Aerobic granular sludge system optimization strategies for Landfill leachate treatmentand co-treatment with domestic Sewage

Abstract

Although the Aerobic Granular Sludge (AGS) technology has been extensively studied and disseminated for domestic sewage and other more complex effluents, its application in the treatment of leachate from sanitary landfills is still very incipient and presents some problems such as high temporal demand for granulation, biomass loss, granules with small dimensions, granule disintegration and nitrite accumulation. Therefore, this work sought to optimize AGS systems for leachate treatment and co-treatment with sewage to minimize or eliminate these commonly reported problems. For this, the impact of real leachate on the granule formation process and on the composition of the microbial structure in conventional AGS systems was initially evaluated. Subsequently, several operational strategies combining oxic (O), anoxic (An), anaerobic (A) periods and step-feeding were evaluated from synthetic effluents with a C:N:P ratio similar to that of a real leachate. After defining the best strategies, the optimized reactor was used for the treatment of real leachate. In the initial investigation, the granulation process was evaluated from two dilutions of real leachate from the inoculum: R1 (25%) and R2 (50%) in 8-hour cycles. In both reactors, the time for granulation was high and there was low retention of solids. Although the granules produced were small (200–300 μm), in R2 the TN removals were greater, especially because it had organic matter available for a longer period of time during the cycle, since in this reactor the COD removal was lower. In addition to the insufficient removal of C, N and P in the two reactors, nitrite accumulation was observed throughout the operation, requiring optimization to reduce the main problems encountered, which corroborate the problems already reported in the literature. In the second investigation, the way in which the AGS systems were fed in 12-hour cycles was investigated. The strategies were: fast (R1), slow (R2) and anoxic/anaerobic step-feeding (R3). R3 showed the highest solids retention (3.4 g/L MLSS) and the best settling (SVI30 = 46.7 mL/g), while R2 had the worst biomass concentration (1.9 g/L MLSS) and the worst settleability (SVI30 = 132.3 mL/g). In addition, compared to fast (R1) and slow (R3) feeding, this staggered mode of operation achieved the best removals of total phosphorus (TP, 53%) and total nitrogen (TN, 92%) without any accumulation of nitrite or nitrate. COD removals were very similar in R2 and R3, but TN and TP removals were significantly higher in R3. In the third investigation, different cycle configurations were investigated: A/O (R1 and R2), O/An with conventional power and well- defined anoxic phase (R3), O/An with step-feeding and well-defined anoxic phase (R4), O/An (R5 and R6). The step-feeding continued to show the lowest biomass losses and the best reactor performance in terms of simultaneous removal of carbon, phosphorus and nitrogen, with a nitrification rate of 99% without nitrite accumulation. Interleaving feast/famine and anoxic periods minimized nitrite accumulation. O/An type reactors with well-defined anoxic phase had the largest granules, with a diameter of around 1mm. In these reactors, the presence of bacillus- dominated coccus is clearly observed, as well as a denser structure around the granule, keeping the microbial communities more aggregated. In contrast, the granules of the O/An type reactors without a well-defined anoxic phase (R5 and R6) were characterized by having empty spaces and cavities, which suggest internal fragmentation. It was also observed that Planctomycetota was the most abundant phylum in reactors with well-defined anoxic phase and step-feeding. Proteobacteria was the most abundant phylum in the anaerobic phase reactors (R1 and R2) and in the fast-feeding reactor (R6). In the final investigation, the leachate co-treatment with sewage occurred gradually in reactors with step-feeding (R1) and with conventional feeding (R2), both with alternating oxic/anoxic phases at the end of the cycle. It was possible to accomplish the domestic sewage co-treatment with leachate in AGS systems. However, feeding strategy, reactor configuration, and methanol supplementation played an important role in process stability and simultaneous carbon, nitrogen, and phosphorus removals. Step-feeding produced an aerobic granular biomass more compact and resistant, resulting in a better operational stability. Moreover, this strategy favored denitrification, especially during methanol supplementation, minimizing one of the main problems reported regarding leachate co- treatment in AGS systems. As a result, higher total nitrogen (TN) removals were obtained. At the end of the last period, in R1, Chemical Oxygen Dissolved (COD), TN, and Dissolved Organic Carbon (DOC) removals were 93%, and phosphorus removal was 54%, reaching values higher or similar to other AGS investigations with sewage, leachate, or co-treatment. Therefore, the presented results bring a good perspective for domestic sewage co-treatment with leachate and other types of industrial wastewater.