The simultaneous nitritation and phenolic compounds removal using aerobic granular reactors in continuous mode was studied in this thesis. The study is divided into two main subjects. The first one is devoted to the modeling of nitritation, while the other part is dedicated to the experimental study of simultaneous nitritation and phenolic compounds removal.
In the modeling study, a mathematical biofilm model was developed to describe nitritation in aerobic granular reactors operating in continuous mode. The model incorporated a [DO]/[TAN] ratio control strategy to maintain the proportion between the concentrations of dissolved oxygen (DO) and total ammonia nitrogen (TAN) in the reactor effluent to a desired value. The model was validated with a large set of experimental results previously reported in the literature, as well as, data gathered from laboratory and pilot plant granular reactors treating reject water. The model was used to study the effect of: a) DO and TAN setpoints, b) operating temperature, c) biofilm characteristics (granules size, density) and d) ammonium concentrations in the influent on the achievement of full nitritation. The results indicated that full nitritation was stably maintained and enhanced by applying the [DO]/[TAN] ratio control strategy in the operation of aerobic granular sludge reactors. Moreover, the model predicted that aerobic granules size larger than 1.5 mm and high ammonium concentrations in the influent enhanced the achievement of stable full nitritation. Furthermore, at low temperature, full nitritation with granular reactors was demonstrated to be possible. On the contrary, poor influence of the biofilm density on the achievement of full nitritation was found with the simulation study.
In the experimental study, an airlift reactor was employed (Figure 1). The airlift reactor was inoculated with granular sludge performing biological nutrient removal. A synthetic wastewater containing a high-strength ammonium concentration (950 ± 25 mg N L-1) was fed into the airlift reactor that was operated until partial nitritation was obtained. Once partial nitritation was achieved, the airlift reactor was bioaugmented with p-nitrophenol (PNP)-degrading activated sludge to enhance the growth of phenols-degraders in the nitrifying granules. Immediately, o-cresol (up to 100 mg L-1) or PNP (up to 15 mg L-1) were progressively added to the high-strength ammonium influent with the objective of studying the simultaneous partial nitritation and phenols removal.
In the study of simultaneous partial nitritation and o-cresol removal, stable partial nitritation process was maintained for more than 100 days of operation. Moreover, full biodegradation of o-cresol was achieved during the whole experimental period. Also, o-cresol shock load events were applied and the partial nitritation process was kept stable and unaffected during these events. The achieved nitrogen loading rate (NLRV) and o-cresol loading rate (oCLRV) were ca. 1.1 g N L-1d-1 and 0.11 g o-cresol L-1d-1, respectively. Analysis of fluorescent in-situ hybridization (FISH) indicated that Acinetobacter genus, betaproteobacterial ammonia-oxidizing bacteria (βAOB)and Nitrobacter sp. were identified into the granules (Figure 2). The operation of the reactor was continued to perform an experiment devoted to assessing its performance under three sequentially alternating pollutant (SAP) scenarios. In each one of the SAP scenarios, 15 mg L-1 of a secondary phenolic compound (i.e. p-nitrophenol (PNP), phenol or 2-chlorophenol (2CP)) was added to the regular influent composed of ammonium and o-cresol (the primary phenolic compound) for a short period of time (between 20 to 25 days). The results illustrated that partial nitritation and o-cresol biodegradation were maintained without exhibiting any sign of inhibition by the presence of PNP or phenol. However, when 2CP was present in the influent, 90 % of the partial nitritation and 25 % of the o-cresol degradation were inhibited within two days. In spite of this massive failure, the reactor could be rapidly re-activated, fully recovering the partial nitritation and o-cresol removal (the primary recalcitrant compound) capacity in 14 and 4 days, respectively. These findings demonstrate that treatment of complex industrial wastewaters with highly variable influent composition could be feasible in a continuous aerobic granular sludge reactor.
In the study of simultaneous nitritation and PNP removal, nitritation was maintained for producing an effluent suitable for heterotrophic denitrification. However, during the first 175 days, PNP biodegradation was unstable and several accumulation episodes occurred. Oxygen limiting condition was found to be the main explanation for these events. The increase in dissolved oxygen (DO) concentration in the reactor from 2 to 4.5 mg O2 L-1 permitted to achieve complete and stable PNP removal till the end of the experimental period. The achieved NLRV and PNP loading rate (PNP-LRV) were ca. 1.0 g N L-1d-1 and 16 mg PNP L-1d-1, respectively. The performance of the reactor was further assessed by performing two starvation studies: i) PNP starvation and ii) total starvation period (temporary stop). Results show that full recovery of PNP degradation was achieved within 2 days after the PNP starvation period ended, while full recovery of simultaneous nitritation and PNP removal was accomplished in just 14 days after the restart of the reactor.
In conclusion, the use of continuous aerobic granular reactors for the simultaneous nitritation and phenolic compounds removal is feasible. Aerobic granules were proven to be resistant and resilient to the shock loads, to the alternating presence of recalcitrant compounds and to the starvation periods that are situations frequently found in industrial wastewater treatment plants due to changes in the industrial production schedules. In the near future, we propose the simultaneous nitritation and phenolic compounds removal should be combined either with heterotrophic denitrification or Anammox for sustainable nitrogen removal.