Environ. Sci. Technol. 2009, 43, 5035–5041
Nitrogen Removal from Digested Black Water by One-Stage Partial Nitritation and Anammox SIEGFRIED E. VLAEMINCK,† AKIHIKO TERADA,‡ BARTH F. SMETS,‡ DAVY VAN DER LINDEN,† NICO BOON,† W I L L Y V E R S T R A E T E , * ,† A N D M A R T A C A R B A L L A †,§ Laboratory of Microbial Ecology and Technology (LabMET), Ghent University, Coupure Links 653, 9000 Gent, Belgium, Department of Environmental Engineering, Technical University of Denmark (DTU), Miljoevej, Building 113, 2800 Kgs. Lyngby, Denmark, and Department of Chemical Engineering, School of Engineering, University of Santiago de Compostela, Rúa Lope Gómez de Marzoa s/n, 15782 Santiago de Compostela, Spain
Received November 20, 2008. Revised manuscript received April 14, 2009. Accepted April 16, 2009.
This study assessed the technical feasibility to treat digested black water from vacuum toilets (>1000 mg NH4+-N L-1) in a labscale oxygen-limited autotrophic nitrification/denitrification (OLAND) rotating biological contactor. After an adaptation period of 2.5 months, a stable nitrogen removal rate of ca. 700 mg N L-1 d-1 was reached over the subsequent 5 months. Suppression of the nitrite oxidizing bacteria at free ammonia levels above 3 mg N L-1 resulted in a nitrogen removal efficiency of 76%. The favorable ratios of both organic and inorganic carbon to nitrogen guaranteed endured anammox activity and sufficient buffering capacity, respectively. Quantitative FISH showed that aerobic and anoxic ammonium-oxidizing bacteria (AerAOB and AnAOB) made up 43 and 8% of the biofilm, respectively. Since a part of the AerAOB was probably present in anoxic biofilm zones, their specific ammonium conversion was very low, in contrast to the high specific AnAOB activity. DGGE analysis showed that the dominant AerAOB and AnAOB species were resistant to the transition from synthetic medium to digested black water. This study demonstrates high-rate nitrogen removal from digested black water by one-stage partial nitritation and anammox, which will allow a significant decrease in operational costs compared to conventional nitrification/ denitrification.
Introduction Over the past decade, the treatment of source-separated domestic wastewater has gained more attention since it can avoid the problems associated with the traditional end-ofpipe treatment of diluted streams. Source separation yields, first, a gray water stream, which contains the washing water from laundry, kitchen, shower, and bath; and second, a black water stream, containing the toilet waste. Black water typically * Corresponding author phone: +32-9-2645976; fax: +32-92646248; e-mail:
[email protected]. † Ghent University. ‡ Technical University of Denmark (DTU). § University of Santiago de Compostela. 10.1021/es803284y CCC: $40.75
Published on Web 05/26/2009
2009 American Chemical Society
contains ca. 70, 90, and 80% of the total domestic chemical oxygen demand (COD), N and P, respectively, and the use of vacuum instead of conventional toilets (1 vs 7 L flush-1) significantly decreases the generated water volumes, and consequently, concentrates the pollutants (1). Anaerobic digestion is the preferred technique to remove COD from high-strength black water from vacuum toilets since this allows for energy recovery, as demonstrated on pilot scale (1). The resultant effluent from digestion at 25 °C was highly nitrogenous with ammonium concentrations exceeding 1000 mg N L-1 and soluble COD/N ratios of around 1.0. In this work, the one-stage oxygen-limited autotrophic nitrification/denitrification (OLAND) process was evaluated to remove nitrogen from digested black water from vacuum toilets. OLAND comprises aerobic ammonium-oxidizing bacteria (AerAOB) oxidizing ammonium to nitrite, and anoxic ammonium-oxidizing or anammox bacteria (AnAOB) combining this nitrite with residual ammonium into dinitrogen gas and some nitrate (2, 3). From an economical point of view, OLAND and related partial nitritation/anammox processes are preferred to conventional nitrification/denitrification for wastewaters with a low biodegradable organic content. An evaluation for the treatment of sludge reject water showed for instance that OLAND can save 30-40% of the overall nitrogen treatment costs (4), due to a lower aeration requirement, sludge production, and organic carbon addition. There are, however, some inherent challenges to obtain good OLAND process performance. First, AnAOB double only every 11 days (5), and this slow growth can result in very long reactor start-up periods (6, 7) and requires a high biomass retention, which must be ensured by growth in biofilms (2) or flocs and granules (8). Second, high nitrogen removal efficiency of the OLAND process relies on limited nitrite accumulation, obtained when the AerAOB activity does not exceed the AnAOB activity. Third, high efficiency requires a limited nitrate production. Due to the anabolic nitrate production of AnAOB (5), OLAND typically converts 11% of the oxidized ammonium into nitrate (8), a value which is not exceeded in case the nitrate consumption by heterotrophic denitrifiers is larger than the nitrate production by nitrite oxidizing bacteria (NOB), i.e., nitratation. So far, autotrophic nitrogen removal from sludge reject water, landfill leachate and specific industrial effluents has been demonstrated at pilot and full scale, as tabulated in refs 6 and 9. With regard to wastewaters with high micropollutant concentrations, promising lab-scale results have been reported for OLAND treatment of urine (10), but it was not clear yet whether the complex mixture of urine and faeces micropollutants (11) or other black water specific compounds would enable long-term AnAOB activity. In view of the economical benefits of the OLAND process, the goal of this study was to examine its technical feasibility to remove nitrogen from digested concentrated black water. Operational simplicity is preferable for decentral wastewater treatment, and therefore a rotating biological contactor (RBC) was applied, a configuration in which high performance stability can be obtained with minimal reactor control (2).
Materials and Methods Black Water. Digested concentrated black water was obtained from a pilot project with 32 houses in Sneek, The Netherlands (1). An overall characterization of two of the received batches of digested black water is presented in Table 1. OLAND Reactor. The lab-scale RBC was based on an airwasher LW14 (Venta, Weingarten, Germany) with a rotor consisting of 40 discs interspaced at 3 mm, resulting in a disk VOL. 43, NO. 13, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5035
TABLE 1. Characterization of Two Batches of Digested Concentrated Black Water (Averages ± Standard Deviations; n = 6)a Parameter
Batch 1
Batch 2
pH (-) NH4+ (mg N L-1) Kjeldahl N (mg N L-1) NO2- (mg N L-1) NO3- (mg N L-1) CODsol (mg O2 L-1) DOC (mg C L-1) DIC (mg C L-1) total VFA (mg acetic acid L-1) conductivity (mS cm-1) Cl- (mg Cl L-1) SO42- (mg S L-1) PO43- (mg P L-1)
7.9 ( 0.1 1065 ( 15 1110 ( 58 0.0 ( 0.0 0.2 ( 0.4 597 ( 49 269 ( 48 803 ( 56 5.9 ( 1.2 7.9 ( 0.6 500 ( 6 45 ( 7 43 ( 1
8.0 ( 0.1 1023 ( 9 NA 10 ( 7 0.4 ( 0.3 581 ( 16 NA NA NA NA 714 ( 31 56 ( 2 76 ( 4
a Because the individual volatile fatty acids (VFA) were present in low concentrations, the acetic, propionic, butyric, isovaleric and caproic acid were summed and expressed as mg acetic acid L-1. Isobutyric, valeric, and isocaproic acid were not detected. CODsol: soluble chemical oxygen demand, DOC: dissolved organic carbon, DIC: dissolved inorganic carbon, NA: not available.
contact surface of 1.32 m2. The rotation speed was ca. 3 rpm and 50% of the disk surface was submerged, resulting in a liquid volume of 2.8 L. The reactor temperature was set at 25.8 ( 0.4 °C, and, only from day 170 on, the reactor pH was adjusted with NaHCO3 to obtain sufficiently high free ammonia levels (see Results section). Dissolved oxygen (DO) levels were not controlled. Reactor Operation. The RBC used in this study was previously used for an OLAND start-up study (data not shown). The reactor was inoculated with OLAND biomass from a previously described RBC (2), and was fed with tap water amended with 0.5 g (NH4)2SO4-N L-1, 3 g NaHCO3 L-1, 0.308 g KH2PO4 L-1, and 2 mL L-1 of a trace elements solution. Over an adaptation period of 2.5 months, the synthetic influent was gradually replaced by digested black water and the hydraulic retention time was gradually increased to prevent a strong increase of the nitrogen loading rate. Finally, during a treatment period of 5 months, the reactor was operated at 100% digested black water. Reactor pH, DO, and temperature were monitored daily, and influent and effluent samples were taken twice or thrice a week for ammonium, nitrite, and nitrate measurements. Soluble chemical oxygen demand (CODsol) was monitored during three weeks of the treatment period. Toward the end of the start-up period (day -52) and at the end of the treatment period (day 220), the volatile suspended solids (VSS) content of the reactor was recalculated from the measured VSS content of the biofilm comprised between two discs. Aerobic and Anoxic Batch Tests. On day 100, in the treatment period, biofilm was harvested to determine the specific activities of AerAOB and AnAOB. Prior to the batch activity tests, the biomass was washed with a phosphate buffer (100 mg P L-1, pH 8) on a sieve (pore size 250 µm) to remove residual dissolved reactor compounds. Aerobic and anoxic ammonium conversion tests were previously described in detail (12) and were performed in triplicate. Fluorescent in Situ Hybridization (FISH). On day 100, in the treatment period, biofilm samples were taken for FISH quantification of the AerAOB and AnAOB. The biofilm was fixed in a 4% paraformaldehyde solution and FISH was performed according to Amann and co-workers (13). To target as many β-proteobacterial AerAOB as possible, an equimolar probe mixture of Nso1225, Nso190 (14), Nmo218 (15), and Cluster 6a192 (16) was used. To target the AnAOB genera “Candidatus Brocadia” and “Candidatus Kuenenia”, probe 5036
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 13, 2009
Amx820 was applied (17). Finally, the biofilm sample was counterstained with a 1:1:1 mixture of EUB338I, EUB338II, and EUB338III for all Bacteria (18, 19). Image acquisition was done on a Leica TCM SP5 confocal laser scanning microscope (CLSM), followed by quantification with Daime software (20), using 10 CLSM image stacks (20 µm depth, five heights), or three-dimensional reconstruction with Imaris software (Bitplane, Zu ¨ rich, Switzerland). Denaturing Gradient Gel Electrophoresis (DGGE). On day -50, in the start-up period, and on day 212, in the treatment period, biofilm was harvested to compare the AerAOB and Planctomycetes communities developed on synthetic medium and digested black water. Conditions and references for DNA extraction, nested PCR and DGGE were previously described in detail (21), and were based on the primers CTO189ABf, CTO189Cf, and CTO653r for AerAOB, and PLA40f and P518r for Planctomycetes. The obtained DGGE patterns were subsequently processed with BioNumerics software (Applied Maths, Sint-Martens-Latem, Belgium) and similarities were calculated as the Pearson correlation coefficient. Chemical Analyses. Nitrite, nitrate, chloride, sulfate and phosphate were determined on a 761 compact ion chromatograph equipped with a conductivity detector (Metrohm, Zofingen, Switzerland). Ammonium (Nessler method), Kjeldahl nitrogen (catalyzed heat degradation, steamdistillation and titration), COD (dichromate method) and VSS (weighing and drying) were determined according to standard methods (22). Free ammonia (FA) levels were calculated based on the reactor ammonium concentration, pH, and temperature (23). Dissolved organic and inorganic carbon were determined with a TOC-5000 (Shimadzu, Kyoto, Japan). Volatile fatty acids were extracted with diethyl ether (22), and analyzed on a Fractovap 4160 gas chromatograph (Carlo Erba, Milan, Italy) equipped with a flame-ionization detector and an ENICA-31 integrator (Delsi-Nermag, Argenteuil, France). Conductivity and pH were measured with electrodes installed on a C833 and C532 meter, respectively (Consort, Turnhout, Belgium), and DO concentration and water temperature with a COM381 DO meter (EndressHauser, Reinach, Switzerland).
Results Reactor Operation. Prior to the addition of digested black water, the synthetically fed OLAND RBC showed a stable nitrogen removal rate of ca. 1300 mg N L-1 d-1 (Figure 1A) and a high biomass concentration (22 g VSS L-1). Evaporation (ca. 1 L d-1) effectively increased effluent concentrations, and thus led to an underestimation of the biological nitrogen removal rate and efficiency. Still, some degree of evaporation is inevitable in rotating contactor practice, and the presented values in Figure 1 and Table 2 were not corrected for evaporation. The start of the adaptation period was called day 0, and over the subsequent 2.5 months, the synthetic influent was gradually replaced with increasing portions of digested black water. Introduction of 10% digested black water in the influent caused a significant decrease in the DO concentration from 1.2 to 0.2 mg O2 L-1 and consequently in the nitrogen removal (Figure 1A). This increased oxygen consumption was most likely due to the aerobic breakdown of the introduced residual biodegradable organic carbon. To increase the oxygen supply to the biofilm, 3.1 g VSS L-1, around 15% of the total biomass, was harvested from the interspaces of all discs, enhancing transport of air and liquid to the biomass between discs. This action quickly restored the DO concentration and nitrogen removal rate at the original level. The further increase of the black water fraction to 100% led to a gradual increase of the effluent ammonium concentration (Figure 1B) and a simultaneous decrease of the nitrogen removal, reaching a stable removal rate around
FIGURE 1. Nitrogen loading and removal rate (panel A) and ammonium and nitrate effluent concentrations and nitrogen removal efficiency (panel B) of the OLAND RBC at the end of the start-up period, and during the adaptation and treatment period. In the adaptation period, the gradual replacement of synthetic influent with digested black water is indicated with the fraction of black water (vol%). The arrow (day 16) indicates an increased oxygen supply resulting from biomass harvesting.
TABLE 2. Summary of the Operational Parameters and Performance Data of the OLAND RBC biofilm at the End of the Start-up (Days -52 to -5; n = 15) and at the End of the Treatment Period with Digested Black Water (Days 196 to 220; n = 7) parameter
end start-up
treatment
volumetric removal rate (mg N L-1 d-1) surfacial removal rate (mg N m-2 d-1) removal efficiency (%) hydraulic residence time (h) reactor pH (-) DO (mg O2 L-1) influent ammonium (mg N L-1) effluent ammonium (mg N L-1) effluent nitrite (mg N L-1) effluent nitrate (mg N L-1) effluent free ammonia (mg N L-1)
1298 ( 197
715 ( 68
2753 ( 418
1517 ( 145
72 ( 5 6.9 ( 0.5 7.2 ( 0.3 1.1 ( 0.3 508 ( 52 50 ( 37 8.9 ( 4.5 86 ( 23 0.9 ( 1.0
76 ( 4 32 ( 2 7.7 ( 0.1 0.7 ( 0.2 1215 ( 54 170 ( 66 36 ( 5 91 ( 19 5.2 ( 3.1
700 mg N L-1 d-1 (Figure 1A) at a high biomass concentration (16 ( 1 g VSS L-1 on day 220).
In the treatment period, nitrate effluent concentrations increased from ca. day 100 on (Figure 1B). To quantify the biological nitrate production stoichiometry, the evaporation effect was corrected for by multiplying effluent concentrations with the inverse of the liquid concentration ratio (ratio of effluent flow to influent flow). Figure 2 depicts the relationship between the biological nitrate production and FA and DO, two well-known parameters for the suppression of NOB (23, 24). The ambient DO levels in the reactor were not sufficiently low to prevent nitratation (Figure 2). For FA, 3 mg N L-1 was adequate to inhibit NOB activity (Figure 2). Considering that the ammonium concentrations in the reactor were of the order of 100 mg NH4+-N L-1, this inhibitory FA level could be reached at pH 7.7. Therefore, in total 38 g NaHCO3 was added to keep the pH between 7.7 and 8.0 from day 170 to 220, decreasing the effluent nitrate concentration (Figure 1B). The combination of suppressed nitrite oxidation and a lower volumetric loading rate resulted in 76% removal efficiency over the last month of operation (Figure 1B; Table 2). The effluent COD concentrations exceeded the influent COD concentrations with a factor 1.30 ( 0.02, and quantification of the biological COD removal required a correction VOL. 43, NO. 13, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5037
FIGURE 2. Relation between the free ammonia (FA), dissolved oxygen (DO) and nitrate production per ammonium removed in the OLAND reactor, corrected for the evaporation effect (day -52 to 220). On each of the planes, all data points are projected to enhance visualization. The expected level of nitrate production in the absence of nitratation and denitrification is indicated with a dashed line at 11%. The plots of DO vs nitrate production (gray squares) and FA vs DO (white hexagons) show scattered data, whereas the FA vs nitrate production plot (black circles) shows that FA levels above 3 mg N L-1 (dashed line) were inhibitory for nitratation. of the effluent levels for the concentration increase due to evaporation (factor 1.49 ( 0.09). This showed that 119 ( 14 mg COD L-1 or 20 ( 2% of the influent COD was biologically removed during OLAND treatment at a consumption ratio of 0.15 ( 0.01 g COD g-1 N. It follows that the residual organic carbon in the digested black water was mainly recalcitrant. Biomass Characterization. FISH was used to study the group morphology of AerAOB and AnAOB of the OLAND biofilm during the treatment period. AnAOB were housed in spherical microcolonies (Figure 3A,D) with diameters ranging from 5 to 100 µm. AerAOB cells displayed different morphologies and arrangements. Typical spherical microcolonies were observed, which were relatively distant from each other (Figure 3B,E) or densely packed (Figure 3C). Within these colonies, all bacterial cells were AerAOB (Figure 3B inset). In addition, parallel bands of AerAOB cells were observed throughout the biofilm (Figure 3D) as well as regions with scattered, ungrouped cells (Figure 3D bottom; Figure 3E and F). In these latter arrangements, non-AerAOB bacterial cells were observed in between the AerAOB cells (Figure 3E; Figure 3F inset). To estimate the specific activity of AerAOB and AnAOB in the treatment period, AerAOB and AnAOB were quantified by FISH, and the aerobic and anoxic biofilm activities were determined in batch tests. AerAOB and AnAOB made up 43 ( 16 and 8.0 ( 7.1% of the biofilm, respectively. The aerobic and anoxic ammonium oxidation rates were 96.9 ( 2.5 and 32.4 ( 3.9 mg N g-1 biofilm-VSS d-1, respectively. For the aerobic test, no nitrate was formed, which corresponds with the reactor performance at the moment biomass was harvested for the activity test (day 100, Figure 1B). For the anoxic test, the ratio of nitrite consumption to ammonium consumption was 0.98 ( 0.02, and the ratio of nitrate production to ammonium consumption was 0.13 ( 0.01. With the derived AnAOB stoichiometry and correction for evaporation, aerobic and anoxic biofilm activities from the continuous reactor performance (days 97 to 104, n ) 4) were 35.5 ( 8.0 and 33.0 ( 6.0 mg NH4+-N g-1 VSS d-1, respectively, calculated at 16 g VSS L-1. DGGE was performed to investigate the stability of the community of AerAOB and Planctomycetes, comprising the AnAOB (25), in relation to the change from a well-defined synthetic medium to a complex matrix of digested black 5038
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 13, 2009
water. Clustering analysis of the DGGE gels showed that the digested black water matrix did not cause major shifts in the functional community (Figure 4). The dominant AerAOB and Planctomycetes species (Figure 4 arrows 1 and 3) were resistant to the influent change. It should be noted that the additional AerAOB bands appearing in the digested black water biofilm pattern (Figure 4 arrow 2) were observed at the same position in synthetically fed OLAND SBR granules (8) (data not shown), so these species were not specific for digested black water.
Discussion This lab-scale study shows the technical feasibility of the OLAND process to remove nitrogen from digested concentrated black water. After a period of 2.5 months in which a well performing OLAND biofilm was gradually adapted to digested black water, stable nitrogen removal of 76% of the nitrogen at 716 mg N L-1 d-1 was reached during the last operational month (Table 2). The obtained nitrogen removal is comparable to full-scale OLAND-type reactors, which remove from 400 to 1700 mg N L-1 d-1 (6), and long-term AnAOB activity was apparently not jeopardized by residual micropollutants. Controlling the FA level at ca. 3 mg N L-1 was demonstrated to be a good NOB inhibition strategy. However, it should be noted that NOB could get acclimated to high FA levels in the longer term (26), and that DO control at a set point of for instance 0.5 mg O2 L-1 could be necessary. The amount of biodegradable organic carbon entering the OLAND reactor depends on the treatment efficiency of the anaerobic digester and, additionally in our study, on temporal storage of the digestate, since soluble COD concentrations were around half of the previously reported values for the same digestate (Table 1 (1)). Although model predictions indicate that heterotrophic bacteria do not outcompete AnAOB at a COD/N influent ratio as high as 5 (27), a high COD/N ratio might lead to process instabilities related to higher denitrification levels, as encountered for the OLAND treatment of diluted urine (COD/N consumption ) 1.9 (10)). Future experiments should confirm the threshold COD/N ratio that an OLAND system can tolerate to ensure long-term stable and high nitrogen removal efficiency. In the presence of 0.93 g HCO3--C g-1 N, the inorganic carbon content of wastewater is sufficient to neutralize the
FIGURE 3. CLSM-FISH images of the OLAND biofilm during the treatment period. Aerobic ammonium-oxidizing bacteria (AerAOB; FLUO-labeled Nso1225, Nso190, Nmo218, and Cluster 6a192) are displayed in green or cyan, anoxic ammonium-oxidizing bacteria (AnAOB; Cy3-labeled Amx820) in red or magenta and all bacteria (Cy5-labeled EUBI, II and III) in blue, as indicated by the color scheme in panel A. (A) Three-dimensional reconstruction of image stack (49 images, 0.5 µm step size) showing AnAOB cells grouped in spherical microcolonies (small grid indents ) 10 µm). Inset: detail of typical torus appearance of hybridized AnAOB cells. (B) Three-dimensional reconstruction of image stack (212 images, 0.1 µm step size) showing AerAOB cells grouped in spherical microcolonies (small grid indents ) 1 µm). Inset: detail of a microcolony showing composed of only AerAOB cells. (C) Densely packed AerAOB microcolonies. (D) AerAOB arranged in parallel bands cells interlarded with AnAOB clusters; scattered AerAOB cells on the bottom. (E) Biofilm zone with both AerAOB microcolonies and scattered AerAOB cells. (F) Biofilm zone with scattered AerAOB cells and neighboring non-AerAOB bacterial cells. Inset: detailed view of dispersed AerAOB organization. VOL. 43, NO. 13, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5039
FIGURE 4. DGGE gels for aerobic ammonium-oxidizing bacteria (AerAOB) and Planctomycetes (Plancto), performed on the OLAND biofilm during start-up (day -50) and treatment period (day 212). Similarities were calculated as the Pearson correlation coefficient and numbered arrows are discussed in the Results section. protons, since these are produced at 1.1 mols of H+ per mol of N removed (8). With the inorganic carbon dissociation constants (28), it was calculated that 97.4% of the dissolved inorganic carbon (Table 1) was present as bicarbonate at 26 °C and pH 8.0, so 0.73 g HCO3--C g-1 N was present in the digested black water. This bicarbonate level could thus neutralize the proton production associated with 78% nitrogen removal, and was around 20% lower than previously reported (29), possibly due to storage of the digestate. Since only 0.06 g HCO3--C g-1 N was added to keep the pH above 7.7, no base addition may be required working with fresh digestate. During the treatment period, the AerAOB activity determined in the batch test overestimated the reactor activity, presumably because disintegration of the biofilm into pieces increased to available surface for oxygen uptake. From the reactor activity and AerAOB abundance, the average AerAOB activity was 83 mg N g-1 AerAOB-VSS d-1, which is at least an order of magnitude below the expected rate (5). This suggests that a considerable part of the detected AerAOB were present in anoxic zones, unable to contribute to aerobic ammonium oxidation. AerAOB can gain energy from anoxic ammonium oxidation (30), a reaction which is around 50 times slower than anammox (5), and therefore considered to contribute minimally to the OLAND nitrogen removal. Also, a number of other physiological traits may enable longterm AerAOB survival under oxygen starvation, as reviewed by Geets and co-workers (31). The AnAOB activity rates from batch and reactor tests corresponded well, and nitrite consumption and nitrate production were comparable to literature reports (32, 33). The AnAOB represented a minority in the biofilm (8%), but exerted a high specific activity, i.e., 411 mg NH4+-N g-1 AnAOB-VSS d-1, based on the reactor activity rate. The latter value is in the order of the reported maximum activity of “Candidatus Kuenenia stuttgartiensis” (33), i.e., 365 mg N g-1 AnAOB-VSS d-1, converted at 0.6 g protein g-1 VSS (32). The grouping of AnAOB in spherical clusters has been observed before (17). In contrast, the spatial arrangement of most AerAOB cells was different from the typical spherical microcolonies described in many reports, e.g., refs 14, 15, 34. Four putative explanations for AerAOB scattering are proposed. First, different AerAOB cell densities may be related to different AerAOB species (34). Also, the cell spreading may result from growth and production of extracellular polymeric substances (EPS) of putative neighboring heterotrophs, which were observed in between scattered AerAOB cells (Figure 3F). Further, lower substrate concentrations deeper in the biofilm may be another factor leading to increased AerAOB cell distances (34), however, spatial data were lacking in our study to confirm this. Finally, disperse, scattered AerAOB cells have been observed after addition of hydroxylamine to a nitrifying biofilm (35). Hydroxylamine is an intermediate of both aerobic and anoxic AerAOB metabolism (30) and putatively of the AnAOB metabolism (5), but it is not clear 5040
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 13, 2009
whether endogenous concentrations inside the OLAND biofilm would reach sufficient levels to influence the AerAOB architecture. DGGE analysis showed that the dominant AerAOB and AnAOB species were not influenced by the transition from synthetic medium to digested black water. In the Supporting Information, the range-weighted richness and functional organization of the communities were given (SI Table S.1) and interpreted according to Marzorati and co-workers (36). The proposed treatment train for decentralized treatment of black water includes the sequence of energy recovery by anaerobic digestion, phosphate recovery by struvite precipitation, nitrogen removal by OLAND and a final polishing step (1). Based on our findings, the order of phosphate and nitrogen treatment could be reversed. First, anammox was not inhibited by the highest measured phosphate levels in the reactor (120 mg P L-1, due to evaporation), despite of the previously reported activity inhibition of 80% at 110 mg P L-1 (2). Second, precipitation of ammonium and phosphate as struvite (MgNH4PO4.6H2O) requires one mole N per mole P, so 120 mg P L-1 would require 54 mg NH4+-N L-1, which was present in the OLAND effluent (Table 2). Furthermore, the application of struvite precipitation after the OLAND step could increase the overall nitrogen removal efficiency from 76 to 80%. Overall, this study demonstrated high-rate nitrogen removal from digested concentrated black water by the OLAND process. In view of the technical feasibility and the economical benefits of this process, the OLAND process is currently being tested at pilot scale (0.5 m3) on site in Sneek, The Netherlands.
Acknowledgments S.E.V. was supported by a grant (Aspirant) from the Fonds voor Wetenschappelijk Onderzoek (FWO) Vlaanderen, M.C. by the Angeles Alvarin ˜ o program (AA-065) from the Xunta de Galicia, and work at DTU by the grant FTP-ReSCoBiR from the Danish Agency for Science Technology and Innovation (Research Council for Technology and Production). We gratefully thank Brendo Meulman (Landustrie NV, The Netherlands) for the kind provision of digested black water, Diederik Vandriessche and Siska Maertens for the molecular analyses, and Hayde´e De Clippeleir, Peter De Schryver and Willem Demuynck for the inspiring scientific discussions.
Supporting Information Available Additional information and Table S.1. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Zeeman, G.; Kujawa, K.; de Mes, T.; Hernandez, L.; de Graaff, M.; Abu-Ghunmi, L.; Mels, A.; Meulman, B.; Temmink, H.; Buisman, C.; van Lier, J.; Lettinga, G. Anaerobic treatment as a core technology for energy, nutrients and water recovery from source-separated domestic waste(water). Water Sci. Technol. 2008, 57 (8), 1207–1212.
(2) Pynaert, K.; Smets, B. F.; Wyffels, S.; Beheydt, D.; Siciliano, S. D.; Verstraete, W. Characterization of an autotrophic nitrogenremoving biofilm from a highly loaded lab-scale rotating biological contactor. Appl. Environ. Microbiol. 2003, 69 (6), 3626– 3635. (3) Kuai, L. P.; Verstraete, W. Ammonium removal by the oxygenlimited autotrophic nitrification-denitrification system. Appl. Environ. Microbiol. 1998, 64 (11), 4500–4506. (4) Fux, C.; Siegrist, H. Nitrogen removal from sludge digester liquids by nitrification/denitrification or partial nitritation/anammox: environmental and economical considerations. Water Sci. Technol. 2004, 50 (10), 19–26. (5) Jetten, M. S. M.; Wagner, M.; Fuerst, J.; van Loosdrecht, M.; Kuenen, G.; Strous, M. Microbiology and application of the anaerobic ammonium oxidation (‘anammox’) process. Curr. Opin. Biotechnol. 2001, 12 (3), 283–288. (6) van der Star, W. R. L.; Abma, W. R.; Blommers, D.; Mulder, J. W.; Tokutomi, T.; Strous, M.; Picioreanu, C.; Van Loosdrecht, M. C. M. Startup of reactors for anoxic ammonium oxidation: Experiences from the first full-scale anammox reactor in Rotterdam. Water Res. 2007, 41 (18), 4149–4163. (7) Wett, B. Solved upscaling problems for implementing deammonification of rejection water. Water Sci. Technol. 2006, 53 (12), 121–128. (8) Vlaeminck, S. E.; Cloetens, L. F. F.; De Clippeleir, H.; Carballa, M.; Verstraete, W. Granular biomass capable of partial nitritation and anammox. Water Sci. Technol. 2009, 59 (3), 610–617. (9) Cema, G.; Szatkowska, B.; Plaza, E.; Trela, J.; Surmacz-Gorska, J. Nitrogen removal rates at a technical-scale pilot plant with the one-stage partial nitritation/Anammox process. Water Sci. Technol. 2006, 54 (8), 209–217. (10) Udert, K. M.; Kind, E.; Teunissen, M.; Jenni, S.; Larsen, T. A. Effect of heterotrophic growth on nitritation/anammox in a single sequencing batch reactor. Water Sci. Technol. 2008, 58 (2), 277–284. (11) Lienert, J.; Burki, T.; Escher, B. I. Reducing micropollutants with source control: substance flow analysis of 212 pharmaceuticals in faeces and urine. Water Sci. Technol. 2007, 56 (5), 87–96. (12) Vlaeminck, S. E.; Geets, J.; Vervaeren, H.; Boon, N.; Verstraete, W. Reactivation of aerobic and anaerobic ammonium oxidizers in OLAND biomass after long-term storage. Appl. Microbiol. Biotechnol. 2007, 74 (6), 1376–1384. (13) Amann, R. I.; Krumholz, L.; Stahl, D. A. Fluorescent-oligonucleotide probing of whole cells for determinative, phylogenetic, and environmental studies in microbiology. J. Bacteriol. 1990, 172 (2), 762–770. (14) Mobarry, B. K.; Wagner, M.; Urbain, V.; Rittmann, B. E.; Stahl, D. A. Phylogenetic probes for analyzing abundance and spatial organization of nitrifying bacteria. Appl. Environ. Microbiol. 1996, 62 (6), 2156–2162. (15) Gieseke, A.; Purkhold, U.; Wagner, M.; Amann, R.; Schramm, A. Community structure and activity dynamics of nitrifying bacteria in a phosphate-removing biofilm. Appl. Environ. Microbiol. 2001, 67 (3), 1351–1362. (16) Adamczyk, J.; Hesselsoe, M.; Iversen, N.; Horn, M.; Lehner, A.; Nielsen, P. H.; Schloter, M.; Roslev, P.; Wagner, M. The isotope array, a new tool that employs substrate-mediated labeling of rRNA for determination of microbial community structure and function. Appl. Environ. Microbiol. 2003, 69 (11), 6875–6887. (17) Schmid, M.; Twachtmann, U.; Klein, M.; Strous, M.; Juretschko, S.; Jetten, M.; Metzger, J. W.; Schleifer, K. H.; Wagner, M. Molecular evidence for genus level diversity of bacteria capable of catalyzing anaerobic ammonium oxidation. Syst. Appl. Microbiol. 2000, 23 (1), 93–106. (18) Amann, R. I.; Binder, B. J.; Olson, R. J.; Chisholm, S. W.; Devereux, R.; Stahl, D. A. Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 1990, 56 (6), 1919–1925. (19) Daims, H.; Bruhl, A.; Amann, R.; Schleifer, K. H.; Wagner, M. The domain-specific probe EUB338 is insufficient for the detection of all Bacteria: Development and evaluation of a more
(20)
(21)
(22)
(23)
(24)
(25)
(26)
(27)
(28)
(29)
(30)
(31)
(32)
(33)
(34)
(35)
(36)
comprehensive probe set. Syst. Appl. Microbiol. 1999, 22 (3), 434–444. Daims, H.; Lucker, S.; Wagner, M. daime, a novel image analysis program for microbial ecology and biofilm research. Environ. Microbiol. 2006, 8 (2), 200–213. Boon, N.; De Windt, W.; Verstraete, W.; Top, E. M. Evaluation of nested PCR-DGGE (denaturing gradient gel electrophoresis) with group-specific 16S rRNA primers for the analysis of bacterial communities from different wastewater treatment plants. FEMS Microbiol. Ecol. 2002, 39 (2), 101–112. Greenberg, A. E.; Clesceri, L. S.; Eaton, A. D. Standard Methods for the Examination of Water and Wastewater; American Public Health Association: Washington DC, 1992. Anthonisen, A. C.; Loehr, R. C.; Prakasam, T. B. S.; Srinath, E. G. Inhibition of nitrification by ammonia and nitrous acid. J. Water Pollut. Control Fed. 1976, 48 (5), 835–852. Laanbroek, H. J.; Gerards, S. Competition for limiting amounts of oxygen between Nitrosomonas europaea and Nitrobacter winogradskyi grown in mixed continuous cultures. Arch. Microbiol. 1993, 159 (5), 453–459. Strous, M.; Fuerst, J. A.; Kramer, E. H. M.; Logemann, S.; Muyzer, G.; van de Pas-Schoonen, K. T.; Webb, R.; Kuenen, J. G.; Jetten, M. S. M. Missing lithotroph identified as new planctomycete. Nature 1999, 400 (6743), 446–449. Turk, O.; Mavinic, D. S. Maintaining nitrite buildup in a system acclimated to free ammonia. Water Res. 1989, 23 (11), 1383– 1388. Lackner, S.; Terada, A.; Smets, B. F. Heterotrophic activity compromises autotrophic nitrogen removal in membraneaerated biofilms: Results of a modeling study. Water Res. 2008, 42 (4-5), 1102–1112. Crittenden, J. C.; Rhodes Trussell, R.; Hand, D. W.; Howe, K. J.; Tchobanoglous, G. Water treatment: Principles and Design, 2nd ed; Wiley: New York, 2005. de Graaff, M. S.; Zeeman, G.; Temmink, H.; Buisman, C. J. N. The challenge to treat concentrated black water: Anaerobic treatment and autotrophic nitrogen removal. In Sanitation Challenge; Wageningen, the Netherlands, 2008. Bock, E.; Schmidt, I.; Stuven, R.; Zart, D. Nitrogen loss caused by denitrifying Nitrosomonas cells using ammonium or hydrogen as electron-donors and nitrite as electron-acceptor. Arch. Microbiol. 1995, 163 (1), 16–20. Geets, J.; Boon, N.; Verstraete, W. Strategies of aerobic ammoniaoxidizing bacteria for coping with nutrient and oxygen fluctuations. FEMS Microbiol. Ecol. 2006, 58 (1), 1–13. Strous, M.; Heijnen, J. J.; Kuenen, J. G.; Jetten, M. S. M. The sequencing batch reactor as a powerful tool for the study of slowly growing anaerobic ammonium-oxidizing microorganisms. Appl. Microbiol. Biotechnol. 1998, 50 (5), 589–596. Egli, K.; Fanger, U.; Alvarez, P. J. J.; Siegrist, H.; van der Meer, J. R.; Zehnder, A. J. B. Enrichment and characterization of an anammox bacterium from a rotating biological contactor treating ammonium-rich leachate. Arch. Microbiol. 2001, 175 (3), 198–207. Okabe, S.; Kindaichi, T.; Ito, T.; Satoh, H. Analysis of size distribution and areal cell density of ammonia-oxidizing bacterial microcolonies in relation to substrate microprofiles in biofilms. Biotechnol. Bioeng. 2004, 85 (1), 86–95. Kindaichi, T.; Okabe, S.; Satoh, H.; Watanabe, Y. Effects of hydroxylamine on microbial community structure and function of autotrophic nitrifying biofilms determined by in situ hybridization and the use of microelectrodes. Water Sci. Technol. 2004, 49 (11-12), 61–68. Marzorati, M.; Wittebolle, L.; Boon, N.; Daffonchio, D.; Verstraete, W. How to get more out of molecular fingerprints: practical tools for microbial ecology. Environ. Microbiol. 2008, 10 (6), 1571–1581.
ES803284Y
VOL. 43, NO. 13, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5041
