Nitrifier decay and recovery in a Moving Bed Biofilm Reactor
(MBBR) treating reverse osmosis concentrate
Liu Ye*, Shihu Hu*, Yvan Poussade**, ***, Jurg Keller* and Zhiguo Yuan*‫٭‬
* Advanced Water Management Centre (AWMC), The University of Queensland, St Lucia, Brisbane, QLD 4072,
Australia
** Veolia Water Australia, Level 15, 127 Creek Street, Brisbane, QLD 4000
*** Seqwater, 240 Margaret Street, Brisbane, QLD 4000
Email addresses: zhiguo@awmc.uq.edu.au; *Corresponding author. Tel: +61 07 33654374; Fax: +61 07
33654726;
Abstract
A two-stage moving bed biofilm reactor (MBBR) is applied at the Bundamba
advanced water treatment plant (AWTP) (Queensland, Australia) to treat the reverse
osmosis concentrate (ROC) for inorganic nutrient removal. One of the operational
challenges for MBBR is to cope with the large fluctuations of the ROC flow. This
study investigated the decay rates of ammonia-oxidizing bacteria (AOB) and nitriteoxidizing bacteria (NOB) and biofilm detachment in MBBR during starvation for up to
one month. An intermittent aeration strategy of 15 min aeration every 6 hr was
applied. This study also evaluated the activity recovery of both AOB and NOB after
normal operation was resumed. The results showed that the activity loss of AOB and
NOB was relatively minor (< 20%) within 10 days of starvation, which ensured
relative quick recovery of ammonium removal when normal operation resumed. In
contrast, the AOB and NOB activity loss reached 60-80% when the starvation time
was longer than 20 days, resulting in slower recovery of ammonium removal after
starvation. Starvation for less than 20 days didn’t result in an apparent biomass
detachment from carriers.
Keywords: AOB; NOB; decay; MBBR; starvation; ROC
Introduction
Large fluctuations of wastewater flow and composition inherent to the industry
activities is one of the big challenges to a lot of wastewater treatment plants
(WWTPs). It’s crucial to maintain the biomass viability and activity during the long
idle or starvation periods due to the interruptions of wastewater flows (E.g. annual
maintenance or seasonal production variations) to the WWTPs for weeks and even
months.
The Bundamba Advanced Water Treatment Plant (AWTP), located in Queensland,
Australia, played a key role in the delivery of purified recycled water supply to the
south east Queensland region. Reverse Osmosis (RO) membrane is used to provide
purified recycled water to secure water supply for the rapidly growing Southeast
Queensland region. The Reverse Osmosis Concentrate (ROC) is treated for
inorganic nutrient removal before its discharge to the Brisbane River. The ROC
treatment train includes nitrification in a two-stage Moving Bed Biofilm Reactor
(MBBR), denitrification through anoxic sand filters, and chemical precipitation for
phosphate removal. However, the ROC flow to the MBBR is sometimes interrupted
due to varying demand for water production.
A lot of studies have been done to investigate the impacts of starvation on the
bacterial population and activities of activated sludge. It has been well-demonstrated
that nitrifiers decayed at a higher rate under aerobic conditions than under anoxic or
anaerobic conditions (Nowak et al., 1994; Siegrist et al., 1999; Morgenroth et al.,
2000; Lee and Oleszkiewicz, 2003; Manser et al., 2006; Salem et al., 2006; Hao et
al., 2009). In a recent study, Munz et al. (2011) further demonstrated that the aerobic
decay rate of nitrifiers increased with the dissolved oxygen concentration.
Slower decay rates of nitrifiers have also been observed under alternating
anaerobic, anoxic and aerobic conditions (Roslev and King, 1996; Morgenroth et al.,
2000; Yuan et al., 2000; Lee and Oleszkiewicz, 2003; Yilmaz et al., 2007). Lee and
Oleszkiewicz (2003) reported that nitrifiers decayed under alternating aerobic, anoxic
conditions at a rate that was 40% lower than that under anoxic conditions.
Morgenroth et al. (2000) observed that there was no or little loss (<5%) of nitrifying
activities when activated sludge was starved under alternating aerobic,
anoxic/anaerobic conditions for a period of up to one week. Yilmaz et al. (2007)
evaluated the effectiveness of a specific operating strategy, which created
alternating anaerobic, anoxic and aerobic conditions, in maintaining the nutrient
removal capacity of activated sludge. Their study also showed that both ammoniumoxidizing bacteria (AOB) and nitrite-oxidizing bacteria (NOB) can actually survive for
weeks and recover their activity rapidly after start-up.
The MBBR technology is well-established for wastewater treatment where bacteria
grow as a biofilm on the protected surfaces of suspended carriers (McQuarrie, et al.
2011). However, how to operate MBBR to cope with large fluctuations of influent flow
is still a big challenge for most plants, and the alternating operational strategy has
not been demonstrated to be effective for biofilm systems yet. Being able to operate
the MBBR process in periods without wastewater feed, ammonium was added at the
Bundamba AWTP to sustain the activities of AOB and NOB. An operational strategy
without requiring ammonium addition during flow interruption is attractive from
operational, environmental and economical perspectives. The aim of this research is
to monitor nitrifier decay and biofilm detachment during starvation with intermittent
aeration, also to evaluate the activity recovery of both AOB and NOB after normal
operation is resumed.
Materials and Methods
Starvation reactor set-up and operation
Carriers with biomass together with wastewater were collected from the 1 st stage
reactor of the full-scale MBBR system at the Bundamba AWTP and were kept in a
25 L starvation reactor (R1) for starvation tests. The volumetric ratio of carriers to the
volume of the reactor was around 40%, which was in line with the full-scale MBBR.
The reactor was aerated intermittently for 15 min every 6 hours. The biomass was
starved for 31 days. The air flow was manually adjusted to keep dissolved oxygen
(DO) between 2-4 mg/L over the aeration period. Solid and liquid samples were
taken from the reactor on a daily basis (weekdays only) at the end of an aeration
period for the analysis of mixed liquor suspended solids (MLSS), NH4+-N, NO2--N,
NO3--N and PO43--P. pH and DO were measured regularly during the period of
starvation.
Batch experiments for measuring activity loss of AOB and NOB
During the starvation period, 2L wastewater containing 26 carriers was taken from
R1 at the end of an aeration period on Days 5, 10, 21 and 31, and transferred into
two 1 L batch reactors (B1 and B2). Batch tests described below were conducted to
determine the AOB and NOB activities. In addition, 2L wastewater along with 26
carriers was collected from the full-scale MBBR for batch test on Day 0 to measure
the AOB and NOB activities of the non-starved biomass.
At the start of each batch test, NH4Cl and NaNO2 were added to B1 and B2,
respectively, which resulted in the initial concentrations of 15 mg/L NH4+-N and 15
mg/L NO2--N, respectively. DO in both batch reactors was maintained at 5±0.5 mg/L
to provide a non-limiting DO concentration. pH was adjusted to around 7.0 by the
addition of 0.1 M NaOH at the beginning of each test, but not controlled in the
remaining time of the test. Each batch experiment lasted for 6-8 hr, during which
NH4+-N, NO2--N and NO3--N were measured hourly. Decrease in the activities of
AOB and NOB under starvation conditions was assessed by comparing specific
ammonium uptake rates (SAUR) (mgN/carrier.hr) and specific nitrite uptake rates
(SNUR) (mgN/carrier.hr) before and after a certain period of starvation.
Recovery tests and reactor operation
1.5 L wastewater together with 20 carriers was taken from R1 at the end of an
aeration period on Day 5, 10, 21 and 31, respectively, and transferred to another
reactor (R2, recovery reactor) for recovery tests.
R2 was continuously fed with real ROC wastewater collected weekly from the
Bundamba AWTP. Samples were analyzed on the day of collection for its NH 4+-N
concentration. Extra NH4Cl was added to the ROC water to maintain the influent
level of NH4+-N at an approximate level of 5-7 mgN/L (wastewater collected
sometimes contained very low levels of ammonium, e.g. < 2 mgN/L, which was not
ideal for activity tests). The hydraulic retention time (HRT) was maintained at 1.5 hr
and DO controlled by a programming logic controller in the range of 2-4 mg/L,
mimicking the operation at the Bundamba AWTP. The DO controller was an on/off
controller with aeration turned on and off when DO reached 2 and 4 mg/L,
respectively. During the recovery period, ammonium in the reactor was analysed
every 4 hours during daytime for 5 days. DO and pH were monitored continuously
over the whole recovery period. Oxygen uptake rate (OUR) was determined as the
slopes of the DO profiles in period when aeration was turned on.
Analytical Methods
The ammonium (N-NH4+), nitrate (N-NO3-), nitrite (N-NO2-) and orthophosphate (PPO43-) concentrations were analyzed using a Lachat QuikChem8000 Flow Injection
Analyzer (Lachat Instrument, Milwaukee, Wisconsin). Total and soluble chemical
oxygen demand (CODt and CODs, respectively), total Kjeldahl nitrogen (TKN), total
phosphorus (TP), MLSS and MLVSS were analysed according to the standard
methods (APHA, 1995).
Results and Discussion
3.0
30
2.5
25
2.0
20
1.5
15
PO43--P
NH4+-N
NO2--N
NO3--N
1.0
0.5
10
NO3--N (mg/L)
NH4+-N, NO2--N, PO43--P (mg/L)
Performance of the starvation reactor
Figure 1 shows the nutrient concentration profiles measured during the starvation
period from R1. The concentration of phosphorus increased gradually, which could
have resulted from cell lysis. Similarly, the rise in the level of nitrate was observed,
which could be attributed to the oxidation of NH4+ released by bacterial decay.
Ammonium was undetectable, indicating all ammonium released as a result of decay
was oxidized by nitrifiers, which provided energy source to nitrifiers to some extent.
The growth of nitrifiers during aerobic period utilising the ammonia released through
the decay of heterotrophic bacteria can largely compensate for the reduction of
nitrifying activity caused by the decay of nitrifiers (Morgenroth et al., 2000). However,
the ROC feed to the full-scale MBBR contained limited amount of organic carbon (<
50 mgCOD/L), and hence the amount of heterotrophic biomass in the biofilms was
expected to be low (not determined).
5
0.0
0
0
5
10
15
20
25
30
Time (day)
Figure 1 The variation of nutrient concentrations in the starvation reactor over the starvation period
During the whole starvation period, pH in the starvation reactor increased
gradually, from 7.1 on Day 0 to 8.0 at Day 31.
400
MLSS (mg/L)
300
200
100
0
0
5
10
15
20
25
30
Time (day)
Figure 2 The variation of stripped biomass concentration in the starvation reactor during the starvation period
The MLSS profile is shown in Figure 2. MLSS didn’t change significantly within the
first 20 days of starvation and had an average concentration of approximate 200
mg/L, suggesting negligible detachment of biofilm from carriers during that period.
After that, during Day 20-31, MLSS increased and reached a concentration of 270
mg/L at the end of starvation. This revealed that biofilm detachment occurred after
25 days of starvation. After a long time starvation, the increased depletion of nutrient
supply to the bacterial growth or maintenance may have resulted in an increase in
the detachment rates (Sawyer et al., 1998).
Activity loss of AOB and NOB
Batch tests revealed that both the NH4+-N and NO2--N oxidation rates decreased
with starvation time, which implied a loss of activity of both AOB and NOB. The
decline in both SAUR and SNUR was less than 20% in the first 10 days of starvation
(shown in Figure.3). However, decreases of 42% and 67% in SAUR occurred after
21 and 31 days of starvation, respectively. SNUR followed the same trend, with a fall
of 40% and 53%, respectively, over the same starvation periods. The average decay
rates of AOB and NOB were determined to be 0.044 d-1 and 0.033 d-1, respectively,
through fitting an exponential decay model. Both AOB and NOB had a relatively slow
decay rate.
Exponential decay
model (SNUR)
Exponential decay
model (SAUR)
Figure 3 SAUR and SNUR measured after 0, 5, 10, 21 and 31 days of starvation. The regression lines
were obtained with an exponential decay model
The decay rates of both AOB and NOB observed in this study were much lower
than those reported in literature. The AOB decay rate in many studies, as
summarized by Salem et al.(2006), ranged between 0.15–0.21, 0.025–0.06 and
0.05–0.2d-1, respectively, for activated sludge systems at 20 °C (similar to the
temperature used in this study) under aerobic, anaerobic and anoxic conditions. The
NOB decay rates under these conditions were 0.21, 0.06 and 0.12d-1 under aerobic,
anaerobic and anoxic conditions, respectively (Salem et al., 2006). It further
confirmed that the intermittent aeration strategy (15 min aeration every 6 hr) is a
suitable strategy for maintaining nitrifier activity during starvation. Yilmaz et al. (2007)
reported that, with this operational strategy, the AOB and NOB decay rates of a
sludge treating abattoir wastewater were even lower, being 0.017d-1 and 0.004d-1,
respectively. The weekly extra addition of ammonium and nitrite into the starvation
reactor during batch tests in the Yilmaz et al. (2007) study likely provided more
electron donors and therefore partially compensated for the nitirifier decay. In this
study, the batch tests were not carried out in the parent reactor, but in separate
batch reactors.
Activity recovery after starvation
Figure 4 Ammonium removal rate in the recovery process after starvation for 5, 10, 21 and 31 days
Figure 5 Ammonium removal in the full-scale two-stage MBBR plant in the last quarter of 2010.
Note that the removal was achieved by two stages. No data were available for the first stage for
direct comparison.
Figure 4 shows the ammonium removal in the recovery reactor following 5, 10, 21
and 31 days of starvation. After normal operational conditions were resumed, the
ammonium consumption quickly improved when the starvation was less than 10
days. Compared with the ammonium removal in the full-scale plant (data shown in
Figure 5), where the results showed the performance of the two-stage MBBR rather
than just the first stage, it is regarded that NH4+-N removal efficiency recovered to
the original level in 24-48 hr when the starvation was less than 10 days, and all
ammonium oxidised was converted to nitrate after 48-72 hr recovery. In contrast, the
ammonium removal recovery was slow in the cases of 21 and 31 days starvation.
These results again suggested that a starvation period of less than 10 days is
preferable, supporting the results obtained for decline in SAUR and SNUR.
Figure 6 presents the results of OUR during the recovery period after starvation for
0, 5, 10, 21 and 31 days, respectively. In the cases of 5 and 10 day starvation, OUR
increased to around 0.006-0.008 mgO2 L-1min-1 carrier-1 (OUR observed on Day 0)
after a recovery period of 1-2 days. However, in the 21 and 31 day cases, OUR went
up slowly and didn’t reach the same value within the 5 day recovery period, which
further confirmed our previous conclusion that a starvation period shorter than 10
days is desirable.
Figure 6 Oxygen uptake rates in the recovery reactor after 0, 5, 10, 21 and 31 days of starvation
Conclusion
The activity loss of both AOB and NOB under starvation conditions and the
recovery of their activities after resuming normal operation were evaluated. The
following conclusions can be drawn.
 The activity loss of AOB and NOB is relatively minor (< 20%) within 10 days of
starvation, which ensures relative quick recovery of ammonium removal when
normal operation resumes;
 The AOB and NOB activities loss is substantial (reaching 60-80%) when the
starvation time is longer than 20 days, resulting in slower recovery of
ammonium removal when normal operation resumes;
 The intermittent aeration strategy of 15 min aeration every 6 hr is a suitable
strategy for maintaining nitrifier activity of MBBR systems during starvation;
Starvation less than 20 days does not result in an apparent biomass
detachment from carriers.
Acknowledgments
This work was financially supported by Seqwater, Australia. Mr. Jason Krzciuk is
acknowledged to the on-site help.
References:
APHA, 1995. Standard Methods for the Examination of Water and Wastewater. American
Public Health Association,Washington, DC.
Hao, X., Wang, Q., Zhang, X., Cao, Y. and Mark Loosdrecht, C.M.v. (2009) Experimental
evaluation of decrease in bacterial activity due to cell death and activity decay in activated
sludge. Water Research 43(14), 3604-3612.
Lee, Y. and Oleszkiewicz, J.A. (2003) Effects of predation and ORP conditions on the
performance of nitrifiers in activated sludge systems. Water Research 37(17), 4202-4210.
Manser, R., Gujer, W., Siegrist, H. (2006) Decay processes of nitrifying bacteria in biological
wastewater treatment systems.Water Reserach 40 (12), 2416–2426.
McQuarrie, J.P. and Boltz, J.P. (2011) Moving Bed Biofilm Reactor Technology: Process
Applications, Design, and Performance. Water Environment Research 83(6), 560-575.
Morgenroth, E., Obermayer, A., Arnold, E., Bruhl, A., Wagner, M. and Wilderer, P.A. (2000)
Effect of long-term idle periods on the performance of sequencing batch reactors, pp. 105113.
Munz, G., Lubello, C. and Oleszkiewicz, J.A. (2011) Modeling the decay of ammonium
oxidizing bacteria. Water Research 45(2), 557-564.
Nowak, O., Schweighofer, P. and Svardal, K. (1994) Nitrification inhibition - a method for
the estimation of actual maximum autotrophic growth rates in activated sludge systems.
Water Science and Technology 30(6), 9-19.
Roslev, P. And King, G.M. (1995) Aerobic and anaerobic starvation metabolism in
methanotrophic bacteria. Applied and Environmental Microbiology 61 (4), 1563–1570.
Salem, S., Moussa, M.S. and van Loosdrecht, M.C.M. (2006) Determination of the decay rate
of nitrifying bacteria. Biotechnology and Bioengineering 94(2), 252-262.
Siegrist, H., Brunner, I., Koch, G., Phan, L.C. and Le, V.C. (1999) Reduction of biomass
decay rate under anoxic and anaerobic conditions. Water Science and Technology 39(1), 129137.
Sawyer, L.K. and Hermanowicz, S.W. (1998) Detachment of biofilm bacteria due to
variations in nutrient supply. Water Science and Technology 37(4-5), 211-214.
Yuan, Z., Bogaert, H., Leten, J. and Verstraete, W. (2000) Reducing the size of a nitrogen
removal activated sludge plant by shortening the retention time of inert solids via sludge
storage. Water Research 34(2), 539-549.
Yilmaz, G., Lemaire, R., Keller, J. and Yuan, Z. (2007) Effectiveness of an alternating
aerobic, anoxic/anaerobic strategy for maintaining biomass activity of BNR sludge during
long-term starvation. Water Research 41(12), 2590-2598.