A full scale nutrient removal study was conducted at the 30-MGD John E. Egan Water Reclamation Plant with a single-stage nitrification activated sludge process and sand filtration. One quarter of the plant’s secondary treatment capacity was used in the study. Simultaneous nitrogen removal through a step feed BNR process and phosphorus (P) removal by chemical precipitation with ferric chloride (FeCl3) were tested to investigate the potential impact of adding FeCl3 to mixed liquor for P removal on the step feed BNR process and on the operation of other unit processes in the plant. During the P removal tests, which were conducted once in early spring and again in early fall of 2005, solutions containing approximately 33% FeCl3 by weight were injected at two different target rates into mixed liquor at the effluent end of the aeration tank. The average daily FeCl3 concentrations applied with respect to 7.5 MGD inflow to the aeration tank were 32.5 to 44.6 mg/L. The average soluble phosphorus (Sol-P) concentrations in 24-hour composite samples of the secondary effluent during the P removal tests ranged from 0.05 to 0.82 mg/L with an arithmetic mean of 0.27 mg/L. The total nitrogen (TN) removal by the step feed BNR process tested averaged 52% and 54% before and during the chemical P removal test in the spring and 52% and 58% in the fall. This indicates that adding ferric chloride to mixed liquor up to 44.6 mg/L for removing particulate FePO4 in the secondary clarifiers has no adverse impact on the step feed BNR process for biological nitrogen removal. The higher percentage TN removal during the P removal test in the fall was probably attributed to the higher MLVSS in the aeration tank resulting from higher solids content in the returned sludge after ferric chloride addition. The accumulation of ferric particles in WAS during the P removal tests appears to have caused the clogging of cloth belt used in the gravity belt thickeners (GBT) for sludge thickening. The sludge thickening performance of the GBT deteriorated during the chemical P removal tests. The performance was improved after the addition of ferric chloride to mixed liquor stopped.
In anticipation of future nutrient removal requirements, technologies that can easily modify existing secondary sewage treatment plants with activated sludge processes to achieve reduction in total nitrogen (TN) and total phosphorus (TP) should be investigated (Pagilla et al., 2006). A full scale nutrient removal study was conducted in 2005 at the John E. Egan Water Reclamation Plant (Egan WRP) of the Metropolitan Water Reclamation District of Greater Chicago (MWRDGC) simultaneously using a step feed biological nitrogen removal (BNR) process for N removal and chemical precipitation with ferric chloride for P removal.
The Egan WRP was built in early 1970s’. It provides tertiary treatment of domestic sewage at a design average flow of 30 MGD. The current treatment processes in the plant include screening, grit removal, primary settling, a single-stage nitrification activated sludge process, and sand filtration for wastewater treatment, gravity belt thickening (GBT) for waste activated sludge (WAS) thichening, and anaerobic digestion and post-digestion centrifugation for biosolids processing. The plant has two secondary treatment batteries, which are operated in parallel, and each battery consists of two identical aeration tanks and four secondary clarifiers. Although step feed to various locations of aeration tanks was designed and constructed before the plant commissioned, this feature has not been employed in the routine operation for BOD5 and ammonia removal.
Chemical P removal is most appropriate when the incremental P removal is relatively small and the ratio of CBOD5 to P in the influent is low (Pagilla et al., 2006). At the Egan WRP, the average concentrations of TP, soluble phosphorus (Sol-P) and CBOD5 in the raw sewage entering the plant were 6.55 mg/L, 3.47 mg/L and 129 mg/L, respectively, in 2004. The average concentrations of TP and Sol-P in the final effluent were 2.93 mg/L and 2.86 mg/L, respectively, with no P removal process in the plant. In order to meet the potential effluent limit of less than 1 mg/L of TP, about 2.0 mg/L of Sol-P should be removed in addition to normal P uptake by the biological community in the activated sludge system. Addition of ferric chloride is commonly used for chemical P removal with effluent limit of 0.5 to 1.0 mg TP/L. Ferric chloride is readily available at the Egan WRP, as the plant uses it for sludge conditioning and struvite control in its
post-digestion dewatering system.
This paper will focus on the chemical P removal tests during the full-scale nutrient removal study. The goals of the teste are (1) to demonstrate the applicability of chemical P removal with ferric chloride addition to the end of the aeration tank to meet potential future total phosphorus (TP) effluent requirements, i.e. TP < 1.0 mg/L, for the plant, (2) to investigate the effects of chemical P removal on the simultaneous application of a step feed BNR process, and (3) to examine any potential impact on the operation of other unit processes in the plant.