Bernard Chaufer

ULTRAFILTRATION OF LYSOZYME AND BOVINE SERUM ALBUMIN WITH POLYSULFONE MEMBRANES MODIFIED WITH QUATERNIZED POLYVINYLIMIDAZOLE

Abstract In order to investigate the influence of ionic interactions on protein ultrafiltration, sulfonated polysulfone and polyethersulfone membranes were coated with polyvinylimidazole (PVI) and quarternized by a cross-linking agent bearing Bisphenol A. The apparent PVI thickness, from ATR-FTIR, was in the range from 10 to 40 nm for the polyethersulfone and sulfonated polysulfone membranes. Significant difference in protein retention (positive lysozyme and negative bovine serum albumin) was observed at low ionic strength for oppositely charged protein and (modified) membrane. At high ionic strength, protein retention and membrane flux with modified and unmodified membranes were similar due to protein adsoption on the membrane and interactions of adsorbed protein to free protein. Hydrophobic interactions between Bisphenol A and protein were demonstrated with similarly modified HPLC silica. Hence, the hydrophobic interactions could partly explain the unexpected high protein retention observed both with the unmodified and modified membrane at high ionic strength.

Preparation of water-soluble chelating aminated starch derivatives and their use for the concentration of metal ions by ultrafiltration

Abstract Water-soluble chelating starch derivatives bearing aminated ligands (diethanolamine, St-DEA; ethylenediamine, St-EDA; tetraethylenepentamine, St-TEPA) were prepared. Stoichiometries and stability constants of Cu 11 -St-EDA complexes were determined by the Fronaeus method. The concentration by ultrafiltration of Cu 2+ ion solutions in the presence of aminated starches was investigated as a function of the molar ratio ligand: metal (retention coefficient, R , 95–99%). Higher retention coefficients were expected when considering the high stability constants of the complexing EDA and TEPA ligands. The amine assays of the filtrates show that the membrane lets small amounts of the polymer pass. This is due to distribution of the pore-size within the mmebrane, as confirmed by the R values for standard dextrans of various membranes. In the case of strong-complexing polymers such as St-EDA and St-TEPA, the measured metal transfer corresponds actually to the metal bound to the macromolecular ligand. These results are discussed in terms of membrane molecular weight cut-off, the relation between metal transfer and stability constants of complexion, and efficiency of metal ion concentration by ultrafiltration.