(658g) Combined Electrocoagulation and Nanofiltration for Removal of Toxic Low Molecular Weight Contaminants from Surface Water

Nanofiltration is a pressure driven membrane based separation process that is capable of rejecting low molecular weight species even as low as 200 Da. It attractive for removing small molecular weight contaminants form surface waters. Here we have investigated the use of nanofiltration for removal of microcystins and PFAS compounds. The former are class of toxic compounds that are produced by some freshwater cyanobacteria or blue-green algae. During algal blooms the release of microcystins can cause serious contamination of drinking water supplies. Here we focus on microcystin-LR (MCLR) which has been found to contaminate local water bodies, where it has been detected at concentrations of up to 50 µg/L. MCLR has a molecular weight of 1000 Da.

PFAS or Per- and Polyfluoroalkyl Substances (PFAS) are becoming increasingly abundant and causing concern in water systems worldwide due to their toxcity. Here we focus on perfluorooctanoic acid (PFOA) which has been detected in local water bodies at concentration of up to 360 µg/L. The high stability, mobility, and toxicity of these compounds has resulted in significant efforts to remove them from surface waters. PFOA has a molecular weight of 400 Da and should be rejected by tighter nanofiltration membranes.

Fouling of nanofiltration membranes, when treating surface waters often limits the viability of nanofiltration. Here we have investigated the use of electrocoagulation as a pretreatment for nanofiltration. This study aimed to answer the following questions.

• How effective is nanofiltration at removal of MCLR and PFOA from Lake Fayetteville water?
• How effective is electrocoagulation as a pretreatment step?
• Can additional MCLR and PFOA removal be obtained during the electrocoagulation step?

Experiments were conducted by spiking MCLR and PFOA into DI water and actual Lake Fayetteville water as needed. The Lake Fayetteville water was characterized and found to contain MCLR naturally, whereas PFOA was spiked into the water. Commercially available NF270 and BW30 (Dow Chemical, Midland, MI, USA) membranes were used. Experiments were run in both dead end (for screening various conditions) and tangential flow mode. Membrane performance (flux and contaminant rejection) was determined with and without electrocoagulation as a pretreatment step. The following ranges of the following variables were investigated. Pressure was varied from 50 - 150 PSI, spiked contaminant concentration was varied between 10- 1000 µg/L, fouling studies were conducted using bovine serum albumin and alginate as model foulants, and experiments were repeated with and without an electrocoagulation pretreatment. Contaminant concentration was measured using liquid chromatography mass spectrometry (LCMS). Membranes were characterized through contact angle testing and atomic force microscopy to identify changes to the membrane’s surface caused by adsorption of rejected species which leads to fouling.

Our results indicate that both processes are effective at removing the low molecular weight contaminants of interest. The nanofiltration membrane tended to reject the small contaminants better from pure water than real water, reaching as high as 95% and 71% for MCLR, respectively. Fouling was also dominant when using real water as expected. NF270 experienced a flux drop of almost 75% and the tighter BW30 immediately lost 99% of its permeability when testing with water from Lake Fayetteville. Electrocoagulation results showed partial removal of both contaminants. For example, maximum removal was commonly recorded at nearly 55% for PFOA. Reducing the initial PFOA concentration was also found to increase the removal percentage. Lowering the initial concentration to 50 µg/L and 10 µg/L increased the removal to 70% and 82%, respectively. This ability to capture these low molecular weight contaminants warrants its use as a functional pretreatment for nanofiltration.