ABSTRACT
A preliminary bench-scale study of parallel University of Cape Town (UCT) biological nutrient removal systems showed improvement in anoxic denitrification rates resulting from prefermentation of a septic (i.e., high volatile fatty acid [VFA] content), phosphorus-limited (i.e., total chemical oxygen demand/total phosphorus [TP] ratio < 40:1) wastewater. Net phosphorus removals due to enhanced biological phosphorus removal (EBPR) were only improved marginally by prefermentation in spite of significant increases in anaerobic phosphorus release, polyhydroxyalkanoate formation, and higher anoxic and aerobic uptakes. This probably was due to the high VFA/TP ratio in the raw influent relative to the VFA requirements for EBPR because enough VFAs were already present for phosphorus removal prior to prefermentation. An additional assessment of prefermentation using parallel UCT systems with step feed of 50% of the influent to the anoxic zone was completed. This second phase quantified the effect of prefermentation in a step-feed scenario, which prioritized prefermentation use to enhance denitrification rather than EBPR. While specific denitrification rates in the anoxic zone were significantly improved by prefermentation, high denitrification in the clarifiers and aerobic zones (simultaneous denitrification) made definitive conclusions concerning the potential improvements in total system nitrogen removal questionable. The prefermented system always showed superior values of the zone settling velocity and sludge volume index and the improvement became increasingly statistically significant when the prefermenter was performing well.
