Statistical Evaluation of the Standard Deviation of Chromatographic Retention Time in the Capillary Column Temperature Programming Mode

Abstract

In the qualitative analysis of complex mixtures of unknown composition, a promising direction for increasing the reliability of the identification of substances is the joint use of several simultaneously determined analytical parameters. A combination of chromatographic methods with FTIR spectroscopy, mass spectrometry, or other methods is used. In solving complex analytical problems of industrial nature, it is economically expedient to use only chromatographic methods. In analyzing complex multicomponent mixtures with a high density of peaks in the chromatogram, the precision of the determination of retention time determines the reliability of the identification of substances and of the result of analysis, using identification both directly by the absolute retention time and by relative parameters derived from it (relative retention times, retention indices). The standard deviation of a measurement of the chromatographic retention time is a measure of precision. Experimental chromatographic retention times were obtained for analyzing a mixture by the following scheme: initial isotherm, temperature programming, final isotherm. A sharp increase (by one order of magnitude) in the numerical value of the standard deviation was noted at the step of temperature programming. A regression equation was obtained linking the point estimate of the standard deviation of the retention time and column temperature. The performance characteristic of the regression equation is obtained as a function that enables calculating the total standard uncertainty in the estimate of the standard deviation of the chromatographic retention time. The results can be used to refine chromatographic retention time and indices that ensure the reliable identification of substances.