| In capillary electrochromatography, the column plays an even more vital role than in high performance liquid chromatography, since it acts as both the separation element and where the flow is generated. Traditional capillary electrochromatography columns have problems stemming from use of frits, which cause bubble formation and band broadening. They also suffer from movement of charged stationary phase particles during separation, leading to spaces in the packed bed. It has been suggested that these problems may be overcome by the use of monoliths for these experiments. The first part of this dissertation describes the formation of a particle fixed monolith, which uses no frit during the separation, and is made by adding a few easy steps to the traditional method for making a capillary electrochromatography column by slurry packing.; The second part of the dissertation discusses band broadening in high performance liquid chromatography. In 1956, van Deemter described the band broadening in a chromatographic system with equation [1]. The A term represents eddy diffusion, the B term represents molecular diffusion, the C term represents resistance to mass transfer and u is the mobile phase velocity. Other equations have since been introduced to model efficiency data, and these include the equations of Giddings, Huber and Hulsman, Horvath and Lin, and Knox. This creates a plight for chromatographers who then need to decide what equation to use since all the equations are based on different theories. H=A+Bu+Cu 1 By choosing solutes that experience known types of broadening, it is possible to chromatographically isolate the different contributions. Such a chromatographic isolation of the band broadening contributions is theoretically better than previously used methods because the experiments would be performed under normal chromatographic conditions. A series of experiments using uracil, benzene, toluene, ethylbenzene, propylbenzene and butylbenzene was completed, and this data has given additional information about these processes. In the final portion of the dissertation, the four equations mentioned above were used to model plate height data, and the resultant fits were compared in order to determine which equation offered the best fit to chromatographic data. |
