.p.s and gas chromatographic retention times of 271 organic compounds of diverse structures was studied using a multivariate regression method and statistical importance evaluation. It was found that satisfactory modelling cannot be obtained using only retention times on a single column. Even dividing the 271 compounds into four groups according to their chemical structures, only hydrocarbons on nonpolar columns gave good results with a linear model. However, if the retention times on two columns of different polarity are used to build the model, a satisfactory two-parameter model can be obtained for the whole data set with a squared correlation coefficient (R2), standard deviation (s) of the prediction, and probability value (P) of: 0.9456, 16.1K and 1.7×10−150, respectively. The results also show that the statistical importance of the two-parameter model is much better than that of the one-parameter model.