当前位置:首页 >> 环境科学/食品科学 >>

Enrichment of anaerobic ammonium oxidizing (Anammox) bacteria


Separation and Puri?cation Technology 104 (2013) 130–137

Contents lists available at SciVerse ScienceDirect

Separation and Puri?cation Technology
journal homepage: www.elsevier.com/locate/seppur

Enrichment of anaerobic ammonium oxidizing (Anammox) bacteria from OLAND and conventional sludge: Features and limitations
A. Monballiu a,?, E. Desmidt a, K. Ghyselbrecht a, H. De Clippeleir b, S.W.H. Van Hulle c,e, W. Verstraete b, B. Meesschaert a,d
a Department of Industrial Sciences and Technology, Katholieke Hogeschool Brugge – Oostende, Associated to the Katholieke Universiteit Leuven as Faculty of Industrial Engineering Sciences, Zeedijk 101, B-8400 Oostende, Belgium b Laboratory of Microbial Ecology and Technology (LabMET), Faculty of Bio-Engineering Sciences, Ghent University, Coupure Links 653, B-9000 Gent, Belgium c Research Group EnBiChem, University College West Flanders, Graaf Karel de Goedelaan 5, B-8500 Kortrijk, Belgium d Centre for Surface Chemistry and Catalysis, Department of Microbial and Molecular Systems (M2S), Faculty of Bio-Engineering Sciences, Katholieke Universiteit Leuven, Kasteelpark Arenberg 23, B-3001 Heverlee, Belgium e Biomath, Department of Applied Mathematics, Biometrics and Process Control, Ghent University, Coupure Links 653, 9000 Gent, Belgium

a r t i c l e

i n f o

a b s t r a c t
The discharge of nitrogen is increasingly restricted. The autotrophic nitrogen removal process, which combines partial nitritation and anaerobic ammonium oxidation (Anammox), is recommended for high ammonium loaded streams with no or low organic carbon content because of the lower operational costs. However, the cultivation and start-up period can be long and time-consuming due to the slow growth rate of Anammox bacteria. The aim of this research was to evaluate the feasibility of Anammox cultivation in an anaerobic continuous stirred tank reactor (CSTR). Monitoring of nitrogen conversion showed Anammox activity with an increasing nitrogen removal rate up to 700 mg N L?1 day?1 in a reactor inoculated with OLAND (Oxygen-Limited Autotrophic Nitri?cation/Denitri?cation) seed sludge. A second lab scale reactor, a modi?ed CSTR which was operated as an anaerobic up?ow ?xed bed reactor (UFBR), was inoculated with nitrifying and denitrifying sludge from a manure treatment plant. The ?xed bed consisted of polyurethane sponge which served as a carrier material for the slowly growing biomass. The UFBR showed removal according to the Anammox metabolism within 3 months and a nitrogen removal rate up to 250 mg N L?1 day?1 was achieved. Primary research on the treatment on pig manure wastewater indicates that organic load had an inhibitory effect on the Anammox activity. The results also indicate that Anammox bacteria can stay dormant or hidden for longer periods, but will reactivate under the suitable conditions. ? 2012 Elsevier B.V. All rights reserved.

Article history: Received 10 July 2012 Received in revised form 29 October 2012 Accepted 30 October 2012 Available online 28 November 2012 Keywords: Anammox Autotrophic nitrogen removal Wastewater treatment Reactor con?guration

1. Introduction Nutrients (N, P) have to be limited in the ef?uent of wastewater treatment plants (WWTPs). The release of these compounds can cause an uncontrolled growth of algae in aquatic ecosystems with a de?cit of oxygen for higher organisms as a result leading to a sharp decrease in biodiversity. This process is known as eutrophication. Conventionally, the removal of nitrogen is achieved by a nitri?cation/denitri?cation process. In an aeration step, ammonium is oxidized to nitrite and nitrate by autotrophic nitrifying bacteria.

? Corresponding author. Tel.: +32 59 56 90 55; fax: +32 59 56 90 53.
E-mail addresses: annick.monballiu@khbo.be (A. Monballiu), evelyn.desmidt@khbo.be (E. Desmidt), karel.ghyselbrecht@khbo.be (K. Ghyselbrecht), haydee.declippeleir@ugent.be (H. De Clippeleir), stijn.van.hulle@howest.be (S.W.H. Van Hulle), willy.verstraete@ugent.be (W. Verstraete), boudewijn.meesschaert@khbo.be (B. Meesschaert). 1383-5866/$ - see front matter ? 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.seppur.2012.10.046

Nitrite and nitrate are further reduced to nitrogen gas by heterotrophic denitrifying bacteria under anoxic conditions. However, the process becomes less sustainable if the wastewaters contain high concentrations of ammonium and lack of organic carbon. Besides the aeration costs, the cost for the addition of an external organic carbon source such as methanol/glycerol is necessary for the denitri?cation stage. This leads to an increase of the total operational costs of the WWTP [1]. An alternative to treat high-loaded ammonium streams with low carbon content, such as ef?uents from anaerobic digestion processes and the thin fraction from piggery manure, is the implementation of autotrophic nitrogen removal (ANR). This process consist of the combination between partial nitritation and the anaerobic ammonium oxidation or Anammox process [2]. The obtained shortcut of the conventional nitrogen removal is illustrated in Fig. 1. Anammox is a relative new promising biotechnological process that was initially discovered in a denitrifying reactor in Delft, The

A. Monballiu et al. / Separation and Puri?cation Technology 104 (2013) 130–137

131

Fig. 1. Nitrogen cycle with the role of anaerobic ammonium oxidation (Anammox) as a metabolic shortcut of nitrogen conversion [3].

Netherlands [4]. Anammox bacteria are able to consume ammonium and nitrite under anaerobic conditions according to the following equation:
? ? ? NH? 4 ? 1:32NO2 ? 0:066HCO3 ? 0:13H

! 1:02N2 ? 0:26NO? 3 ? 0:066CH2 O0:5 N0:15 ? 2:03H2 O

? 1?

Ammonium is oxidized to nitrogen gas using nitrite as the electron acceptor. A small fraction of nitrite is anaerobically oxidized to nitrate and yields electrons for the reduction of carbon dioxide for cell growth. Hence, the molar ratio of ammonia and nitrite in Anammox catabolism is 1:1.32 [5]. Therefore, a partial nitritation process is necessary to produce a suited Anammox in?uent. Advanced techniques to reach this ±50% conversion of ammonia into nitrite are (1) manipulating the temperature and the hydraulic retention time (HRT), (2) inhibition of nitrite oxidizing bacteria by free ammonium or (3) manipulating the dissolved oxygen concentration [6,7]. In comparison with conventional nitri?cation/denitri?cation, energy costs for aeration can be reduced up to 63% and organic carbon is no longer required in this fully autotrophic process, as shown in Fig. 1. Furthermore, emission of the greenhouse gasses NO and N2O, emitted by the nitri?cation and denitri?cation process, are signi?cantly reduced [8]. Moreover, CO2 is consumed because of the autotrophic character of the process. The latter results in less sludge production which could reduce the dimensions of current WWTPs. Previous studies showed the effectiveness of autotrophic nitrogen removal [3]. Moreover, since Strous et al. [9] proved the reversible inhibition of oxygen to the Anammox process, research was dedicated to the possibility of the combination of partial nitritation and Anammox in one single process. This had given rise to processes such as OLAND (Oxygen-Limited Autotrophic Nitri?cation/ Denitri?cation) [10] and CANON (Completely Autotrophic Nitrogen-removal over Nitrite) [11]. However, the slow growth rate of the Anammox bacteria – which have a doubling time of 9 days under optimal conditions [12] – creates long start-up times, de?nitely at full scale implementation [13]. Once suf?cient amounts of Anammox sludge are available, upscaling can be carried out. Thus, it is important to optimize the start-up parameters for Anammox cultivation in respect to reactor con?guration and feed-regime [14,15]. A fundamental problem is to retain the slow-growing biomass in the reactor. A ?rst successful enrichment was achieved with a ?uidized bed reactor (FBR) where the microbial community grew as bio?lms on sand particles [16]. However, the bio?lm structure was not constant over de reactor because of the absence of complete bulk mixing which led to a decreased

Anammox activity. Anammox sludge easily attaches to any solid surface such as the inside surface of the reactor. The sequencing batch reactor (SBR) was shown to be a powerful experimental set-up to combine the major strategies of bio?lm improvement and biomass retention [17]. Moreover, it provides opportunities for the one-stage reactor which combines the partial nitritation and the Anammox process. The ammonium oxidizers are active in the outer layers of the bio?lm, producing a suitable amount of nitrite for the Anammox organisms active in the inner layers. In this way, the Anammox bacteria are protected from oxygen. Rapidly settling biomass is opportune and excellent biomass retention can be achieved by granulation [18,19]. The granulation of the Anammox microorganisms is an important contribution to obtain a high nitrogen removal, in combination with a well-controlled strategy to avoid nitrite accumulation [20]. Liquid-induced shear forces in?uenced this granulation process signi?cantly [21]. Nevertheless, due to the fact that Anammox bacteria are inhibited by their own substrates, continuous reactors are prevailing. Isaka et al. [22] worked with a nonwoven biomass carrier in a continuously anaerobic biological ?ltrated (ABF) reactor to enrich Anammox seed sludge. Whereas nitrite inhibition was considered to be the key factor in the instability of the Anammox process [12], more attention was paid to the inhibitory effect of pH and free ammonium by Tang et al. [23]. It was shown that anaerobic granular sludge could be successfully used for the start-up of the Anammox process in a continuously up?ow bio?lm reactor (UBF). A relative low and stable pH (8.30) and free ammonium (FA) concentration (below 30 mg L?1 NH3) ensured that the UBF achieved a better nitrogen removal. Jung et al. [24] investigated the factors affecting the cultivation of Anammox by using an up?ow anaerobic sludge blanket reactor (UASB). The reactor was inoculated with granular sludge and showed Anammox activity after 3 months of continuous operation. Here, it was reported to keep dissolved oxygen (DO) below 0.2 mg L?1; FA below 0.2 mg L?1 and nitrite nitrogen below 35 mg L?1. Another promising tool for cultivating slowgrowing Anammox bacteria, is the membrane bioreactor (MBR) where biomass retention is not based on settling of biomass but where the ef?uent is withdrawn via a membrane which is impermeable for microbial cells. So, full biomass retention can be achieved without a selection on settling ability [25]. Within 2 months of operation, Wang et al. [26] noticed Anammox activity in a continuously fed lab scale MBR inoculated with conventional activated sludge. So, literature is extensive, nevertheless cultivation of Anammox from conventional sludge is still cumbersome. One major bottleneck is the use of a suitable reactor design. In this context, the main objective of this study was the assessment of an anaerobic continuous

132

A. Monballiu et al. / Separation and Puri?cation Technology 104 (2013) 130–137

stirred tank reactor (CSTR) towards its enrichment properties of slowly growing microorganisms. The features and limitations during the successful cultivation of Anammox in an anaerobic CSTR inoculated with OLAND seed sludge were linked back to the startup of the Anammox process based on conventional activated sludge from a pig manure treatment plant. Within this part of the research, the anaerobic CSTR was modi?ed in an anaerobic up?ow ?xed bed reactor (UFBR) by providing a CSTR with carrier material (polyurethane sponge) in order to enhance biomass retention. Both reactors processed synthetic wastewater. Furthermore pig manure wastewater was treated by the anaerobic UFBR to investigate the in?uence of organic load on Anammox performance. 2. Materials and methods

?7H2O 0.2 and 1 mL L?1 of both trace element solutions I and II as described by van de Graaf et al. [16]. Nitrogen compounds were added to the synthetic wastewater as (NH4)2SO4, NaNO2 and NaNO3. The concentrations used varied depending on the experimental results. The synthetic feed consisted of ammonium:nitrite in a molar ratio of about 1:1.32 in order to meet the requirements for Anammox metabolism (Eq. (1)). To investigate the in?uence of organic load on the Anammox performance, thin fraction of pig manure was diluted and supplemented with nitrite to obtain the desired ratio of ammonia and nitrite. The synthetic wastewater and the pig manure wastewater, having a pH between 7.5 and 8, were both fed to the reactors with a hydraulic retention time (HRT) of 40 h. 2.3. Analytical methods

2.1. Reactor con?gurations The ?rst reactor (Reactor A, Fig. 2a) consisted of a lab scale CSTR with a working volume of 4 L, and was built according to BNB EN ISO 11733. The reactor was inoculated with sludge (250 mL) obtained from a lab scale Rotating Biological Contactor (RBC) with OLAND activity (LabMET, University of Ghent, Belgium) [10,27]. The reactor was continuously mixed with a mechanical stirrer and kept at 35 °C with an aquarium heater. Sludge was recycled to the reactor via a compressed air lift system during 20 s every 30 min. A layer of paraf?n was placed on the top of the reactor to improve anaerobic conditions. Fig. 2b shows a schematic overview of the second lab scale reactor (Reactor B). A CSTR was operated as an anaerobic UFBR with a working volume of 3.7 L. The reactor contained a carrier material of polyurethane sponge to retain biomass. A glass tube in the middle of the sponge, made it possible to mix the lower part of the reactor continuously and to feed the in?uent below the sponge, while the ef?uent was decanted at the upper side. The reactor was also kept at a temperature of 35 °C. The UFBR was inoculated with a mixture of nitrifying and denitrifying sludge from a manure treatment plant (DANIS NV, Izegem, Belgium). When Anammox activity was detected in this reactor (Section 3), it was scaled up to a similar 16 L reactor by transferring the sponges with biomass and adding extra sponges. Light has a negative effect on Anammox metabolism. Van de Graaf et al. [16] reported a decrease of 30–50% in Anammox activity due to light. Therefore, both reactors were shielded from visible light with aluminum foil. 2.2. Composition of the feed The reactors were fed with synthetic wastewater which contained (g/L): KHCO3 1.25, KH2PO4 0.025, CaCl2?6H2O 0.45, MgSO4Samples of in?uent and ef?uent were taken two to three times a week. The pH was determined with a pH meter from Mettler Toledo. Inorganic carbon (IC) was measured with a Total Carbon Analyzer (TOC–VCPN, Shimadzu, Japan). Ammonia was measured with a DIONEX (USA) DX-100 ion chromatograph supplied with an IonPac?CS14 analytical column with 10 mM CH3SO2OH (methane sulfonic acid) as eluent. Nitrite and nitrate were determined by using a DIONEX series 4500i ion chromatograph provided with an IonPac?AS14 analytical column with 3.5 mM Na2CO3 and 1 mM NaHCO3 as eluent. Both ion chromatographs were equipped with a conductivity detector. 2.4. Fluorescent In Situ Hybridisation (FISH) FISH assays of the biomass were implemented to detect nitrifying (aerobic ammonium and nitrite oxidizing bacteria (AOB and NOB)) and Anammox bacteria. Cell ?xation and FISH analysis were carried according to the standard hybridization protocol [28]. The probe combinations, labeled with ?uorescein or Cy3, used were Nso1225 and Nso190 which represent for b-proteobacterial AOB; Amx820 for Anammox bacteria ‘‘Candidatus Brocadia’’ and ‘‘Candidatus Kuenenia’’; NIT3 and its competitor for Nitrobacter and Ntspa662 and its competitor for Nitrospira [29]. Image acquisition was done on a Zeiss Axioskop 2 Plus epi?uorescence microscope. 3. Results and discussion 3.1. Reactor A (inoculated with sludge from OLAND RBC) This CSTR was inoculated with sludge from a lab scale RBC with OLAND activity. Initially, after inoculation several attempts to enhance Anammox activity failed due to the feeding of non-optimal ef?uent from a lab scale partial nitritation reactor. Therefore, it

Fig. 2. Schematic representation of (A) Reactor A, a CSTR sealed with paraf?n and (B) Reactor B, modi?ed to an UFBR by the use of a carrier material.

A. Monballiu et al. / Separation and Puri?cation Technology 104 (2013) 130–137

133

was decided to switch to synthetic wastewater which initially con?1 sisted of 50 mg L?1 NH? NO? 2 —N. Feeding syn4 —N and 50 mg L thetic water to the reactor gave rise to the removal of both ammonia and nitrite. Moreover, there was a simultaneous production of nitrate which was an indication for Anammox activity. When the ef?uent had the desired quali?cations (neither ammonia nor nitrite were present), in?uent concentrations for both nitrogen compounds were gradually increased. Figs. 3 and 4 show the results obtained with this reactor one year after the inoculation. The nitrogen removal rate (NRR) reached up to 700 mg ? ?1 (NH? day?1 with almost complete removal of 4 —N ? NO2 —N) L ammonia and nitrite. The achieved NRR was in good accordance with previous studies where nitrogen removals between 500 and 1000 mg L?1 day?1 were reported for Anammox reactors which operated under different circumstances. Biological treatment by autotrophic nitrogen removal is preferred for concentrated wastewater streams with ammonium concentrations in the range of 100–5000 mg N L?1 [3], so the cultivated Anammox is de?nitely employable. A molar removal ratio (RR) of ammonia and nitrite of 1.05 ± 0.08 was achieved. This is somewhat less than the theoretical 1.32 found in Anammox metabolism and is in agreement with the amount of NO? 3 —N found, which is also somewhat lower than excepted form Eq. (1). Moreover, changing stoichiometry was already noticed at higher nitrite concentrations and at modi?ed temperatures [12]. Both evolutions of NRR and RR throughout the experimental period can be seen in Fig. 3. Total nitrogen removal reached 90% while nitrate (NO? 3 —N) was formed in a range of 0.17 ± 0.04 for each consumed NH? 4 —N (Fig. 4). In Anammox metabolism, for each mole of consumed ammonia, theoretically 0.26 mol of nitrate is formed (Eq. (1)). According to our results, less nitrate was formed. One possible explanation is the occurrence of denitri?cation with carbon from the lysis of the biomass. Another reason for this limited nitrate formation is given in the study of Lee et al. [30] where it is hypothesized that lower nitrate yields may result from non-growth associated ammonium and nitrite consumption when longer sludge retention times (SRT) are applied. The anaerobic CSTR exhibited stable and reliable Anammox performance for more than 2 years of operation. The Anammox sludge has the characteristic brown–reddish color and is distributed homogeneous over the reactor as ?ne granules, although bio?lm structures are apparently attached on the inner surface of the reactor and on the reactor elements. No pH inhibition was seen. In?uent pH was 7.8 ± 0.2 while ef?uent pH was 8.3 ± 0.1 for the Anammox CSTR. Also, no nitrite inhibition was detected, although a properly adjusted in?uent was essential.

Fig. 4. Total Nitrogen Removal (TNR) (-e-) and NO? 3 —N production per consumed NH? 4 —N (-N-) for the anaerobic CSTR A.

As mentioned, the reactor system contained an air lift system to pump sedimented sludge from the decanter to the reactor. As the Anammox bacteria are anaerobic this system may have in?uenced the outgrow of this bacteria, especially in the ?rst part where the feeding also was not optimal. The switch to a synthetic wastewater resulted in a simultaneous removal of ammonia and nitrite due to Anammox. At that moment, no new sludge was inoculated to the reactor and because sludge washout was minimized since the inoculation of the reactor with OLAND sludge by the sludge recycle we therefore concluded that the Anammox bacteria remained hidden or in some kind of a ‘‘dormant phase’’ in the reactor during the unstable feeding period. In conclusion, Anammox cultivation could be performed in an anaerobic continuous stirred tank reactor (CSTR) at an elevated temperature of 37 ± 2 °C, a hydraulic retention time (HRT) of 1.5 days, a maximum sludge retention time (SRT) (no sludge washout) and a strict feeding regime. A small disturbance in the in?uent resulted in an Anammox inhibition, however this inactivation was completely eliminated when a stable in?uent was applied. The stable in?uent consisted of a synthetic wastewater with a concentra? tion of NH? 4 and NO2 in a molar ratio of about 1:1.32; the nitrogen concentration was gradually increased according to the results. 3.2. Reactor B (conventional activated sludge from manure treatment plant) The second reactor, an anaerobic UFBR, was started up with nitrifying and denitrifying sludge from a manure processing plant. Fig. 5 shows the nitrogen pro?le for the in?uent and the ef?uent concentrations of ammonia, nitrite and nitrate during the whole experiment. The experimental phase can be divided into 3 periods: start-up (I), up scaling (II) and Anammox cultivation (III). Start-up (Period I) was performed in a 3.7 L reactor ?lled with polyurethane sponges as carrier material. The same synthetic feeding media as for Reactor A was used. Initial nitrogen concentrations ?1 ?1 were 50 mg L?1 NH? NO? 2 —N and 140 mg L 4 —N, 50 mg L NO? —N. Nitrate was added to prevent anaerobic sulfate reduction 3 that results in the release of sul?de which is toxic to Anammox bacteria [2]. Initially, denitri?cation of nitrite and nitrate occurred. Degradation of the aerobic biomass fraction resulted in a slight increase of ammonia. Denitri?cation declined when nitrate was lowered until zero and so nitri?cation was the only process established. Anammox activity was con?rmed within 3 months of operation by the simultaneous removal of ammonia and nitrite and the production of nitrate. The biomass characteristics also changed from dark brown dense sludge to brown ?ocs distributed

? Fig. 3. Nitrogen Removal Rate (NRR) (-d-) and NO? 2 —N=NH4 —N Removal Ratio (RR) (-e-) for the anaerobic CSTR A.

134

A. Monballiu et al. / Separation and Puri?cation Technology 104 (2013) 130–137

200 180 160 140 120 100 80 60 40 20 0 200 180 160 140 120 100 80 60 40 20 0 200 180 160 140 120 100 80 60 40 20 0

Period I

Period II

Period III

Concentration NO2--N (mg L-1)

Concentration NH4+-N (mg L-1)

? Fig. 6. Nitrogen Removal Rate (NRR) (-?-) and NO? 2 —N=NH4 —N Removal Ratio (RR) (-s-) for the anaerobic UFBR B.

Time (days)

Fig. 5. In?uent (-?-) and ef?uent (-h-) concentrations for nitrogen (ammonia, nitrite and nitrate) in the anaerobic UFBR B started with conventional activated sludge; Period I = start-up with synthetic water in 3.7 L reactor, Period II = up scaling to 16 L and feeding with wastewater based on the thin fraction of pig manure; Period III = feeding with synthetic water.

and woven in the sponge. At the end of this period, nitrogen removal rate (NRR) reached 150 mg N L?1 day?1 with an acceptable ? removal ratio (RR) of NO? 2 —N/NH4 —N near 1.2, as shown in Fig. 6. After 250 days of operation, the sponges with biomass were transferred to a similar 16 L UFBR and the reactor was ?lled with more carrier material. In Fig. 5, this is denoted as Period II. During this period, the in?uent of the reactor was prepared by diluting thin fraction of pig manure to a concentration in the range of 50– 80 mg L?1 NH? 4 —N while nitrite was manually added to the same concentration. The data show a decrease in ammonia removal and a total removal of nitrite. Nitrate was not formed during this period. By feeding the pig manure wastewater, organic carbon was introduced into the reactor which resulted in a competition of denitri?ers against the Anammox bacteria. The COD levels were around 500 mg O2 L?1 whereas the synthetic in?uent did not contain any COD. At a certain point, (day 280–300) ammonia removal decreased to zero. Apparently heterotrophic denitri?ers had overgrown the Anammox bacteria. However after day 300 ammonium removal slowly restarted. Since no aerobic activity was present and nitrite was available this ammonium removal could be the result of Anammox activity. The nitrite formed by denitri?cation of the nitrate formed by the Anammox bacteria probably also could serve as substrate for the latter. However, because of the turbid feed, the sludge became almost invisible since the sponge operated as a kind of ?lter for the suspended solids in the pig manure wastewater.

In Period III, the pig manure wastewater was replaced by synthetic water because of the negative effect of the former on the Anammox activity. Nitrate was not fed to the reactor and concentrations of ammonia and nitrite were gradually increased depending on the removal results. Within 2 months, the Anammox activity was restored in the reactor. This is con?rmed by the removal of both ammonia and nitrite, nitrate production and visible as a brown–reddish sludge culture. At the end of the experimental period, NRR reached 250 mg N L?1 day?1 with a stable nitrate production ? (NO? 3 —N per consumed NH4 —N) in the range of 10–15% (Fig. 7). An established Anammox culture could thus be cultivated from conventional activated sludge in an UFBR with a polyurethane sponge as carrier material. Although the relative low NRR, there can be concluded that Anammox cultivation from conventional sludge was possible under controlled strategy of feed-regime and biomass retention within 3 months. By providing the reactor with the sponge, two bottlenecks for optimal Anammox growth were tackled: (1) minimization of sludge wash-out (maximum sludge retention) and (2) creation of anaerobe zones because of bio?lm formation (avoidance of oxygen inhibition). The sludge is retained in the sponges and up scaling is perfectly possible by transferring the sponges in their entirety to a larger reactor and ?lling it up with additional sponges. A third bottleneck includes the feeding of the reactor with in?uent containing ammonia and nitrite. The latter is also a growth limiting factor. Nitrite inhibition was not detected in the reactor, but as already mentioned, a well-adjusted in?uent must be available.

Concentration NO3--N (mg L-1)

Fig. 7. Total Nitrogen Removal (TNR) (-s-) and NO? 3 —N production per consumed NH? 4 —N (-N-) for the anaerobic UFBR B.

A. Monballiu et al. / Separation and Puri?cation Technology 104 (2013) 130–137

135

An increase of pH was detected; in?uent pH was 7.7 ± 0.2 while an ef?uent pH of 8.3 ± 0.1 was measured, the latter seems not to inhibit the process. Finally, the organic load of the wastewater should be taken into account in order to prevent that denitri?cation suppresses the Anammox process. As the Anammox process releases small amounts of nitrate, conventional denitri?cation must be incorporated in the overall process if one wants to remove all nitrogen. In the presence of reasonable amounts of organic carbon denitrifying bacteria probably can coexist with the Anammox bacteria. During the feeding periods with synthetic water, it was however observed that the NO? 3 —N level formed was lower than excepted. As already mentioned before, the reduced nitrate formation may be due to the large SRT that reduces net cell growth [30]. This assumption is further strengthened by the fact that the synthetic wastewater contained insuf?cient organic matter for denitri?cation. Since NO? 3 formation is the result of the autotrophic growth of the Anammox ? bacteria, the relative lack of NO3 formation in combination with ? NH? 4 and NO2 consumption thus indicates that growth of Anammox bacteria and Anammox activity probably can be uncoupled. When one thus wants to accumulate Anammox bacteria we not only have to avoid small SRTs in order to prevent washing out of the Anammox bacteria, we thus also have to avoid large SRTs to prevent growth uncoupled metabolisms and so an optimal SRT exists. When the anaerobic UFBR was fed with pig manure wastewater, denitri?cation of nitrite prevented the simultaneous removal of ammonia by Anammox bacteria. One drawback to the use of the carrier material is that the retention time of the organic matter is increased because the sponge operated like a kind of ?lter. Because of this COD accumulation, the denitri?ers probably were favored. It is also worth mentioning that the pig manure wastewater was not pretreated by the partial nitritation which is also responsible for the removal of some COD. When the reactor was switched to the treatment of the pig manure, the NRR amounted only 150 mg L?1 day?1. The Anammox bacteria present probably could not compete with the denitri?ers. A pre?ltration step is recommended for streams highly loaded with suspended solids, such as pig manure in order to reduce organic matter in the Anammox reactor.

We assume that the ratio of in?uent COD to nitrogen will have a signi?cant in?uence on the Anammox and denitri?ers community. According to a modeling study of Veys et al. [15], the Anammox step is a stable process even if the COD/N ratio is high provided that the conditions are optimal. However some reports mentioned the negatively effect of organic matter on the Anammox process. The threshold ratio for organic carbon to nitrogen for Anammox inhibition by denitri?cation was determined as 1–3 [3,31,32]. The effect of the ratio C/N needs further investigation, also taking into account the biomass concentration, biomass structure and speci?c Anammox activity. An important point to note is the fact that the Anammox bacteria did not die nor were washed out from the reactor during the feeding with the pig manure wastewater especially in the period when virtually no Anammox activity was detected (day 280–300) since the process was restored when a suitable in?uent was processed by the UFBR. In agreement with the results of reactor A, which was operated as a CSTR, we can thus conclude for reactor B, which was operated as an UFBR, that the Anammox bacteria can stay dormant until they encounter better conditions; in reactor B this apparently was already the case during the second period of feeding manure based wastewater. Moreover, as already reported by Van Hulle et al. [14], it was revealed that the feeding regime is an important factor in the successful start-up of the Anammox process, but also in the reactivation of the process when it was subjected to unsuitable conditions. In both reactors we were confronted with the phenomenon that the Anammox bacteria could survive during periods without detectable Anammox activity when non-optimal in?uent conditions were applied. In the ?rst reactor this condition was a vary? ing molar ratio of NH? 4 to NO2 due to an unstable partial nitritation process. In the second reactor, the Anammox bacteria had to compete with denitri?ers. This also showed that the Anammox process is less sensitive than previously thought: when a suited in?uent was applied in optimal conditions (37 ± 2 °C and maximum sludge retention), Anammox activity clearly restored in both reactors.

Fig. 8. FISH images of a sludge sample from Reactor A (scale 1/1000); (a) visible biomass; (b) total DNA staining with DAPI, displayed in blue; (c) aerobic ammonium oxidizing bacteria (AerAOB) (FLUO-labeled Nso1225 and Nso190) appear green and (d) anaerobic AOB (AnAOB) (Cy3-labeled Amx820) appear red. (For interpretation of the references to color in this ?gure legend, the reader is referred to the web version of this article.)

136

A. Monballiu et al. / Separation and Puri?cation Technology 104 (2013) 130–137

3.3. FISH characterization Fluorescent in situ hybridization was applied on a sludge sample from the anaerobic reactor A at day 435 in order to characterize the biomass. Total biomass is visualized in Fig. 8a and total DNA staining was done with DAPI (Fig. 8b). No aerobic ammonium oxidizing bacteria (FLUO-labeled probes Nso1225 and Nso190) were detected, as shown in Fig. 8c. In Fig. 8d, a cluster of Anammox cells hybridized with the speci?c probe Amx820-Cy3, is visualized. Sludge from reactor B was subjected to the FISH assessment on day 249 and day 354. On both moments, no visualization of Anammox structures was made. Few AOB were distributed among the sludge on day 354 which were not yet detected on day 249. One of the dif?culties is the fact that the sludge is attached to the sponge as granules. The removal from the sponge resulted in the destruction of the granules and no good FISH compositions could be made. Another aspect that could give an explanation for the negative FISH test, is the low NRR for the sludge at the day of sampling for the FISH assay. On day 249, NRR was about 150 mg N L?1 day?1 while on day 354 almost no Anammox activity was detected. This could mean that the percentage of Anammox bacteria in the sludge was too low to detect. 4. Conclusions Nowadays, the traditional nitri?cation–denitri?cation is no longer bene?cial from an economic point of view, especially for high ammonium loaded streams in absence of organic carbon. The autotrophic nitrogen removal process is recommended for such streams since this process demands less oxygen and no organic carbon. However, full scale operations are delayed because of the low growth rate of Anammox bacteria. On the basis of the cultivation of Anammox bacteria there is a choice of a suitable reactor con?guration together with a controlled feed-regime. This study describes the successful enrichment of Anammox bacteria from OLAND sludge in an anaerobic continuous stirred tank reactor (CSTR) at lab scale. A nitrogen removal rate up to 700 mg N L?1 day?1 was obtained and ?uorescent in situ hybridization (FISH) results revealed that Anammox bacteria were distributed in the CSTR. In a second lab scale reactor, an anaerobic up?ow ?xed bed reactor (UFBR), within 3 months of inoculation with sludge from a manure treatment plant, a nitrogen removal rate up to 250 mg N L?1 day?1 was achieved due to Anammox activity. The implementation of polyurethane sponge as carrier material enhanced sludge retention, however it hampered the FISH test. The results from both reactors indicate that Anammox activity can stay inactive for longer times, that these bacteria will reactivate as soon as the environmental conditions are suited for their growth, and that growth of Anammox bacteria and Anammox activity apparently can be uncoupled. The results further show that the Anammox process can be a good alternative – in combination with partial nitritation – to treat high ammonium loaded streams with low carbon content. In the presence of organic carbon, denitrifying bacteria can coexist with the Anammox bacteria to remove the nitrate formed during the Anammox process. It must however be avoided that this heterotrophic denitri?ers crowd out the Anammox bacteria. Acknowledgements This research was funded by the Institute for the Promotion of Innovation by Science and Technology in Flanders (Agentschap voor Innovatie door Wetenschap & Technologie) (IWT Project 80126). The participating companies are gratefully acknowledged.

Helge Vandeweyer and Wim Audenaert are thanked for the discussion on the topic of nitrogen removal and wastewater treatment in general.

References
[1] D. Paredes, P. Kuschk, T.S.A. Mbwette, F. Stange, R.A. Muller, H. Koser, New aspects of microbial nitrogen transformations in the context of wastewater treatment – a review, Engineering in Life Sciences 7 (2007) 13–25. [2] L.G.J.M. Van Dongen, S.M. Jetten, M.C.M. van Loosdrecht, C.M. Van Loosdrecht, The Combined Sharon/Anammox Process: A Sustainable Method for NRemoval from Sludge Water, IWA Publishing, 2001. [3] S.W.H. Van Hulle, H.J.P. Vandeweyer, B.D. Meesschaert, P.A. Vanrolleghem, P. Dejans, A. Dumoulin, Engineering aspects and practical application of autotrophic nitrogen removal from nitrogen rich streams, Chemical Engineering Journal 162 (2010) 1–20. [4] A. Mulder, A.A. van de Graaf, L.A. Robertson, J.G. Kuenen, Anaerobic ammonium oxidation discovered in a denitrifying ?uidized-bed reactor, Fems Microbiology Ecology 16 (1995) 177–183. [5] A. van de Graaf, P. de Bruijn, L. Robertson, M. Jetten, J. Kuenen, Metabolic pathway of anaerobic ammonium oxidation on the basis of N-15 studies in a ?uidized bed reactor, Microbiology-Uk 143 (1997) 2415–2421. [6] C. Trigo, J.L. Campos, J.M. Garrido, R. Mendez, Start-up of the Anammox process in a membrane bioreactor, Journal of Biotechnology 126 (2006) 475–487. [7] E. Volcke, M. van Loosdrecht, P. Vanrolleghem, Controlling the nitrite: ammonium ratio in a SHARON reactor in view of its coupling with an Anammox process, Water Science and Technology 53 (2006) 45–54. [8] M.J. Kampschreur, R. Poldermans, R. Kleerebezem, W.R.L. van der Star, R. Haarhuis, W.R. Abma, M.S.M. Jetten, M.C.M. van Loosdrecht, Emission of nitrous oxide and nitric oxide from a full-scale single-stage nitritation– Anammox reactor, Water Science and Technology 60 (2009) 3211–3217. [9] M. Strous, E. van Gerven, J.G. Kuenen, M. Jetten, Effects of aerobic and microaerobic conditions on anaerobic ammonium-oxidizing (Anammox) sludge, Applied and Environmental Microbiology 63 (1997) 2446–2448. [10] L.P. Kuai, W. Verstraete, Ammonium removal by the oxygen-limited autotrophic nitri?cation–denitri?cation system, Applied and Environmental Microbiology 64 (1998) 4500–4506. [11] A.O. Sliekers, N. Derwort, J.L. Campos-Gomez, M. Strous, J.G. Kuenen, M.S.M. Jetten, Completely autotrophic nitrogen removal over nitrite in one single reactor, Water Research 36 (2002) 2475–2482. [12] M. Strous, J.G. Kuenen, M.S.M. Jetten, Key physiology of anaerobic ammonium oxidation, Applied and Environmental Microbiology 65 (1999) 3248–3250. [13] W.R.L. van der Star, W.R. Abma, D. Blommers, J.W. Mulder, T. Tokutomi, M. Strous, C. Picioreanu, M.C.M. Van Loosdrecht, Startup of reactors for anoxic ammonium oxidation: experiences from the ?rst full-scale Anammox reactor in rotterdam, Water Research 41 (2007) 4149–4163. [14] S. Van Hulle, H. Vandeweyer, W. Audenaert, A. Monballiu, B. Meesschaert, In?uence of the feeding regime on the start-up and operation of the autotrophic nitrogen removal process, Water Sa 37 (2011) 289–294. [15] P. Veys, H. Vandeweyer, W. Audenaert, A. Monballiu, P. Dejans, E. Jooken, A. Dumoulin, B. Meesschaert, S. Van Hulle, Performance analysis and optimization of autotrophic nitrogen removal in different reactor con?gurations: a modelling study, Environmental Technology 31 (2010) 1311–1324. [16] A.A. van de Graaf, P. de Bruijn, L.A. Robertson, M.S.M. Jetten, J.G. Kuenen, Autotrophic growth of anaerobic ammonium-oxidizing micro-organisms in a ?uidized bed reactor, Microbiology-Uk 142 (1996) 2187–2196. [17] M. Strous, J.J. Heijnen, J.G. Kuenen, M.S.M. Jetten, The sequencing batch reactor as a powerful tool for the study of slowly growing anaerobic ammoniumoxidizing microorganisms, Applied Microbiology and Biotechnology 50 (1998) 589–596. [18] R. Jin, P. Zheng, Q. Mahmood, L. Zhang, Performance of a nitrifying airlift reactor using granular sludge, Separation and Puri?cation Technology 63 (2008) 670–675. [19] S. Vlaeminck, L. Cloetens, M. Carballa, N. Boon, W. Verstraete, Granular biomass capable of partial nitritation and Anammox, Water Science and Technology 58 (2008) 1113–1120. [20] H. De Clippeleir, S. Vlaeminck, M. Carballa, W. Verstraete, A low volumetric exchange ratio allows high autotrophic nitrogen removal in a sequencing batch reactor, Bioresource Technology 100 (2009) 5010–5015. [21] C. Tang, P. Zheng, Q. Mahmood, The shear force amendments on the slugging behavior of up?ow Anammox granular sludge bed reactor, Separation and Puri?cation Technology 69 (2009) 262–268. [22] K. Isaka, Y. Date, T. Sumino, S. Yoshie, S. Tsuneda, Growth characteristic of anaerobic ammonium-oxidizing bacteria in an anaerobic biological ?ltrated reactor, Applied Microbiology and Biotechnology 70 (2006) 47–52. [23] C.J. Tang, P. Zheng, Q. Mahmood, J.W. Chen, Start-up and inhibition analysis of the Anammox process seeded with anaerobic granular sludge, Journal of Industrial Microbiology and Biotechnology 36 (2009) 1093–1100. [24] J.Y. Jung, S.H. Kang, Y.C. Chung, D.H. Ahn, Factors affecting the activity of Anammox bacteria during start up in the continuous culture reactor, Water Science and Technology 55 (2007) 10.

A. Monballiu et al. / Separation and Puri?cation Technology 104 (2013) 130–137 [25] W. van der Star, A. Miclea, U. van Dongen, G. Muyzer, C. Picioreanu, M. van Loosdrecht, The membrane bioreactor: a novel tool to grow Anammox bacteria as free cells, Biotechnology and Bioengineering 101 (2008) 286–294. [26] T. Wang, H.M. Zhang, F.L. Yang, S.T. Liu, Z.M. Fu, H.H. Chen, Start-up of the Anammox process from the conventional activated sludge in a membrane bioreactor, Bioresource Technology 100 (2009) 2501–2506. [27] K. Pynaert, R. Sprengers, J. Laenen, W. Verstraete, Oxygen-limited nitri?cation and denitri?cation in a lab-scale rotating biological contactor, Environmental Technology 23 (2002) 353–362. [28] R.I. Amann, L. Krumholz, D.A. Stahl, Fluorescent–oligonucleotide probing of whole cells for determinative, phylogenetic, and environmental studies in microbiology, Journal of Bacteriology 172 (1990) 762–770. [29] S.E. Vlaeminck, A. Terada, B.F. Smets, H. De Clippeleir, T. Schaubroeck, S. Bolca, L. Demeestere, J. Mast, N. Boon, M. Carballa, W. Verstraete, Aggregate size and

137

architecture determine microbial activity balance for one-stage partial nitritation and Anammox, Applied and Environmental Microbiology 76 (2010) 900–909. [30] P.-H. Lee, W. Kwak, J. Bae, P.L. McCarty, The effect of SRT on nitrate formation during autotrophic nitrogen removal of anaerobically treated wastewater, in: IWA World Water Congress & Exhibition, Busan Korea, 2012 (IWA-8473R1). [31] B. Molinuevo, M. Garcia, D. Karakashev, I. Angelidaki, Anammox for ammonia removal from pig manure ef?uents: effect of organic matter content on process performance, Bioresource Technology 100 (2009) 2171–2175. [32] S. Ni, J. Ni, D. Hu, S. Sung, Effect of organic matter on the performance of granular Anammox process, Bioresource Technology 110 (2012) 701–705.


赞助商链接
相关文章:
更多相关标签: