Haloacetic acid-degrading bacterial communities in drinking water systems as determined by cultivation and by terminal restriction fragment length polymorphism of PCR-amplified haloacid dehalogenase gene fragments.

AIMS: To characterize the HAA-degrading bacteria in drinking water systems. METHODS AND RESULTS: Haloacetic acid (HAA)-degrading bacteria were analysed in drinking water systems by cultivation and by a novel application of terminal restriction fragment length polymorphism (tRFLP). Substantial similarities were observed among the tRFLP patterns of dehI and dehII gene fragments in drinking water samples obtained from three different cities (Minneapolis, MN; St Paul, MN; Bucharest, Romania) and from one biologically active granular activated carbon filter (Hershey, PA). The dominant fragment in the tRFLP profiles of dehI genes from the drinking water samples matched the pattern from an Afipia sp. that was previously isolated from drinking water. In contrast, the dominant fragment in the tRFLP profiles of dehII genes did not match any previously characterized dehII gene fragment. PCR cloning was used to characterize this gene fragment, which had <65% nucleotide sequence identity with any previously characterized dehII gene. CONCLUSIONS: Afipia spp. are an appropriate model organism for studying the biodegradation of HAAs in drinking water distribution systems as encoded by dehI genes; the organism that harbours the most prominent dehII gene in drinking water has yet to be cultivated and identified. SIGNIFICANCE AND IMPACT OF THE STUDY: The development of a novel application of tRFLP targeting dehI and dehII genes could be broadly useful in understanding HAA-degrading bacteria in numerous environments.

Detection and enumeration of haloacetic acid-degrading bacteria in drinking water distribution systems using dehalogenase genes.

AIMS: To develop a PCR-based tracking method for the detection of a subset of bacteria in drinking water distribution systems capable of degrading haloacetic acids (HAAs). METHODS AND RESULTS: Published degenerate PCR primers were used to determine that 54% of tap water samples (7/13) were positive for a deh gene, indicating that drinking water distribution systems may harbour bacteria capable of HAA degradation. As the published primer sets were not sufficiently specific for quantitative PCR, new primers were designed to amplify dehII genes from selected indicator strains. The developed primer sets were effective in directly amplifying dehII genes from enriched consortia samples, and the DNA extracted from tap water provided that an additional nested PCR step for detection of the dehII gene was used. CONCLUSIONS: This study demonstrates that drinking water distribution systems harbour microbes capable of degrading HAAs. In addition, a quantitative PCR method was developed to detect and quantify dehII genes in drinking water systems. SIGNIFICANCE AND IMPACT OF THE STUDY: The development of a technique to rapidly screen for the presence of dehalogenase genes in drinking water distribution systems could help water utilities determine if HAA biodegradation is occurring in the distribution system.

Non-steady state simulation of BOM removal in drinking water biofilters: model development.

A numerical model was developed to simulate the non-steady-state behavior of biologically-active filters used for drinking water treatment. The biofilter simulation model called "BIOFILT" simulates the substrate (biodegradable organic matter or BOM) and biomass (both attached and suspended) profiles in a biofilter as a function of time. One of the innovative features of BIOFILT compared to previous biofilm models is the ability to simulate the effects of a sudden loss in attached biomass or biofilm due to filter backwash on substrate removal performance. A sensitivity analysis of the model input parameters indicated that the model simulations were most sensitive to the values of parameters that controlled substrate degradation and biofilm growth and accumulation including the substrate diffusion coefficient, the maximum rate of substrate degradation, the microbial yield coefficient, and a dimensionless shear loss coefficient. Variation of the hydraulic loading rate or other parameters that controlled the deposition of biomass via filtration did not significantly impact the simulation results.

Non-steady state simulation of BOM removal in drinking water biofilters: applications and full-scale validation.

A biofilter model called "BIOFILT" was used to simulate the removal of biodegradable organic matter (BOM) in full-scale biofilters subjected to a wide range of operating conditions. Parameters that were varied included BOM composition, water temperature (3.0-22.5 degrees C), and biomass removal during backwashing (0-100%). Results from biofilter simulations suggest a strong dependence of BOM removal on BOM composition. BOM with a greater diffusivity or with faster degradation kinetics was removed to a greater extent and also contributed to shorter biofilter start-up times. In addition, in simulations involving mixtures of BOM (i.e. readily degradable and slowly degradable components), the presence of readily degradable substrate significantly enhanced the removal of slowly degradable material primarily due to the ability to maintain greater biomass levels in the biofilters. Declines in pseudo-steady state BOM removal were observed as temperature was decreased from 22.5 to 3 degrees C and the magnitude of the change was significantly affected by BOM composition. However, significant removals of BOM are possible at low temperatures (3-6 degrees C). Concerning the impact of backwashing on biofilter performance, BOM removal was not affected by backwash resulting in biomass removals of 60% or less. This suggests that periodic backwashing should not significantly impact biofilter performance as observed biomass removals from full-scale biofilters were negligible. In general, the simulation results were in good qualitative and quantitative agreement with experimental results obtained from full-scale biofilters.

Passive dissolution of hydrogen gas into groundwater using hollow-fiber membranes.

A new hollow-fiber membrane remediation system has recently been developed to passively supply groundwater with dissolved hydrogen (H2) to stimulate the biodegradation of chlorinated solvents. Understanding the mass transfer behavior of membranes under conditions of creeping flow is critical for the design of such systems. Therefore, the objectives of this research were to evaluate the gas transfer behavior of hollow-fiber membranes under conditions typical of groundwater flow and to assess the effect of membrane configuration on gas transfer performance. Membrane gas transfer was evaluated using laboratory-scale glass columns operated at low flow velocities (8.6-12,973 cm/d). H2 was supplied to the inside of the membrane fibers while water flowed on the outside and normal to the fibers (i.e. cross-flow). Membrane configuration (single fiber and fabric) and membrane spacing for the fabric modules did not affect gas transfer performance. Therefore, the results from all of the experiments were combined to obtain the following dimensionless Sherwood number (Sh) correlation expressed as a function of Reynolds number (Re) and Schmidt number (Sc): Sh = 0.824Re(0.39)Sc(0.33) (0.0004

Evaluation of polyethylene hollow-fiber membranes for hydrogen delivery to support reductive dechlorination in a soil column.

Engineered systems are often needed to supply an electron donor, such as hydrogen (H(2)), to the subsurface to stimulate the biological dehalogenation of perchloroethene (PCE) to ethene. A column study was performed to evaluate the ability of gas permeable hollow-fiber membranes to supply H(2) directly to PCE-contaminated groundwater to facilitate bioremediation. Two glass columns were packed with soil obtained from a trichloroethene-contaminated site at Cape Canaveral, Florida, and were fed a minimal medium spiked with PCE (7 microM) for 391 days. The columns were operated in parallel, with one column receiving H(2) via polyethylene hollow-fiber membranes (lumen H(2) pressure of approximately 1atm) and a control column receiving no H(2). PCE was initially dechlorinated at a similar rate and to a similar extent in both columns, likely due to the presence of soil organic matter that was able to support dechlorination. After 265 days of operation, dechlorination performance declined in the control column and the benefits of membrane-supplied H(2) became evident. Although the membrane-supplied H(2) effectively stimulated PCE dechlorination at the end of the experiment (days 359-391), the system was inefficient in that only 5% of the supplied H(2) was used for dechlorination. Most of the remainder was used to support methanogenesis (94%). Despite the dominance of methanogens, nearly complete dechlorination of PCE to ethene was observed in the H(2)-fed column. In addition to the inefficient use of H(2), operational problems included excessive foulant accumulation on the outside of the membrane fibers and water condensation inside the fibers. Use of alternative membrane materials and changes to the operating approach (e.g. pulsing or supplying H(2) at low partial pressures) may help to overcome these problems so that this technology can provide effective and stable remediation of aquifers contaminated with chlorinated ethenes.

Membrane gas transfer under conditions of creeping flow: modeling gas composition effects.

A computational model was developed to predict gas transfer and gas composition changes in membrane modules designed for addition of gases to groundwater. The model was verified using pilot-scale gas transfer experiments. The modeling and experimental results suggest that back diffusion of dissolved gases into the membrane has a significant effect on gas transfer via hollow-fiber membrane. In the experimental study, N(2) back-diffusion reduced the partial pressure of O(2) within the membrane and decreased the concentration gradient for gas transfer. The model was able to simulate both the dynamic and steady-state gas transfer behavior of the membranes under a variety of operating conditions. This model can be used to estimate gas transfer as a function of different membrane module design and operating conditions.

Reduction of haloacetic acids by Fe0: implications for treatment and fate.

To predict the fate of haloacetic acids (HAAs) in natural or engineered systems, information is needed concerning the types of reactions that these compounds undergo, the rates of those reactions, and the products that are formed. Given that many drinking water distribution systems consist of unlined cast iron pipe, reactions of HAAs with elemental iron (Fe0) may play a role in determining the fate of HAAs in these systems. In addition, zerovalent iron may prove to be an effective treatment technology for the removal of HAAs from chlorinated drinking water and wastewater. Thus, batch experiments were used to investigate reactions of four trihaloacetic acids, trichloroacetic acid (TCAA), tribromoacetic acid (TBAA), chlorodibromoacetic acid (CDBAA), and bromodichloroacetic acid (BDCAA), with Fe0. All compounds readily reacted with Fe0, and investigation of product formation and subsequent disappearance revealed that the reactions proceeded via sequential hydrogenolysis. Bromine was preferentially removed over chlorine, and TBAA was the only compound completely dehalogenated to acetic acid. In compounds containing chlorine, the final product of reactions with Fe0 was monochloroacetic acid. Halogen mass balances were 95-112%, and carbon mass balances were 62.6-112%. The pseudo-first-order rate constants for trihaloacetic acid degradation were as follows: BDCAA (10.6 +/- 3.1 h-1) > CDBAA (1.43 +/- 0.32 h-1) approximately TBAA (1.41 +/- 0.28 h-1) >> TCAA (0.08 +/- 0.02 h-1).