Expanded bed chromatography using ion-exchange media is a powerful first step in purification processes. Expanded bed chromatography can be used to extract components from complex and viscous solution. This can be achieved because of the void created between adsorbent particles where as in packed bed chromatography, the adsorbent is too compact and dense for a complex feed stock to flow through. Expanded bed chromatography was used to purify bovine serum albumin (BSA) from chicken egg white (CEW). The high viscosity of CEW presents a unique challenge for efficient large-scale protein purification. This project aimed to optimize and evaluate a separation method that is believed to be particularly suitable for high viscosity solutions: expanded-bed ion exchange chromatography. The BSA was admixed into the CEW and the solution was pumped through the column for purification. The media used in the column was Streamline DEAE which is an anion-exchanger. The yield obtained was 85% and the purity was 57%. A mathematical model to understand and predict the behavior of expanded bed chromatography was developed to provide an estimation of the breakthrough curves obtained for BSA. A small sized porous dense adsorbent was also synthesized to enhance the purification process. This zirconia-based adsorbent allows use of higher flow velocities that is a key factor when working with viscous fluids such as chicken egg white.

"Expanded Bed Chromatography (EBC) is rapidly becoming the preferred choice for initial product recovery from crude process streams as it enables direct protein recovery from culture broths after appropriate dilution. However, the process is time intensive, and there are still some difficulties with very high cell density cultures in the 500 g/L range. Problems include in-column clogging and poor column efficiency. With the development of a new prototype EBC column capable of product recovery from undiluted culture broth, it is proposed in this study to combine the fermentation with EBC recovery. This strategy was tested using a wild type, non-producing strain of Pichia pastoris. Culture broths were spiked with 200 mg/L lysozyme to mimic an actual production fermentation. Key parameters for the process were identified and tested independently to better assess system performance: potential toxic effects of the resin on the culture, nutrient deprivation of the cells as they pass through the column and binding of the target protein from whole broth. The cation exchanger had a negligible effect on cell proliferation in shake flask studies using YNB Medium. Isolation of the culture from the fermenter for up to two hours appeared to have minimal effect on overall cell viability and the ability to metabolize methanol. The dynamic binding capacity for lysozyme was 50 mg/mL in buffer, and 20 mg/mL in undiluted fermentation broth containing 500 g/L cells. When harvested undiluted fermentation broth was allowed to recirculate through the EBC column, the binding capacity was increased to 30 mg/mL. The combination of the fermentation and recovery process allowed for a binding capacity of 30-40 mg/mL, with no dramatic effects on biomass accumulation or metabolic rate."