вход по аккаунту


Biological treatment of milk processing wastewater in a sequencing batch flexible fibre biofilm reactor.

код для вставкиСкачать
Asia-Pac. J. Chem. Eng. 2009; 4: 698–703
Published online 26 June 2009 in Wiley InterScience
( DOI:10.1002/apj.320
Special Theme Research Article
Biological treatment of milk processing wastewater in a
sequencing batch flexible fibre biofilm reactor
Mohamed Abdulgader,1 * Qiming Jimmy Yu,1 Ali Zinatizadeh2 and Philip Williams1
Griffith School of Engineering, Griffith University Nathan Campus, Brisbane, Queensland, 4111, Australia
Department of Civil Engineering, Faculty of Engineering, Razi University, Kermanshah, Iran
Received 29 October 2008; Revised 4 March 2009; Accepted 4 March 2009
ABSTRACT: Biological treatment of dairy wastewater was investigated using a laboratory scale aerobic sequencing
batch flexible fibre biofilm reactor (SBFFBR). The SBFFBR system was modified from a typical sequencing batch
reactor system by using eight flexible fibre bundles with a very high specific surface area, which served as support
for microorganisms. The reactor was operated under different influent chemical oxygen demand (COD) concentrations
(610, 2041 and 4382 mg l−1 ) and constant hydraulic retention times of 1.6 days. The results have shown successful
applicability of the SBFFBR system to treat this dairy wastewater. High COD removal efficiencies between 89.7 and
97% were achieved at average organic loading rates of 0.4 and 2.74 kg COD m−3 d−1 , respectively.  2009 Curtin
University of Technology and John Wiley & Sons, Ltd.
KEYWORDS: biofilm reactor; flexible fibre packing; milk industry wastewater; sequencing batch reactor; COD removal
The high cost of wastewater treatment for food industry wastes and increasingly stringent effluent regulations
have increased interest in alternative treatment methods. The selection of a biological treatment process is
normally based on the quality of effluent needed to
meet environmental protection agency standards, the
characteristics and strength of the particular wastewater and operational costs.[1,2] Food processing industries
such as milk processing produce considerable quantities of wastewaters, characterised by high concentration of chemical oxygen demand (COD), biochemical
oxygen demand (BOD), and total Kjeldahl nitrogen
up to 11 000, 5900 and 720 mg l−1 , respectively.[2]
Consequently, these wastewaters are commonly treated
biologically.[3,4] High-rate anaerobic systems seem to
be cost effective and efficient in terms of sludge production and energy consumption, but the effluent quality does not comply with standards for discharge into
natural water environments.[5] Therefore, aerobic methods are generally considered to be more cost effective
*Correspondence to: Mohamed Abdulgader, Griffith School of
Engineering, Griffith University, Nathan Campus, Brisbane, Queensland, 4111, Australia. E-mail:
 2009 Curtin University of Technology and John Wiley & Sons, Ltd.
These processes are classified as either attached
growth (biofilm) or suspended growth system. Traditional continuous flow systems, such as the conventional
activated sludge process, aerated lagoon and oxidation
pond, have been extensively utilised for various types of
industrial wastewaters.[3,4,7] However, these processes
have difficulties in meeting effluent standards and have
high operational costs and sludge production.[1,4,7,8] An
alternative approach using a sequencing batch reactor (SBR) process has exhibited an effective performance for organic carbon, nitrogen and phosphorus
removal within a single reactor for domestic and industrial wastewaters.[1,2,9] Furthermore, the SBRs flexible operation allows mimicking of many processes in
conventional continuous flow reactors and can yield
superior performance.[10] SBRs have overcome many
drawbacks and also differ from other suspended growth
processes. The clearest difference is the reactor volume variation with time, whereas it remains constant
in the traditional continuous flow system.[11] Some
studies[2,12] have reported an acceptable performance of
SBRs in treating dairy industrial wastewater with varied concentration of COD and Total suspended solids
(TSS). However, the SBR process still has some drawbacks such as the high excess sludge produced under
high organic loading, high sludge volume index and
lower removal efficiency due to the increase in biosludge.[1,2]
Asia-Pacific Journal of Chemical Engineering
The applications of aerobic biofilm processes in
food processing wastewater treatment have been developed to avoid the drawbacks in suspended growth
processes.[1,2,7,13,14,19] In the attached growth process,
microorganisms are attached on an inert support media.
The reactor biomass concentration is high due to a
longer sludge age. Hence, the biofilm process can be
much more resistant to shock organic loads.[7,10] Consequently, the treatment efficiency is enhanced for the
retained film of biomass. Furthermore, bulking problems for many suspended growth systems limit their
Performance of a laboratory scale rotating biological disk bioreactor has been evaluated for treatment of
dairy wastewater. A COD removal efficiency of 97%
has been achieved at organic loading rate (OLR) of
14.5 g COD m−2 d−1 without any nutrient correction.
The reactor performance was sharply decreased to 69%
due to the higher loading rate applied.[15] Treatability
of dairy wastewater with influent COD concentration
of 427–1384 mg l−1 was evaluated with the help of a
cross flow medium trickling filter. A decrease in COD
removal efficiency was observed when hydraulic loading rate increased.[16] In another study[2] , a sequencing
batch reactor (MSBR) biofilm system was examined for
treatment of milk industry wastewater. The effect of
OLR on COD and BOD removal was clearly demonstrated. The optimal removal efficiency of the MSBR
system with milk industry wastewater was noted at a
low OLR of 80 g BOD5 m−3 d−1 . The COD and BOD
removal efficiencies of the MSBR system were also
about 5–7% higher than conventional SBR under the
same organic loading condition.[2]
A two-stage continuous flexible fibre biofilm reactor
has been developed for the treatment of food processing
wastewater at 7.7 kg COD m−3 d−1 and 8 h hydraulic
retention times (HRT).[7] Overall, 96% COD removal
efficiency was achieved. However, a literature search
showed that the performance of a sequencing batch
flexible fibre biofilm reactor (SBFFBR) has not been
investigated with dairy industrial wastewater.
This study aims to determine the treatability of milk
processing wastewater in a SBFFBR and to evaluate the
effect of OLR on the reactor performance.
Schematic diagram of experimental
setup. This figure is available in colour online at
Figure 1.
filled with eight bundles of simple flexible fibre made
of rayon fibre (Fig. 2). The fibre is 75 mm long and
0.07 mm in diameter, when straightened, with a specific
growth surface area of 2200 m2 m−3 and avoid fraction
of >90%. An 885-mm-long rope was fixed vertically to
a support frame, and the bundles were attached with an
interval of about 80 mm and installed in the centre of
the reactor.
Wastewater and seed culture
Milk processing wastewater from National Food Ltd,
Brisbane, Australia was used in the experiments.
Wastewater samples were usually collected from the
factory wastewater effluent stream and transported
immediately to the laboratory and stored in a cool room
at 0–4 ◦ C. The wastewater samples’ characteristics are
shown in Table 1. The raw wastewater samples were
Experimental setup
A laboratory scale SBFFBR Perspex system was used.
The experimental setup is shown in Fig. 1. The fabricated SBFFBR had a working volume of 8 l with
a cylindrical shape (120 mm in diameter, 900 mm
height). Mixing and aeration were provided by ceramic
porous diffusers at the bottom of the reactor. Air was
provided via a compressed air source. The SBFFBR was
 2009 Curtin University of Technology and John Wiley & Sons, Ltd.
Figure 2. Schematic of flexible fibre bundles. This figure is
available in colour online at
Asia-Pac. J. Chem. Eng. 2009; 4: 698–703
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
Table 1. Characteristics of milk processing wastewater.
(mg l−1 )
Total solids
Total suspended solid
Volatile suspended solid
Total nitrogen (as N)
Total phosphate (as P)
Oil and grease
N/A, Not Analysed.
diluted with tap water to obtain the desired COD concentrations in each experiment. Before feeding into the
reactor, the pH was regularly checked and adjusted to
neutral. The SBFFBR was inoculated with a mixed
activated sludge sample from Oxley Creek Wastewater
Treatment Plant and Loganholme Wastewater Treatment
Plant. A volume of 1 l mixed culture was fed to the
SBFFBR as inoculum.
Reactor operation
The sequence for different operations during each
treatment cycle of the SBFFBR consisted of five steps.
A fresh raw dairy wastewater was pumped using a
peristaltic pump (master flex model 7523-60 at 42 ml
min−1 ) for a final volume of 8 l (replacement volume
5 l) for a period of 2 h (Fill). The system was
continuously aerated and the oxidation reaction was
carried out for periods of 20 h. The air was cut off and
the reactor was allowed to settle for approximately 1.5 h
(Settle). Then, for a period of 0.5 h (Draw), the effluent
was discharged and supernatant was removed from
the effluent port based on the operational strategy in
Table 2. An idle step was neglected in this experiment.
After each cycle, fresh raw wastewater was fed into
the reactor and the above sequence was repeated. This
experiment was operated under different influent COD
concentrations ranging between 610 and 4382 mg l−1 ,
and a constant HRT (1.6 day), days until a pseudosteady-state condition, was achieved at each cycle.
Analytical procedures
The performance of SBFFBR with milk processing
wastewater was assessed by monitoring carbon removal
(COD) throughout the reactor operation. Samples were
withdrawn from each influent and effluent cycle for
analysis. The following parameters were analysed:
 2009 Curtin University of Technology and John Wiley & Sons, Ltd.
Table 2. Operation strategies of SBFFBR.
Wastewater concentration
610 mg 2041 mg 4382 mg
HRT (d)
Working volume, (l)
Flow rate (l d−1 )
Replacement volume (l d−1 )
Operating cycle time (d)
Fill up (h)
Aeration (h)
Settling (h)
Draw (h)
Operating period (d)
OLR kg COD m−3 d−1
AFR/WFR ratio
AFR, air flow rate; WFR water flow rate.
COD, TSS, volatile suspended solids (VSS), pH, dissolved oxygen (DO) and Turbidity. All analyses were
determined based on standard methods.[17] The total
COD was determined by the dichromate method (digestion method) using a HACH DR/2000 spectrophotometer (HACH Company, USA). TSS and VSS were measured in accordance with the standard American Public
Health Association methods.[17] Samples were filtered
through Whatman GF/C glass microfiber filter paper
(pore size 1.2 µm). The DO concentration in wastewater was monitored by a DO probe model YSI 5010. The
digital DO meter model YSI/5000 was supplied by the
YSI Company, USA. The pH was monitored by a pH
meter model 90-FL provided from TPS Pty Ltd, Australia. Turbidity was also measured by a turbidity meter
model 2100A, supplied by HACH Company, USA,
based on the nephelometric method and the results of
the measurements were reported as nephelometric turbidity units (NTU). The experiment was operated at
room temperature (22 ± 2 ◦ C). All electrodes and instruments were calibrated on the day of each measurement
in accordance with the manufacturer’s procedures.
The reactor retains a high biomass concentration, due
to the packing media high specific surface area for
microorganism attachment. Therefore, the system has
the ability to treat high strength wastewaters due to its
high substrate and product mass transfer provided by
the very high specific surface. In addition, it can also
resist high organic load and hydraulic shock loading
due to synergistic relations among different microbial
species grown in the biofilm.
Asia-Pac. J. Chem. Eng. 2009; 4: 698–703
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
It is evident from the previous results that with
increase in OLR, the COD removal rate of the system
is increased. To achieve steady reactor performance at
low concentrations, the reactor required a few days
to reach steady state between each operating cycle, as
indicated by the COD removal efficiency, typically less
than 14 days, but it may require longer at high OLRs,
as the microorganisms need more time to adjust to the
higher loads.
The SBFFBR was operated with a 24-h cycle period
to assess the performance of the reactor for treating
milk industry wastewater. The variation of COD concentrations in the influent and effluent with respect
to operation time are depicted in Fig. 3. The reactor
was run at three different COD concentrations. First,
a set of experiments was carried out at low influent
COD with an average value of about 610 mg l−1 (corresponding to an OLR of 0.4 kg COD m−3 d−1 ). At
this stage, the reactor attained a very high performance
with an average effluent COD concentration of 15.4 mg
l−1 (97.2% COD removal). In the second trial, influent COD concentration of the reactor was increased to
an average 2041 mg l−1 (corresponding to an OLR of
1.27 kg COD m−3 d−1 ). An average of 70 mg l−1 of
effluent COD (96.6% removal) was achieved. Finally,
the SBFFBR was operated at a high influent COD concentration with an average of 4380 mg l−1 (OLR of
2.74 kg COD m−3 d−1 ). An effluent with a COD concentration of 460 mg l−1 was obtained. However, COD
effluent, 1350 mg l−1 , was initially obtained because
of the sudden increase in influent COD. Generally, a
variation in the influent COD concentration may be
attributed to the degradation in the equalisation tank.
Similarly, the average of 750 mg l−1 of COD effluent (89.3% removal efficiency), obtained by using the
SBR biofilm for treatment of milk industry wastewater at 3 days HRT, showed an effluent concentration
decrease with increased HRT.[6] About 35 mg l−1 of
effluent COD was obtained with 773 mg l−1 of a synthetic wastewater treated by a sequencing batch biofilm
reactor.[19] Findings reported by Bandpi and Bazari[12]
for a SBR system treating dairy wastewater achieved
COD removal efficiency around 90% with COD concentrations varying from 400 to 2500 mg l−1 .
The SBFFBR system achieved a higher COD removal
compared with the previous studies, due to the increased
amount of biofilm mass in the reactor volume provided
by the flexible fibres. This increased biomass concentration was achieved by the combination of two factors.
First, the fibre packing media used has a very high specific surface area (in the order of 2200 m2 m−3 ), which
can provide a much higher support area per unit reactor volume compared with most solid support materials,
for which the specific surface area was in the order of
lower hundreds, e.g. 300 m2 m−3 . Effectively, it is the
higher specific surface areas of the fibre that has led
to the increased biomass concentration in the reactor.
 2009 Curtin University of Technology and John Wiley & Sons, Ltd.
Figure 3. Variation of Influent and Effluent COD in SBFFBR.
Second, the bulk of the biomass in the SBFFBR system is attached to the fibre support. This is in contrast
to the case of a conventional SBR system, where the
biomass is suspended in the wastewater medium. The
attachment of the biomass to the fibre seemed to have
significantly increased the ability of the reactor to retain
the biomass in the reactor discharge phase of the operation. This increased retention of the biomass through
the attachment mechanism in turn may have further
increased the total biomass concentration in the reactor
for substrate degradation. Furthermore, the fibre was
fixed within the reactor and part of the fibre, and its
attached biomass was exposed to the air for a short
period of time before the reactor is completed filled.
The exposure to air may provide some benefits to the
microorganisms, e.g. increased aeration effects similar
to the case of a rotating biological disc. Therefore, these
combined effects are believed to be the theoretical reasons behind the reactor performance observed.
In this paper, the reactor performance was evaluated
by carrying out a number of experiments at a fixed HRT
(1.6 day) and COD (610–4380 mg l−1 ) (corresponding to OLRs ranging between 0.4 and 2.74 kg COD
m−3 d−1 ). Figure 4 presents variations of average COD
removal efficiency and COD removal rate as a function
of the COD OLRs. At low OLRs of 0.4 and 1.3 kg
COD m−3 d−1 , high COD removal efficiencies of 97
and 96% were achieved, respectively. Whereas, 89%
COD removal efficiency was obtained at an OLR of
2.74 kg COD m−3 d−1 . A similar finding was reported
by Yu et al .[7] with a removal efficiency of 90% at an
OLR of 1.04 kg COD m−3 d−1 treating food processing
wastewater in a comparable reactor. Sirianuntapiboon
and his co-workers[2] also achieved an similar performance (89.3%) at OLR of 2.5 kg COD m−3 d−1 . However, a high COD removal (97.4%) was obtained in a
rotating biological contactor (RBC) at 18.44 g m−2 d−1 ,
indicating relatively high removal rate in such a biofilm
reactor.[14] In this RBC reactor, less COD removal efficiency (85.4%) was obtained when the OLRs increased
to 36.89 g m−2 d−1 .
Asia-Pac. J. Chem. Eng. 2009; 4: 698–703
DOI: 10.1002/apj
Figure 4. Reactor performance at HRT of 1.6 day.
In contrast, in this research, the rate of COD removal
showed an increasing tendency from 0.37 to 2.45 kg
COD m−3 d−1 with increase in OLR, indicating high
treatment capacity of the system. With an OLR of
2.75 kg COD m−3 d−1 , specific substrate utilisation
rate was calculated to be 4.64 g CODrem gVSS−1 d−1 .
Note that with increasing COD removal rate, oxygen
uptake rate was also increased (to the level below 1 mg
O2 l−1 ), so that it correlated with the increasing air flow
Figure 5 illustrates the overall removal efficiency
(%) of the reactor as a function of time (days).
Generally, it can be seen that SBFFR achieved a
high level of COD effectively for all different COD
concentrations studied. It attained an average of 94.4%
of COD removal efficiency. Basically, a high removal
efficiency of COD can be achieved at a low COD
concentration.[1] The COD removal efficiency of the
SBFFBR approached 97.2 and 96.7% when the reactor
operated at influent COD concentrations of 610 and
2040 mg l−1 , respectively. This was quite consistent
over the time interval up to 14 days. However, the
removal efficiency decreased to 89.3% as the influent
Asia-Pacific Journal of Chemical Engineering
concentration of COD increased to 4382 mg l−1 . The
COD removal efficiency showed more variations and
was generally lower, but towards the end of the 14-day
period, the removal efficiency was comparable with that
of the two lower COD concentrations. Findings reported
by Bandpi and Bazari[12] for a SBR system treating
dairy wastewater achieved a COD removal efficiency
of around 90% with COD concentrations varying from
400 to 2500 mg l−1 . This illustrates the capability of a
SBFFBR for treating industrial wastewater due to the
increase in the biomass concentration by using a high
specific surface area support media.
Other effluent quality parameters are shown in
Table 3. The amount of COD applied to the system
has a significant effect on the reactor performance; the
DO level in the treated effluent is decreased by about
75% from 5.9 to 1.32 mg O2 l−1 , which is attributed
to the increase of the activity of biomass as the substrate concentration increased. However, the turbidity
of the effluent showed a remarkable increase as influent COD concentration increased, which is due to a
high solid loading rate (SLR) into the system. TSS values in the effluent were due to biomass loss in the
beginning of the experiment. It was observed that the
sloughed biomass from the first run was reattached
on the biofilm. When the influent COD concentration
increased to 4382 mg l−1 , the performance of the reactor was reduced. This can also be seen in Table 3 where
effluent TSS increased to 186 mg l−1 due to insufficient
time for TSS hydrolysis and increase in biomass growth
rate as well as biomass sloughing proved by dropping
in DO level (5.9–1.32 mg l−1 ).
The influence of the solids loading rate on the
SBFFBR system performance is shown in Fig. 6. The
rate of TSS removal was strictly dependent on the SLR
and rises fairly uniformly over the SLR studied. The
percentage of TSS removal is fairly uniform at about
98% in the first and second points of SLR. However, it
dropped off to 85% when the SLRs increased to 0.81 kg
SS m−3 d−1 . It showed that such a SLR needs longer
time to be hydrolysed and biologically consumed.
A SBFFBR, which has been developed for this research,
has illustrated effective treatment of dairy processing
Table 3. SBFFBR effluent quality at HRT of 1.6 day.
Effluent characteristics
Influent waste strength
(mg l−1 ) (mg l−1 ) pH (NTU)
(mg COD l−1 )
Figure 5. COD removal efficiencies of SBFFBR.
 2009 Curtin University of Technology and John Wiley & Sons, Ltd.
Asia-Pac. J. Chem. Eng. 2009; 4: 698–703
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
appreciation goes to the industry for their full cooperation. Special thanks to the plant managers of the
National Food Milk Ltd for their excellent cooperation.
Figure 6. TSS removal efficiency and TSS removal rate.
wastewater. The COD removal efficiencies for a range
of OLRs have been investigated and an inverse relationship between OLR and COD removal efficiency was
observed. Conversely, a positive relationship between
OLR and COD removal rate was observed. At OLRs of
0.4 and 1.27 kg COD m−3 d−1 , higher COD removal
efficiencies of 97.2 and 96.6% were achieved, respectively. However, COD removal efficiency decreased to
89.3% when the influent COD increased to 4382 mg
l−1 with an OLR of 2.74 kg COD m−3 d−1 . The overall COD removal efficiency was 94%. Therefore, the
SBFFBR system could be attractive for use in the treatment of milk processing wastewater because of high
carbon removal efficiency and good effluent quality.
The authors would like to acknowledge the Libyan
Government for their financial support. The greatest
 2009 Curtin University of Technology and John Wiley & Sons, Ltd.
[1] M. Eddy. Wastewater Engineering: Treatment and Reuse,
McGraw-Hill: New York, 2003.
[2] S. Sirianuntapilboon, N. Jeeyachok, R. Larplai. J. Environ.
Manage., 2005; 76, 177–183.
[3] H. Xu, Q. Yu. Review of methods of biological treatment
of food processing wastewater. In Proceedings of 28th
Australian Chemical Engineering Conference, July 9–12 ,
Perth, Australia, 2000.
[4] M.E. Abdulgader, Q.J. Yu, P. Williams, A.A.L. Zinatizadeh.
A review of the performance of aerobic bioreactors for
treatment of food processing wastewater. In Proceedings of
1st Conferenceon Environmental Management, Engineering,
Planning and Economics, Greece, 2007; pp.1131–1136.
[5] A. Tawfik, M. Sobhey, M. Badawy. Desalination, 2008; 227,
[6] M. Gavrilescu, M. Macoveanu. Acta Biotechnol., 1999; 19,
[7] Q.J. Yu, H. Xu, D. Yao, P. Williams. Water Sci. Technol.,
2003; 47, 189–194.
[8] J.M. Garrido, F. Omil, B. Arrojo, R. Mendez, J.M. Lema.
Water Sci. Technol., 2001; 43, 249–256.
[9] J. Keller, S. Watts, W. Battye-Smith, R. Chong. Water Sci.
Technol., 2001; 43, 355–362.
[10] P.L.J. Grady, G.T. Daigger, H.C. Lim. Biological wastewater
treatment, 2nd edn, Marcel Dekker: New York, 1999.
[11] S.V. Mohan, N.C. Rao, K.K. Prasad, B.T.V. Madhavi, P.N.
Sharma. Process Biochem., 2005; 40, 1501–1508.
[12] A.M. Bandpi, H. Bazari. Iran. J. Environ. Health Sci. Eng.,
2004; 1, 65–69.
[13] E. Casey, B. Glennon, G. Hamer. Biotechnol. Bioeng., 1999;
62, 183–192.
[14] G.D. Najafpour, A.A.L. Zinatizadeh, L.K. Lee. Biochem. Eng.
J., 2006; 30, 297–302.
[15] A. Duarte, F.M.M. Oliveira. Environ. Technol. Lett., 1984; 5,
[16] S.A. Raj, D.V.S. Murthy. J. Environ. Sci. Health, 1999; A34,
[17] APHA, AWWA, WEF. Standard methods for the examination
of water and wastewater, 20th edn, American Public
Health Association/American Water Works Association/Water
Environmental Federation, Washington, DC, 1999.
[18] M. Rodgers, M. Zhan, M.D. Burke. Environ. Technol., 2004;
25, 211–218.
[19] G. Andreottola, P. Foladori, M. Ragazzi, R. Villa. Water Sci.
Technol., 2002; 45, 347–354.
Asia-Pac. J. Chem. Eng. 2009; 4: 698–703
DOI: 10.1002/apj
Без категории
Размер файла
175 Кб
batch, treatment, flexible, wastewater, biological, reactor, fibre, biofilm, sequencing, processing, milk
Пожаловаться на содержимое документа