Abstract in Englisch:

The technological progress to more sensitive detection techniques such as mass spectrometry reveals the complexity of samples from fields as diverse as proteomics, genomics, environmental protection and food science. One dimensional liquid chromatography coupled to mass spectrometric detection is a powerful tool for the analysis of such complex samples, but it has limitations in terms of peak capacity. An alternative separation technique with higher peak capacity is comprehensive two dimensional liquid chromatography. This thesis deals with the development and optimization of multidimensional methods based on microscale comprehensive online liquid chromatography coupled with mass spectrometry. The performance of the microscale two-dimensional system was evaluated by a comparison to a conventional one-dimensional system. For the comparison multi-reference standards as well as samples with a complex matrix (influent of a wastewater treatment plant) were used. An automatic workflow based on Excel macro programming was developed and applied for the visualization of the two- or three-dimensional data sets. The comparison of the reference standards showed that it is possible to detect more compounds with the conventional one-dimensional approach. In contrast, for the complex sample it was possible to detect and identify two times more compounds with the microscale two-dimensional approach than with the conventional one-dimensional approach. The comparison showed not only the enormous potential of the two-dimensional system, but also significant technical and methodological weaknesses, including dilution caused by the system, the selection of stationary phases and data evaluation/data visualization. These aspects were addressed in this thesis and possible solutions are presented and discussed in detail. To counteract dilution caused by the system, a 50 mm long stationary phase based on porous graphitic carbon was used in the first dimension of the microscale system at the beginning. This allowed the injection of large volumes of aqueous samples, which resulted in a better sensitivity. However, significant disadvantages relating to the peak shape of non-polar analytes could be observed. Therefore, a pre-column concept was developed in this work that is based on a serial coupling of a 1 cm short pre-column filled with porous graphitic carbon and an additional main column. The concept evaluation revealed that it is possible to inject large volumes of aqueous samples without significant peak broadening for non-polar analytes. To evaluate the pre-column concept in the first dimension of the microscale two-dimensional system, a new method for the calculation of the orthogonality and selectivity based on the compound distribution across the separation space was developed. Additionally, the new method was compared with three established methods reported in literature. The orthogonality study of this work showed that the additional selectivity of the serially coupled columns in the first dimension results in a higher orthogonality of the two-dimensional system. In addition, there is now a simple and automatable tool available for method development in the field of two-dimensional separation techniques. The best column combination can be determined based on a few basic measurements and a representative reference standard. Further technical developments in the field of microscale two-dimensional liquid chromatography could focus on the coupling with ion mobility spectrometry to further increase the analytical information gained from measurement of unknown samples. In that regard, the evaluation of new algorithms for automatic data processing should also be pursued.