Applying coagulation, flocculation and discfiltration in tertiary treatment

Reducing eutrophication in our surface waters caused by nutrient overload is of importance in order to ensure an environment in ecological balance for future generations. Municipal wastewater treatment plants are the main point source of nutrients emissions. Effluent water can be treated by introducing a final advanced treatment step to existing wastewater treatment plants. The present thesis concerns addition of coagulation, polymer aided flocculation and discfiltration using a 10 μm media in tertiary treatment to reduce total phosphorous concentrations in municipal wastewater treatment plant effluents to < 0.1mg/l. Experiments were conducted in laboratory scale, using modified jar test and test tube filtration methodology, and in pilot scale. The pilot experiments were conducted as a two year study at Ruhleben WWTP in Berlin, Germany (OXERAM project), followed by two months at Lundåkraverket WWTP, Landskrona, Sweden, and three weeks at Sjölunda WWTP, Malmö, Sweden. Various chemicals and combination of chemicals were tested at various doses. Furthermore diverse process conditions regarding hydraulic retention time and mixing intensity were applied to study the impact and to optimize chemical utilization, effluent water quality and discfilter performance. The preferred chemicals were polyaluminumchloride (PACl) as coagulant and as flocculant aid, polyacrylamide based cationic synthetic powder polymer of high molecular weight and medium to high charge density. Both iron chloride (FeCl3) and anionic powder polymers of polyacrylamide was producing an effluent similar in quality but filtration rate was reduced. Residual iron (Fe3+) in the effluent and polymer dosing was higher when dosing anionic polymer. The secondary effluents were on average containing 0.3 mg/l total phosphorus of which about 0.1 mg/l was identified as orthophosphate, 10 mg/l of suspended solids and 30-50 mg COD/l. Treating these effluents, the required dose to obtain an effluent containing < 0.1 mg/l total phosphorus was on average about 1.9 mg Al3+/l and 0.6 mg/l synthetic cationic polymer as active material for an optimized coagulation/flocculation process. Using iron instead of aluminum as a coagulant, a dose of about 5 mgFe3+/l and an increased polymer dosing to about 1-1.5 mg/l was required. Experiments were conducted on wastewater treatment plants with various process configurations including plants with activated sludge, BIODENIPHO ® and plants with activated sludge and biofilm systems. The required dose to achieve < 0.1 mg/l total phosphorous in the effluent was on a molar basis identified to be similar for the three pilot plant experiments. A molar ratio of 5-7 moleAl3+/mole influent total phosphorus was shown to be applicable and 0.07-0.1 mg polymer / mg influent suspended solids was to be applied. These findings are of interest if load proportional dosing relying on online measurement of influent phosphorus and suspended solids is to be considered. Mixing intensity in the coagulation and flocculating stage was shown to influence the discfiltration process in achieving optimum chemical utilization, effluent water quality and discfilter performance. Mixing intensity, defined as the mean velocity gradient G, of up to 270 s-1 in the coagulation stage and 150 s-1 in the flocculation stage was found to be applicable and an improvement in chemical utilization and increased filtration rates without loss in effluent water quality was observed. This is higher than normally recommended for other processes such as sedimentation or dissolved air flotation and a tradeoff between decreased chemical usage, improved filtration rate and an increased energy demand for mixing has to be considered. Furthermore for improved performance, a useful mixing intensity was identified to be G=150-250 s-1 for coagulation and G=120-170 s-1 for flocculation and to be combined with a hydraulic retention time around 2-3 minutes in coagulation and 6-8 minutes in the flocculation. Moreover, it is also argued that a hydraulic retention time of 1.5 minutes in the coagulation and 4 minutes in flocculation could be applied when designing for peak flow conditions and this would not have an impact on overall performance to obtain an average TP of < 0.1 mg/l. It was observed that the chemicals and dosages was affecting the performance of the pilot in a similar way as observed in the laboratory experiments regarding effluent water quality and filtration rates, thus the laboratory jar and test tube filtration experiment were qualitatively estimating performance of a pilot plant and therefore it can reduce time for optimization and the duration of pilot experiments would be shortened. It was also shown that the measuring of total phosphorus with the cuvette method relying on the formation of molybdenum blue was comparable with the more advanced ICP-OES (inductively coupled plasma-optical emission spectrometry) method. Furthermore, the general improvement in effluent water quality from the coagulation, flocculation and discfiltration process was shown to give secondary benefits by improving the ozone utilization in reducing micro pollutants in secondary effluents. Applying coagulation, flocculation and discfiltration prior to ozonation the applicable ozone dose could be reduced by about 1.5 mg/l for the same micro pollutant reduction