ARTICLE IN PRESS

0043-1354/$ - se

doi:10.1016/j.w

�CorrespondCollege of Sci

Kanda-Surugad

3259-0672; fax:

E-mail addr

Water Research 38 (2004) 3760–3768

www.elsevier.com/locate/watres

Effect of nitrite on phosphate uptake by phosphateaccumulating organisms

T. Saitoa,c,�, D. Brdjanovicb, M.C.M. van Loosdrechtc

aDepartment of Civil Engineering, College of Science and Technology, Nihon University, 1-8 Kanda-Surugadai,

Chiyoda-ku, Tokyo, JapanbDepartment of Municipal Infrastructure, UNESCO-IHE Institute for Water Education, Westvest, 7, PO Box 3015,

2601 DA Delft, The NetherlandscDepartment of Biochemical Engineering, Delft University of Technology, Julianalaan 67, 2628 BC Delft, The Netherlands

Received 11 November 2003; received in revised form 12 May 2004; accepted 28 May 2004

Abstract

In biological nitrogen removal processes, nitrite can be formed and accumulated through both nitrification and

denitrification. Despite the fact that, in practice, biological phosphate removal (BPR) is often combined with biological

nitrogen removal, there are only a few publications reporting the effect of nitrite on BPR. In this study, phosphate-

accumulating organisms (PAO) were cultivated in an anaerobic–anoxic–aerobic sequencing batch reactor (SBR). The

effect of nitrite on the enrichment of the sludge with PAO, the phosphate uptake rates and the sludge respiration was

investigated. The results indicate that (1) presence of nitrite inhibits both aerobic and anoxic (denitrifying) phosphate

uptake, (2) aerobic phosphate uptake was more affected than anoxic phosphate uptake, (3) presence of nitrite could be

one of the factors enhancing the presence of glycogen accumulating organisms (GAO)—competitors to PAO for

substrate in the anaerobic phase, and (4) it is required to monitor and control nitrite accumulation in a full-scale

wastewater treatment plants.

r 2004 Elsevier Ltd. All rights reserved.

Keywords: Biological phosphate removal; Phosphate accumulating organisms; Nitrite; Inhibition; Denitrification; Glycogen

accumulating organisms

1. Introduction

Biological phosphate removal (BPR) has been intro-

duced into municipal wastewater treatment to control

eutrophication of surface waters and is presently a well-

established process in practice. In BPR processes,

e front matter r 2004 Elsevier Ltd. All rights reserve

atres.2004.05.023

ing author. Department of Civil Engineering,

ence and Technology, Nihon University, 1-8

ai, Chiyoda-ku, Tokyo, Japan. Tel.: +81-3-

+81-3-3293-3319.

ess: saito@civil.cst.nihon-u.ac.jp (T. Saito).

phosphate-accumulating organisms (PAO) are respon-

sible for this complex process. PAO have as character-

istic the potential to use intracellular polyphosphate as

energy reserve in the absence of electron acceptors such

as nitrate or oxygen. In an anaerobic compartment, this

can be used to accumulate substrate (preferential volatile

fatty acids) inside the cell as polyhydroxyalkanoates

(PHA) (Sato et al., 1992; Smolders et al., 1994). The

consumption of polyphosphate leads to release of

phosphate into the liquid media. Under aerobic or

anoxic conditions, PHA is used for growth and

phosphate uptake, leading to a net phosphate removal

d.

ARTICLE IN PRESST. Saito et al. / Water Research 38 (2004) 3760–3768 3761

from the wastewater. By wasting phosphate-rich sludge,

phosphate is removed from the water line and could

eventually be further efficiently recovered by chemical

coagulation (van Loosdrecht et al., 1997). Nowadays,

full BPR or combination of BPR and chemical

phosphate removal replace pure chemical processes

more and more, mainly because of the absence of

chemicals addition and the lower sludge production.

However, performance of BPR can become unstable,

especially when it is applied in combination with

biological nitrogen removal process. This instability of

BPR has been explained by several mechanisms, namely:

(a) competition with glycogen accumulating organisms

(GAO) (Liu et al., 1997), (b) introduction of nitrate into

anaerobic phase (Hascoet and Florentz, 1985), (c)

excessive aeration (Brdjanovic et al., 1998). So far, there

is sufficient evidence to believe that these mechanisms

can be responsible for lowering the phosphate removal

capacity at full-scale plants. Still unsolved dilemmas

regarding BPR exist, such as the factors favoring

presence of GAO over PAO.

Recently, the existence of PAO that can utilize nitrate

as an electron acceptor instead of oxygen has been

confirmed experimentally and practically (Kerrn-Jesper-

sen and Henze, 1993; Kuba et al., 1993). The merits of

the anoxic phosphate removal is that both the amount

of COD and oxygen required for nutrients removal can

be significantly reduced, since stored PHA is used

simultaneously for denitrification and phosphate uptake

(Kuba et al., 1996b). Several processes have been

developed making specific use of these organisms (Kuba

et al., 1996b; Bortone et al., 1996; van Loosdrecht et al.,

1998; Hu et al., 2001). Nitrite is a normal intermediate in

the denitrification process and is therefore not a strange

electron acceptor for PAO that can denitrify. The

possibility of utilizing nitrite as an electron acceptor

has been recently examined (Ahn et al., 2001). However,

if long term studies with nitrite fed systems have, as yet

not been reported in the literature, feasibility of nitrite

utilization is still unclear. In fact, high concentration of

nitrite in the mixed liquor is reported to inhibit anoxic

phosphate uptake (Kuba et al., 1996a; Meinhold et al.,

1999). While there are a few reports investigating the

effect of nitrite on anoxic phosphate uptake (Meinhold

et al., 1999; Ahn et al., 2001), no reports on the effect on

aerobic phosphate uptake is available. In general, nitrite

exhibits strong toxicity against bacteria on their growth

and respiration process (Rowe et al., 1979; Yarbrough et

al., 1980) so that elevated nitrite concentrations are

expected to negatively affect BPR process. In order to

establish more stable performance of BPR process, the

effect of nitrite should be adequately evaluated. This

research was, therefore, conducted with the aim to

evaluate the effect of nitrite on phosphate uptake under

aerobic and anoxic conditions. Hereto an enriched

culture of PAO bacteria was used.

2. Experimental setup

2.1. Cultivation of sludge enriched by PAO

An anaerobic–anoxic–aerobic sequencing batch reac-

tor (SBR) with 2L working volume and equipped with

dissolved oxygen electrode and temperature (20–251C)

and pH control (7.470.1), was used to cultivate sludge

enriched with PAO. The cycle was a 6-hour cycle

consisting of 2 h of anaerobic phase, 2.5 h of anoxic

phase, 0.5 h of aerobic phase and 1 h of settling and

discharging phase. During the initial 3min of the first

(anaerobic) phase, 1L of influent solution containing

acetate and phosphate (the composition is listed below)

was fed into the reactor. After 2 h of anaerobic phase,

50ml of a concentrated nitrate solution (1,875mgN/L)

was introduced during the first 90min of the anoxic

phase. The influent COD/NO3–N ratio (the ratio of the

amount of COD added in the anaerobic phase to the

amount of nitrate added in the anoxic phase) and COD/

PO4–P ratio (the ratio of the amount of COD to the

amount of phosphate in the anaerobic feed solution)

were 4.3 (gCOD/gN) and 26.7 (gCOD/gP), respectively.

Anaerobic and anoxic conditions were maintained by

slowly sparging nitrogen gas through the system. After

2.5 h of anoxic phase, aeration was provided for 30min.

Aeration was sufficient to keep DO concentration above

4.0mgO2/L throughout the aerobic phase. At the end of

the 6-hour cycle, 1.05L of supernatant was discharged

(resulting in HRT of 12 h). Sludge retention time was

controlled at average 12.5 days by sludge wasting at the

end of aerobic phase. The system was inoculated with

fresh activated sludge from a nitrogen and phosphate

removing full-scale plant in the Netherlands. The SBR

was continuously operated for more than 90 days. In

order to study in a controlled way, 8mg/l of allylthiour-

ea was regularly added to suppress nitrifying activity.

The influent solution consisted of NaAc (400mgCOD/

L), K2HPO4 (49mg/L), KH2PO4 (28mg/L), NH4Cl

(107mg/L), MgSO4 � 7H2O (180mg/L), CaCl2 � 2H2O

(14mg/L), and trace elements 0.3mL/L. Composition

of trace elements: EDTA (10 g/L), FeSO4 � 7H2O (1.54 g/

L), H3BO3 (150mg/L), CuSO4 � 5H2O (30mg/L),

MnCl2 � 4H2O (120mg/L), KI (180mg/L), Na2-MoO4 � 2H2O (60mg/L), ZnSO4 � 7H2O (120mg/L),

CoCl2 � 6H2O (150mg/L).

2.2. Batch experiments for inhibition of BPR by nitrite

2.2.1. Inhibition of phosphate uptake

Phosphate uptake tests were conducted with various

nitrite concentrations under three different electron

acceptor conditions (oxygen, nitrite and mixture of

nitrite and nitrate) to examine the effect of nitrite on

both anoxic and aerobic phosphate uptake. Sludge

sample was taken from SBR at the end of anaerobic

ARTICLE IN PRESS

Fig. 1. Cycle behaviour of soluble compounds at day 21 in SBR

operation.

T. Saito et al. / Water Research 38 (2004) 3760–37683762

phase and washed by phosphate-free mineral solution.

Tris-buffer solution was used to control pH around 7.0.

For aerobic test, aeration was provided throughout the

experiments. Initial concentration of phosphate and

nitrite were 45mgP/L and 0–12mgN/L, respectively.

For anoxic test, nitrogen gas was provided throughout

the experiment. Initial concentrations of phosphate and

nitrite were 9.0mgP/L and 0–12mgN/L, respectively.

For the anoxic test using mixture of nitrate and nitrite,

initial phosphate, nitrate and nitrite concentration were

9.0mgP/L, 2.6 and 0–12mgN/L, respectively.

2.2.2. Inhibition of oxygen respiration

Off-line respiration measurements in a biological

oxygen monitor (BOM) were done to examine the effect

of nitrite on oxygen respiration. Sludge was taken from

SBR at the end of anaerobic phase. Oxygen utilization

rate was measured under various nitrite concentrations

(0–7mgN/L).

2.3. Analytical procedures

The performance of SBR system was monitored on a

daily basis by sampling at the end of each SBR cycle

phase. On several occasions, a more detailed, so-called,

SBR cycle measurements was performed. An extensive

sampling program was designed for each of the batch

experiments performed. Phosphate, COD (HAc), nitrate

and nitrite were analyzed by Dr. Lange measurement

kits. MLSS and MLVSS were measured according to

Dutch standard method.

Fluorescence in situ hybridization (FISH) analysis

was conducted using modified method of Amann (1995).

Sludge was taken from SBR at the end of the aerobic

phase and was fixed by 4% paraformaldehyde solution

in phosphate buffer. Hybridization was performed with

35% formamide solution at 46 1C for 1.5 h. PAOMIX

(Crocetti et al., 2000) labeled by Fluos, GAOMIX

(Crocetti et al., 2002) labeled by Cy-3 and EUB338

probe labeled by Cy-5 were used to detect PAO, GAO

and eubacteria, respectively.

3. Results

3.1. Cycle measurements

Typical patterns of soluble compounds are shown in

Fig. 1. As expected, the characteristic PAO activity was

observed. During the anaerobic phase, acetate (COD)

was taken up (150mg/L) and phosphate was released

(30mg/L). In anoxic and subsequent aerobic phases,

phosphate was taken up, 20 and 10mgP/L, respectively.

During the anoxic nitrate-feeding phase, nitrite accu-

mulated, although it was completely denitrified at the

end of the anoxic phase and no nitrite was introduced

into the subsequent aerobic phase. During the first 40

days of operation, the observed P/C ratio (the ratio of

the phosphate released to the acetate consumed under

anaerobic condition) in the anaerobic phase was still low

and was not more than 0.2 gP/gCOD. Net phosphate

removal was only 33% of the influent concentration.

After 40 days of operation, nitrite accumulated only in

small quantities. No nitrification was observed through-

out the operation, because of regular addition of

allylthiourea.

3.2. Sludge acclimatization

The accumulation of PAO in the system was relatively

slow compared to previous experiments in our labora-

tory (Smolders et al., 1995; Kuba et al., 1993). This is

illustrated by the development of phosphate concentra-

tion at the end of anaerobic and aerobic phases as

shown in Fig. 2. The phosphate concentration at the end

of anaerobic phase gradually increased from 30mgP/L

at day 4 to 80mgP/L at day 80 and the phosphate

concentration at the end of cycle decreased from

13mgP/L at day 4 to 4mgP/L at day 80. These results

strongly indicate that PAO gradually accumulated in the

system. However, the gradual and rather slow improve-

ment of PAO performance suggests that the activity had

been suppressed for some reason. Hascoet and Florentz,

(1985) suggested that introduction of nitrate into

anaerobic zone reduces phosphate removal activity.

Kuba et al. (1994) also reported that in the presence of

both nitrate and acetate, lower P/C ratio is observed,

because of direct use of acetate through TCA cycle.

Since in this study, neither nitrate nor nitrite was

ARTICLE IN PRESS

Experimental period (days)

Pho

spha

te (

mgP

/l)

0

20

40

60

80

100

0 20 40 60 80 100

phase

End of aerobic phase

End of anaerobic Nitriteaccident!

Fig. 2. Course of phosphate concentration at the end of

anaerobic and aerobic phase in SBR operation.

Nitr

ate

and

nitr

ite(m

gN/l)

0 20 40 60 80 100

Pho

spha

te u

ptak

e in

one

cyc

le (

mgP

/l)

Nitrite

Nitrate

Anoxic phosphate uptake

0

5

10

15

20

Experimental periods (days)

60

40

0

20

20

40

Aerobic phosphate uptake

Fig. 3. Phosphate uptake under aerobic/anoxic condition in

one cycle (upper graph) and nitrate/nitrite concentration at the

end of anoxic phase (lower graph) in SBR operation.

Initial nitrite concentration (mgN/l)

Pho

spha

te u

ptak

e ra

te(m

gP/g

VS

S.h

)

0

5

10

15

20

0 4 10 12

without nitratewith nitrate

862

Fig. 4. The effect of initial nitrite concentration on anoxic

phosphate uptake rate—the results of the batch experiments on

anoxic phosphate uptake rate.

T. Saito et al. / Water Research 38 (2004) 3760–3768 3763

entering the anaerobic phase, and HAc was not entering

the anoxic phase, these possibilities are excluded. One of

the reasons could be the competition with GAO.

Actually, FISH probes verified the inferred presence of

GAO (see below). The low P/HAc ratio also indicates

the presence of GAO (Zeng et al., 2003).

Nitrite and nitrate concentration at the end of anoxic

phase and the amount of phosphate taken up under

anoxic and aerobic conditions are shown in Fig. 3.

Interesting observation is that during the initial 40 days,

nitrite accumulated in the anoxic phase (up to 20mgN/

L), but did not accumulate afterwards. Correspondingly,

total phosphate uptake (the sum of anoxic and aerobic

phosphate uptake) after day 40 increased much faster in

comparison with the preceding period. Furthermore, a

significant increase of aerobic phosphate uptake was

observed (relative to that of anoxic). In addition, a two-

step progress was observed in terms of phosphate

release: a slow increase until 40 days and faster increase

afterwards (as documented by Fig. 2). It strongly looks

as if nitrite was limiting the accumulation of PAO in the

system. Moreover it seems that the limiting effect of

nitrite is mainly exerted during the aerobic period,

because the aerobic phosphate uptake was mainly

improved. The faster increase in anaerobic phosphate

release after day 40 indicates that not only the

phosphate uptake but most likely also the growth of

PAO was negatively affected. The effect of nitrite was

underlined by an experimental accident at day 80.

Malfunctioning of the acetate-feeding pump caused

nitrite accumulation: no acetate was added in the

anaerobic phase during 6 cycles and therefore nitrite

accumulated. As a result of the nitrite accumulation, the

phosphate removal activity drastically decreased (Figs. 2

and 3). The observed pattern of phosphate and nitrite

concentration strongly indicates that nitrite could have

severe effect on the activity of PAO. Moreover, FISH

probing showed a strong decrease in the number of PAO

after the incident (see below).

3.3. Effect of nitrite on anoxic phosphate uptake

Two types of anoxic phosphate uptake tests with

different electron acceptors were conducted. In the first

test, both nitrate and nitrite were present in the mixed

liquor to test the effect on ordinary anoxic phosphate

uptake with nitrate. In the second test, nitrate was not

present and nitrite was both the sole electron acceptor

and the suspected inhibitor for anoxic phosphate uptake

simultaneously. The initial phosphate uptake rates at

varying initial nitrite concentrations are shown in Fig. 4.

ARTICLE IN PRESS

P/N

rat

io (

mol

P/m

ole- )

Phosphate uptake rate (mgP/gVSS.h)

0.00

0.04

0.08

0.12

0.16

0.20

0 12 15

with nitratewithout nitrate

3 96

Fig. 6. Relationship between P/N ratio and phosphate uptake

rate—the results of the batch experiments on anoxic phosphate

uptake rate.

Time (min)

Pho

spha

te (

mgP

/gV

SS

)

20

16

12

8

4

00 10 20 30 40

12 mgN/l 6 mgN/l2 mgN/l0 mgN/l

Nitrite

Fig. 7. Results of aerobic phosphate uptake tests in the

presence of nitrite.

T. Saito et al. / Water Research 38 (2004) 3760–37683764

The figure illustrates a typical substrate inhibition curve

obtained in the test with nitrite only. Highest phosphate

uptake rate was observed around 3mgN/L of initial

nitrite concentration. At nitrite concentration in excess

of 3mgN/L, phosphate uptake rate gradually decreased.

At nitrite concentration below 3mgN/L, phosphate

uptake rate decreased because of electron acceptor

limitation. In the presence of nitrate, there is always

enough electron acceptor. Therefore, at low nitrite

concentrations, the maximal phosphate uptake rate is

not influenced by nitrite. At higher nitrite concentra-

tions, the inhibition effect of nitrite was not influenced

(compensated) by the presence of nitrate.

Fig. 5 shows the effect of the initial nitrite concentra-

tion on the denitrification rate in the form of nitrate-

equivalents by considering the oxidation state of nitrite

relative to nitrate. Results show that denitrification rates

in both tests were almost constant at around 8mgN/

gVSS h. The only exception is the test without nitrate at

initial nitrite concentration less than 2.5mgN/L, which

results in lower denitrification rate due to electron

acceptor limitation. No inhibition effect was observed

for denitrification up to 12mgN/L of initial nitrite

concentration. These results suggest that nitrite does not

have negative impact on the enzyme system related to

denitrification, but rather on the enzyme system related

to phosphate uptake and potentially polyphosphate

formation.

From these results, P/N ratio (the ratio of the

phosphate taken up to the denitrified nitrate-equiva-

lents), which expresses efficiency of denitrification for

phosphate uptake, were calculated. Fig. 6 shows the

relationship between the observed P/N ratio and

phosphate uptake rate. It can be clearly stated that P/

N ratio is strongly dependent on the phosphate uptake

rate. According to measurements by Kuba et al. (1996c),

the maximal P/N ratio would be around 3, if ATP

gained in denitrification was only used for phosphate

Den

itrifi

catio

n ra

te

(mgN

-nitr

ate

eq/g

VS

S.h

)

Initial nitrite concentration (mgN/l)

0

2

4

6

8

10

0 4 6 8 10 12

with nitrate

without nitrate

2

Fig. 5. The effect of initial nitrite concentration on denitrifica-

tion rate—the results of the batch experiments on anoxic

phosphate uptake rate.

uptake. The total contribution of phosphate uptake in

the energy conversions is clearly limited. It is therefore

logical that P/N ratio increases linearly with increasing

phosphate uptake rate. It is also clear that there is no

real difference between nitrite or nitrate in this case,

indicating that the other cell processes are not strongly

influenced by nitrite in the studied concentration range.

This confirms the observation that denitrification as

such is not affected by the presence of nitrite.

3.4. Effect of nitrite on aerobic phosphate uptake

From the SBR experiments, it was already concluded

that aerobic phosphate uptake might be most sensitive

to nitrite. Therefore the effect of nitrite on aerobic

phosphate uptake was investigated. Batch experiments

were conducted in aerobic conditions with various

nitrite concentrations. Parts of the results are shown in

Fig. 7. 2mgN/L of nitrite causes already a severe inhibi-

tion of aerobic phosphate uptake (2.4mgP/gVSSh), and

ARTICLE IN PRESS

0

20

40

60

80

100

0 10 15 20Initial nitrite concentration* (mgN/l)

% A

ctiv

ity Anoxic uptake

Denitrification

Oxygen respiraion

Aerobic uptake

5

Fig 9. Effect of nitrite on activity of PAO. (*Initial nitrite

concentrations of oxygen respiration tests are the calculated

value from the amount of nitrite added in BOM measurements

shown in Fig. 8.)

T. Saito et al. / Water Research 38 (2004) 3760–3768 3765

more than 6mgN/L of nitrite results in almost complete

inhibition (less than 0.6mgP/gVSSh), whereas the

maximal phosphate uptake rate (in absence of nitrite)

was 24mgP/gVSSh.

In addition, to assess the effect of nitrite on

respiration, oxygen utilization rates were measured at

various nitrite levels. Parts of the results are shown in

Fig. 8. After addition of nitrite, oxygen utilization rates

suddenly dropped. Observed oxygen utilization rates are

55, 25 and 8 gO2/m3 h in the presence 0, 2 and 6mgN/L

of nitrite, respectively. The loss of oxygen respiration

activity is lower than the loss of phosphate uptake rate.

This again shows a lower tolerability of phosphate

uptake than respiration against nitrite exposure.

Sensitiveness of metabolic activities (phosphate up-

take and respiration) to nitrite exposure was further

compared between aerobic and anoxic activities. Re-

lative activity was calculated by dividing the observed

activity with the potential activity, which was obtained

from each batch test as the highest activity with neither

inhibition nor electron acceptor limitation. Fig. 9

illustrates the sensitivity of phosphate uptake and

respiration (which is expressed as % activity) to initial

nitrite concentration. Nitrite up to 4mgN/l is not

influential for anoxic phosphate uptake, while %

activity of aerobic phosphate uptake at 1mgN/l of

nitrite is already only 20%. Even at low concentration

nitrite strongly inhibits aerobic activity of PAO. The

effect is clearly stronger than under anoxic conditions.

At 12mgN/L, aerobic phosphate uptake was almost

completely lost, while anoxic phosphate uptake was still

at 65%. Similar effect is noticeable for the respiration

activity. Oxygen utilization rates are strongly affected by

the presence of nitrite. Approximately 2 and 5mgN/L of

nitrite in the system results in 50 and 20% of respiration

activity, respectively, whereas nitrate respiration still

maintained normal at these concentrations. Potentially,

this is due to the fact that nitrite is metabolized under

0

2

4

6

8

10

0 12 15 18Time (min)

DO

(m

gO2

/l)

0 mgN/l

6 mgN/l

0 mgN/l 2 mgN/l

Nitrite

Nitriteaddition

63 9

Fig. 8. Results of BOM measurement with enriched PAO

sludge.

anoxic condition. It could mean that the concentrations

near the cell are lower than in the bulk liquid.

4. Discussion

4.1. Nitrite inhibition of phosphate uptake

From SBR operation, we observed the interesting

findings that (1) aerobic phosphate uptake and most

likely aerobic growth was severely suppressed even when

nitrite was only present during the anoxic phase, and 2)

after a nitrite accident, bio-P activity was clearly

damaged. These findings suggest that nitrite has toxicity

rather than inhibition against PAO, so that specific

concentration or dose of nitrite (mgN/gVSS) might be a

more appropriate expression than nitrite concentration

(mgN/l) to access the nitrite effect on PAO activity. In

this research, less than 0.4mgN/gVSS and around

2mgN/gVSS were observed as threshold for aerobic

and anoxic phosphate uptake, respectively. A relatively

higher tolerance of PAO’s anoxic activity against nitrite

probably comes from anoxic metabolism of nitrite

giving lower concentration near the cell.

While several researchers have reported widely

distributed concentrations of nitrite as threshold on

anoxic phosphate uptake, nitrite effect on PAO activity

was not evaluated properly. Meinhold et al. (1999) used

sludge from a BIODENIPHOs pilot plant and reported

that at concentration in excess of 8mgN/L of nitrite,

anoxic phosphate uptake was strongly inhibited. Ahn

et al. (2001) cultivated sludge containing denitrifying

fraction of PAO in an anaerobic–aerobic SBR and

reported that even 40mgN/L of nitrite did not show any

ARTICLE IN PRESS

Table 1

Proportion of PAO and GAO among general gram-positive

bacteria

(Day) 0 65 85a

PAOs (%) 20–30 30–40 20

GAOs (%) 5 50–60 60

aAfter nitrite accident on day 80.

T. Saito et al. / Water Research 38 (2004) 3760–37683766

detrimental effect on anoxic phosphate uptake. Both of

them used mixed sludge so that the nitrite level on PAO

cannot be well evaluated. Lee et al. (2001) reported that

sludge cultivated in an anaerobic–aerobic–anoxic–aero-

bic SBR, which was experiencing nitrite on daily basis,

was tolerant to 10mgN/L of nitrite. They explained

widely reported nitrite level as the results of degree of

acclimatization. However, considering that they used

mixed cultures growing on wastewater, the effect of

acclimatization cannot be clearly evaluated. Hu et al.

(2003) used enriched sludge with PAO cultivated under

anaerobic and anoxic conditions. They reported that

35mgN/l of nitrite did not show negative effect on

anoxic phosphate uptake. However, in their case, pH

was not controlled (7.0–8.3). This range of pH could

easily have chemical precipitation (Maurer et al., 1999).

So far, there is only one published paper (Kuba et al.,

1996a) using enriched PAO culture with properly

controlled pH around 7. They reported that

5–10mgN/L of nitrite strongly inhibited anoxic phos-

phate uptake. However, the value was not obtained

from batch test but from SBR operation. This paper is

the first report evaluating nitrite effect on PAO activity

by the specific batch test using enriched PAO culture

and proper pH control. The experiments clearly indicate

that especially the phosphate uptake process was

vulnerable for nitrite. The exact biochemical back-

ground is unknown and deserves attention in future

research.

4.2. Effect of nitrite on PAO and GAO competition

Inhibition of phosphate transport may not be fatal for

PAO, but obviously has severe effect on their metabo-

lism and competitiveness. If phosphate uptake is

reduced, phosphate release and PHA storage under

subsequent anaerobic condition will decrease. Lower

PHA storage will result in less phosphate uptake under

subsequent anoxic or aerobic condition. If non-poly-P

bacteria, like GAO, exist in the system and if they are

less sensitive to nitrite as compared to PAO, the

competition for organic carbon could be much more

severe for PAO. Kuba et al. (1996a) reported that

enrichment culture of denitrifying PAO was affected by

nitrite accumulation, P/C ratio dropped from 0.45 to

0.30 (gP/gCOD) and the lower P/C ratio lasted for

relatively long period. This observation suggests that

inhibition of phosphate uptake gave competitive ad-

vantage to GAO in utilizing organic carbon over PAO

(Zeng et al., 2003). In our study, low P/C ratios were

indeed observed for a long period, supporting the

suggestion of GAO in the system. Results of FISH

analysis, expressed as relative abundance to total gram-

negative bacteria (Table 1), show that not only PAO but

also GAO accumulated in the system. It is remarkable

that GAO accumulated in the system much more than

PAO, which must be potentially the result of nitrite

inhibition to PAO. The situation worsened further on

expense of PAO after the nitrite accident on day 80, after

which the PAO population decreased to 20% on day 85.

These observations clearly indicate that nitrite exposure

strongly inhibits PAO activity and enhances activity of

GAO. However, the effect of nitrite on GAO is not clear

yet, and needs further study. From the present study it

seems that the hypothesis that BPR can be negatively

influenced by the presence of elevated nitrite needs to be

taken seriously in evaluating full-scale failures of BPR.

4.3. Effect of nitrite on phosphate removal in actual

WWTPs

In the combined process of phosphate and nitrogen

removal such as A2O process, it is well-known that

phosphate removal could become unstable and less

efficient than A/O process. This is for a large part due to

the direct competition for substrate between PAO and

normal denitrifying heterotrophs, when nitrate is enter-

ing the anaerobic compartment (Hascoet and Florentz,

1985). However, it is also clear that nitrite accumulation

in the aerobic phase is highly detrimental for PAO. This

can be either due to the low oxygen concentration that

effects the nitrite oxidizers more strongly than the

ammonium oxidizers (Garrido et al., 1998). It could

also be the result of elevated temperatures because

ammonium oxidizers have a comparative higher growth

rate at temperatures above 20–251C than nitrite oxidizer

(Hunik, 1993; Hellinga et al., 1998). This would imply

that in warmer climates the risk for deterioration of

BPR due to competition of GAO is higher. It is

interesting to note that occurrence of GAO in European

plants always seem to be low, whereas from Australia

regular reports, significant GAO populations occur.

Also when BPR needs to be implemented in (warmer)

industrial wastewater treatment systems, this effect

should be evaluated.

5. Conclusions

The results of this study show that:

(1)

Nitrite strongly inhibits aerobic phosphate uptake

and respiration rate. Therefore, more attention

ARTICLE IN PRESST. Saito et al. / Water Research 38 (2004) 3760–3768 3767

should be paid at full-scale wastewater treatment

plants to nitrite accumulation from the perspective

of stability of BPR process.

(2)

Aerobic phosphate uptake is more sensitive to nitrite

in comparison to anoxic phosphate uptake. Systems

which are characterized by high contribution of

anoxic phosphate uptake to total phosphate uptake,

could have comparably more stable performance

regarding BPR, especially when it is applied at

installations in which nitrite easily accumulates in

mixed liquor.

(3)

Nitrite could be an important factor in the competi-

tion between PAO and GAO bacteria.

References

Ahn, J., Daidou, T., Tsuneda, S., Hirata, A., 2001. Metabolic

behavior of denitrifying phosphate-accumulating organisms

under nitrate and nitrite electron acceptor conditions.

J. Biosci. Bioeng. 92 (5), 442–446.

Amann, R.I., 1995. In situ identification of microorganisms by

whole cell hybridization with rRNA-targeted nucleic acid

probes. In Molecular Microbial Ecology Manual. Kluwer

Academic Publications, Dordrecht.

Bortone, G., Saltarelli, R., Alonso, V., Sorm, R., Wanner, J.,

Tilche, A., 1996. Biological anoxic phosphorus removal—

the DEPHANOX process. Water Sci. Technol. 34 (1–2),

119–128.

Brdjanovic, D., Slamet, A., van Loosdrecht, M.C.M., Hooij-

mans, C.M., Alaerts, G.J., Heijnen, J.J., 1998. Impact of

excessive aeration on biological phosphorus removal from

wastewater. Water. Res. 32 (1), 200–208.

Crocetti, G.R., Banfield, J.F., Keller, J., Bond, P.L., Blackall,

L.L., 2002. Glycogen- accumulating organisms in labora-

tory- scale and full-scale wastewater treatment processes.

Microbiol.-SGM 148, 3353–3364 Part II.

Garrido, J.M., van Benthum, W.A.J., van Loosdrecht,

M.C.M., Heijnen, J.J., 1998. Influence of dissolved

oxygen concentration on nitrite accumulation in a

biofilm airlift suspension reactor. Biotechnol. Bioeng. 53,

168–178.

Hascoet, M.C., Florentz, M., 1985. Influence of nitrates on

biological phosphorus removal from wastewater. Water SA

11 (1), 1–8.

Hellinga, C., Schellen, A.A.J.C., van Loosdrecht, M.C.M.,

Heijnen, J.J., 1998. The SHARON process: an innovative

method for nitrogen removal from ammonium-rich waste.

Water. Sci. Tech. 37 (9), 135–142.

Hu, J.Y., Ong, S.L., Ng, W.J., Lu, F., Fan, X.J., 2003. A new

method for characterizing denitrifying phosphorus removal

bacteria by using three different types of electron acceptors.

Water Res. 37, 3463–3471.

Hu, Z.R., Wentzel, M.C., Ekama, G.A., 2001. External

nitrification in biological nutrient removal activated sludge

systems. Water Sci. Technol. 43 (1), 251–260.

Hunik, J.H., 1993. Engineering aspects of nitrification with

immobilized cells. Ph.D. Thesis. Wageningen Agricultural

University.

Kerrn-Jespersen, J.P., Henze, M., 1993. Biological phosphorus

uptake under anoxic and aerobic conditions. Water. Res. 27

(4), 617–624.

Kuba, T., Smolders, G.J.F., van Loosdrecht, M.C.M., Heijnen,

J.J., 1993. Biological phosphorus removal from wastewater

by anaerobic–anoxic sequencing batch reactor. Water Sci.

Technol. 27, 241–252.

Kuba, T., Wachtmeister, A., van Loosdrecht, M.C.M.,

Heijnen, J.J., 1994. Effect of nitrate on phosphorus release

in biological phosphorus removal systems. Water Sci.

Technol. 30 (6), 263–269.

Kuba, T., van Loosdrecht, M.C.M., Heijnen, J.J., 1996a. Effect

of cyclic oxygen exposure on the activity of denitrifying

phosphorus removal bacteria. Water Sci. Technol. 34 (1–2),

33–40.

Kuba, T., van Loosdrecht, M.C.M., Heijnen, J.J., 1996b.

Phosphorus and nitrogen removal with minimal COD

requirement by integration of denitrifying dephosphatation

and nitrification in a two-sludge system. Water Res. 30 (7),

1702–1710.

Kuba, T., Murnleitner, E., van Loosdrecht, M.C.M., Heijnen,

J.J., 1996c. A metabolic model for biological phosphorus

removal by denitrifying organisms. Biotechnol. Bioeng. 52,

685–695.

Lee, D.-S., Jeon, C.-O., Park, J.-M., 2001. Biological nitrogen

removal with enhanced phosphate uptake in a sequencing

batch reactor using single sludge system. Water Res. 35 (16),

3968–3976.

Liu, W.T., Nakamura, K., Matsuo, T., Mino, T., 1997. Internal

energy-based competition between polyphosphate and

glycogen-accumulating bacteria in biological phosphorus

removal reactors—effect of P/C feeding ratio. Water Res. 31

(6), 1430–1438.

Maurer, M., Abramovich, D., Siegrist, H., Gujer, W., 1999.

Kinetics of biologically induced phosphorus precipita-

tion in waste-water treatment. Water Res. 33 (2),

484–493.

Meinhold, J., Arnold, E., Isaacs, S., 1999. Effect of nitrite on

anoxic phosphate uptake in biological phosphorus removal

activated sludge. Water Res. 33 (8), 1871–1883.

Rowe, J.J., Yarbrough, J.M., Rake, J.B., Eagon, R.G., 1979.

Nitrite inhibition of aerobic bacteria. Curr. Microbiol. 2,

51–54.

Sato, H., Mino, T., Matsuo, T., 1992. Uptake of organic

substrates and accumulation of polyhydroxyalkanoates

linked with glycolysis of intracellular carbonhydrates

under anaerobic conditions in the biological excess

phosphate removal process. Water Sci. Tech. 26 (5/6),

933–942.

Smolders, G.J.F., Van Loosdrecht, M.C.M., Heinen, J.J., 1994.

Model of the anaerobic metabolism of the biological

phosphorus removal process: stoichiometry and pH influ-

ence. Biotechnol. Bioeng. 43, 461–470.

Smolders, G.J.F., Bulstra, D.J., Jacobs, R., Van Loosdrecht,

M.C.M., Heinen, J.J., 1995. A metabolic model of

the biological phosphorus removal process: valida-

tion during start-up conditions. Biotechnol. Bioeng. 48,

234–245.

Van Loosdrecht, M.C.M., Hooijmans, C.M., Brdjanovic, D.,

Heijnen, J.J., 1997. Biological phosphate removal processes.

Appl.Microbiol.Biotechnol. 48, 289–296.

ARTICLE IN PRESST. Saito et al. / Water Research 38 (2004) 3760–37683768

Van Loosdrecht, M.C.M., Brandse, F.A., de Vries, A.C., 1998.

Upgrading of wastewater treatment processes for integrated

nutrient removal—the BCFSs process. Water Sci. Tech. 37

(9), 209–217.

Yarbrough, J.M., Rake, J.B., Eagon, R.G., 1980. Bacterial

inhibitory effects of nitrite: inhibition of active transport,

but not of group translocation, and of intracellular enzymes.

Appl. Environ. Microbiol. 39, 831–834.

Zeng, R.J., van Loosdrecht, M.C.M., Yuan, Z.G., Keller, J.,

2003. Metabolic model for glycogen-accumulating organ-

isms in anaerobic/aerobic activated sludge systems. Bio-

technol. Bioeng. 81 (1), 92–105.