The increasing concern over potential health effects of pharmaceutically active compounds (PhACs) and endocrine disrupting compounds (EDCs) in drinking water has led to an increase in assessment of drinking water treatment plant efficiencies at removing these emerging micropollutants. For the most part, tight commercial membrane processes such as nanofiltration (NF) and reverse osmosis (RO) successfully eliminate PhACs and EDCs, however these are costly processes and infrequently implemented in North American treatment facilities. The more frequently used microfiltration (MF) and loose ultrafiltration (UF) membranes are ineffective in the removal of these compounds.
This thesis focuses on developing tight charged ultrafiltration (UF) membranes which could effectively remove PhACs and EDCs from drinking water without compromising flux and cost. The approach centers on developing the membrane surface charge by incorporating charged surface modifying macromolecules (CSMMs) as additives. Four CSMMs (MDI-PPG-HBS, MDI-PEG200-HBS, MDI-PEG400-HBS, MDI-DEG-NDS) were evaluated at three different casting conditions for poly(ether sulfone) (PES) based membranes. The modified membranes were compared to controls (without CSMMs) and one commercial membrane (NF270, Dow/Filmtec). Membrane properties including flux, molecular weight cut off, surface porosity, charge and hydrophilicity were evaluated and compared to the removal of four representative PhACs and EDCs (sulfamethazine, carbamazepine, bisphenol A and ibuprofen) at the mg/L-level.
The experimental membranes only achieved a temporary partial removal of the PhACs and EDC tested, thus further development is required. Given the temporary target compound removal and the large membrane pores, size exclusion and charge repulsion are not the dominant removal mechanism. From the removal pattern, and the fact that removal increased with increasing solute hydrophobicity, it is assumed that initial removal is caused by adsorption to the membrane. The membranes developed were tight by conventional ultrafiltration standards but did not achieve the performance desired.
In general, it was found that the CSMM-modified membranes did not significantly outperform the control membranes. CSMM-modified membranes tested generally produced less hydrophilic membranes with increasing pre-gelation time or PES concentration, in comparison to the control membranes. Pre-gelation time (i.e., three minutes versus no pre-gelation time) increases membrane porosity, and therefore flux is increased, without compromising removal. Increased PES concentration (i.e., 20% PES in comparison to 18% PES) yields more distinct effects from the different CSMMs. From these results, the most promising casting condition appears to be 20% PES and the CSMMs achieving the best removal are MDI-PEG 400-HBS and MDI-PPG-HBS. As increased surface porosity was achieved, continuing this line of research by optimizing membrane preparation conditions to decrease the pore size may produce the desired characteristics. It is recommended that further tests be performed at increased PES concentrations and with pre-gelation time to achieve better results.
Identifer | oai:union.ndltd.org:uottawa.ca/oai:ruor.uottawa.ca:10393/27686 |
Date | January 2008 |
Creators | Garand-Sheridan, Anne-Marie |
Publisher | University of Ottawa (Canada) |
Source Sets | Université d’Ottawa |
Language | English |
Detected Language | English |
Type | Thesis |
Format | 168 p. |
Page generated in 0.0016 seconds