THE EFFECT OF HEAVY METALS ON THE ACTIVATED SLUDGE PROCESS

Abstract

This study examines the effects of heavy metal on the performance of the activated sludge process. The distribution of the major metals, Cd, Cr, Cu, Fe, Mn, Ni, Pb and Zn within the Jurong Sewage Treatment Works, treating combined industrial and domestic waste, was first examined. Fe, Zn and Cu .were found to be the most abundant metals present in crude sewage as well as the returned liquor. Cr, Mn, Ni and Pb were present in much lesser quantities. Cd was not detectable in most of the liquid streams but was detected in the solids streams. Except for Ni and Mn, most of the metals were effectively removed from the liquid streams through adsorption and accumulated in the sludges. The sludge streams had significantly higher metal contents and were 1,000 to 10,000 times more than those in the corresponding liquid streams. The sludge in the primary stage had removed 50% of the total metals content from the liquid stream while 51 % of the remaining metals was removed by the secondary activated-sludge, giving a total of 76% removal of metal from the crude sewage. Almost all the metals removed by the sludge were retained in the sludge cake following dewatering. The data on metal distribution in primary and secondary sludges appear to suggest that metal affinity of both sludges are about the same and surface adsorption played the major role in the removal of these metals from the liquid streams. At the levels of metals present in the raw sewage, the performance of both the aerobic and anaerobic treatment processes at the Works was found to be little affected. Laboratory-scale sequencing batch reactors (SBRs) were subsequently used to examine the effects of Cu on aerobic treatment of synthetic waste. The SBRs were operated with a loading factor of 0.3 mg-BOD/mg-VSS-d and 3 operation cycles per day, with a react time of 6.5 hours per cycle. The percent COD removal, VSS, effluent SS, sludge settleability and nitrite and nitrate levels were evaluated with respect to reactor performance and stability. Upon reactor failure, copper dosage was terminated and reactor recovery was then studied. Tracking experiments were carried out at 5 days interval during both operating and recovery phases to examine the effects of Cu on the rates of COD depletion, biomass production and nitrification. In the absence of Cu in feed, the control reactor operating at the desired loading factor, had maintained stable daily COD removal efficiency of 90% for 214 days. The performance of other SBRs treating-Cu-laden waste was affected in varying degree depending on the level of Cu and duration of operation. The reactor fed with 1 mg-Cu/L had the least effect and daily %COD removal had only marginally declined to 87% during the 21 days of operating phase. For the reactors dosed with 3 and 5 mg-Cu/L, daily %COD removal was lower, but still fairly stable at 80% and 82 % respectively throughout the operating phase. The 5 mg-Cu/L reactor, however, had a faster rate of decline and failed after 10 days of operation as compared to 21 days for the 3 mg-Cu/L reactor. The concentrations of biomass in these reactors were observed to decline gradually with increasing days of operation. The performance of reactors dosed with 10 and 30 mg-Cu/L had deteriorated rapidly and %COD removal fell sharply to 48% and 27% respectively. The levels of biomass and the viability also decline rapidly due to the fast build-up of Cu concentration in the biomass. Biomass deflocculation was also observed in these reactors resulting in higher effluent SS concentration and poor %COD removal efficiency. Microbial activity in terms of rates of COD depletion, growth of biomass and nitrification was reduced in all reactors dosed with copper. This was because of the reduction in the biomass substrate adsorption capacity when specific adsorption of Cu increased. The uptake of Cu by biomass proceeded by a rapid surface adsorption process followed by a much slower metabolic assimilation mechanism. Both Cu specific adsorption and accumulation within the cell increased with increasing Cu dosage and duration of operation. This, in turn, had reduced the COD adsorption capacity and degenerated the microbial activity which eventually led to the failure of the reactors. When Cu feeding was terminated, desorption of Cu from the biomass occurred and this had improved COD adsorption capacity and regenerated microbial activity. Recovery of reactor performance was good, particularly for reactors previously dosed with 5 mg-Cu/L and lower. Improvement of %COD removal was also observed for reactors charged with 10 and 30 mg-Cu/L; however, the biomass concentration remained low and required a much longer period for recovery.