Abstract
Purification of proteins is an increasingly important process for the biotechnology industry. Separation of the desired high value protein from other proteins produced by the cell is usually attempted using a combination of different chromatographic techniques. These techniques separate mixtures of proteins on the basis of their charge, degree of hydrophobicity, affinity or size. Adequate purity is often not achieved unless several purification steps are combined thereby increasing cost and reducing product yield. Conventional fractionation of proteins using ultrafiltration membranes is limited to the variation in size of the proteins and a reasonable separation factor can be observed only when the size difference is in the order of 10 or more. This is partly caused by concentration polarization and membrane fouling which hinders an effective separation of the proteins. Application of an electric field across the porous membrane has been demonstrated to be an effective way to reduce concentration polarization and membrane fouling. In addition, this technique can also be used to separate the proteins based on difference in charge, which to some extent overcome the limitations of size difference. In this thesis, separations using crossflow elecro-membrane filtration (EMF) of amino acids, bovine serum albumin (BSA) and industrial enzymes from Novozymes were performed. The main objective of this study was to investigate the technological feasibility of EMF in the application of industrial enzyme fractionation, such as removal of a side activity from the main enzyme activity. As a proof-of-concept, amino acids were used as model solution to test the feasibility of EMF in the application of amphoteric molecule separation. A single amino acid was used to illustrate the effect of an electric field on the transport of a charged amino acid; the mass transport can be enhanced or decreased enormously when an electric field is applied in the same direction with convective transport or opposite to the direction of convective transport. Water splitting caused by limiting current density situation was observed at polarity +UF- (anode at ultrafiltration membrane side) due to the depletion of ions in the permeate compartment. By applying the electric field in UF filtration, it was possible to uncouple the transport between the charged Glutamic acid (Glu) and neutral Leucine (Leu) due to the fact that mass transport of Glu was enormously decreased because of electrophoretic force and that of Leu was not affected. The separation performance can be tuned by choosing different combinations of current density and TMP. The highest selectivity value (Leu separation from Glu) was achieved at nearly 90 in the condition of 60 A/m2 current density and TMP 0.3bar. The effect of electric field was also investigated and verified with EMF filtration of BSA solution. EMF filtration of BSA both with ultrafitration (UF) membrane and more open microfiltration (MF) membrane was studied and compared with normal UF and MF filtration in terms of flux and transmission. It was found that the flux and BSA transmission can be well manipulated and predicted based on the knowledge of solution pH and the polarity of electric field. However, the membrane-protein and protein-protein interactions caused by electrostatic interactions have to be taken into account and should be considered for optimization purpose. Finally the separation experiments with a binary mixture of Lipase (LP) and Phospholipase (PLA) were performed. Results have shown that separation of LP (side activity) from PLA (main activity) which is not possible to achieve with normal MF has been successfully performed with EMF filtration using MF membrane. The highest selectivity value (LP separation from PLA) of around 5 was obtained when operating with EMF. The effects of feed concentration, solution pH, property of porous membrane TMP and electric field strength have been investigated in the EMF experiments. It has been found that the separation performance in terms of selectivity and Lipase purity in permeate was dependent on the feed concentration, solution pH and membrane properties. The effects of increasing electric field strength and TMP on the separation performance were very small in the investigated range. The mass transport of each enzyme can be well explained by the Extended-Nernst-Planck equation. Better separation was observed at lower feed concentration, higher solution pH in the investigated range and with a polysulfone (PS) MF membrane. It can be concluded that EMF has been successfully demonstrated for the separation of enzymes which normal pressure-driven membrane process could not achieve. However, in order to achieve better separation a holistic optimization procedure is needed for future work.