engleski

BENEFITS OF MATHEMATICAL MODELING IN SOLVING REAL CHROMATOGRAPHIC PROBLEMS

Water can be loaded with diversity of pollutants what makes each sample specific and requires frequent modification of applied analytical methods or development of completely new ones. Implementation of mathematical modeling can reduce costs and experimental effort for the method development. In this presentation, the focus will be on chromatography and modeling of chromatographic retention. Most of retention models are focusing on isocratic elution, while only few cases are considering gradient elution as well. Moreover, some of those gradient models are limited to predict only for highly predefined elution conditions. One of the models that overcome those deficiencies is a so-called “iso-tograd” model. This model includes matrix effects and is capable to predict retention for any isocratic or gradient elution program. Even more, if this model is accompanied by an appropriate peak-shape function, it becomes a superior tool for optimization of chromatographic separation. A reasonable step for further reduction of experimental effort and costs is Quantitative Structure-Activity Relationship (QSAR) modeling. QSAR methodology assumes that structure of a molecule contains information related with its properties. Therefore, quantitative relationship between numerical interpretation of the molecular structure and some property of interest – in our case the retention, has to be established. Once developed, QSAR model should be able to predict the value of the same property for other molecules using their structural information only. In this research, QSAR methodology was tested in combination with “iso-to-grad” modeling to create models for analytes for which no experimental data were collected. Although the application of QSAR methodology reduced somewhat the accuracy of the retention prediction, it can still be applied to approximate the optimal separation and thus to reduce time and effort in method development. Moreover, QSAR methodology does not require additional experiments and therefore can be considered as more environmental friendly than other approaches.

MATHEMATICAL MODELING, CHROMATOGRAPHY, METHOD DEVELOPMENT

nije evidentirano