Water Quality Impacts of Pure Chlorine Dioxide Pretreatment at the Roanoke County (Virginia) Water Treatment Plant
by Christine S. Ellenberger Dr. Robert C. Hoehn, Chairman
Chlorine dioxide (ClO2) was included in the Spring Hollow Water Treatment Plant (Roanoke County, Virginia) to oxidize manganese and iron, prevent tastes and odors, and avoid the formation of excessive halogenated disinfection by-products. A state-of-the-art, gas:solid ClO2 generation system manufactured by CDG Technology, Inc. was installed at the plant and is the first full-scale use of this technology in the world. The ClO2 generator produces a feed stream free of chlorine, chlorite ion (ClO2-), and chlorate ion (ClO3-), resulting in lower by-product concentrations in the treatment system
The objectives of this project were to study ClO2 persistence and by-product concentrations throughout the treatment plant and distribution system and to evaluate granular activated carbon (GAC) columns for removing ClO2- from the finished water.
The ClO2 dosages applied during this study were relatively low (<0.75 mg/L), and, as a result, ClO2- concentrations never approached the maximum contaminant level (MCL) (1.0 mg/L). Likewise, the plant effluent ClO2 concentration never approached the maximum residual disinfectant level (MRDL) (0.80 mg/L), but concentrations as high as 0.15 mg/L reformed in the distribution system by ClO2- reaction with chlorine.
Chlorate ion was monitored despite the fact that no ClO3- MCL has been proposed, and concentrations were quite low (never greater than 0.10 mg/L) throughout the treatment plant and in the distribution system. The reasons for the low concentrations are that ClO3- is not produced by the gas-solid generator used at the facility and ClO2- concentrations in the clearwell prior to chlorination were uniformly low.
The average ClO2- reduction upon passage of treated water through the GAC contactor was approximately 64 percent, but the GAC effectiveness was declining over the six-month study period. Apparently, GAC effectiveness, as shown by others, is short-lived, and if higher ClO2 dosages are ever applied at the Roanoke County facility, the ClO2- concentrations will have to be reduced by either ferrous coagulants or reduced-sulfur compounds.
Regenerated ClO2 concentrations in the distribution system were below 0.2 mg/L, but concentrations as low as 0.03 mg/L were found at homes of customers who complained of odors. During this study, twelve complaints were received from eight customers, and each complainant had recently installed new carpeting, which has been shown to contribute volatile organics that react with ClO2 to produce odors similar to kerosene and cat urine. While meeting the ClO2- MCL likely will be no problem if the ClO2 dose at the plant remains below 1.0 mg/L, the problem of offensive odors in the distribution system will likely continue as long as any ClO2- is in the finished water when chlorine is present.
Advisor:Dr. Andrea M. Dietrich; Dr. Daniel L. Gallagher; Dr. Robert C. Hoehn
School Location:USA - Virginia
Source Type:Master's Thesis
Date of Publication:01/08/2000