Trials and Tribulations of Operating a Full-Scale Post-Filter Chlorine Contact Basin In a Large Conventional Treatment Plant

摘要

Delaying the point of initial chlorine application is a proven method of reducing disinfection by-product (DBP) formation. One way to do this while still achieving Ct requirements is through the use of a post-filter chlorine contact basin. Building upon pilot scale and short-term full-scale studies, the Philadelphia Water Department conducted a lengthy full-scale trial of post-filter chlorine contact basin operation at its Belmont Water Treatment Plant. Though this operation achieved significant success in the area of DBP reduction, it also presented operational challenges that had not surfaced in the previous studies. The Belmont Water Treatment Plant is a conventional treatment facility, producing an average of 50 MGD. Typically, sodium hypochlorite is dosed for disinfection at the rapid mix and post-sedimentation. Ammonia is dosed after filtration to form monochloramine in the clearwell. Coagulation pH is maintained at 6.5, and pH is raised to 7.2 after filtration but before clearwell storage. From May through October 2010, the clearwell was converted to a post-filter chlorine contact basin. During this time, rapid mix chlorination ceased, making the post-sedimentation dosing location the initial point of chlorination. Ammonia and hydroxide addition were moved from the clearwell influent to effluent. This now made the clearwell, maintained at pH 6.5, a free chlorine contact basin. At first, costs associated with the clearwell conversion were minimal and included adding some baffling and additional chemical application points and on-line instrumentation. The benefits of the clearwell conversion were tremendous, as DBP levels decreased by more than 50% within hours. However, within weeks of post-filter chlorine contact basin operation, treatment issues began unfolding. An attached algae mat developed on the walls and launders of the sedimentation basins. This was accompanied by significantly increased filter head loss accumulation rates and high post-backwash filter effluent turbidities. Numerous strategies were employed throughout the summer months to battle the symptoms, including increased algae sampling, additional application of potassium permanganate at the raw water basin, partially covering the sedimentation basins, periodically shocking the basins with chlorine and application of cationic polymer prior to filtration. The periodic shock alleviated the symptoms, but they returned soon after the shock was terminated. None of the other strategies, alone or in combination with others, was able to solve the problems. In addition, the plant was plagued by a summer of drought-like conditions and record-high temperatures. By late August, a combination of cationic polymer at the rapid mix along with partial basin coverage was found to be effective. The polymer continued to be effective into September after the covers were removed. An additional trial was planned for the spring and summer of 2011, to fine-tune the operational strategies that seemed to work best. The trial proved more costly than originally anticipated, due to additional labor hours, chemical costs and the purchase of materials to cover four large basins. Still, the ability to leverage existing infrastructure to bring about substantial DBP reduction far outweighed the costs.