CALIBRATION OF CHROMATOGRAPHY MODELS

Sammanfattning: AbstractThis thesis is devoted to calibration of mathematical chromatography models. Chromatography models include unknown parameters that must be determined before the models can be applied to optimization or scale-up of a chromatography process. The aim of the work presented in the thesis was to develop methods for determining these unknown model parameters.The fundamental approach adopted to determine model parameters was fitting of the mathematical chromatography model in question to experimental data by estimating the parameters using regression methods.Several factors, as parameter dependencies, the order in which the parameters should be estimated, the generation of experimental data required, suitable fitting procedures, etc., must be taken into account in the methodology developed for the determination of the unknown model parameters.This thesis presents a methodology for calibration of a detailed chromatography model. Fitted empirical correlations for different band broadening behaviours taking the flow rate and bead size into account, have also been developed and presented. Model simplifications are discussed, and the use of simpler models to describe continuous monolithic beds.The methodology was applied to chromatography columns packed with different cross linked Sepharose gels and columns containing continuous supermacroporous monolithic beds, using acetone, latex particles, blue dextran, E. coli cells, lysozyme, immuno globulin g and bovine serum albumin, at different flow rates. This was done under both non-retained and sorption conditions. The results show that it is possible to determine unknown parameters included in chromatography models describing both packed beds and continuous monolithic beds using the methodology developed.