Investigating the Enhancement of Biological Filtration with Capping Material Designs and Nutrient Amendments

Abstract

Biologically active filters, or biological filters, remove particles and harness the metabolic capacity of bacteria attached to filtration media, in the form of a biofilm, to metabolize biodegradable organic matter (BOM). Pilot-scale biological filtration experiments were carried out at the Mannheim Water Treatment Plant in Kitchener, Ontario, Canada to evaluate the impact of capping material selection and nutrient amendments for granular activated carbon (GAC) filters, on both traditional and biological filtration performance parameters. Traditional filtration parameters included filter effluent turbidity, head loss development, and filter run time. Biological filtration performance was evaluated by total organic carbon (TOC), dissolved organic carbon (DOC), ammonia-nitrogen (NH3-N), and soluble reactive phosphorus (SRP) removal. The top 20 cm layer of GAC (d10 = 1.3 mm) was replaced by a capping material with a larger effective size in three of the five pilot-scale filter columns—such use of capping layers in rapid biological filtration for drinking water treatment has not been reported previously. The capping materials that were investigated were an expanded clay (EC) aggregate (d10 = 1.7 mm) and a plastic “pinwheel” style medium (diameter = 2.5 cm). A stoichiometric carbon, nitrogen, and phosphorus (C:N:P) ratio of 100:10:1 is most commonly referenced in the drinking water industry as being ideal for microbial growth in distribution systems and biological filters. The nutrient amendment experiments studied the impact of amending the influent stoichiometric C:N:P ratio to 100:10:1 and 100:20:2, in a systematic and controlled manner. The monitoring and experimental program was conducted over 14 months to account for seasonal water quality and temperature effects. The results of this study have several implications for optimizing the design and operation of biological filters for drinking water treatment. The capping materials delayed terminal head loss by 10-40 hours, compared to the control GAC filter, and significantly reduced the rate of head loss accumulation at all temperature ranges without negatively impacting filter effluent turbidity or BOM removal. There were no significant differences in filter run time at cold water conditions between each of the filter configurations; however, both capping layers extended filter run time at warm water conditions. Replacing a relatively small layer of media with one that has a larger effective size can lead to more robust filter operation. At cold water conditions, amending the influent stoichiometric C:N:P ratio to 100:10:1 or 100:20:2 of the GAC or EC capped filters did not yield significant differences in either traditional or biological filtration performance. The observed reduction of SRP and no reduction in NH3-N concentrations suggest that the system was phosphorus limited but not nitrogen limited; however, the performance of the filters was not nutrient limited. The maximum stoichiometric C:N:P ratio of consumed nutrients by the biological filters was 100:0:10; thus, it was concluded that a C:N:P ratio of 100:10:1 was not optimal for performance enhancement at cold water conditions. At warm water conditions, amending the influent stoichiometric C:N:P ratio of the GAC filter to either 100:10:1 or 100:20:2 did not yield any improvements in traditional or biological filtration performance. Reductions in the NH3-N and SRP concentrations at the effluent of the nutrient-amended GAC filter suggests that it was both nitrogen and phosphorus limited, but not with respect to operational performance or BOM removal. Amending the influent stoichiometric C:N:P ratio to 100:10:1 of the EC capped filter led to a significant increase in its filter run time, while increasing the influent ratio to 100:20:2 improved both filter run time and rate of head loss accumulation; however, no improvements in BOM removal were observed. The long length of time required to observe improvements in filter performance at warm water conditions indicates that nutrient enhancement strategies may not be suitable for biological filters that operate in climates that experience short, or no periods of warm water conditions. Similar to the nutrient-amended GAC filter, reductions in the NH3-N and SRP concentrations at the effluent of the nutrient-amended EC capped filter suggest that it was also nitrogen and phosphorus limited. The observed improvements in performance of the nutrient-amended EC capped filters, but not the GAC filter, suggests that nutrient enhancement strategies can be beneficial but at certain conditions only. The stoichiometric C:N:P ratio of consumed nutrients by the biological filters ranged between 100:67.3:6.0 to 100:153.3:7.4; thus, it was concluded that a C:N:P ratio of 100:10:1 was not optimal for performance enhancement at warm water conditions. Residual amounts of SRP measured at the effluent of the nutrient-amended filters at all temperature ranges and nutrient dosing rates, suggests that there is a maximum amount of phosphorus can be metabolized by the biological filters. The plastic capped filter outperformed or matched the performance of the nutrient-amended filters in terms of the rate of head loss accumulation and filter run time, without any loss in performance in terms of turbidity trends or DOC removal at cold or warm water conditions. This suggests that using capping materials can be a cost effective way to improve biological filtration hydraulic performance, and is operationally less complicated than a nutrient addition system. However, adding capping layers to existing filters may require modifications to their operation.