| Recently, there has been an increased concern of the potential effects of pharmaceuticals, personal care products (PPCPs) and endocrine disrupting compounds (EDCs) in drinking water. Their presence in surface waters has resulted in the skewing of sex ratios in aquatic biota and the effect on humans, as yet, remains unknown. Investigation into the effective removal of these compounds by water treatment plants (WTPs) has shown that conventional treatment processes are not very effective in removing these trace compounds. Studies have shown PPCPs and EDCs have been successfully removed by commercial nanofiltration (NF) and reverse osmosis (RO) membranes, but have low flux and high cost. North American WTPs, using membrane separation processes, are typically equipped with microfiltration (MF) or loose ultrafiltration (UF) membranes which, thus far, have proven ineffective for the removal of these target compounds.;This thesis focuses on the development of a tight charged UF membrane that effectively removes PPCPs and EDCs from drinking water while still maintaining a high flux and is cost effective. Novel membranes were developed by incorporating charged surface modifying macromolecules (CSMMs) in the manufacturing of polyether sulfone (PES) based membranes. The charged additives were expected to enhance the removal of PPCPs and EDCs by charge repulsion. Controls and three different CSMM (DEG-HBS, DEG-HBC and PPG-HBC) blended membranes were prepared at three different casting conditions and subsequently evaluated for various properties: flux, molecular weight cut-off (MWCO), porosity, charge and contact angle. Experimental membranes were further evaluated for the removal of four representative target compounds, sulfamethazine (SMZ), carbamazepine (Carb), bisphenol A (BPA) and ibuprofen (IB). Removal by a commercial nanofiltration membrane, NF270 (DOW/FilmTec) was compared to the experimental membranes.;Removal results from the experimental membranes indicate membranes were unable to sustain effective removal of the target compounds. Typically, removal was initially high but decreased over the run. Membrane characteristics showed membranes had significantly larger pores than the target compounds indicating size exclusion was not the removal mechanism. Charge results indicated CSMM blended membranes were generally unchanged from the control membrane indicating, in addition to the unsustained removal, that charge repulsion was not the removal mechanism. From the shape of the removal curves, it is assumed the removal mechanism is the result of membrane adsorption.;The CSMMs were found to have modified the membranes, though not sufficiently, to be considered significantly different than the controls in many respects. Membrane characteristics varied as a result of each CSMM incorporated and depending on each casting condition. Contact angle results for both PES-DEG-HBS and PES-PPG-HBC membranes at all three casting conditions increased in comparison to the controls, presumably because of changes in surface roughness. PES-DEG-HBC, on the other hand, decreased in contact angle at 18%, and increased in contact angle at 20% in comparison to the respective controls. Incorporation of migration time, particularly in the case of DEG-HBC, increased membrane flux without affecting MWCO. Increased PES concentration (from 18 to 20%) saw an increased target compound removal. With the success of the DEG-HBC CSMM, incorporation of migration time at higher PES concentrations appears promising for achieving the desired characteristics. It is recommended that further optimization using CSMM DEG-HBC at increased PES concentrations with migration time be investigated for this application. |
