Access to clean water is one of the world's greatest concerns. Because 97% of global water resources are seawater, desalination via reverse osmosis (RO) membrane process has become a vital technology to obtain drinkable water. At the same time, the discharge of industrial waste effluents containing heavy metal ions to the available water resources (seawater and brackish water) without adequate pre-treatment is a major cause of water pollution. Heavy metal rejection using nanofiltration (NF) membrane process is a recognized water treatment methodology. Thin-film nanocomposite (TFN) membranes have shown vast performance enhancement using both RO and NF processes. However, TFN membrane fabrication has been limited due to poor dispersion of the nanoparticles in the polyamide (PA) layer of the membrane, and the leaching of the often-hazardous nanoparticles from the TFN membranes.
For various reasons such as their dispersibility in aqueous media, safety, high aspect ratio, and functionality, cellulose nanocrystals (CNCs) are an ideal nanoparticle for inclusion in TFN membranes. Because of their hydrophilicity, CNCs have more commonly been dispersed in the aqueous monomer solution during PA interfacial polymerization. In this thesis, we investigated two different CNC modification routes to improve CNC dispersion within the trimesoyl chloride (TMC)/n-hexane (non-aqueous) monomer solution. In one case, we acetylated the CNCs (ACNCs) using a straightforward, efficient, solvent-free method to achieve a more uniform CNC dispersion in the PA layer. The resulting ACNCs were less hydrophilic, which allowed increased nanoparticle loading and improved dispersion in the PA layer. In an RO desalination process, compared to unmodified CNC-TFN membranes, the NaCl rejection of the ACNC-TFN membranes remained stable (at 98-99%) up to a 0.4 wt% loading, while water permeability increased by up to 40%.
For the second case, we synthesized L-cysteine functionalized CNCs (CysCNCs) and incorporated them into the PA layer for testing in an NF wastewater treatment process. The amine functional groups of L-cysteine covalently bonded with the acyl chloride groups of the TMC monomer. This resulted in improved nanoparticle dispersion but could also have prevented nanoparticle leaching. Moreover, because L-cysteine contains strong chelating groups, their inclusion in the PA layer led to improved heavy metal rejection. A loading of 0.1 wt% CysCNCs in the TFN membranes provided high rejection of both copper and lead ions, 98.1 and 95.2%, respectively. The CysCNCs were also evaluated in an NF desalination process resulting in a 40% increase in water permeability with almost no decline in Na₂SO₄ (97-98%), MgCl₂ and NaCl rejection. The modified CNCs enabled us to overcome the water permeability/selectivity trade-off in CNC-TFN membranes for both RO and NF membrane desalination.
Finally, we developed an experimental protocol to investigate the effect of the adsorption of heavy metal ions (if any) on the performance of thin film composite (TFC) and TFN membranes in NF. We confirmed that adsorption occurred, and the equilibrium capacity of the membranes was reached after 8 - 12 h of the experiment. Despite reaching the equilibrium capacity, the water permeability and heavy metal rejection remained at their highest values. This led to the conclusion that the adsorbed heavy metals altered the membrane surface, thereby improving the performance of both TFC and TFN membranes.
The ability to modify CNCs enables one to achieve a controlled range of hydrophilicity/ hydrophobicity. This allows one to fine-tune CNC compatibility with the TMC/n-hexane non-aqueous monomer solution and enable improved dispersion in the PA layer, eventually leading to improved TFN membrane performance for both RO and NF processes.
Identifer | oai:union.ndltd.org:uottawa.ca/oai:ruor.uottawa.ca:10393/45239 |
Date | 10 August 2023 |
Creators | Abedi, Fatemeh |
Contributors | Dubé, Marc, Kruczek, Boguslaw |
Publisher | Université d'Ottawa / University of Ottawa |
Source Sets | Université d’Ottawa |
Language | English |
Detected Language | English |
Type | Thesis |
Format | application/pdf |
Page generated in 0.0627 seconds