1 Center for Phase Equilibria and Separation Processes, Department of Chemical and Biochemical Engineering, Technical University of Denmark2 Department of Chemical and Biochemical Engineering, Technical University of Denmark3 Center for Energy Resources Engineering, Center, Technical University of Denmark4 Office for Study Programmes and Student Affairs, Administration, Technical University of Denmark
The plate heights for the amino acid tyrosine (anion exchange) and the polypeptide aprotinin (cation exchange) were determined on a porous media (Resource 15) and a get filled media (HyperD 20) at salt concentrations ranging from weak to strong retention. At a constant velocity, measurements showed that the plate height increase with increasing retention, went through a maximum, and finally, decreased as the retention increased, i.e., when the salt concentration was lowered further. The band broadening of a chromatographic peak in the column was caused by the axial dispersion and mass transfer. In this article, the rate of mass transfer in the particles is described by three different rate mechanisms, pore diffusion, solid diffusion, and parallel diffusion. The van Deemter equation was used to model the data to determine the mass-transfer properties. The development of the plate height with increasing retention revealed a characteristic behavior for each rate mechanism. In the pore diffusion model, the plate height increased toward a constant value at strong retention, while the plate height in the solid diffusion model decreased, approaching a constant value at strong retention. In the parallel diffusion model, both pore and solid diffusion took place. Therefore, the parallel diffusion model coincides with the pore diffusion model at weak retention and with the solid diffusion model at strong retention, while a maximum is reached at intermediate retention, resulting in a bell-shaped curve. This behavior corresponds to the observed variation of the plate height at constant velocity. Neither the pore nor the solid diffusion model can describe the experimental data while a satisfactory fit was obtained using the parallel diffusion model.
Separation Science and Technology, 2004, Vol 39, Issue 9, p. 2011-2030