Determination of chromatographic peak parameters by non-linear curve fitting using statistical moments

Abstract

There are many methods (manual, computer), which make possible the detailed analysis of chromatographic peak shape. Curve fitting seems to be one of the successful applications of the evaluation of chromatographic signals. In this case, a mathematical function is fitted to the digitized measuring points. Chromatographic elution curves are generally asymmetric, so it is necessary to find the best mathematical function, which corresponds perfectly to the signal form, and then to determine the parameters of the found equation. In this work a modified mathematical function is presented and used for curve fitting, in which the starting parameters are calculated from the statistical moments (retention time, variance, skewness, excess) of the peak shape. The fitting process gives a more accurate and rapid determination of peak shape characteristics, noise filtering and correction of baseline. The suggested modified Gaussian function, the simulated chromatographic peak shapes, the application of the function to the description of real signals and the parameters obtained by the fitting process will be demonstrated.

Development of a Solid-Phase Extraction (SPE) Cartridge Based on Chitosan-Metal Oxide Nanoparticles (Ch-MO NPs) for Extraction of Pesticides from Water and Determination by HPLC

“…The LOD of the pesticides determined by HPLC ranged from 0.002 to 0.046 μ g/mL and the LOQ was in the range of 0.006 to 0.154 μ g/mL. The asymmetry factor (A s ) is an indication for the peak tailing [ 51 , 52 ] being in the range of 0.870 to 1.070.…”

Relationship between Hpt polymorphisms and serum protein electropherogram

“…This gave a single peak or two unresolved peaks in case of 1α2 profile and two well‐resolved peaks for 2α2 profile. This classification was based on a definition of peak shape 17. The statistically significant association between the α2 profile in the PROTEIN 6 electropherogram and the Hpt phenotypes permitted an exclusion of the Hpt 1–1 phenotype from the 1α2 profile (LR of 24.2), and exclusion of the Hpt 2–2 phenotype from the 2α2 profile (LR of 19.9).…”

The utility of statistical moments in chromatography using trapezoidal and Simpson's rules of peak integration

“…Empirical models to characterize peaks have also been suggested in the literature. Some of these empirical models are bi‐Gaussian model , models of Torres Lapasio , Li , Chesler and Cram , Lan and Jorgenson , and Nikitas , Frazer‐Suzuki model , Gram‐Charlier series , Edgeworth‐Cramer series , and Pap‐Papai model . Although these various functions and empirical models may provide some improvement over Gaussian model in peak characterization for some of the unusually shaped peaks, these cannot be all‐encompassing for accurately characterizing the vast number of possible chromatographic peak shapes that result from various physical and chemical processes taking place inside the column.…”