Appl Microbiol Biotechnol (2002) 59:344–352 DOI 10.1007/s00253-002-1007-7
O R I G I N A L PA P E R
S.K. Toh · N.J. Ashbolt
Adaptation of anaerobic ammonium-oxidising consortium to synthetic coke-ovens wastewater
Received: 24 July 2001 / Revised: 10 March 2002 / Accepted: 22 March 2002 / Published online: 15 May 2002 © Springer-Verlag 2002
Abstract A consortium with autotrophic anaerobic ammonium oxidising (AAAO) activity was developed from municipal sludge, and its ability to remove high ammonium concentrations in a toxic wastewater such as coke ovens wastewater is presented here. The enriched AAAO consortium was acclimatised to a synthetic coke ovens wastewater to establish anaerobic ammonium oxidation (AAO) activity. Phenol was the main carbon component of the synthetic wastewater whereby it was added stepwise from 50±10 to 550±10 mg l–1 into an anammox enrichment medium. Ammonium-N removal was initially impaired; however, it gradually recovered. After 15 months of further selection and enrichment, the ammonium removal rate reached 62±2 mg NH4+-N l–1 day–1, i.e. 1.5 times the rate in the original AAAO reactor. The new consortium demonstrated higher ammonium and nitrite removal rates, even under phenol perturbation (up to 330±10 mg l–1). It is therefore concluded that the AAO activity in the consortium was resistant to high phenol and has potential for treating coke-ovens wastewater.
Introduction Coke-ovens wastewaters contain high concentrations of conventional inorganic and organic contaminants, such as ammonium, thiocyanate, cyanides, sulphites, and phenolics, as well as trace amounts of polynuclear aromatic hydrocarbons (PAHs) and heterocyclic nitrogenous compounds (Table 1). The presence of these potentially carS.K. Toh (✉) · N.J. Ashbolt School of Civil and Environmental Engineering, University of New South Wales, Sydney, Australia e-mail:
[email protected] Tel.: +65-7904853, Fax: +65-8615254 S.K. Toh Environmental Engineering Research Centre, Civil and Environmental Engineering School, Nanyang Technological University, Nanyang Ave., Singapore 640261
cinogenic and toxic compounds, including high concentration of ammonium, has a detrimental effect on the environment. Biological treatment processes plus physico-chemical pre- or post-treatments (Yi et al. 1994; Yu et al. 1997) are widely used to remove the majority of the nitrogenous compounds indigenous to coke-processing wastewater. However, due to the toxic and inhibitory characteristics of many of the organic and inorganic compounds in coke plant wastewater, coke-ovens wastewater is far more difficult to treat than municipal wastewater with biological nitrogen removal (BNR), and conventional BNR treatment often fails to fully nitrify the N-compounds in this wastewater (Blum and Speece 1991; Melcer et al. 1984). Therefore, to reduce the inhibitory and toxic effects of these compounds, as well as to prevent over-shock of ammonium and temperature for BNR, raw coke-ovens wastewater is generally diluted in an equalisation tank with cool freshwater (Bridle et al. 1980), or with side-streams from a coking plant (Melcer et al. 1984). However, the dilution method involves the handling of large amounts of liquid and sludge. In other words, it increases capital costs with the use of more and larger tanks as well as higher-capacity pumps. Steam stripping, another traditional solution to lower ammonium levels prior to the BNR processes (Luthy 1981; Luthy and Jones 1980), requires addition of lime to elevate the pH up to ~10, while efficient nitrification only occurs in a pH range of 7–8. As a result, the addiTable 1 Major components in various coke-ovens wastewaters Components (mg l–1)
Port Kemblaa
Newcastlea US EPAb
Ammonia-N Cyanides Thiocyanates Phenols Chemical oxygen demand
330 93 184 333 2,200
650 10 400 330 2,000
a Average of six plants (Nutt and Marvan 1984) b BHP coke-ovens plants in New South Wales, Australia c Data not available
5,100 69 535 1,040 Nac
345
tion of phosphoric or sulphuric acids is essential to readjust the pH of the wastewater. Other methods used to minimise ammonium concentration include alkaline breakpoint chlorination, ozonation and ion exchange using synthetic zeolite (Prasad and Singh 1991; Singh and Prasad 1997). Nonetheless, the use of caustic solutions, ozone or other chemicals during the pre-treatment stage is costly. Since the requirements for optimum reaction are different for nitrification and denitrification, elimination of nitrogen in coke-ovens wastewater is commonly undertaken in two (or more)-step biological plants (Melcer et al. 1984; Paris and Jell 1993; Zhang et al. 1997, 1998). A single sludge system is, however, preferred because of its simplicity and attractive economic benefits, with the need for intermediate clarifiers and reactors eliminated. Consequently, single sludge systems yielding more efficient removal of nitrogen have also been studied, but either an extended sludge age (Shaw 1993; Zacharias and Kayser 1995) or a sequence of preconditioning processes is mandatory (Prasad and Singh 1991; Wong-Chong and Fittipaldo 1994; Yi et al. 1994). BNR using a single sludge system, without any costly pre- and post-treatment has yet to be achieved in fullstrength coke-ovens wastewater. To fulfill this objective, the functioning microbial populations in a BNR treatment plant must be able to withstand the toxicity of fullstrength coke-ovens wastewater, which is mainly attributed to phenol, cyanides and high ammonium. Furthermore, it would be most advantageous if these microbial populations consisted of autotrophic anaerobic ammonium-oxidising (AAAO) microorganisms so that no external carbon source is required. Thus far, a number of groups have reported the existence of AAAO microorganisms, but only in nature and municipal wastewater treatment systems (Abeliovich and Vonshak 1992; Helmer and Kunst 1998; Helmer et al. 1999; Hippen et al. 1997; Kuai and Verstraete 1998; Schmidt and Bock 1997; Siegrist et al. 1998; Van de Graaf et al. 1991). These findings reported that ammonium diminished in the systems with a concurrent depletion of certain nitrogen oxides in limited organic-carbon source environments. Although the nitrogen oxides present in their systems varied, the depletions of NH4+ and nitrogen oxides were both found in molar ratios that ranged from 1:1 to 1:1.5 (Abeliovich and Vonshak 1992; Helmer et al. 1999; Schmidt and Bock 1997; Van de Graaf et al. 1991, 1996). In addition, Broda (1977) had postulated two thermodynamically possible stoichiometries for ammonium oxidation in the absence of oxygen, in which the molar ratios of ammonium to nitrogen oxides were 1:1 and 3:5. Hence the ratio of NH4+ consumed to NO2– consumed was used as an indicator for anaerobic ammonium oxidation activities in this and other studies (Egli et al. 2001; Helmer et al. 2001; Schmidt and Bock 1997, 1998; Sliekers et al. 2002). The aim of this study was to adapt an AAAO microorganism consortium enriched previously from municipal sludge (Toh et al. 2002) to synthetic coke-ovens
wastewater in order to investigate if these organisms can be applied in coke-ovens wastewater treatment plants. The enriched cultures from municipal biomass were slowly acclimatised to chemo-litho-autotrophic basic salts (CLABs) medium, with additional high phenol and ammonium, which are the two key contaminants in coke-ovens wastewater. The effect of phenol on ammonium removal is also reported. In order to avoid the outgrowth of heterotrophic denitrifiers in the AAAO consortium after phenol addition, chloramphenicol inhibition was applied in the initial stage (Toh et al. 2002). Since the enriched culture obtained in this study was supplemented with phenol, the consortium was termed anaerobic ammonium oxidizing (AAO) instead of AAAO, despite the fact that phenol might not be used for the growth of the enriched AAO bacteria.
Materials and methods Biomass seed The details of culture selection and enrichment were described elsewhere (Toh et al. 2002). A fixed-bed AAO reactor (set up as shown in Fig. 1) was seeded with 50% mesophilic coke-ovens wastewater activated sludge and 50% AAAO-enriched reactor biomass (after 4 months selection and enrichment) (Toh et al. 2002). The latter, i.e. the parent reactor, was also run in parallel as a control. Adaptation to phenol The reactor was modified from a 1 l volume conical flask. Four baffles were created inside the flask, 2.5 cm up from the bottom, to give better medium mixing. The reactor was placed on a heated stirrer to control temperature and agitation (Fig. 1). All tubing was black butyl rubber, which has low air permeability and light transmission. Peristaltic pumps (Watson Marlow, 500 series) were used to control the feed inflow rate whereas the outflow was driven by gravity. Initially, the AAO reactor was run in a feed batch mode from a starting volume of 500 ml to 1 l reactor volume using CLABs (Van de Graaf et al. 1996), with nitrite (initial concentration 5 mM) as the electron acceptor rather than nitrate (Van de Graaf et al. 1991). Later, a fixed bed continuous mode of operation was adopted and anammox enrichment medium (AEM) was used. The volumetric flow rate was manipulated (within acceptable dilution rates) during the running of the reactor to change the supply of substrates. The flow rate was changed gradually from 12 ml h–1 to 30 ml h–1, which gave dilution rates of 0.012–0.030 h–1. AEM consisted of (g/l): (NH4)2SO4 (0.33), NaNO2 (0.345), NaHCO3 (1.25), KH2PO4 (0.0272), MgSO4·7H2O (0.3), CaCl2·2H2O (0.18), trace element solution I (1 ml) and trace element solution II (1 ml) (Van de Graaf et al. 1991); however, it was separated into two media bags; one contained 2× concentration of trace element solution I, the other contained phenol but no trace element solution I. It was found that phenol reacted with Fe2+ in solution I, turning the medium to a greenish colour. Hence, Fe2+ and phenol feeds were introduced separately into the reactor by different lines. Samples (5 ml) were taken from the combined feed, reactor and effluent once every 2–3 days and spun down for nitrite-N, ammonium-N and phenol analysis. After the AAO consortium was established (~10 weeks), when a stable depletion of ammonium and nitrite was recovered back to the ratio of 1:1, phenol was added into the feed. The phenol concentration was increased from a starting concentration of 50 mg l–1 gradually up to the concentration normally present in coke-ovens wastewater: ~550 mg l–1. The volumetric flow rate was changed from 10 ml h–1 to 30 ml h–1, i.e. adjusted according to the re-
346
Results Initial acclimatisation of AAAO-enriched culture to phenol
Fig. 1 A Anaerobic ammonium oxidizing (AAO) biomass in phenol-adapted reactor, developed on the support medium and wall of the tubing. B The biomass was pinkish in colour, especially where attached on the wall of the tubing
sponses of AAO activity to phenol additions, giving retention times of 3.78 and 1.38 days, respectively. Chloramphenicol (100 mg l–1) was used at the initial stage of phenol acclimatisation, to suppress heterotrophic denitrifiers. Without chloramphenicol, nitrite was rapidly consumed without significant ammonium removal, implying that denitrifiers were active. Analysis methods An HPLC method was developed to quantify the phenol concentration in the culture and synthetic coke-ovens wastewater. The HPLC set-up consisted of an HP1050 series auto sampler and isocratic pump with an on-line degassing unit, and a thermostat column (Hypersil BDS-C18, Hewlett Packard) set at 40°C. The mobile phase applied consisted of MilliQ water:methanol:acetic acid in the ratio of 75:25:0.25. The flow rate was 1.5 ml min–1 with a column pressure of ~250 kPa. Phenol was detected as a single peak at 280 nm with a diode array detector (HP1100 series, Hewlett Packard, USA). The sample was spun down (15,000 rpm, 10 min) in a 1.5 ml Eppendorf tube and filtered through a 0.45 µm cellulose acetate membrane (Millipore, Australia) prior to HPLC injection. The detection limit of the method was 50 µM with a 10-µl injection volume. This method was also able to detect chloramphenicol. Ammonium-N was assayed colorimetrically by the direct Nesslerization method (American Public Health Association 1989), while Nitrite-N was quantified by HPLC (Toh et al. 2002). The mobile phase consisted of 14% acetonitrile and 86% MilliQ water, adjusted to pH 8.6. The mobile phase was run through an Ashipak column (HP part no. 799231C-564) at 1.5 ml min–1 with a backpressure of ~200 kPa. Indirect UV measurement was used with the detection wavelength set at 360 nm and a reference wavelength of 266 nm. Chemicals Phenol was of reagent grade and purchased from Sigma, Australia. Methanol, acetonitrile and acetic acid were of HPLC grade and purchased from Merck, Australia.
As shown in Fig. 2A, once phenol was added to the AAAO reactor, nitrite consumption instantly increased. Instead of a proportional increase as displayed by the AAAO control reactor, ammonium consumption fluctuated and decreased at various stages. The control reactor was giving a stable ratio of NH4+ consumed:NO2– consumed of 1:1–1:1.3, which falls within the range displayed by other AAAO consortia (Helmer et al. 1999, 2001; Schmidt and Bock 1997; Sliekers et al. 2002; Toh et al. 2002; Van de Graaf et al. 1991, 1996). Phenol at 50 mg l–1 clearly perturbed AAO activities, and stimulated a different microbial population. In order to investigate if the AAO activity was recoverable, phenol was removed from the feed from 13– 22 October 1997 (Fig. 2A). Consequently, nitrite consumed reduced from 24 mg day–1 l–1 to 15 mg day–1 l–1, whereas the ammonium removal rate showed an instant increase of 5±2 mg day–1 l–1. To reduce the likelihood of heterotrophic denitrifiers growing during the acclimatisation of the AAAO consortium to phenol, chloramphenicol was added to inhibit nitrite reduction via denitrification. With a 100 mg l–1 chloramphenicol supplementation (from 1–28 November 1997), the proportion of ammonium to nitrite consumption was slowly brought back to, or close to, a ratio of 1:1–1:1.3 (Fig. 2B), which was the range given by the AAAO consortium selected in previous studies (Toh 2000) and the control. Considering the report of Van de Graaf et al. (1995) that chloramphenicol reduced the ammonium removal rate by 68±10% in their anammox biomass, supplementation of chloramphenicol was withdrawn (on 28 November 1997) from the feed after the heterotrophic denitrifiers had been suppressed for 1 month. This was to avoid unnecessary chloramphenicol suppression that could hinder AAO consortium enrichment and cause other side-effects (Dendooven et al. 1994). However, this move re-established a small increase in the consumption of nitrite, an additional 5±2 mg day–1 l–1. The incremental phenol feed clearly impacted on the AAO consortium, as shown by the fluctuation in the consumption of NO2– over that of ammonium (Fig. 2B). On the other hand, the AAO biomass was able to adapt slowly to phenol, even after discontinuation of chloramphenicol inhibition. After 3–4 months acclimatisation of the AAO culture to phenol at 50±10 mg l–1, phenol concentration in the feed was doubled to 100±10 mg l–1 (29 December 1997). This increment of phenol concentration again caused an initial drop in ammonium removal rate and thus affected the amount of nitrite consumed over ammonium consumed. Nevertheless, after being fed with CLABs plus 100±10 mg l–1 phenol for 3 months, the AAO consortium managed to readjust the consumption of NO2– over the consumption of NH4+
347 Fig. 2A, B Initial phenol acclimatisation, from 30 September 1997 to 30 March 1998. Chloramphenicol at 100 mg l–1 was added on 1 November 1997 and withdrawn on 28 November 1997 (shown in the graphs as upward and downward facing arrows). Concentrations of phenol added into reactor affected ammonium and nitrite consumption (mg day–1 reactor volume–1) (A), and also resulted in fluctuation in the consumption of nitrite over a unit of ammonium depletion (NO–:NH4+) (B)
Fig. 3 The consumption of nitrite over a unit of ammonium depletion (NO2–:NH4+) at various phenol concentrations is shown from 14 March 1998 to 29 November 1998. The arrows indicate series of phenol shock-loads
348 Fig. 4 A Trendlines for substrates consumed were drawn in each period of phenol shockload to the AAO reactor. Gradients of the nitrite-consumed trendlines decreased with subsequent high shock-load of phenol. However, ammonium consumption rate showed a constant increase. B Removal of ammonium and nitrite (mg substrate day–1 reactor volume–1) for the whole operation period (30 September 1997–30 December 1998) of the phenol-acclimated AAO reactor
back to the range of 1–1.3, showing a gradual increase in the depletion of ammonium and nitrite. Over a period of 6 months (from 30 September 1997 to 30 March 1998), the overall ammonium removal was increased from an average of 12 mg day–1 l–1 to 31 mg day–1 l–1 (Fig. 2A), i.e. an increment of 19 mg day–1 l–1. Compared to the AAAO control reactor, which demonstrated an increment of only 7.5 mg day–1 l–1 in the same interval of time (data not shown), it seemed that AAOs were either enriched at a greater rate, or that the enriched AAO consortium possessed a higher ammonium removal rate, despite the perturbation caused by phenol. Acclimatisation of AAO consortium to high phenol shock loads When the AAO consortium had been acclimatised with 100 mg l–1 phenol, different concentrations of phenol
were successively fed to investigate the tolerance of the AAO biomass towards phenol. The results are summarised in Fig. 3. Phenol concentration was increased stepwise from 100±10 to 350±10 mg l–1 and then dropped back to 250±10 mg l–1 for the first phenol shock load (14 March–1 May 1998); NH4+ consumed:NO2– consumed changed accordingly from a range of 1:1–1.5 to 1:2.5–3.6. When the ratio recovered back to 1:1.5 at 250 mg l–1 phenol (9 May 1998), phenol shock-loading was introduced gradually up to 400±10 mg l–1 (15 May 1998) and 550±10 mg l–1 (1 June 1998), then reduced to 250±10 mg l–1 again and brought down to zero. During these changes, NH4+ consumed: NO2– consumed increased to 1:1.9–2 and fell back to an average of 1:1.5 as the phenol concentration was dropping. However, there was little fluctuation in the ratio when the phenol feed was decreased to zero. A second series with the same pattern of phenol feed concentrations was repeated from 20 June to 7 October
349 Fig. 5 The enrichment of the AAO consortium in the phenolacclimatised reactor (A) and AAAO control reactor (B) (Toh et al. 2002). The increases in ammonium removal (reactor volume–1) in both systems were almost equivalent, whereas the gradient of ammonium consumption rates were 0.221 and 0.108 for the former and latter, respectively
1998, but the increment was steeper, i.e. phenol concentration in the feed was changed abruptly from 100±10 to 250±10 mg l–1 (20 July 1998), 180±10 to 400±10 mg l–1 (15 August 1998) and 100±10 to 550±10 mg l–1 (20 September 1998). During this second series of high-phenol shock-loads, the NH4+ consumed: NO2– consumed ratio was largely unaffected, rising to an average of 1:1.4–1.7. This implied that the AAO consortium had built up some immunity in the first shock exposure and AAO activity became less susceptible to high phenol concentrations. With the phenol feed was maintained at 330±10 mg l–1 for 2 months, NH4+ and NO2– were consumed in a stable ratio of 1:1–1.3. Thus the AAO consortium was confirmed to have quite a high tolerance towards phenol, up to 550 mg l–1, from which the AAO activities were still recoverable. After a couple of series of high concentration phenol shock loading (14 March–7 October 1998), the AAO consortium attained a stable condition and was able to remove 52±2 mg (NH4+-N) day–1 l–1, as shown in Fig. 4A. It was clearly seen that ammonium-N removal
gradually increased, even though it fluctuated with every increment in phenol concentration. In Fig. 4A, trendlines of the substrates consumed were drawn for two successive series of phenol shock loading (14 March–15 June 1998; 20 June–7 October 1998); the increases of substrate consumption rate, i.e. gradients of the trendlines, were calculated. The gradients for nitrite-consumption trendlines decreased in subsequent exposure to the same phenol concentration. For the phenol shock loads ranged from 200–360±10 mg l–1, the gradient of first exposure (14 March–10 May 1998) was 0.2700 and the second (20 June–15 August 1998) was 0.1608. Likewise, for the phenol shock loads ranged up to 550 mg l–1, the gradients were 0.2735 for the first exposure of 180–550±10 mg l–1 (10 May–15 June 1998), and 0.1755 for the second exposure of 100–550± 10 mg l–1 (1 September–7 October 1998), respectively. In constrast, the gradient of ammonium consumption rate was more positive, at 0.0983 for the first series of shockloadings (14 March–15 June 1998) and 0.1244 for the second (20 June–7 October 1998). From Fig. 4B, it is
350 Fig. 6 Phenol concentrations in the feed and effluent of the phenol-acclimatised AAO reactor. Chloramphenicol was added to the feed from 1–28 November 1997 to suppress denitrifiers
clearly seen that the removal of ammonium for the whole phenol-acclimatisation operation (30 March 1997–30 December 1998) was also very encouraging, reaching 62±2 mg l–1 NH4+-N, with a nitrite removal rate of 78±2 mg l–1 NO2–1-N. Compared to the control reactor, the increase of NH4+-N removal rate was greater in the phenol-acclimatised AAO reactor. Over the same period of time (1 October–30 December 1998), the amount of NH4+-N consumed in 1 day per litre volume of the AAO reactor (Fig. 5A) increased at the rate of 0.221 in the presence of 330±10 mg l–1 phenol, whereas the control reactor (Fig. 5B) was only at the rate of 0.108. This implied that after 15 months of phenol acclimatisation (30 September 1997–30 December 1998), the bacterial consortium enriched in the AAO reactor possessed quite a different ammonium-N removal efficacy from the control reactor. After 15 months of enrichment-acclimatisation of the AAO consortium to high phenol concentrations, the biomass obtained in the reactor displayed a pinkish colour like that in the control reactor. Phenol degradation Phenol was not detected in the effluent from 30 September to 7 November 97, when the AAO reactor was first fed with phenol at a concentration of 50 mg l–1 (Fig. 6). When chloramphenicol was supplemented in the feed at 100 mg l–1 for a month (from 1–28 November 1997), phenol concentration in the effluent increased to almost the feed concentration, i.e. 50 mg l–1. After the withdrawal of chloramphenicol, phenol degradation was still observed, but at a slower rate. At the hydraulic retention time of 3.2±0.5 days, complete degradation was only achieved at a phenol feed concentration below 100 mg l–1 and the efficacy declined to 40±10% if phenol feed was elevated above 200±50 mg l–1 (data not shown). At the steady state (30 September–30 December 1998), when the phenol feed concentration was main-
tained at 330 mg l–1 and above, the AAO reactor showed only less than 20% phenol degradation efficiency. Neither an increase in the feed concentration nor acclimatisation for a period of 15 months significantly enhanced phenol consumption, further supporting the view that phenol-degrading heterotrophs were not a significant component of the AAO consortium.
Discussion Several researchers have successfully enriched the AAO consortia from wastewater treatment plants (Egli et al. 2001; Toh et al. 2002; Van de Graaf et al. 1996) and proved their application in treating ammonium-rich wastewaters (Hellinga et al. 1998); nonetheless, these studies were limited to municipal wastewater. The ability of these consortia to treat industrial wastewater with high toxicity has not been investigated so far. It was therefore important to demonstrate that this consortium could actually tolerate toxic compounds that commonly exist in industrial wastewater prior to its application in a real wastewater treatment plant. This is the first report demonstrating that an AAO consortium was able to adapt to a high strength of phenol toxicity, without losing its anaerobic ammonium-removing activity. This finding will be a great contribution in treating ammonium-rich industrial wastewater, coke-ovens wastewater in particular, whereby not only could the nitrification/denitrification be accomplished in a single stage without any aeration and addition of external carbon source, the costly pre- and post-treatment could also be omitted. In the study, the results of initial acclimatisation of the AAO consortium to phenol suggested that the short pulse of phenol encouraged activities of other nitriteconsuming groups where nitrite was consumed at a greater rate but ammonium removal was not correspondingly affected. The deviation of NH4+ consumed: NO2– consumed from the range of 1:1–1.3, which was the ratio observed in anaerobic ammonium oxidation consortia
351
(Abeliovich and Vonshak 1992; Helmer et al. 1999, 2001; Schmidt and Bock 1997, 1998; Van de Graaf et al. 1991), suggested that the AAAO activity was greatly affected by the addition of phenol. However, the initial perturbation of AAO activity was recoverable if phenol was excluded from the feed. It is therefore confirmed that the short pulse of phenol did not impose permanent detrimental effects on the apparent anaerobic ammonium oxidation reaction. Based on the fact that, under anoxic conditions, heterotrophic denitrifiers could degrade phenol by using nitrite as an electron acceptor (Brackmann and Fuchs 1993; O'Connor and Young 1996; Sanford and Tiedje 1997; Van Schie and Young 1998), this initial adaptation of the AAAO consortium to phenol could be explained by the existence of denitrifiers. Therefore, chloramphenicol, which has been used and proved effective (Toh et al. 2002) in inhibiting denitrification but impacts little on the anaerobic ammonium oxidation reaction (Van de Graaf et al. 1996), was used to aid in the elimination of these undesired denitrifiers. Figures 2 and 6 clearly displayed the results of the success of chloramphenicol application. Initially, over the time period from 30 September to 7 November 1997, phenol diminished instantly to zero once it was fed into the reactor at 50 mg l–1, while nitrite consumption increased sharply. Later, when 100 mg l–1 of chloramphenicol was supplemented in the reactor (1 November 1997), phenol depletion was retarded and depletion of nitrite again corresponded to ammonium depletion, indicating that AAO activities gradually recovered when chloramphenicol was added. Chloramphenicol suppressed phenol-degrading heterotrophs, which may also have been linked to, or were, denitrifiers, and thus impeded the enrichment of heterotrophic denitrifiers, ensuring the AAO consortium was not overgrown. The increase of nitrite consumption rate was less with subsequent phenol shock loads; this also suggested that the heterotrophic denitrifiers did not engage themselves in this anaerobic N-removal reaction as the main activity, even with the increase of phenol or time. Therefore, the NH4+-N depletion observed was considered to be largely due to AAO activities and only insignificantly to the N-assimilation for biomass growth (Kuenen and Robertson 1994; Strous 2000). Although a small amount of phenol degradation was still observed after the apparent elimination of heterotrophic denitrifiers by chloramphenicol, nitrite consumption rates dropped and the NH4+ consumed: NO2– consumed ratio readjusted back to the range 1:1–1.3 at steady state, with phenol degradation maintained at an average of 20% efficacy, which was just 60±10 mg consumed out of 330 mg phenol per reactor volume. This observation could be explained by two possibilities: (1) other phenol degrading microorganisms or denitrifiers might have been present but were not enriched over the time of phenol acclimatisation (30 September 1997–30 December 1998). However, their presence did not affect the overall consumption of ammonium and nitrite as they probably
used phenol as a substrate in another catabolic pathway; or (2) phenol was consumed by the AAO microorganisms. It was reported by Keener and Arp (1994) that Nitrosomonas europaea, which was also found to be able to nitrify and denitrify simultaneously under anoxic conditions (Abeliovich and Vonshak 1992; Helmer and Kunst 1998; Schmidt and Bock 1997), was capable of degrading phenol using its ammonium monooxygenase (AMO). In fact, AMO was reported to oxidize a wide range of substrates (Hooper et al. 1997) and able to oxidize ammonium using nitrogen oxides, namely NO, NO2, N2O4 as oxidant (Hooper et al. 1997; Schmidt and Bock 1998). From the results, the AAO activity was still observed and increased after acclimatisation to high concentrations of phenol. Even though phenol was added to the AAO reactor at concentrations up to ~550 mg l–1, ammonium-N removal increased from an initial rate of 0.09 to 0.124 mg day–1 l–1 after acclimatisation (shown as the gradients in Fig. 4). Consequently, the total consumption of 62±2 mg (NH4+-N) day–1 l–1 was achieved after 15 months of operation, an increment of 57 mg day–1 l–1 from the initial consumption of 5±2 mg day–1 l–1. This affirmed that the enriched AAO consortium has a high tolerance towards phenol. Furthermore, the consortium showed higher AAO activity than the parent reactor without phenol perturbation. Therefore the AAO consortium, which is enriched under phenol addition, probably contained different bacterial members or a different AAO pathway where phenol can stimulate the AAO reaction. In summary, the AAO consortium that was selected and enriched was able to remove ammonium-N under anoxic conditions, even in the presence of phenol up to 330±10 mg l–1, which is the average concentration present in the influent of coke-ovens wastewaters treatment plants (Paris and Jell 1993; Suschka et al. 1994; Wong-Chong and Fittipaldo 1994). In addition, the AAO process was recoverable after high phenol shock loads up to 550 mg l–1. Therefore, this enriched AAO consortium, which was originally selected from an AAAO reactor (Toh 2000), certainly has significant potential in the BNR process for coke-ovens wastewater treatment plant. For further interest, the tolerances of AAO consortium towards cyanides and other toxic compounds in coke-ovens wastewaters shall also be investigated.
References Abeliovich A, Vonshak A (1992) Anaerobic metabolism of Nitrosomonas europaea. Arch Microbiol 158:267–270 American Public Health Association (1989) Standard methods for the examination of water and wastewater. American Public Health Association, Washington D.C. Blum DJW, Speece RE (1991) A database of chemical toxicity to environmental bacteria and its use in interspecies comparisons and correlations. J Water Pollut Control Fed 63:198–207 Brackmann R, Fuchs G (1993) Enzymes of anaerobic metabolism of phenolic compounds 4-hydroxybenzoyl-CoA reductase (dehydroxylating) from a denitrifying Pseudomonas species. Eur J Biochem 213:563–571
352 Bridle TR, Bedford WK, Jank BE (1980) Biological treatment of coke plant wastewaters for control of nitrogen and trace organics. Proceedings of the 53rd annual conference of water pollution control federation. June 23–27, 1980, Las Vegas, Nev. Water Sci Technol 13:667-680 Broda E (1977) Two kinds of lithotrophs missing in nature. Z Allg Mikrobiol 17:491–493 Dendooven L, Splatt P, Anderson JM (1994) The use of chloramphenicol in the study of the denitrification process: some sideeffects. Soil Biol Biochem 26:925–927 Egli K, Fanger U, Alvarez PJJ, Siegrist H, Meer JR, Zehnder AJB (2001) Enrichment and characterisation of an anammox bacterium from a rotating biological contactor treating ammoniumrich leachate. Arch Microbiol 175:198–207 Hellinga C, Schellen AAJC, Mulder JW, van Loosdrecht MCM, Heijnen JJ (1998) The SHARON process: an innovative method for nitrogen removal from ammonium-rich waste water. Water Sci Technol 37:135–142 Helmer C, Kunst S (1998) Simultaneous nitrification/denitrification in aerobic biofilm system. Water Sci Technol 37:183–188 Helmer C, Kunst S, Juretschko S, Schmid MC, Schleifer KH, Wagner M (1999) Nitrogen loss in a nitrifying biofilm system. Water Sci Technol 39:13–21 Helmer C, Tromm C, Hippen A, Rosenwinkel, KH, Seyfried CF, Kunst S (2001) Single stage biological nitrogen removal by nitritation and anaerobic ammonium oxidation in biofilm systems. Water Sci Technol 43:311–320 Hippen A, Rosenwinkel KH, Baumgarten G, Seyfried CF (1997) Aerobic deammonification: a new experience in the treatment of wastewaters. Water Sci Technol 35:111–120 Hooper AB, Vannelli T, Bergmann DJ, Arciero DM (1997) Enzymology of the oxidation of ammonia to nitrite by bacteria. Antonie van Leeuwenhoek 71:59–67 Keener WK, Arp DJ (1994) Transformations of aromatic compounds by Nitrosomonas europaea. Appl Environ Microbiol 60:1914–1920 Kuai L, Verstraete W (1998) Ammonium removal by the oxygenlimited autotrophic nitrification-denitrification system. Appl Environ Microbiol 64:4500–4506 Kuenen JG, Robertson LA (1994) Combined nitrification and denitrification processes. FEMS Microbiol Rev 15:109–117 Luthy RG (1981) Treatment of coal coking and coal gasification wastewaters. J Water Pollut Control Fed 53:325–330 Luthy RG, Jones LD (1980) Biological oxidation of coke plant effluent. J Environ Eng 106:847–851 Melcer H, Nutt S, Marvan I, Sutton P (1984) Combined treatment of coke plant wastewater and blast furnace blowdown water in a coupled biological fluidized bed system. J Water Pollut Control Fed 56:192–198 Nutt SG, Marvan IJ (1984) Biological fluidised bed treatment of coke oven wastewater and blast furnace scrubber blowdown. Canada Environmental Protection Service, Ontario O'Connor OA, Young LY (1996) Effect of six-different functional groups and their position on the bacterial metabolism of monosubstituted phenols under anaerobic conditions. Environ Sci Technol 30:1419–1428 Paris D, Jell A (1993) Theiss wastewater treatment plant for BHP coke ovens by-products at Port Kembla. Waste Disposal Water Manage 20:23–30 Prasad B, Singh G (1991). Ammonium removal from wastewater with special emphasis on coke-oven effluent. Environ Prot Eng 16(3–4):39–43 Sanford RA, Tiedje JM (1997) Chlorophenol dechlorination and subsequent degradation in denitrifying microcosms fed low concentration of nitrate. Biodegradation 7:425–434
Schmidt I, Bock E (1997) Anaerobic ammonia oxidation with nitrogen dioxide by Nitrosomonas eutropha. Arch Microbiol 167:106–111 Schmidt I, Bock E (1998) Anaerobic ammonia oxidation by cell-free extracts of Nitrosomonas eutropha. Antonie van Leeuwenhoek 73:271–278 Shaw KC (1993) Biological treatment of full-strength coke plant wastewater at Geneva Steel. Iron Steel Eng 7:29–32 Siegrist H, Reithaar S, Koch G, Lais P (1998) Nitrogen loss in a nitrifying rotating contactor treating ammonium-rich wastewater without organic carbon. Water Sci Technol 38:241– 248 Singh G, Prasad B (1997) Removal of ammonium from coke-plant wastewater by using synthetic zeolite. Water Environ Res 69(2):157–161 Sliekers AO, Derwort N, Campos Gomez JL, Strous M, Kuenen JG, Jetten MSM (2002) Completely autotrophic nitrogen removal over nitrite in one single reactor. Water Res (in press) Strous M (2000) Microbiology of anaerobic ammonium oxidation. PhD Thesis, The Technology University Delft, Delft, The Netherlands Suschka J, Morel J, Mierzwinski S, Januszek R (1994) Full scale treatment of phenolic coke coking wastewater under unsteady conditions. Water Sci Technol 29(8):69–79 Toh SK (2000) Ammonium removal in synthetic coke ovens wastewater. PhD Thesis. University of New South Wales, Sydney, Australia Toh SK, Webb RI, Ashbolt NJ (2002) Enrichment of autotrophic anaerobic ammonium-oxidising consortia from various wastewaters. Microb Ecol DOI 10.1007/s00248-001-0033-9 Van de Graaf AA, De Bruijn P, Robertson LA, Kuenen JG, Mulder A (1991) Biological oxidation of ammonium under anoxic conditions: Anammox process. ISEB:667–669 Van de Graaf AA, Mulder A, De Bruijn P, Jetten MSM, Robertson LA, Kuenen JG (1995) Anaerobic oxidation of ammonium is a biologically mediated process. Appl Environ Microbiol 61:1246–1251 Van de Graaf AA, Bruijn P, Robertson LA, Jetten MSM, Kuenen JG (1996) Autotrophy growth of anaerobic ammonium-oxidising microorganisms in a fluidized bed reactor. Microbiology 142:2187–2196 Van Schie PM, Young LY (1998) Isolation and characterization of phenol-degrading denitrifying bacteria. Appl Environ Microbiol 64:2432–2438 Wong-Chong GM, Fittipaldo JJ (1994) Retrofitting LTV coke plant wastewater treatment system to comply with pretreatment discharge limits. Iron Steel Eng 71:26–28 Yi Q, Wen Y, Zhang H (1994) Efficacy of pre-treatment methods in the activated sludge removal of refractory compounds in coke-plant wastewater. Water Res 28:701–707 Yu H, Gu G, Song L (1997) Post treatment of effluent from cokeplant wastewater treatment system in sequencing batch reactors. J Environ Eng 123(3):305–308 Zacharias B, Kayser R (1995) Treatment of a steel works effluent with a conventional single-sludge system built in cascades. In: Proceedings of the 50th Purdue University Industrial Waste Conference. Ann Arbor Press, Chelsea, Mich. pp 705– 715 Zhang M, Tay JH, Qian Y, Gu XS (1997) Comparison between anaerobic anoxic-oxic and anoxic-oxic systems for coke plant wastewater treatment. J Environ Eng 123(9):876–883 Zhang M, Tay JH, Qian Y, Gu XS (1998) Coke plant wastewater treatment by fixed biofilm system for COD and NH3-N removal. Water Res 32:519–527
