Prasad Babu Kakarla

• Expanded Bed Adsorption (EBA) is an ingenious approach to reducing the steps required in primary purification while simultaneously allowing the processing of fermentation broths with extremely high cell densities. Most of the downsides of the first generation EBA have been negated with optimized flow distribution as well as the development of new chromatographic materials of sufficient weight to ensure an adequately fluidized and stable bed. This technology has become one of the most promising ways into the future of downstream processing. Nevertheless, a major problem yet to be solved for EBA are the unfavorable biomass-adsorbent interactions as well as the cell aggregation phenomenon, i.e. adhesive interactions between the adsorbent and crude biomass, which are the major cause of poor process performance, low product recovery, and in the worst case, bed collapse. The research work reported in this Ph.D. thesis is aimed to develop a predictive framework for optimized EBA process conditions, for which we have coined the term IntEP (Interaction Energy Predictor). This framework employs classical colloid theory, namely extended Derjaguin, Landau, Verwey and Overbeek (xDLVO) theory, to determine (within specified operation windows) best-case scenarios of mobile phase composition that will result in the maximization of repulsive interactions between biomass and adsorbent. Based on IntEP-optimized conditions, lab-scale EBA experiments using MabDirect® Protein-A adsorbent beads to purify a proprietary mAb from high-density CHO cells feedstock performed. This experimental work leads to the conclusion that these optimized conditions not only improve process efficiency and hydrodynamics but also increase product recovery. With IntEP, it is possible to circumvent the need for empirical determination of these optimal process conditions in EBA, thus leading to quicker method development and better purification.