Estimation of uncertainty of measurements of linear retention indices with increasing temperature of the capillary chromatographic column

Abstract

A brief overview of analysis methods and analytical identification parameters of components of complex mixtures is given. The parameters of identification of substances in the chromatographic analysis of complex multicomponent mixtures – linear retention indices – are considered, and it is shown that to ensure the necessary level of reliability of the analysis results, it is necessary to assess the uncertainty of measuring these indices. A general equation for determining the variance of the result of indirect measurements of linear retention indices is obtained. For the case of the analysis of mixtures containing both assays and n-alkanes, and mixtures with a separate content of these components, its particular formulations are derived from this general equation. A formula for determining the effective number of degrees of freedom of standard uncertainty is presented, taking into account the correlation of input variables in the measurement equation. Mathematical expressions are obtained for calculating the extended uncertainties of estimating linear retention indices, which take into account correlations of retention times of neighboring chromatographic peaks. By methods of interval mathematics, an upper estimate of the standard uncertainty of linear retention indices in the case of the joint presence of analytes and n-alkanes in the analyzed mixture was obtained and its significant decrease in comparison with the specified uncertainty in the analysis of mixtures with a separate content of these components was shown. The correspondence of theoretical and experimental results has been established. The results obtained can be used to identify the components of the analyzed mixture when standardizing chromatographic analysis techniques.