To evaluate continuous and sporadic nitrification inhibition at the HRSD Nansemond Wastewater Treatment Plant, which has a history of nitrification upsets, continuous sequencing batch reactors (SBRs) were operated to simulate the full-scale plant. Four reactors were operated in this study. One reactor was fed with raw influent (RWI) from the Nansemond Wastewater Treatment Plant (NP). Another was fed with NP primary clarifier influent (PCI), which includes the raw influent, as well as plant recycle streams and truck delivered septage, grease, and chemical toilet waste. The remaining two SBRs were fed with RWI from the VIP Wastewater Treatment Plant, which achieves reliable nitrification year-round. One of these VIP SBRs would remain a control at all times, while the other would be used to evaluate suspected inhibitors to nitrification.

The first phase of this project was to determine whether NP was inhibited when compared to VIP, which would be ascertained through a comparison of nitrification performance. The next step was to determine whether the source of inhibition was an industry within the collection system or plant recycles and delivered wastes, which would be ascertained based on comparison of the NR RWI and NP PCI reactor performance. If nitrification performance was comparable between the two SBRs, then it would indicate that the source of inhibition is somewhere within the collection system, whereas if the NP PCI reactor was inhibited when compared to the NP RWI reactor, it would mean that the inhibition is a result of plant recycles or delivered wastes. The next phase would be to determine the specific source by either working back up the collection system or by testing the plant recycles and delivered wastes.

After approximately 27 weeks of SBR sampling and monitoring, there was no statistical difference between nitrification rates in reactors A and B, and no signs of nitrification inhibition in either reactor when compared to the VIP control

Simulation modeling of reactors A, B, and D (control) was performed with BioWin 3.1 (EnviroSim, Ltd.) as a means for comparison and to ensure reactors were performing as intended. Results suggest that there was some level of continuous inhibition for both NP RWI and PCI reactors, however no sporadic inhibition events were observed. It also appeared that the VIP RWI control reactor experienced some level of continuous nitrification inhibition, although BioWin modeling results indicated that both NP RWI and NP PCI were more inhibitory than VIP RWI. Conclusions drawn from modeling results conflict with those drawn from nitrification rate comparisons. Since solids retention time (SRT) was maintained at exactly 15 days for all reactors, it was assumed that a direct comparison of corrected maximum nitrification rates could be used to compare nitrification performance between SBRs, however the significantly higher influent COD, TKN, and TSS loading to the NP reactors resulted in higher nitrification rates when compared to the VIP RWI control reactors. This was confirmed with BioWin modeling, which also showed consistently higher nitrification rates for NP when compared to VIP RWI, however BioWin also showed that maximum specific growth rates for ammonia-oxidizing bacteria (?maxAOB) in NP RWI and PCI were consistently lower than the ?maxAOB for VIP RWI. This indicates that NP RWI and NP PCI are slightly inhibitory to nitrification, with ?maxAOB values between 0.65 and 0.75 days??, and the fact that both NP RWI and NP PCI are both inhibitory suggests that the source of inhibition is somewhere within the collection system.

In a simultaneous study using the reactors fed with raw influent from the VIP Wastewater Treatment Plant, reactor C was spiked with aqueous Film Forming Foam (AFFF) such as that used in methanol feed facility fire suppression systems, while reactor D was left as a control. AFFF was initially added at a concentration of 20 ppm with no effect on either nitrification or denitrification performance. When increased to 40 ppm, the AFFF reactor experienced a complete loss of denitrification, while nitrification rates were not affected when compared with the control reactor. Reactor C took 31 days to fully acclimate to the AFFF feed and fully regain denitrification, and then exhibited no other performance problem throughout this acclimation period. This result was completely unexpected, appears to be repeatable, and is one of very few cases of selective denitrification (and COD uptake) inhibition, as opposed to more commonly observed nitrification inhibition.