HYDROGEN-BASED, HOLLOW-FIBER MEMBRANE BIOFILM REACTOR FOR REDUCTION OF PERCHLORATE AND OTHER OXIDIZED CONTAMINANTS
R. Nerenberg1 and B. E. Rittmann2 Department of Civil and Environmental Engineering, Northwestern University 2145 Sheridan Road, Evanston, Illinois 60208-3109, USA. (E-mail: r-nerenberg@northwestern.edu) 2 Department of Civil and Environmental Engineering, Northwestern University, (E-mail: b-rittmann@northwestern.edu) Abstract Many oxidized pollutants, such as nitrate, perchlorate, bromate, and chlorinated solvents, can be microbially reduced to less toxic or less soluble forms. For drinking water treatment, an electron donor must be added. Hydrogen is an ideal electron donor, as it is non-toxic, inexpensive, and sparsely soluble. We tested a hydrogen-based, hollow-fiber membrane biofilm reactor (MBfR) for reduction of perchlorate, bromate, chlorate, chlorite, chromate, selenate, selenite, and dichloromethane. The influent included 5-mg/L nitrate or 8-mg/L oxygen as a primary electron accepting substrate, plus 1 mg/L of the contaminant. The mixed-culture reactor was operated at a pH of 7 and with a 25-minute hydraulic detention time. High recirculation rates provided completely mixed conditions. The objective was to screen for the reduction of each contaminant. The tests were short-term, without allowing time for the reactor to adapt to the contaminants. Nitrate and oxygen were reduced by over 99 percent for all tests. Removals for the contaminants ranged from a minimum of 29% for chlorate to over 95% for bromate. Results show that the tested contaminants can be removed as secondary substrates in an MBfR, and that the MBfR may be suitable for treating these and other oxidized contaminants in drinking water. Key Words: Biofilm reactor, hollow-fiber membrane, hydrogen, secondary substrate, denitrification.
1
INTRODUCTION In recent years, several oxidized contaminants have emerged as drinking water pollutants, including arsenate (H2AsO4-), chromate (CrO42-), selenate (SeO42-), bromate (BrO3-), and, most recently, perchlorate (ClO4-). In many cases, conventional water treatment processes, as well as oxidative processes such as chlorine-oxidation or ozonation, are ineffective. Advanced separation processes, such as reverse osmosis, ion exchange, membrane filtration, and electrodialysis, can be effective, but are expensive and generate concentrated wastes that require proper disposal. Biological reduction may provide a more suitable treatment alternative, especially when the oxidized contaminant is reduced to a less toxic species (Lovley and Coates, 1997). Many oxidized contaminants are reduced in thermodynamically favorable reactions (Table 1) and have been shown to support bacterial growth. However, in some cases the treatment standards may be below bacterial growth thresholds (Smin) (Rittmann and McCarty, 2001). In such cases, reduction may occur in parallel to reduction of more amply available "primary" electron acceptors, such as nitrate or oxygen. In this research, we tested a hollow-fiber membrane biofilm reactor (MBfR) for reduction and removal of several oxidized contaminants when nitrate or oxygen served as primary electron acceptors. The MBfR is ideal for treating oxidized compounds in drinking water, as it uses hydrogen as an electron donor (Ergas and Reuss, 2001; Lee and Rittmann, 2002; Nerenberg et al., 2002). Hydrogen is non-toxic, inexpensive compared to organic donors, and leaves no residuals that could cause bacterial re-growth. Table 1 lists important oxidized contaminants, their reduction reactions with hydrogen as the electron donor, and their Gibbs free energy. Oxygen and sulfate are shown for reference.

下一页