Systems that use bacteria to treat drinking water have been shown to be highly efficient and environmentally sustainable.
Microbial biomass has been used since the early 1900s to degrade contaminants, nutrients, and organics in wastewater. Until recently, the biological treatment of drinking water was limited, particularly in the United States, but recent developments may mean that biological drinking water treatment may become more feasible and more likely to be accepted by the public. These developments include (1) the rising costs and increasing complexities of handling water-treatment residuals (e.g., membrane concentrates); (2) the emergence of new contaminants that are particularly amenable to biological degradation (e.g., perchlorate); (3) the push for green technologies (i.e., processes that efficiently destroy contaminants instead of concentrating them); (4) regulations limiting the formation of disinfection by-products (DBPs); and (5) the emergence of membrane-based treatment systems, which are highly susceptible to biological fouling.
Bacteria gain energy and reproduce by mediating the transfer of electrons from reduced compounds (i.e., compounds that readily donate electrons) to oxidized compounds (i.e., compounds that readily accept electrons). Once electrons are donated by a reduced compound, they travel back and forth across a cell’s mitochondrial membrane in a series of internal oxidation-reduction reactions. Ultimately, the electrons are donated to the terminal electron-accepting compound. This series of reactions, which is cumulatively known as the electron-transport chain, creates an electrochemical gradient across the cell membrane that bacteria use to generate adenosine triphosphate, also known as energy (Madigan et al., 1997).
As compounds gain or lose electrons, they are converted to different, often innocuous, forms that are thermodynamically more stable than the original compounds. The example below illustrates the microbially mediated oxidation-reduction reaction between acetate (an electron donor) and dissolved oxygen and nitrate (two environmental electron acceptors).
- CH3COO- + 2O2 → 2HCO3- + H+
ΔG° = -844 KJ/mol acetate
- CH3COO- + 3/5NO3- + 13/5H+ → 2HCO3- + 4/5H20 + 4/5 N2
ΔG° = -792 KJ/mol acetate
Notice that nitrate, a common contaminant in drinking water, is converted to innocuous nitrogen gas. The Gibb’s free-energy values for the overall reactions are shown below the equations (Rikken et al., 1996). The more negative the Gibb’s free-energy value, the more thermo-dynamically unstable the reaction and the greater the energy yield for the bacteria mediating the reaction. Electron transfer in the overall reactions can be observed only by evaluating the oxidation states of individual atoms.
FIGURE 1 - Microbially mediated oxidation-reduction reactions.
Biological drinking water treatment processes are based on the growth of bacterial communities capable of mediating oxidation-reduction reactions involving at least one target contaminant (Figure 1). Heterotrophic biological processes use an organic electron donor (e.g., acetic acid). Autotrophic biological processes use an inorganic electron donor (e.g., hydrogen).
Biological processes can be used for a wide range of organic and inorganic contaminants (Table 1) in both surface water and groundwater.
Numerous forms and configurations of biological treatment processes are used to degrade contaminants in drinking water. Most are operated as fixed biofilm systems, meaning that the process includes a biogrowth support medium on which bacterial communities attach and grow (e.g., granular media). A smaller number of technologies operate as suspended growth systems, in which free-floating bacteria are hydraulically maintained within a reactor. Biological reactors can be inoculated with an enriched bacterial community or can simply be acclimated by the organisms indigenous to the water source being treated. Examples of biological treatment configurations are described below.
In fixed-bed (FXB) biological processes, biofilms develop on a stationary bed of media, such as sand, plastic, or granular activated carbon (Figure 2). The granular media bed can be contained in pressure vessels or open basins. In pressure-vessel systems, water is pumped up-flow or down-flow across the biological bed; in open-basin systems up-flow requires pumping, but down-flow occurs by gravity. As water is treated, biofilms increasingly restrict flow and cause head loss across the bed. If unchecked, the loss eventually exceeds the available driving pressure or causes short-circuiting through the bed. To avoid these complications, FXB systems are routinely taken off line and backwashed to remove excess biomass from the system (Brown et al., 2005; Kim and Logan, 2000). FXB is often coupled with pre-ozonation to improve the removal of organic material, which reduces regrowth potential and DBP formation in distribution systems.
FIGURE 2 - Electron micrographs showing (a) the surface of virgin granular activated-carbon media and (b) and (c) biofilm-covered granular activated carbon that was used for three years in a biological drinking water treatment reactor. Source: Reprinted with permission from WaterWorld, April 2007
Fluidized-bed reactors (FBRs) also use granular media to support biogrowth. Contaminated water is pumped up-flow through the reactor at a high rate to fluidize the granular media bed and reduce resistance to flow. Typically, the fluidization rate is controlled to maintain a 25 to 30 percent bed expansion over the resting bed height. Feed flow is supplemented with recycle flow to provide the appropriate up-flow velocity for fluidization (Green and Pitre, 1999; Guarini and Webster, 2004). Excess biomass is removed from FBR systems by (1) shear forces generated by the high feed-pumping rates and/or (2) in-line mechanical shearing devices. Therefore, although FBRs require higher feed-flow capacity, they do not require an off-line backwashing step.
Membranes can also be coupled with biological systems to improve the treatment of drinking water. In one approach, ultrafiltration membranes are submerged in a reactor basin that contains suspended biomass. The reactor basin provides the detention time necessary to achieve effective biological treatment. Treated water is drawn through the membranes by vacuum and pumped out to permeate pumps for further processing.
Airflow introduced at the bottom of the reactor basin performs several functions. First, it creates turbulence that scrubs and cleans the outside of the membranes and reduces the accumulation of solids on the membrane surface. Thus the membrane can operate for extended periods of time at high permeate fluxes. Second, the air has the beneficial side effect of oxidizing iron, manganese, and some organic compounds that may be present. Third, air ensures mixing in the process tank to maintain suspension of the biomass. Periodic backwashing of the membranes is done by passing permeate through the membranes in the reverse direction to dislodge solids from the membrane surface.
Biofilm Reactor Systems
A different approach to the “conventional” membrane bioreactor (MBR) uses hollow-fiber membranes to deliver hydrogen gas (an electron donor) to biofilms that grow on the outside of the fibers. When the hollow-fiber membranes are submerged in a reactor vessel through which contaminated water passes, contaminants diffuse from the bulk water into the biofilms and are degraded (Nerenburg et al., 2002). Occasionally, the membranes are chemically cleaned to remove excess biomass.
Ion-Exchange Membrane Systems
Yet another MBR method involves a reactor with two treatment chambers separated by an ion-exchange membrane. One chamber contains suspended biomass plus nutrients; the other chamber contains raw water. As raw water enters the system and moves through one chamber, ionic contaminants diffuse across the membrane into the biological treatment chamber where they are degraded. The objective of this approach is to separate the active biomass from the raw and treated water (Liu and Batista, 2000).
Bank Filtration Systems
Bank filtration wells, drilled near rivers and lakes, draw surface water through soil and aquifer material, which act as a passive treatment reactor. As the surface water moves through the aquifer, it is subject to filtration, dilution, sorption, and biodegradation processes (Gollnitz et al., 2003; Ray et al., 2002; Weiss et al., 2003a,b). Bank filtration, which has been used for more than 130 years in Europe, has aroused a great deal of global interest for use in reducing organic and particulate loads to drinking water treatment plants.
One of the oldest bank filtration systems draws water from the Rhine River in Germany and is part of the D?sseldorf Waterworks. This system has been in operation since 1870, and until about 1950, it was the only treatment process used at that facility.
FIGURE 3 - Bench-scale biological-treatment testing apparatus.
Various tools are available to facilitate the optimization of engineered biological treatment systems. Bench-scale reactors (Figure 3) and pilot-scale reactors (Figure 4) are often used in conjunction with mathematical models to isolate the impacts of various water quality conditions and operating parameters on overall system performance.
FIGURE 4 - Illustration of a pilot-scale biological-treatment testing apparatus.
Available commercial models can be tailored to a specific treatment application and process configuration. Typically calibrated using results from bench- and/or pilot-scale testing, these models can simulate steady-state or dynamic conditions and account for hydraulic-flow regimes from plug-flow to complete mixing.
The models incorporate a wide range of parameters, such as feed-water quality and temperature, substrate and nutrient loading rates, contact time, biofilm thickness, specific surface area of reactor media, and biomass detachment. Not only can they predict bioreactor performance for a given set of environmental conditions, they can also elucidate observed phenomena in bioreactor systems. In other words, they can eliminate the “black box” perception of bioreactor processes.
Culture-Based Microbiological Analyses
Microbiological analyses provide another optimization tool. Using a targeted nutrient medium in conjunction with specific incubation conditions, pure cultures can be isolated from the mixed community of bacteria comprising a bioreactor. An enrichment of each pure culture can then be tested to identify optimal environmental conditions for that classification of bacteria. The information can then be used to tweak the operation of or nutrient loading to a given bioreactor to favor the activity and growth of key contaminant-degrading microorganisms.
Molecular Microbiological Techniques
A complement to culture-based techniques, molecular microbiological techniques can be used to identify, quantify, locate, and track specific classes, families, genera, or species of bacteria. These techniques rely on the extraction, amplification, and sequencing (i.e., order of nucleic acid bases A, T, C, and G) of bacterial community DNA.
Once DNA sequences have been identified for a mixed community, they can be compared against large libraries of known bacterial DNA sequences to identify specific bacteria in a given treatment system to provide a “fingerprint” of the mixed microbial community. Nucleic acid probes, which are constructed using DNA sequence data to target specific bacteria, can then be used to quantify and track changes in a microbial community’s fingerprint as a function of operational conditions or water-quality characteristics.
As environmental conditions change, the composition of a microbial community also changes. Molecular microbiological techniques can rapidly identify the environmental conditions that favor the growth and activity of the key contaminant-degrading bacteria in a bioreactor.
Given that a key objective of treating drinking water is the inactivation or removal of microorganisms from raw water, using bacteria to help produce potable water would seem to fly in the face of conventional wisdom. However, biological drinking water treatment pro-cesses, which use indigenous, nonpathogenic bacteria, are always followed by downstream processes, such as final disinfection. Consequently, well designed biological treatment systems pose no significant, inherent threats to the health or safety of distributed water. On the contrary, they can often provide an alternative to conventional processes that has several potential advantages:
- low operating costs
- high water-recovery rates
- destruction, rather than sequestration or concentration, of contaminants
- simultaneous removal of multiple contaminants
- minimal sludge production
- no hazardous waste streams
- minimal or no added chemicals
- robustness over a wide range of operating conditions and water qualities
Overall, biological drinking water treatment is highly efficient and environmentally sustainable. As green water treatment philosophies gain traction and as regulatory and residuals-handling constraints continue to tighten, the use of biological drinking water treatment technologies and processes will likely continue to expand around the globe.
Bouwer, E.J., and P.B. Crowe. 1988. Biological processes in drinking water treatment. Journal of the American Water Works Association 80(9): 82–93.
Brown, J.C. 2006. Simultaneous Destruction of Multiple Drinking Water Contaminants Using Biological Filtration. Proceedings of the 2006 AWWA Annual Conference and Exhibition, San Antonio, Texas. CD available from AWWA, 6666 W. Quincy Avenue, Denver, Colorado 80235.
Brown, J.C., R.D. Anderson, J.H. Min, L. Boulos, D. Prasifka, and G.J.G. Juby. 2005. Fixed-bed biological treatment of perchlorate-contaminated drinking water. Journal of the American Water Works Association 97(9): 70–81.
Dahab, M.F., and B.L. Woodbury. 1998. Biological Options for Nitrate Removal from Drinking Water. Paper presented at the AWWA Inorganic Contaminants Workshop, San Antonio, Texas; February 22–24, 1998. CD available from AWWA, 6666 W. Quincy Avenue, Denver, Colorado 80235.
Gollnitz, W.D., J.L. Clancy, B.L. Whitteberry, and J.A. Vogt. 2003. RBF as a microbial treatment process. Journal of the American Water Works Association 95(12): 56–66.
Greene, M.R., and M.P. Pitre. 2000. Treatment of groundwater containing perchlorate using biological fluidized bed reactors with GAC or sand media. Pp. 241–257 in Perchlorate in the Environment, edited by E.T. Urbansky. New York: Kluwer Academic/Plenum.
Guarini, W.J., and T. Webster. 2004. Ex Situ Biological Treatment of Perchlorate Using Biological Fluidized-Bed Reactors—An Update. Proceedings of the East Valley Water District/AWWARF Water Quality Conference, Ontario, California. CD available from AWWA, 6666 W. Quincy Avenue, Denver, Colorado 80235.
Herman, D.C., and W.T. Frankenberger Jr. 1999. Bacterial reduction of perchlorate and nitrate in water. Journal of Environmental Quality 28(3): 1018–1024.
Kim, K., and B.E. Logan. 2000. Fixed-bed bioreactor treating perchlorate-contaminated waters. Environmental Engineering Science 17(5): 257–265.
Kirisits, M.J., V.L. Snoeyink, J.C. Chee-Sanford, B.J. Daugherty, J.C. Brown, and L.M. Raskin. 2002. Effects of operating conditions on bromate removal efficiency in BAC filters. Journal of the American Water Works Association 94(4): 182–193.
Lauderdale, C.V., J.C. Brown, B. MacLeod, M. Simpson, and K. Petitte. 2006. A Novel Biological Treatment Approach for the Removal of Algal Metabolites. Proceedings of the 2006 AWWA Water Quality and Technology Conference, Denver, Colorado. CD available from AWWA, 6666 W. Quincy Avenue, Denver, Colorado 80235.
Liu, J., and J. Batista. 2000. A Hybrid (Membrane/Biological) System to Remove Perchlorate from Drinking Waters. Paper presented at Perchlorate Treatment Technology Workshop, 5th Annual Joint Services Pollution Prevention and Hazardous Waste Management Conference and Exhibition, August 21–24, 2000, San Antonio, Texas.
Madigan, M.T., J.M. Martinko, and J. Parker. 1997. Brock Biology of Microorganisms, 8th ed. Upper Saddle River, N.J.: Prentice Hall.
Nerenberg, R., B.E. Rittmann, and I.N. Najm. 2002. Perchlorate reduction in a hydrogen-based membrane-biofilm reactor. Journal of the American Water Works Association 94(11): 103–114.
Ray, C., G. Melin, and R.B. Linsky. 2002. Riverbank Filtration: Improving Source-Water Quality. The Netherlands: Kluwer Academic Publishers.
Rikken, G.B., A. G. M. Kroon, and C. G. van Ginkel. 1996. Transformation of (per)chlorate into chloride by a newly isolated bacterium: reduction and dismutation. Applied Microbiology and Biotechnology 45(3): 420–426.
Weiss, W.J., E.J. Bouwer, W.P. Ball, C.R. O’Melia, M.W. LeChevallier, H. Arora, and T.F. Speth. 2003a. Riverbank filtration—fate of DBP precursors and selected microorganisms. Journal of the American Water Works Association 95(10): 68–81.
Weiss, W.J., E.J. Bouwer, W.P. Ball, C.R. O’Melia, H. Arora, and T.F. Speth. 2003b. Comparing RBF with bench-scale conventional treatment for precursor reduction. Journal of the American Water Works Association 95(12): 67–80.
Rittmann, B.E., and P.L. McCarty. 2001. Environmental Biotechnology. Boston, Mass.: McGraw-Hill.
TABLE 1 Contaminants Amenable to Biological Treatment1
|Natural organic matter
- Regrowth substrate
- DBP precursors
- Membrane foulants
- The biological oxidation of carbonaceous organic matter to CO2 can minimize potential regrowth in distribution systems, decrease the production of DBPs, remove color, and improve transmembrane fluxes without chemical additives. Ozone is often used before a biological process to increase removal of NOM.
- 2-methyl-isoborneol (MIB)
- Algal toxins
- Endocrine disruptors and pharmaceutically active compounds
- Biological oxidation to CO2, often degraded as a secondary electron donor (i.e., does not yield the requisite energy to support cell maintenance and growth), requires the presence of a primary substrate, such as NOM.
- Methyl tertiary-butyl ether (MTBE)
- Biological oxidation to CO2.
- Perchloroethylene (PCE)
- Trichloroethylene (TCE)
- Dibromochloro-propane (DBCP)
- Biological reductive dechlorination produces innocuous ethane or CO2.
- Biological reduction produces innocuous end products (Cl–, N2, Br–, H2O), thus eliminating the generation of a contaminated concentrate stream.
- Biological reduction produces insoluble species that can be readily filtered or settled out of water, thus eliminating the need for chemical reduction.
- Biological oxidation of ammonia to nitrate provides an alternative to chemically intensive break-point chlorination.
- Biological oxidation of soluble species (Fe2+, Mn2+) to insoluble species (Fe3+, Mn4+) eliminates the need for chemical oxidation prior to filtration or settling.
Source: Adapted from personal experience and Bouwer and Crowe, 1988; Brown, 2006; Brown et al., 2005; Dahab and Woodbury, 1998; Herman and Frankenberger Jr., 1999; Kirisits et al., 2002; and Lauderdale et al., 2007.