Inoculum-free start-up of biofilm- and sludge-based deammonification systems in pilot scale

Undiluted reject water from the dewatering of anaerobic sludge with an average total nitrogen content of 718 ± 117 mg L−1 (n = 63) was used to start-up autotrophic nitrogen removal in three different pilot-scale (3 m3) deammonification configurations: (1) biofilm; (2) activated sludge sequence batch; and (3) two-staged (nitritation–anammox). Time- and concentration-based aeration control with alternating aerobic/anaerobic phases was applied for all reactor configurations. All reactors were initiated without anammox-specific inoculum, and biofilm was grown onto blank carriers. During the initial start-up period, biological nitrogen removal was found to be inhibited by an excessive free ammonia content (>10 mg-N L−1), resulting from the use of high-strength reject water as the process feed. After implementation of free ammonia control by pH adjustment to 6.5–7.5, propagation of the deammonification process was observed with increased nitrogen removal with slight accumulation of NO3−–N. The highest total nitrogen removal rates were achieved with the single-reactor biofilm- and sludge-based deammonification processes (1.04 and 0.30 kg-N m−3 day−1, respectively). The critical factors for successful start-up and stable operation of deammonification reactors turned out to be control of pH below 7.5, dissolved oxygen at 0.3–0.8 mg-O2 L−1 and influent solids values below 1000 nephelometric turbidity units. Microbial analysis demonstrated that highest anammox enrichment was achieved in the biofilm reactor (9.40 × 108 copies g−1 total suspended solids). These data demonstrate the potential of an in-situ grown sludge- or biofilm-based concept for the development and propagation of deammonification process.