Resolution Of Overlapping Chromatographs By Vertically Iterative Transition And Modified Vertically Iterative Transition

Nowadays, Chromatographic technique has been extensively used in the field of atmosphere, water quality, soil, living things, food, and other monitoring. It has become a dominating means in environmental monitoring and analysis. It has been proved be a powerful technique in separating the mixed compounds of a sample taking advantage of compound specific parameters such as boiling point, molecular structure, mass, charge, as well as diffusivity. However, with the increase of complexity of the analyte complete resolution of all compounds can rarely be achieved even when extremely high performance separation columns are employed and an optimum selectivity was applied, especially when the analyte is a mixture of compounds belonging to the same family, and undergoing similar interactions with the separation environment. As a result, coelution or even a cluster of peeks will be acquired. Since analysis of complex mixtures and multicomponent analyte is the main character of modern analytical chemistry, the resolution of overlapping peaks is now a significant problem to be solved.Perpendicular-drop and tangent-skim methods are traditional means for the calculation of peak areas and heights of overlapping peaks. Such methods are known to lead to substantial errors in the calculation, thus more accurate means are needed. As the development of, more and more algorithms based on chemometrics are used for resolution of overlapping peaks, where the curve fitting, fast Fourier transform, wavelet transform, and artificial neural networks have been widely used in the literature. And these methods are proved to be effective. However, they are too complex and time consumable for on-line analysis.In this paper a new approach based on the reason of the formation of overlapping peaks namely vertically iterative transition was proposed to resolve peak overlaps. The theory is based on the fact that only when the analyte is a mixture of compounds having similar character, and thus undergoing similar with the separation environment the overlapping peaks are required, and the similar interactions result in similar peak shape. Thereby a reasonable hypothesis is proposed, that is the individual peaks which composing the overlapping peaks are similar to each other and can lap over each other by multiplying a certain coefficient and processing a parallel translation.In this work, a mathematical model for resolution of overlapping peaks was founded based on the hypothesis mentioned above; the method of was described; two modified methods namely dichotomizing iterative transition and accelerated vertically iterative transition which are more accurate and fast were established; a comparison of the performance of those methods was carried out and the result was presented and and accelerated vertically iterative transition was carried out; the infection of scale coefficient in the accelerated vertically iterative transition to the result is calculated and discussed; and finally a method of heuristic scan was test for finding the individual peaks in complete overlapping peaks. It is found that the new algorithms are much faster in calculating than former at the same precision. Besides, the seriously overlapping peaks can also be resolved precisely by the new algorithms. This finding is of great impotance offering a brandly new approach for the resolution of the seriously overlapping peaks.The result showed that the modified vertically iterative transition is simple, convenient, accurate and well performed, thus is more appropriate for online analysis.