Journal of Environmental Sciences 20(2008) 933–939
Effects of COD/N ratio and DO concentration on simultaneous nitrification
and denitrification in an airlift internal circulation
MENG Qingjuan1, YANG Fenglin1,∗, LIU Lifen1, MENG Fangang2
1. Key Laboratory of Industrial Ecology and Environmental Engineering, MOE, School of Environmental and Biological Science and Technology,
Dalian University of Technology, Dalian 116024, China. E-mail: email@example.com
2. Department of Chemical Engineering, Technical University of Berlin, Berlin 10623, Germany
Received 26 September 2007; revised 14 November 2007; accepted 27 November 2007
The effects of chemical oxygen demand and nitrogen (COD/N) ratio and dissolved oxygen concentration (DO) on simultaneous
nitrification and denitrification (SND) were investigated using an airlift internal circulation membrane bioreactor (AIC-MBR) with
synthetic wastewater. The results showed that the COD efficiencies were consistently greater than 90% regardless of changes in the
COD/N ratio. At the COD/N ratio of 4.77 and 10.04, the system nitrogen removal efficiency became higher than 70%. However, the
nitrogen removal efficiency decreased to less than 50%, as the COD/N ratio shifted to 15.11. When the operating DO concentration was
maintained at 1.0 mg/L in AIC-MBR, a satisfying SND was achieved. Either low or high DO concentration could restrain SND.
Key words: simultaneous nitrification and denitrification (SND); chemical oxygen demand and nitrogen (COD/N) ratio; dissolved
oxygen (DO) concentration
In recent years, pollution and eutrophication of aquatic
environment has become worse. As the discharge standard
of nitrogen and phosphorus in the effluent has become
stricter, a membrane bioreactor (MBR) has been proposed
as an alternative for the conventional activated sludge
process (Brindle and Stephendon, 1996). The MBRs have
higher biomass concentrations and better retain slow grow-
ing microorganisms (such as nitrifiers), which enhance
biological nutrient removal (de Silva et al., 1998).
Wang et al. (2005) reported that a laboratory scale anox-
ic/aerobic MBR was capable of achieving 94% chemical
oxygen demand (COD), 91% ammonia nitrogen (NH+
and 74% total nitrogen (TN) removal, by continuous
runs under appropriate operational conditions. Yeom et al.
(1999), applied an intermittently aerated membrane biore-
actor with a submerged fiber hollow membrane to treat
household wastewater. With 8–15 h hydraulic retention
time (HRT) and long solid retention time (SRT), COD and
TN removal efficiency was 96% and 83%, respectively.
Rosenbergera et al. (2002) evaluated the performance
of a bioreactor with submerged membranes for aerobic
treatment of municipal wastewater. Although many MBR
systems could efficiently eliminate nitrogen, the elimina-
tion of nitrogen was always time-consuming, requiring
* Corresponding author. E-mail: firstname.lastname@example.org.
complex configuration, higher energy, and operation cost.
On the basis of these considerations, simultaneous ni-
trification and denitrification (SND) has gained significant
attention, because of its potential to eliminate the need for
separate tanks required in conventional treatment plants,
which consequently simplifies the plant design, saving
space and time. Pochana and Keller (1999) have investi-
gated the efficiency of nitrogen removal from wastewater
by an SND-based sequencing batch reactor (SBR). They
have observed that higher DO concentrations enhance the
nitrification rates. Simultaneously, high DO concentrations
inhibit the denitrification process, causing an accumula-
tion of nitrite and nitrate in the reactor. On the other
hand, at lower DO concentrations the nitrification process
is inhibited and the denitrification process is enhanced.
Therefore, the DO level is a factor critical to the SND
process. It must be maintained at an appropriate level in
the SND reactor. Zhao et al. (1999) has pointed out that
DO must be available for the nitrifiers, but must not exceed
a certain level at the same time for the denitrifiers and
SND. Chiu et al. (2007) has studied how the C/N ratio
controls the simultaneous nitrification and denitrification
in a sequencing batch reactor. They have proposed that
the SBR system can achieve nearly complete removal of
both organic matter and NH4+-N, with no accumulation
of intermediate NO2−-N when the initial C/N ratio is
controlled at 11.1.
Most of the SND studies have been conducted in
934MENG Qingjuan et al.Vol. 20
sequential batch reactors (Pochana et al., 1999), contin-
uous flow extended aeration plants (Collivignarelli and
Bertanza, 1999), and oxidation ditches (Rittmann and
Langeland, 1985). There are limited studies on the achiev-
ability of SND in continuously aerated MBRs. A novel
MBR, airlift internal circulation membrane bioreactor
(AIC-MBR) has been established and applied for nitrogen
removal through the SND process. Unlike the reactors
mentioned earlier, this reactor has a simple configuration
and is convenient to operate. On one hand, the config-
uration of internal circulation can use the energy that
the aeration produces, for a nitrified liquid cycle. The
energy consumption and operational cost will be reduced
significantly. On the other hand, the special configuration
generates the aeration and anoxic zone in the AIC-MBR.
It overcomes the difficulty in DO control in conventional
MBRs and provides an advantageous environment for the
The available literature shows that under suitable con-
ditions, the SND process will occur in the wastewater
treatment system. However, the conditions that lead to
efficient SND in the AIC-MBR process are not yet well
established. In this investigation, an SND in the AIC-MBR
was studied, to resolve some of the conditions under which
significant SND activities can occur.
1 Material and methods
1.1 Reactor setup and operation
Figure 1 shows the schematic diagram of the experi-
mental setup, consisting of a Plexiglas reactor in which
a membrane module is submerged in the upper part. The
AIC-MBR, with a working volume of 18 L, is seeded with
sludge from a local municipal wastewater treatment plant
(Dalian, China). The membrane module used in this study
is a bundle of hollow-fiber membranes of polyethylene
(Daiki, Japan) with a pore size of 0.1 µm and a filtration
area of 0.01 m2. An air diffuser, an 8-mm inner diam-
eter tube with many 1 mm openings, is fitted under the
membrane, for aeration. Two baffle plates are fixed on
both sides of the membrane to form an upflow aerobic
zone between the membrane and plates, and downflow
Fig. 1 Schematic diagram of the experimental setup. (1) feed tank; (2)
balance tank; (3) airlift internal circulation membrane bioreactor (AIC-
MBR); (4) membrane module; (5) baffle plates; (6) electric heater; (7)
air compressor; (8) rotameter; (9) peristaltic pump; (10) effluent vacuum
gauger; (11) time relay.
anoxic zone between the plates and reactor walls. The
reactor temperature is maintained at about 25°C with an
electric heater. To control membrane fouling, the suction
mode is 6 min on, 2 min off by time relay. When the
operating pressure increases to over 0.02 MPa, hydraulic
cleaning is adopted. If hydraulic cleaning cannot improve
the situation of membrane fouling, chemical cleaning is
used with NaClO solution to recover the membrane flux.
The operating parameters are summarized in Table 1.
Operating parameters in AIC-MBR system
Mixed liquor suspended solids (MLSS) (mg/L)
Mixed liquor volatile suspended solids (MLVSS) (mg/L)
Hydraulic retention time (HRT) (h)
Solid retention time (SRT) (d)
* Mean ± SE.
1.2 Feed medium
Synthetic wastewater consisting of sugar, NH4Cl,
KH2PO4,a mineral solution
MgSO4·7H2O, FeSO4·7H2O, and EDTA, and a trace
CuSO4·5H2O, KI, MnCl2·4H2O, ZnCl2, Na2MoO4·2H2O,
and CoCl2·6H2O, was used to simulate the composition
of real domestic wastewater. The pH of the synthetic
wastewater was controlled at 7.8 ± 0.2 by the addition of
The COD, NH4+-N, nitrate (NO3−-N), nitrite (NO2−-N),
total suspended solids (TSS), and volatile suspended solid
(VSS) contents were determined according to the standard
methods (APHA/AWWA/WEF, 1995). TN concentration
was evaluated by a TOC analyzer equipped with a total
nitrogen-measuring unit (TOC-VCPH, Shimadzu, Japan).
The DO concentration was measured using a DO meter
(Model 55,YSI, USA).
1.4 Experimental procedure
The operating conditions of the AIC-MBR are shown
in Table 2. Holding the influent COD concentration, the
influent TN concentration was increased gradually, to
change the COD/N ratio during run I. Then the influent
COD/N ratio was kept constant, and a gradual increase of
DO concentration was adopted in run II. DO concentration
monitored during the whole experiment mainly aimed
at the aerobic zone. Generally, the DO concentration in
the anoxic zone was 40%–60% of the aerobic zone. The
specific values of COD and nitrogen compounds for the
influent, membrane effluent, and supernatant were moni-
tored daily. The sludge concentration in the reactor was
1.5 Batch tests
The batch tests were conducted in a 400-ml beaker,
which was placed in a water bath at 25°C. The batch
reactor was equipped with compressed air installation.
No. 8Effects of COD/N ratio and DO concentration on simultaneous nitrification and denitrification······
Operating conditions of the AIC-MBR
RunOperational time (d) Influent COD (mg/L)Influent TN (mg/L)COD/N ratioDO (mg/L)
The numbers in parentheses are standard deviation, n = 150.
Sludge samples of 200 ml were added to the beaker, which
was fed with synthetic domestic wastewater.
Two experiment series were conducted. Series 1 evalu-
ated the effect of COD/N ratio on organics and nitrogen
removal, and series 2 was concerned with the effect of
DO concentration on nitrogen removal. In series 1 the
influent TN concentration fixed at about 40 mg/L, whereas,
the influent COD was increased stepwise to obtain three
different COD/N ratios (4.77, 10.04, and 15.11). The DO
concentration of batch reactors was maintained at about
1.0 mg/L throughout the tests. In series 2 the same quality
of synthetic domestic wastewater was added to the beaker
to achieve initial COD and TN concentrations of 350
and 35 mg/L, respectively. Oxygen was supplied and the
concentration was controlled at 0.5, 1.0, 1.5, and 3.0 mg/L,
respectively. All the batch tests lasted for 6 h. During the
period, samples were collected regularly and immediately
analyzed for COD, TN, NH4+-N, NO3−-N, and NO2−-N.
2 Results and discussion
2.1 Influence of nitrogen loading rate
Figure 2 presents the COD concentration in the effluent
and COD removal efficiency during the whole opera-
tion period. The system showed excellent performance
in organic carbon removal. An average organic removal
of over 90% was achieved throughout the experiment.
Although the supernatant COD fluctuated greatly from 30
to 120 mg/L, the membrane effluent COD stabilized at low
levels. This was attributed to the complete retention of all
particulate COD and macromolecular COD components
Fig. 2 Profiles of COD concentrations in the system when the nitrogen
loading rate was adjusted at 0.013 (phase I), 0.020 (phase II) , or 0.040
gN/(g VSS·d) (phase III).
by the membrane.
Figure 3 shows the removal performance of nitrogen
in the three phases. During the whole operation period,
with the given influent COD of 400 mg/L, the influent
COD/N ratio was shifted, by varying the nitrogen loading
rate supplied to the system.
During phase I, from day 1 to day 48, the nitrogen
loading rate was maintained at 0.013 gN/(g VSS·d).
NH4+-N was about 8 mg/L. NO3−-N and NO2−-N in the
effluent were detected at low concentration level (0.59
and 0.08 mg/L, respectively). Good nitrification was not
achieved because the heterotrophic microorganism had
dominated the growth competition between autotrophic
and heterotrophic microorganism populations and utilized
the available DO. TN removal efficiency was only 68% as
a result of the inhibition of nitrification.
In phase II, although the nitrogen loading rate was
increased to 0.020 gN/(g VSS·d), NH4+-N removal effi-
ciency gradually increased to 73%. This was mainly the
result of the growth and enrichment of the nitrifiers. The
products of nitrification were reduced, and the effluent
NO3−-N and NO2−-N was maintained at a low residual
level. TN removal efficiency was increased to 73% because
of the promotion of nitrification. Therefore, under the
condition of phases I and II, nitrification was the limiting
process and the efficiency of TN removal mainly depended
on the degree of the nitrification in the system.
During phase III, from day 97 to day 144, the nitrogen
loading rate increased to 0.040 gN/(g VSS·d). NH4+-N
removal efficiency increased rapidly to over 89% and the
effluent NH4+-N concentration remained at a low level.
However, this condition was disadvantageous for aerobic
denitrification. The effluent NO3−-N residual concentra-
tion was high, up to 37 mg/L. As the nitrogen loading rate
increasing, the amount of COD could not supply enough
carbon sources for denitrification. This situation resulted in
the accumulation of nitrates and limited the SND process.
The TN removal efficiency decreased to 45% because of
the deterioration of denitrification. Zhang et al. (2007)
pointed out that the key to achieving high total nitrogen
removal lay in controlling the balance of nitrification
and denitrification, when nitrification and denitrification
occurred simultaneously in one reactor.
2.2 Influence of DO concentration
The TN removal wasgreatly influenced by the
operating DO concentration. In phase I, when the DO
936 MENG Qingjuan et al.Vol. 20
was maintained at 0.5 mg/L, nitrification was hampered
because of oxygen limitation, with an average effluent
NH4+-N of 14 mg/L. TN removal efficiency was 55%. In
phase II, to improve NH4+-N removal, the operating DO
was increased to 1.0 mg/L. Along with the improvement
of nitrification, the efficiency of TN removal increased to
78%. The operating DO was further increased to 1.5 mg/L,
and nitrification was enhanced, with an average effluent
NH4+-N of 3 mg/L. However, the average TN removal
decreased to 55% (Fig.4).
SND in the system was realized from the analysis of
the effluent nitrogen compounds. When the operating DO
was maintained between 0.5 and 1.0 mg/L, no significant
NO3−-N and NO2−-N were detected in the effluent. These
results showed that the efficiency of aerobic denitrification
was high when the DO was in this range. However, when
the DO concentration was at 0.5 mg/L or lower, NH4+-N
became the main form of the effluent nitrogen compounds.
Under this operating condition, the sludge floc tended to
be anoxic, even anaerobic. This was capable of increasing
the ability of aerobic denitrification of the whole system.
Simultaneously, it also inhibited the nitrifiers existing in
the sludge floc. Therefore, the removal of NH4+-N and
TN was affected significantly. When the operating DO was
increased to 1.0 mg/L, the effluent nitrogen compounds
were all at a low level. The results indicated that the pro-
motion of nitrification and maintaining of denitrification
were the main factors that affected the SND of the system.
Nevertheless, when the DO concentration was increased to
1.5 mg/L or higher, considerable NO3−-N was measured
in the effluent, and TN removal was greatly decreased
as a result of the reduced denitrification rate. The higher
DO concentration restrained the activity of the denitrifiers
that were preferentially active in areas of very low DO
concentration. The denitrification ability of the system was
decreased greatly.In addition, the higher DO concentration
enhanced the activity of heterotrophic aerobic bacteria and
the organics could be biodegraded rapidly and thoroughly.
Thus, although an anoxic zone could be formed in part of
the sludge floc, the denitrification process was hampered
because of lack of organics. Pochana and Keller (1999)
explained that increased DO concentration in the reactor
bulk liquid negatively affected SND.
These results reflected the relationship between the DO
concentration and SND efficiency. The optimal DO con-
centrations for maximum efficiencies of SND presented a
certain extent of difference among the previous studies.
Katie et al. (2003) found that the optimal range of DO
concentration was between 0.8–1.2 mg/L when they in-
vestigated SND that used reserved PHB as an electronic
Fig. 3 Profiles of NH4+-N, NO3−-N, NO2−-N, and TN concentrations in the system when the nitrogen loading rate was adjusted at 0.013 (phase I),
0.020 (phase II) or 0.040 gN/(g VSS·d) (phase III).
Fig. 4 Profiles of NH4+-N, NO3−-N, NO2−-N, and TN concentrations in the system when the DO concentration was controlled at 0.5 (phase I), 1.0
(phase II), 1.5 (phase III), and 3.0 mg/L (phase IV).
No. 8Effects of COD/N ratio and DO concentration on simultaneous nitrification and denitrification······
acceptor in an SBR. Pochana and Keller (1999) achieved
more than 80% SND under DO conditions between 0.3 and
0.8 mg/L. Von M¨ unch et al. (1996) required a DO con-
centration of 0.5 mg/L to achieve 100% SND. In general,
the comparison of the studies suggested that an ideal SND
process was achievable with low DO concentrations.
2.3 Batch tests
Under an initial COD/N ratio of 4.77, there was a
noticeable decrease in the NH4+-N removal. At the same
time, peaking of NO3−-N was observed (Fig.5). This
phenomenon indicated that nitrification by autotrophic
bacteria was the major process occurring at low COD/N
ratio, which is in agreement with Hooper et al. (1997).
This also implied that the COD/N ratio was inadequate to
provide enough degradable carbon sources as an electronic
donor, for denitrification. Chiu et al. (2007) pointed out
carbon substrate, to provide the power demanded by the
denitrification reaction. Therefore this experimental condi-
tion was not fit for establishing an efficient SND process.
The efficiency of the SND process (ESND,%) was calcu-
lated by Eq.(1):
× 100% (1)
and NO2−-N present when the batch test ended and
during the batch test.
When the initial COD/N ratio was increased to a high
level (10.04 and 15.11), there was no obvious NO3−-N
and NO2−-N detected in the effluent (Fig.5). von M¨ unch et
al. (1996) proposed that an efficient SND process should
not produce abundant NO3−-N or NO2−-N. The results of
this experiment were consistent with the previous reports
(von M¨ unch et al., 1996; Zeng et al., 2003). Although the
efficient SND process was attained and left trace amount
of NO3−-N and NO2−-N, at a high COD/N ratio TN
removal was not continuously enhanced along with the
increase of COD/N ratio. When the COD/N ratio got to
15.11, TN removal showed a decrease to some extent,
because of the limitation of nitrification. Denitrification
was also affected by the lack of NOx−-N. The COD/N ratio
x- produced (mg/L) was the sum of NO3−-N
4-oxidized(mg/L) was the amount of NH4+-N oxidized
Profiles of NH4+-N, NO3−-N, and NO2−-N concentrations and COD during the batch tests.
Performance of SND process under different initial COD/N ratios
TN removal (%)
of SND (%)
938MENG Qingjuan et al. Vol. 20
exerted some control over how heterotrophs and nitrifiers
consumed common resources DO and space (Chu et al.,
2006). With the increase of COD/N ratio, heterotrophs
gradually dominated and utilized the available DO and
sludge floc space.
The nitrification and denitrification rates and efficiency
of SND are summarized in Table 3. Chiu et al. (2007)
proposed that an efficient SND process occurred when the
nitrification and denitrification rates were in a balanced
equilibrium. Therefore, 10.04 was considered as the opti-
mum COD/N ratio value for SND in a single reactor in this
cation rates were calculated as 4.99 mg NH4+-N/(L·h) and
4.90 mg NOx−-N/(L·h), respectively, basically reaching
equilibrium. Furthermore, the nitrogen and organic carbon
in the system were both optimally removed.
Figure 6 shows the impact of DO concentrations of 0.5,
1.0, 1.5, and 3.0 mg/L on SND. Running at 0.5 mg/L
DO, almost all products from the nitrification process were
converted to nitrogen gas. However, under this operating
condition, the sludge floc tended to be anoxic even in
an anaerobic environment that resulted in the limitation
of nitrification. Moreover, the denitrification rate did not
achieve a higher level as expected because of the lack of
NOx−-N. With the increase of DO concentration, a more
significant increase in NH4+-N removal was observed.
Theoretically, the denitrification rate should be increasing-
ly inhibited by high DO levels. Nevertheless, when the DO
concentration increased from 0.5 to 1.0 mg/L, the deni-
trification rate presented a certain extent of improvement.
It could be explained that improved nitrification supplied
a sufficient electronic acceptor for denitrification. At full
aeration (DO concentration >1.5 mg/L), complete nitrifi-
cation was noticed, but the percentage of SND decreased.
for significant denitrification to occur simultaneously.
As shown in Fig.7, with the reduction in DO concentra-
tion, the percentage of SND could be improved. This result
was in accordance with a previous research (Pochana et al.,
Katie et al. (2003) proposed that for satisfactory nutrient
removal from wastewater not only the percentage of SND,
but also the rate was important.
From the results, the maximum RSND was achieved
at a DO concentration of 1.0 mg/L, yet the percentage
of SND at this DO concentration was not the optimal.
The DO concentration was higher than 1.0 mg/L, yet
both the rate and quantity of SND decreased; whereas,
by lowering the oxygen supply (DO concentration < 1.0
mg/L), the percentage of SND greatly increased, but the
rate decreased. Hence, there was a compromise between
the rate and quantity at which SND occurred and total
nitrogen removal. Katie et al. (2003) adopted the method
where the rate of SND was multiplied by the fraction of the
Profiles of NH4+-N, NO3−-N, and NO2−-N concentrations during the batch tests when the DO concentration was controlled.
jesc.ac.cn Download full-text
No. 8Effects of COD/N ratio and DO concentration on simultaneous nitrification and denitrification······
Fig. 7 ProfilesofpercentageofSND(ESND)andtherateofSND(RSND),
and the SND rate multiplied by SND (%) when the DO concentration was
controlled at 0.5, 1.0, 1.5 or 3.0 mg/L.
SND profile of DO concentration, to assess the area of DO
concentration where the optimum SND rate and quantity
occurred between 0.75–1.0 mg/L.
This study presents the manner in which the COD/N
ratio and DO concentration affect the occurrence of the
SND process in the AIC-MBR system.
Fixing the organic loading rate, the nitrogen loading rate
was increased gradually, to investigate nitrogen removal.
The study showed that it administered nitrification, but TN
removal did not present an increasing trend along with
the promotion of nitrification. The results indicated that
there was an optimum COD/N ratio for nitrogen removal
through the SND process.
At significant high or low DO concentration, the SND
of aerobic or anoxic zones within the floc, by oxygen
and denitrification rates reached equilibrium and resulted
in nearly complete SND when the COD/N ratio was
controlled at 10.04. With this COD/N ratio, nitrogen and
organic carbon were both optimally removed. From the
batch tests, 0.75–1.0 mg/L was taken as the optimum
range of DO concentration for SND. This range of DO
concentration decided the quantity and the rate of SND at
which high levels of SND were achieved.
This work was financially supported by the Key Labora-
tory of Industrial Ecology and Environmental Engineering
ofthe Schoolof Environmentaland BiologicalScience and
Technology, Dalian University of Technology.
APHA/AWWA/WEF, 1995. Standard Methods for the Examina-
tion of Water and Wastewater (19th ed.). Washington DC:
American Public Health Association.
a granular sludge sequencing batch airlift reactor. Biotech-
nol Bioeng, 75: 82–92.
Brindle K, Stephenson T, 1996. The application of mem-
brane biological reactors for the treatment of wastewaters.
Biotechnol Bioeng, 49: 601–610.
Chiu Y C, Lee L L, Chang C N, Chao A C, 2007. Control of
carbon and ammonium ratio for simultaneous nitrification
and denitrification in a sequencing batch bioreactor. Int
Biodeterior Biodegrad, 59: 1–7.
Chu L B, Zhang X W, Yang F L, Li X H, 2006. Treatment of
domestic wastewater by using a microaerobic membrane
bioreactor. Desalination, 189: 181–192.
Collivignarelli C, Bertanza G, 1999. Simultaneous nitrification-
denitrification process in activated sludge plants: perfor-
mance and applicability. Water Sci Technol, 40: 187–194.
de Silva D G V, Urbain V, Abeysinghe D H, Rittmann B E, 1998.
Advanced analysis of membrane bioreactor performance
with aerobic-anoxic cycling. Water Sci Technol, 38: 505–
Hooper A B, Vannelli T, Bergmann D J, Arciero D, 1997.
Enzymology of the oxidation of ammonia to nitrite by
bacteria. Antonie van Leeuwenhoek, 71: 59–67.
Katie A T, Natalie B, Ralf C R, 2003. Simultaneous nitrification
and denitrification using stored substrate (PHB) as the
electron donor in an SBR. Int Biodeterior Biodegrad, 83:
Pochana K, Keller J, 1999. Study of factors affecting simul-
taneous nitrification and denitrification (SND). Water Sci
Technol, 39: 61–68.
Pochana K, Keller J, Lant P, 1999. Model development for simul-
taneous nitrification and denitrification. Water Sci Technol,
Rittmann B E, Langeland W E, 1985. Simultaneous denitrifica-
tion with nitrification in single-channel oxidation ditches. J
Water Pollut Control Fed, 57: 300–308.
Rosenbergera S, Kr¨ uger U, Witzig R, 2002. Performance of a
bioreactor with submerged membranes for aerobic treat-
ment of municipal waste water. Water Res, 36: 413–420.
von M¨ unch E, Lant P, Keller J, 1996. Simultaneous nitrification
and denitrification in bench-scale sequencing batch reac-
tors. Water Res, 30: 277–284.
Wang Y, Huang X, Yuan Q P, 2005. Nitrogen and carbon
removals from food processing wastewater by an anox-
ic/aerobic membrane bioreactor. Process Biochem, 40:
Yeom I T, Nah Y M, Ahn K H, 1999. Treatment of household
wastewaterusingan intermittentlyaeratedmembrane biore-
actor. Desalination, 124: 193–204.
Zeng R J, Lemaire Z, Yuan Z, Keller J, 2003. Simultaneous
nitrification and denitrification, and phosphorus removal in
a lab scale sequencing batch reactor. Biotechnol Bioeng, 84:
Zhang X H, Zhou J T, Guo H Y, Qu Y Y, Liu G F, Zhao L H,
2007. Nitrogen removal performance in a novel combined
biofilm reactor. Process Biochem, 42: 620–626.
Zhao H W, Mavinic D S, Oldham W K, Koch F A, 1999.
Controlling factors for simultaneous nitrification and den-
itrification in a two stage intermittent aeration process
treating domestic sewage. Water Res, 33: 961–970.