Enhanced biological phosphorus removal in a full-scale activated sludge system and phosphorus release in a new high-rate anaerobic reactor

Low-cost modifications to promote enhanced biological phosphorus removal (EBPR) in an existing nitrification-only activated sludge system were evaluated at full-scale with a focus on dry and wet weather dynamics. EBPR was successfully established with average phosphorus content of the waste sludge 3.2±0.2% (95% confidence) compared to 1.6±0.1% in the control basin. Microbiological investigations showed a significant increase in relative abundance of bacteria phyla Chloroflexi, Nitrospirae, and Verrucomicrobia in the modified basins, but their abundance had no correlation to treatment performance. Quantitative polymerase chain reaction (qPCR) indicated significant increase in relative quantity of Accumulibacter/16S, but not for Actinetobacter-like phosphorus accumulating organisms (PAOs), which includes the PAO Tetrasphaera. Significant changes in some Accumulibacter clades were observed including consistent higher than expected dominance of clade I and increased relative quantities of clades IIB and IIC during extended wet weather. A BioWin wastewater process model was developed to predict EBPR performance during wet and dry weather conditions. The default model significantly over-predicted EBPR performance during wet weather. A sensitivity analysis determined several stoichiometric model parameters related to aerobic phosphorus sequestration and anaerobic release and the yield coefficient of ordinary heterotrophic organisms (OHOs) were significant and used to develop a calibrated model. In the EBPR system, phosphorus release and uptake rates were approximately 4 and 3 times lower respectively during wet weather compared to dry weather conditions. Two high-rate anaerobic reactors were developed and compared with a conventional complete mix reactor at bench-scale to evaluate as the phosphorus release component of a struvite recovery process. The high-rate reactors featured internal sludge thickening that permitted decoupling the hydraulic and solids retention times. The hydraulic and solids retention time (HRT and SRT) of the complete mix reactor ranged from 6-48 hours. The HRT of the plug flow reactors were reduced to between 6-18 hours while the SRT was maintained in the range of 18-54 hours, resulting in a significant reduction in reactor size required and associated costs. Phosphorus release for all reactors was correlated with SRT ranging from 0.063±0.75 to 0.37±1.8 mg effluent orthophosphate (OP)/mg influent total phosphorus (TP) at SRTs of 6 and 54 hours respectively. An economic analysis of the three reactors determined the three-zone plug flow reactor had the lowest present worth cost. A pilot-scale version of the three-zone reactor was evaluated with average HRT and SRT of 21 and 39 hours respectively. Phosphorus release correlated to SRT (R2=0.62), with release of about 0.18 mg effluent OP/mg influent TP at SRT of 20 hours and about 0.35 mg effluent OP/mg influent TP at SRT of 60 hours. Total suspended solids averaged 9,186 mg/L influent, 28,422 mg/L reactor waste and 335 mg/L effluent. A BioWin process simulation model was developed and sensitivity of kinetic parameters were evaluated to create a calibrated model. Sensitive parameters for phosphorus release were anaerobic decay rates for OHOs and PAOs; hydrolysis rate; half saturation rate for hydrolysis of slowly degradable organics; and half saturation rate for residual soluble substrate for OHOs. DNA sequencing showed a significant increase in relative abundance of phyla Chloroflexi and Firmicutes between the influent and waste sludge from the pilot reactor, but no correlation to performance. qPCR indicated no significant differences in quantity of Accumulibacter or Actinetobacter-like PAOs