Kong, Darryl Yung Chiu; (2019) Influence of different methods of bed compression on protein separation in process chromatography. Doctoral thesis (Eng.D), UCL (University College London).

Preview

Text Kong-D thesis.pdf Download (2MB) | Preview

Abstract

This thesis examines the effects of hydrodynamic and mechanical compression on process chromatography. Poorly packed columns can have serious consequences on the performance of chromatography and efforts to understand the impact of different methods of compression on protein separation is limited in the literature. By understanding this impact, there is significant potential for facilitating better decisions in chromatographic operations, minimising batch failures while achieving high chromatographic performance. Conventional methods were used to quantify five different structural properties of chromatographic resins; the column efficiency was used to assess the quality of packing and porosity tests was used to determine the interparticle and intraparticle porosity of the packed bed. The influence of bed length, column diameter, and average particle size on the extent of bed compression were examined. The results showed better asymmetry and reduced plate height with both increasing levels of hydrodynamic and mechanical compression. There were practical limitations in using the conventional approaches to investigate the quality of packing; column efficiency and porosity tests only provide an overall indication of the whole column. The reverse-flow technique using an acetone tracer was therefore developed as a novel technique in this field to quantify the microscopic dispersion effects due to bed compression on defined axial sections within a packed bed. This technique allows reversible macroscopic factors to be separated from irreversible microscopic factors. The results showed higher levels of compression towards the bottom of the column with hydrodynamic compression and higher levels of compression towards the top of the column with mechanical compression. This technique has shown to be a simple, non-intrusive method for investigating microscopic factors along the different sections of the column. The breakthrough curves were used to determine the dynamic binding capacities for three different anion exchange resins using BSA as a model protein. Q Sepharose Fast Flow, Q Sepharose High Performance and Capto Q were selected to cover a range of bead rigidity and particles sizes. The shape and position of the breakthrough curves were used to assess quantitatively the impact of bed compression on binding capacity and mass transfer properties. In particular, a range of rigidity and particle sizes of AEX chromatography resins were examined. The results showed the overall impact of compression on breakthrough performance depended heavily on the method of compression applied to the bed. For both hydrodynamic and mechanical compression, the dynamic binding capacity (DBC) increased by 60% for Capto Q. However, when Q Sepharose FF, a softer resin was hydrodynamically compressed the DBC decreased by 10% at 0.15 CF. By contrast, when Q Sepharose HP (2 – 3 times smaller than Q Sepharose FF) was hydrodynamically compressed to the equivalent compression factor, the DBC increased by 20%. This suggests that the particle size distribution (PSD) also influenced changes in breakthrough behaviour when compressed. For all three resins tested, mechanical compression produced the largest increases in DBC. Finally, purification factor vs. yield (PFY) diagrams were used to relate directly the effects of bed compression to the maximum purity and yield at each compression factor. In this study, fractionation diagrams were adapted to describe the elution profiles of the product and its various impurities to show the relationships between bed compression and overall chromatographic performance. A protein mixture was used to challenge three AEX resins (Q Sepharose Fast Flow, Q Sepharose High Performance, and Capto Q) and subsequently a gel filtration resin (Sepharose CL-6B). In particular, the effects of one-step vs. multiple incremental step compression were also examined. The results showed one-step hydrodynamic compression caused flow instability, due to the formation of regions of higher compaction towards the bottom of the packed bed which together resulted in poor protein separation. With mechanical compression via multiple incremental steps, an even distribution of pressure was applied from the top column diameter which gave greater levels of product purity and yield for all resins.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as CSVXMLJSON

Archive Staff Only

View Item