No description available.

This thesis describes the mechanism of exclusion chromatography in terms of conventional liquid chromatography terminology. The various theories for sample retention are discussed and it is concluded that elution under ordinary conditions with a rigid pore structure is governed by the equilibrium theory of steric exclusion. Using the steric exclusion mechanism, calibration curves were computed for various pore geometries and spherical macromolecules which gave an exceptionally good fit to experimental data. In exclusion chromatography, band broadening or dispersion of a polymer sample arises from both the polydispersity, P, of the sample and from the kinetic processes occurring within the column. It is shown that the plate height due to kinetic processes within the column, H, is given by H = H - L (P-1) (1 + Kapp) 2x where H is the apparent plate height calculated from the experimental peak, L is the column length, a is a function of (P-1) and x is a measure of the relative molecular mass range covered by the packing material. By varying the column length and extrapolating to zero length, the true plate height (due to kinetic effects) was obtained as a function of velocity. This E (incidence of plate height upon velocity) followed the well known trend in liquid crystallography giving a minimum plate height of around 2 particle diameters. This was verified by measuring the plate height for monodisperse polymer fractions obtained by adsorption chromatography. By manipulation of this plate height versus velocity data to account for longitudinal diffusion and flow effects, the stationary phase mass transfer coefficients were obtained. Interpretation of those coefficients in terms of the non-equilibrium theory of chromatography leads to the conclusion that within the pores of the material, diffusion of molecules is restricted with the degree of restriction increasing molecular size approaches the pore size. However, this restriction is small enough and mass transfer fast enough so as not to negate the near-equilibrium assumption of chromatography which is shown by the narrow and symmetrical peaks obtained for monodisperse polymers. These results finally confirm that the theory of retentive chromatography applies with only minor adjustments to exclusion chromatography, thereby solving a longstanding argument as to the nature of the processes occurring in exclusion chromatography. Sample loading is discussed and it is shown that sample loads must be kept below about 10µg for analytical columns: larger loads lead to loss of efficiency and change in retention volume.

No description available.

No description available.

No description available.

Chromatography is an increasingly important separation technique in the fine chemical, pharmaceutical and biotechnological industries. The development of accurate, reliable mathematical models for chromatography is necessary for an efficient optimisation of the process. The past decade has seen a tremendous advance in the mathematical modelling and optimisation of the chromatographic processes, with the increase in computational power that is now available. The purposes of this work are to (1) compare chromatographic models, (2) examine the choice of chromatographic models employed and (3) compare different chromatography configurations applied to the same process. With the variety of chromatography models available, there is a need to decide which model is best suited to a given process and the means by which the model parameters can be determined. A novel approach is proposed in this thesis for the model parameter and model selection for chromatographic processes to address both these issues and is illustrated using three case studies. This work highlights the differences in using simulated, theoretical data (which most modelling work commonly illustrated with) and experimental data, particularly data of complex bio-mixtures. Model selection is conducted using a recent graphical interpretation method, discussing the advantages and disadvantages of this method. Over the years, the operation of the chromatographic process in these industries has undergone some changes and it is no longer limited to batch processing. Whilst the single column is still popular in preparative chromatography, multi-column processes are now becoming increasingly popular in industrial-scale chromatography to produce large amounts of highly purified products. In light of the diversity of operational policies now available to chromatography, the second half of this thesis addresses examines the differences between the single column and the multicolumn chromatographic processes. Preliminary work is done on developing a detailed model of the simulated moving bed (SMB) chromatographic process, presenting both a dynamic model and two cyclic steady state (CSS) models. A theoretical case study is then optimised for the operation of the SMB process. The simulated moving bed (SMB) process and its recent variation, the Varicol process, are particularly well known. Such processes are continuous and are able to produce large quantities of high purified products. The decision of choosing either a single column or multi-column process for a separation is not a clear cut one. As the configurations and process operations in these processes are vastly different, an economic comparison between the optimised process alternatives is thus necessary to properly assess the strengths and weaknesses of each system, particularly from an industrial point of view. In column chromatography, a single column, as well as a single column with recycle and peak shaving operations, are considered, whilst for the multi-column alternative, the SMB process and its variations (Varicol, Powerfeed etc.) are examined.

No description available.

No description available.

No description available.

No description available.