Isolation and identification of the bacteria from a hydrogenotrophic reactor for the denitrification of drinking water revealed that several microorganisms are involved. Acinetobacter sp., Aeromonas sp., Pseudomonas sp. and Shewanella putrefaciens were repeatedly isolated from the hydrogenotrophic sludge and postulated to be of primary importance in the process. Nitrate reduction to nitrite appears to be a property of a diverse group of organisms. Nitrite reduction was found to be stimulated by the presence of organic growth factors. Thus, in a mixed culture, hydrogenotrophic denitrification reactor, NO inf2 (sup-) formed by NO inf3 (sup-) -reducers can be converted by true denitrifiers thriving on organic growth factors either present in the raw water, or excreted by the microbial community. Mixotrophic growth also contributes to NO inf2 (sup-) reduction. Finally, chemolithotrophic bacteria participate in the nitrite to nitrogen gas conversion.

In SBR plants for nutrient removal it is often necessary to add supplementary rbCOD during the anoxic phase to obtain complete nitrogen removal. In addition to the aeration, this supply of high-quality BOD is a non-negligible part in the operating costs. Because of the complexity of the bighly interconnected biological processes a heuristic approach for process optimization is hardly possible. Therefore the Nitrification Denitrification Biological Excess Phosphorus Removal (NDBEPR) model of Wentzel et al. and a numerical optimization a1goritbm were used to optimize SBR time scheduling, i.e. minimize both effluent concentrations and operating costs. It was found that a sequence of short aerobic/anoxic phases appears to be better than the usual sequence (one aerobic phase followed by one anoxic phase). This result was validated on a 500 I scale SBR. The optimized process saves up to 50% on extra BOD supply and up to 30% on aeration time. Moreover, it was shown that these cost savings were not at the expense of the phosphorus removal efficiency or the nitrification rate. From an additional numerical optimization it was seen that the ideal SBR time scheduling may depend on the loading. Therefore. a control strategy hased on OUR and ORP measurements is proposed.

No abstract

A pilot biological fluidized‐bed plant with a capacity of 40 m3/h has been in operation since January 1988 at De Blankaart drinking water production center for removing nitrate from surface water. Methanol is used as the reductant. With a nitrate removal efficiency of 9.0 kg NO3−/m3 reactor.day at 3.5°C, the system has shown superior performance compared with conventional fixed‐bed biofilm reactors. With an influent concentration of 75 mg NO3−/L, complete nitrate removal was achieved at an empty bed contact time of 15 min. Nitrite was not detected in the effluent, provided there was a slight excess of methanol (1–2 mg/L). Residual methanol was easily removed by the existing downstream drinking water treatment processes.

Fluidized‐bed denitrification of surface water with methanol as the carbon source is currently being studied in Belgium. Residual methanol and microbial excretion products increased the dissolved assimilable organic carbon in the reactor effluent. Hyphomicrobium sp. was isolated as a methylotrophic denitrifier of primary importance. The anoxic denitrifying environment apparently provides a favorable niche for a large diversity of associated bacterial strains. Hydraulic shear resulted in the constant washout of these indicator species from the reactor. Thus, the microbiological quality of the effluent is altered by the denitrifying process, and the treated water has to be further subjected to filtration and disinfection in order to guarantee the removal of residual organic carbon and prevent breakthrough of indicator organisms.