Lake Erie is as a source of drinking water for over 13 million people, with annual water use of approximately 2 billion gallons. In recent years, blooms of Microcystis algae have occurred that produce microcystin-LR, and other toxins, that have both acute and chronic effects in humans. Although no maximum contaminant level (MCL) has been set in the United States, the World Health Organization (WHO) has established a safe level of microcystin-LR in drinking water of 1 mg/L. Because conventional treatment processes are ineffective at removing microcystins, approaches for the removal of these toxins during drinking water treatment are needed. In this research, we will examine the use of powdered activated carbon-ultrafiltration (PAC-UF) for the removal of microcystin-LR during drinking water treatment. PAC-UF effectively removes pathogens, is less prone to fouling compared to nanofiltration, and recent studies indicate this technology can also remove trace organic contaminants. Despite the potential of PAC-UF, no research has been carried out examining the effectiveness of this technology for removing cyanobacterial toxins. The objectives of this research include
Determine the most effective PAC-UF process configurations for the removal of microcystins from drinking water
Examine the effect of pretreatment processes (i.e., coagulation, softening, anti-scalant addition) on microcystin removal by PAC-UF, and
mechanisms by which natural organic matter (NOM) influences the removal of microcystins by this advanced treatment process.

Blooms of Microcystis, and subsequent release of the associated toxin microcystin-LR, may result in both acute and chronic health effects if present in drinking water. The research proposed here would provide both practical and fundamental information on the effectiveness of advanced treatment processes for removing microcystin-LR from drinking water. Successful completion of this project will generate important information water utilities can use to alter current unit processes during Microcystis blooms, and also provide information for the design of new, more effective treatment operations in the future.

To determine the best process configurations for the removal of microcystin-LR by PAC-UF, we will use a cross-flow flat-sheet membrane filtration cell coupled to a PAC slurry reactor and a variable pressure pump. The system setup will accommodate different points of PAC addition (slurry reactor versus immediately before UF) and different filtration modes (cross-flow versus dead-end). Source water spiked with microcystin-LR will be passed through the system and the amount of microcystin-LR measured in feed, permeate and concentrate lines measured by HPLC. We will also measure turbidity, dissolved organic carbon, and total organic carbon to evaluate the ability of PAC-UF to meet multiple treatment objectives. To examine how source water constituent affect microcystin removal, we will examine different concentrations of inorganic elements (e.g., Ca, Mg, Si, Fe, Al) and anions (nitrate and sulfate).

Pretreatment may affect the adsorption capacity of PAC as well as influence the interactions of microcystin and activated carbon with the membrane surface. To determine the influence of pretreatment, we will carry out PAC-UF experiments at varying lime/soda ash, coagulant, and anti-scalant dosages using the bench-scale system described above and a standard jar test apparatus. In addition to elucidating the impacts of pretreatment on PAC-UF, these experiments will also provide useful information for utilities currently using PAC with conventional treatment processes such as the coagulation and lime-soda softening. SEM and TEM-EDS will be used to examine the formation of precipitates on PAC and membrane surfaces. The adsorption of NOM to activated carbon and/or membranes will influence both the physical and chemical interactions between microcystin and these surfaces.

To examine the mechanisms by which natural organic matter influences the removal of microcystin by PAC-UF we will utilize a host of powerful analytical tools, including ATR-FTIR, to probe interactions between natural organic matter and microcystin, activated carbon, and membrane surfaces. In this proposal, we plan to examine two mechanistic questions in particular: (1) How do the physical and chemical characteristics of NOM influence the adsorption of microcystin on the surface of PAC? and (2) What influence does NOM have on the interactions between microcystin and UF membrane surfaces?