Comparison of RBC and SBR systems for ammonia removal from landfill leachate

• Main Author: Besler, David A.
• Language: English
• Published: 2009
• Online Access: http://hdl.handle.net/2429/4293

Pilot scale RBC and SBR systems were compared in order to find out which system would be more cost effective for full scale treatment of Vancouver Landfill leachate. The leachate is an older leachate with NHX-N concentrations between 83 and 336 mg/L, BOD5 concentrations between 20 and 89 mg/L, and BODs:COD ratios between .06 and .19. Therefore, the primary objective of the study was to investigate NHX-N removal through nitrification. In general, the RBC recovered from nitrogen loading increases and process upsets more quickly and completely than the SBR did, and RBC process upsets were less frequent and less severe. This was attributed to the RBC's superior solids retention capabilities. The RBC was operated successfully at an average nitrogen loading of 4.15 g/m2 /d at an HRT of 0.35 days. However, loadings of 2 g/m2/d or less were required in order for effluent NOJ-N levels to remain below 4 mg/L. The RBC was unable to acclimatize to several of the applied loadings due primarily to provision of insufficient HRT. HRTs greater than 0.3 days were required for NHX-N removals between 2 and 8 g/m2/d, and P limitation at the higher loadings inhibited complete nitrification. The SBR achieved complete nitrification at an average loading of 107 g/m3/d at an HRT of 1.93 days. Complete NHX-N removals were achieved at an average loading of 331 g/m3 /d at an HRT 0.71 days, but complete nitrification was not. It appeared that the lack of complete nitrification in the SBR system was also caused by P limitation. Both systems were able to perform at low temperatures. The RBC achieved an NHX-N removal of 1.51 g/m2/d at 2.5°C, and the SBR was able to remove 23.3 g/m3 /d at 3°C. Nitrogen balance calculations were performed for both systems, but neither system exhibited significant N disappearances. The two systems performed similarily with respect to BOD5 removal, but the RBC was superior in terms of COD and colour removal. While neither system produced large amounts of excess solids, the RBC solids had better settling characteristics. Evidence of precipitation of inorganic solids was found for both systems, but neither system had scale problems. No differences between the two systems were found in terms of Cd or Co removal, but the RBC tended to remove about 40% of Fe and Mn while the SBR removed 30%. Both systems tended to add rather than remove Zn, but the RBC was more likely to achieve positive removals. Overall, the performance of the RBC system was superior to the performance of the SBR system. Toxicity studies were carried out in order to determine whether a substitute could be found for the traditional 96 hour rainbow trout LC 50 tests. Rainbow trout LC 50 was found to vary with leachate NHX - N according to the relationship T = 2259/N, where T = 96 hour LC 50 (%) and N = NHX-N concentration (mg/L), R2 = 0.90. Daphnia magna 48 hour LC 50 results correlated well with fish results based on the above relationship, but Microtox EC 50 results did not, due to the high tolerance of the Microtox test for NHX-N. The sucrose modified Microtox test is supposed to be more sensitive to NHX-N, but it did not produce meaningful results in this experiment, as sucrose solutions tended to be toxic. The 2259/N relationship was found to be the best way to predict fish toxicity.