Perchlorate is a groundwater contaminant that has recently received heightened attention. Its presence is often associated with facilities that once manufactured, handled, or stored ammonium perchlorate, a solid-rocket fuel oxidant. The severity and extent of perchlorate contamination was difficult to assess until 1997, when a new ion chromatographic method was developed to decrease the limit of detection (LOD) for perchlorate from 400 µg/L to 4 µg/L. Since then, perchlorate has been detected in drinking water sources in 25 states.
Both abiotic and biotic processes have been developed and evaluated for treating perchlorate-contaminated drinking water. Typical abiotic perchlorate treatment processes include ion exchange, reverse osmosis/nanofiltration, electrodialysis reversal, and tailored granular activated carbon (GAC). These processes separate perchlorate from the bulk solution by adsorption or diffusion-limited filtration.
The main drawback with abiotic approaches is that they each create a concentrated perchlorate waste stream that must be further treated or disposed. On the other hand, biological processes convert perchlorate to innocuous chloride and oxygen, thereby eliminating perchlorate from the environment. Of the various available biological perchlorate treatment technologies, none has been tested more extensively on drinking water and been demonstrated to be as simple, efficient, robust, and cost-effective as GAC-based heterotrophic (i.e., uses organic carbon sources) fixed-bed (FXB) bioreactors. The main advantages of FXB biological processes relative to conventional perchlorate treatment processes include:
The overall objective of this work was to evaluate the efficacy of using FXB bioreactors and post-treatment to remove perchlorate from drinking water and to produce water that meets all regulations. Specific project emphases included the demonstration of sustained perchlorate removal capabilities, the identification and evaluation of process limitations and potential failure scenarios, and the development of realistic designs and cost estimates for full-scale, potable FXB biological perchlorate treatment.
In February 2007, a 10-month demonstration was initiated in Rialto, California, to treat perchlorate-contaminated groundwater using FXB bioreactor technology. Two first-stage, parallel FXB bioreactors (F120 with a 3.9-ft bed depth and a 2-ft diameter, and F130 with a 4.7-ft bed depth and a 2-ft diameter) treated groundwater to remove perchlorate. Effluent from these reactors was dosed with hydrogen peroxide (i.e., reoxygenate + oxidize residual organics and hydrogen sulfide). The reoxygenated water was then passed through an FXB biofilter (F150) to oxidize any remaining organics and sulfide and to remove turbidity. Chlorine was then dosed to the effluent of the biofilter as a final disinfection step. In parallel with the pilot testing, a mathematical model was developed and calibrated, which can be used to elucidate observed phenomena during pilot testing and to predict the perchlorate removal performance of an FXB bioreactor system at other sites. Additionally, molecular microbiological analyses were performed to quantify the relative abundance of specific bacteria within the mixed microbial community in the bioreactor bed. A bench-scale FXB bioreactor was also constructed to test how nutrient addition and intermittent electron donor addition patterns affect the performance and microbial community of a bioreactor. Tests were run using the bench-scale bioreactor that could not be easily conducted using the demonstration-scale system. The bench-scale system also provided “replicates” for the tests that were performed with both systems.
The results showed that (1) as FXB bioreactor treatment systems scale up, process efficiencies also go up (i.e., the required contact time to achieve sustained, robust perchlorate removal to below detection was one-third the contact time required during previous, smaller scale studies); (2) hydrogen peroxide reoxygenation, polishing filtration, and chlorination provide effective post-treatment; (3) system operation is straightforward, requiring no specialized training or extraordinary maintenance procedures; (4) the bacterial communities in these systems are largely gram-negative Proteobacteria; (5) site-specific performance of these systems can be predicted using a mathematical model; and (6) costs for FXB biological perchlorate treatment systems can be low.
Any full-scale, potable FXB biological perchlorate treatment process would be subject to all federal and state drinking water regulations. In addition to these established and emerging drinking water regulations, which primarily apply to distributed water quality, utilities will also have to consider how to handle the backwash wastewater. This waste stream should be less than or equal to 3% of the total water treated and have backwash wastewater of low strength. Therefore, it is expected that it can be discharged to the local sewer in many instances, though this would have to be confirmed on a site-specific basis. If no sewer discharge is allowable at a given site, a wastewater clarification and recycle process would need to be considered.
Lastly, a permit for full-scale installation and operation of a potable FXB biological perchlorate treatment system must be applied for and received from the California Department of Public Health (CDPH). Conditional CDPH approval for full-scale implementation of the FXB process was granted to Carollo Engineers in 2004, and discussions with CDPH in February 2008 indicated that, based on the performance data from various FXB biological perchlorate and nitrate treatment pilot studies, full-scale FXB biological treatment facility permitting should follow the standard schedule and protocol for any new water treatment facility in California. Information gained from this project was used to inform ESTCP project ER-201169.