Nitrogen removal in a foam media biofilter for on-site wastewater treatment systems : a thesis presented in partial fulfilment of the requirements for the degree of Master of Engineering in Environmental Engineering at Massey University

Abstract

Discharges of nitrogen can contaminate groundwater, and cause algal blooms or eutrophication in surface waters. On-site wastewater treatment systems (OWTS) have been identified as significant sources of nitrogen. Homeowners and manufacturers are under increasing pressure to install OWTS capable of effective nitrogen removal. Biological nitrogen removal in OWTS usually takes place in a fixed growth biofilter, following primary treatment in a septic tank arrangement. Three configurations of OWTS using foam media biofilters were assessed in the field. Foam media has advantages over sand, as high porosity and large air gaps allow the simultaneous flow of wastewater and air, thus reducing clogging and allowing higher loading rates. Septic tank effluent had lower concentrations of TSS, COD and TN in configurations with larger tank volume. Biofilters provided additional removal of TSS and COD to give effluent concentrations as low as 9 mg/L and 36 mg/L respectively. TN concentration in the effluent varied from 41-53 mg/L depending on configuration. The least nitrogen removal occurred in the configuration with the highest loading rate (in terms of L/m2/d). A bench-scale biofilter constructed using a single foam block (200 x 160 x 60 mm) achieved TN removal up to 10.7 mg/L (0.024 g-N/d at a dosing rate of 2.2 L/d). It was observed that nitrification and denitrification can both occur in a single foam block. Assimilation was also a significant nitrogen removal mechanism, accounting for up to 49 % of total removal. DO concentrations at microenvironments within the bench-scale biofilter were determined using a miniature membrane electrode. A syringe needle and custom-made plunger with the electrode fitted inside allowed DO concentration to be determined in sample volumes as small as 1 mL. The empirical equation derived to calculate DO concentration was accurate to within ± 2.9 %. The extent of nitrification was greatest after an overnight rest period. At microenvironments within the bench-scale biofilter, nitrification increased at longer hydraulic residence time. Nitrification increased at high feed concentrations of carbon, which was not expected, and did not decrease at DO concentrations as low as 0.88 mg/L. Denitrification was greatest when feed was high in carbon and low in DO, but was not affected by DO concentrations as high as 2.70 mg/L. The effects of loading rate, biofilter depth, recirculation ratio and flooding need to be investigated further to optimise the design of biofilters in the field.