The presence of cyanobacteria and associated cyanotoxins in surface water is of increasing concern. Microcystins are one of the most dangerous and commonly occurring classes of cyanotoxins. Ingestion of microcystin-LR can lead to liver damage and the promotion of liver tumors. Due to adverse health effects, the World Health Organization set a guideline level of 1 part per billion (ppb) for microcystin-LR in drinking water. However, current water treatment facilities may not specifically treat drinking water for microcystins.
The overall goal of this research was to develop an advanced and effective process for the removal of microcystins from drinking water. To achieve this goal, powdered activated carbon (PAC), iron oxide nanoparticles, and ultrafiltration (UF) membranes were explored as promising treatment technologies.
The use of ultrafiltration was investigated for the rejection of microcystin-LR from drinking water. Adsorption dominated rejection for most UF membranes, at least at early filtration times, while both size exclusion and adsorption were important in removing microcystin-LR by the tight thin-film membranes. The extent of membrane adsorption was generally related to membrane hydrophobicity.
The application of ultrafiltration coupled with powdered activated carbon (PAC-UF) was also investigated. Of the two different PAC materials, wood-based activated carbon was more effective at removing microcystin-LR than coconut-based carbon due to greater mesopore volume. The PAC-UF system had the highest removal efficiency among the three processes (i.e., PAC adsorption, ultrafiltration, and PAC-UF) for both hydrophobic polyethersulfone (PES) and hydrophilic cellulose acetate (CA) membranes. When PAC was coupled to UF using PES membranes, greater removal of microcystin-LR occurred compared to when CA membranes were used, due to sorption of the toxin to the PES membrane surface.
In further studies, Suwannee River Fulvic Acid (SRFA) was used to examine the effect of natural organic matter on the removal of microcystin-LR during UF or PAC-UF. When PES membranes were previously fouled by SRFA, increased size exclusion and reduced adsorption of microcystin-LR were observed, probably due to pore blockage and fewer available adsorption sites as a result of SRFA sorption. However, simultaneous addition of both microcystin and SRFA resulted in no change in microcystin-LR adsorption. The presence of SRFA reduced microcystin-LR removal by PAC-UF, primarily due to competition between SRFA and microcystin-LR for adsorption sites on the PAC surface.
Finally, an adsorption study was performed on microcystin-LR using iron oxide (maghemite) nanoparticles. Adsorption was primarily attributed to electrostatic interactions, although hydrophobic interactions may also play a role. The adsorption of microcystin-LR decreased with increasing pH. The ionic strength affected microcystin adsorption by screening the electrostatic interactions. Adsorption decreased at higher SRFA concentrations (above 2.5 mg/L) due to competitive adsorption between SRFA and microcystin-LR for limited sorption sites.
This laboratory-scale work is an initial step in developing an advanced treatment system that could be easily incorporated into drinking water treatment facilities. It is expected that this research can provide both practical and fundamental information for more efficient process design, leading to effective removal of harmful cyanotoxins and improved water quality and safety.