Combined effects of EPS and HRT 
enhanced biofouling on a submerged and 
hybrid PAC-MF membrane bioreactor
Authors: Mohiuddin Md. Taimur Khan, Satoshi 
Takizawa, Zbigniew Lewandowski, M. Habibur 
Rahman, Kazuhiro Komatsu, Sara E. Nelson, Futoshi 
Kurisu, Anne K. Camper, Hiroyuki Katayama, & 
Shinichiro Ohgaki.
NOTICE: this is the author’s version of a work that was accepted for publication in Water 
Research. Changes resulting from the publishing process, such as peer review, editing, 
corrections, structural formatting, and other quality control mechanisms may not be reflected in 
this document. Changes may have been made to this work since it was submitted for publication. 
A definitive version was subsequently published in Water Research, 47, 2, February 2013.  
DOI#10.1016/j.watres.2012.10.048. 
Khan M,Takizawa S, Lewandowski Z, Rahman MH, Komatsu K, Nelson SE, Kurisu F, Camper 
AK, Katayama H, Ohgaki S, "Combined effects of EPS and HRT enhanced biofouling on a 
submerged and hybrid PAC-MF membrane bioreactor," Water Research February 2013 
47(2):747-757. 
  Made available through Montana State University’s ScholarWorks 
     scholarworks.montana.edu 
Combined effects of EPS and HRT enhanced biofouling on 
a submerged and hybrid PAC-MF membrane bioreactor
Mohiuddin Md. Taimur Khan a,b,*, Satoshi Takizawa b, Zbigniew 
Lewa  Kazuhiro Komatsu e, Sara E. Nelson 
c, Fu e K
Ohgaki e
a Center for Molecular Discovery, University of New Mexi
b Department of Urban Engineering, University of Tokyo, T
c Center for Biofilm Engineering, Montana State University, 
d Department of Civil Engineering, Bangladesh University o
e National Institute for Environmental Studies, 16-2 Onogaw
Keywords:
Biofouling
Microfiltration membrane 
Powdere
Extracel
Hydrauli
Natural o
A B S
The g tify T), 
organi ents ms 
on the hollo ted 
carbon eat 
surfac nic 
carbon (þ) 
manno se), 
protein ent 
reacto nd 
remov ng. 
However, HRT influenced the length of the filtration cycle and had less effect on organic carbon and EPS 
compo as 
more t  on 
memb t a 
potential adsorbent of carbohydrates. The abundance of N-acetyl-D-galactosamine and D-galactose was the 
highest in the foulants on membranes receiving biofilter-treated river water. Most of the biological fouling 
compounds were produced inside the reactors due to biodegradation. PAC inside the reactor enhanced the 
biodegra-dation of polysaccharides up to 97% and that of proteins by more than 95%. This real-time extensive 
and nonent removal and/or biodegradation. The abun-dance of carbohydrates in the foulants on MF surfaces w
han 40 times higher than in the bulk phase, which demonstrates that the accumulation of carbohydrates
rane surfaces contributed to the increase in transmembrane pressure significantly and PAC was no (PAC)-MF membrane bioreactor (MBR). The reactors were operated continuously for 45 days to tr
e (river) water before and after pretreatment using a biofiltration unit. The real-time levels of orga
 and the major components of EPS including five different carbo-hydrates (D(þ) glucose and D
se, D(þ) galactose, N-acetyl-D-galactosamine and D-galactose, oligosaccharides and L() fuco
s, and polysaccharides were quantified in the influent water, foulants, and in the bulk phases of differ
rs. The presence of PAC extended the filtration cycle and enhanced the organic carbon adsorption a
al more than two fold. Biological filtration improved the filtrate quality and decreased membrane foulid activated carbon 
lular polymeric substances 
c retention time
rganic matter
T R A C T
oal of this study was to quan
c carbon and various compon
 performance of submersed vel study demonstrates that the PAco, Albuquerque, NM 87131, USA
okyo 113-8656, Japan
Bozeman, MT 59717, USA
f Engineering & Technology, Dhaka 1000, Bangladesh
a, Tsukuba, Ibaraki 305-8506, Japan
and demonstrate the dynamic effects of hydraulic retention time (HR
of extracellular polymeric substances (EPS) produced by microorganis
w-fiber microfiltration (MF) membrane in a hybrid powdered activatoshi Kurisu b, Ann . Camper c, Hiroyuki Katayama b, Shinichiro 
ndowski c, M. Habibur Rahman d,C-MF hybrid MBR is a sustainable technology for treating river water.
wastewater (Kim et al., 2006). To further improve this process,results from wastewater treatment showed that PAC dosing
resulted in lower concentrations of soluble EPS and colloidal
2.1. Site location1. Introduction
In the past decades,microfiltration (MF) has emerged as one of
themost reliable, cost-effective and sustainable unit processes
for surface water treatment separating macromolecules,
bacteria and discrete particles (Lebeau et al., 1998; Khan et al.,
2001, 2011; Fabris et al., 2007). In this process, biofouling has
been identified as the key factor affecting long-term MF
membrane separation processes and worsening overall plant
performance. The biofouling process initiated by bacterial
attachment to the surface of amembrane gradually reduces its
flux andmay lead to structural failure of themembrane (Khan
et al., 2010). Bacteria imbedded in extracellular polymeric
substances (EPS) form a highly hydrated gel layer on the MF
membrane surface that affects its performance in severalways
(Jarusutthirak and Amy, 2006; Fonseca et al., 2007).
EPS are large molecular weight compounds excreted by
bacteria, such as polysaccharides, proteins, lipids and DNA
(Barker and Stuckey, 1999;Wingender et al., 1999; Barker et al.,
2000; Ye et al., 2005). These substances affect the development
of biofouling deposits on membranes and alter physicochem-
ical characteristics of themembrane such as electrical charge,
hydrophobicity, and chemical properties (Nagaoka et al., 2000;
Go´mez-Sua´rez et al., 2002; Mikkelsen and Keiding, 2002; Khan
et al., 2010, 2011). Moreover, polysaccharide and metaleion
specific interactions have a major effect on the mechanical
stability of biofoulants (Turakhia and Characklis, 1989). The
components of EPS also create scaffolds with physical char-
acteristics and interconnected pore structures that promote
membrane flux decline and further cell attachment, prolifer-
ation and differentiation (Chen and Ma, 2004; Wotton, 2004;
Park et al., 2005). Although the initial adhesion mechanism is
still not entirely clear, it is known that over time the bonding
strengthof theEPS to themembrane increases,mostlybecause
of the flexibility of the gel layer and the cross-linking of the EPS
on themembrane surface (Tansel et al., 2006). In general,when
the EPS concentration increases, cell adhesion is enhanced by
polymeric interactions (Tsuneda et al., 2003).
To counteract the effects of membrane fouling, mainte-
nancebyperiodicchemical cleaningormembrane replacement
needs tobecarriedoutwhichwill inevitablyaccruehigher costs
of the MBR process (Ng et al., 2006). Therefore, operational
adjustments that can reduce the effects of foulingwill increase
productivityofmembraneprocesses. Theprocessof adsorption
using powdered activated carbon (PAC) can be hybridized with
theMFmembranes to remove organic compounds (Basar et al.,
2004; Thiruvenkatachari et al., 2006), disinfection byproducts
(Khan et al., 2009), 17 b-estradiol (an endocrine disrupting
compound) (Lee et al., 2009), microorganisms like Norovirus
(27e40nmindiameter) (Ohet al., 2007), atrazine (Jia et al., 2009),
and other substances. In our previous studies (Khan et al., 2001;
Kim et al., 2005) on PAC-MF systems for surface water treat-
ment, in which the PAC dose was varied from 0 to 50 g/L of the
reactor, we observed that at 40 g of PAC/L of the reactor, the
membrane fouling frequency was the lowest. Experimentaltotal organic carbon (TOC) in the PACeMBR sludge (Yang et al.,
2010). Such a hybrid process improvesmembrane performancebetter and important understandings of the differences in
biofouling between MF filtration with and without PAC are
needed. One of themain factors affecting biofouling, the fate of
EPS in hybrid MF MBRs, is not well understood and explored.
Furthermore, biofiltration pretreatment of surface water also
improves PAC-MF system performance (Kim et al., 2005; Khan
et al., 2009), but its role in EPS removal from surface water has
not been demonstrated yet.
It was further demonstrated that a decrease in hydraulic
retention time (HRT) enhanced growth of biomass and accu-
mulation of soluble microbial products (SMP), which acceler-
ated membrane fouling rate (Huang et al., 2011). Other work
showed that low HRT caused excessive growth of filamentous
bacteria which resulted in high EPS concentration, high solid
concentration and sludge viscosity (Meng et al., 2007). We
hypothesized that PAC particles inside a MF membrane
reactor would adsorb and enhance the biodegradation of EPS
regardless of whether the EPS was from the influent river
water or produced inside the reactor as well as reduce the
membrane cleaning frequency. The presence of PAC would
interfere with the biofouling process and increase the filtra-
tion period at higher HRT; however, this presence and
pretreatment of river water would not affect the number of
filtration cycles for the membrane modules operated at
reduced HRT. In this study, we monitored the concentrations
of the major EPS components: carbohydrates, proteins and
polysaccharides, and their effects on MFmembrane fouling in
the presence and absence of PAC and biofiltration pretreat-
ment of surface water at two different HRT.
Bench-scale experiments on a PAC-MF MBRs were carried
out using settled surface water (Tama River, Tokyo, Japan)
either before or after treatment by a biofilter system,
depending on the experimental protocol. More than 80% of the
total flow of this river consists of the treated effluent from
wastewater treatment plants located upstream. As a result,
dissolved organic matter (DOM) and other biodegradable
contaminants in the river water exist at much higher
concentrations than in typical surface waters. Implementing
the PAC-MF MBR process may improve the quality of the
treated water considerably. The objectives of this study were
to: (1) quantify the effect of HRT in the presence and absence
of PAC and biofiltration pretreatment of surface water on MF
membrane performance, and (2) quantify the kinetics of real-
time adsorption and/or biodegradation of EPS components
and their effects on MF membrane fouling at varied opera-
tional conditions in the PAC-MF MBR.
2. Experimental sectionand extends the filtration periods, likely because the presence
of PAC interferes with the process of fouling. Due to the pres-
ence of PAC, the total cost formembranemaintenance could be
reduced by 25% in a PACeMBR system treating municipalThe Tokyo Metropolitan Authority has a water treatment
plant to treat water from the Tama River, which is located in
the southwest region of Tokyo, around 15 km from the main
city. The treated water is supplied only for industrial use. The
bench-scale experiments on the submerged PAC-MF MBRs
were performed in this treatment plant. River water was
pumped from the intake point to a series of primary and
secondary sedimentation ponds. A stream of water taken
from the secondary pondswas split into four streams; three of
these streams were used directly as feed into three reactors,
while the fourth stream was directed after pretreatment by
a biofilter into two reactors. The reactor set-up used in this
study is shown schematically in Fig. 1.
2.2. Experimental design and operational conditions of
the reactors
Five bench-scale reactors (Table 1) were operated continu-
ously for 45 days. Each reactor consisted of a hollow-fiber MF
membrane module operated in suction mode to maintain
constant flux. The biofilter consisted of a column packed with
polypropylene pellets with a length of 5 mm, an inner diam-
eter of 3 mm and an outer diameter of 4 mm. The filtration
velocity and retention time of this biofilter were 320 m/d and
w10 min, respectively. Discharges from the biofilter and raw
water were stored separately in two 100-L reservoirs and
stirred continuously prior to being fed into the reactors. The
Milli-Q water when received and soaked in freshMilli-Q water
prior to use. The experiment was carried out under ambient
conditions (17e28 C).
Reactors R1eR3 received settled river water, while reactors
R4 and R5 received the biofilter system effluent. PAC (coconut
shell origin, JWWA K 113-1985, Shirosagi-C, Takeda Chemical
Co., Japan) was used as received and the dose was 40 g of PAC/
L of the reactor was added into reactors R2eR5 at the begin-
ning of operation. Reactor R1 was operated without PAC as
a control. No modifications were made to the method of
operation, and no additional PAC was added.
While maintaining the same filtration flux, the number of
membrane fibers in the reactors operated at lower HRT (1.2 h)
was twice higher (640 fibers) than that of those operated at
higher HRT. It is not possible to calculate the effective
membrane surface areas during each suction and back-
washing cycle; therefore, it was assumed that the entire
membrane surface area was effectively used in the process.
Aeration (see Fig. 1) was maintained underneath the fiber
module to disturb cake formation on the membrane surface
and to prevent rapid flux decline of the submergedmembrane
modules. The airflow rate underneath the module that con-
tained a higher number (640 fibers) of membrane fibers (R3
and R5) was twice than that of the other modules with lower
numbers (320 fibers) of fibers (R1, R2 and R4) (see Table 1).
G ispolyethylene hydrophilic membrane used in this study was
made by Mitsubishi Rayon Co. Ltd., Japan. The membrane
properties, and reactor configurations are described else-
where (Khan et al., 2009, 2011). Briefly, the nominal pore size
was 0.1 mm, and the outer and inner diameters were 0.41 mm
and 0.27 mm, respectively. The reactors were made of 5-mm-
thick polyvinyl chloride plates, and each had an effective
volume of 5 L. The membrane modules were cleaned with
Fig. 1 e Schematic diagram of a single PAC-MF unit. Here, Pelectric valve for retardation of water during backwash, M2 the
circulation, SP the suction pump (connected to controller) and BLevel sensors were used to ensure a constant reactor
volume, and suction through the membrane module main-
tained a constant flow rate. All systems were equipped with
a backwash mechanism, providing 2 min of backwash after
every 20 min of filtration. Fouled membranes were back-
washed with the stored filtrate (Fig. 1). The backwash-water
reservoirs were cleaned once a week and refilled with fresh
filtrate, and the backwashing flux was twice as high as the
the pressure gauge (connected with computer), M1 theelectric valve for backwash, M3 the electric valve for water
P the backwash pump (not connected to controller).
5 d
n o
ack
me
20 a
20 a
20 a
20 a
20 afiltration flux. Transmembrane pressure (TMP) and back-
washing pressure (BWP) were automatically recorded using
a data logger. Once TMP values reached to 50 kPa (0.5 bar), the
fouled membrane modules were removed from the reactors
and washed with the Milli-Q water using a soft brush to
remove the foulants from the membrane surfaces. Pure water
flux was measured after membrane cleaning.
2.3. Analytical techniques
2.3.1. Measurement of total organic carbon (TOC) and
dissolved organic carbon (DOC)
To measure the TOC and DOC, the influents and effluents of
each reactor were homogenized (Branson Sonifier 450,
Yamato, Japan) for 10 min using specific protocols developed
in our previous study (Khan et al., 2009). The homogenized
samples were filtered through 0.45 mm pore size poly-
vinylidene fluoride (PVDF, Millipore, USA) filters for DOC
Table 1 e Operating conditions of various reactors during 4
was 20.8 L/m2/h for all reactors.
Reactor PAC
(g/L of the reactor)
Influents Duratio
and b
ti
R1 0 River water
R2 40 River water
R3 40 River water
R4 40 Biofiltered water
R5 40 Biofiltered watermeasurement. A total organic carbon analyzer (TOC-5000A,
Shimadzu Co., Osaka, Japan) was used to measure TOC and
DOC of the samples.
2.3.2. Extraction of EPS from reactor and membrane samples
EPS extracted from the bulk liquid and from the foulants on
membrane surfaces were stained with lectins specific to
particular carbohydrate moieties. The EPS extraction protocol
used here is modified from Zhang et al. (1999). They extracted
EPS in various ways from different activated sludge samples
and proposed a method for extracting EPS based on the
highest recovery of carbohydrates, proteins and DNA. We
introduced a newmethod of EPS extraction from sampleswith
low to very high concentration of inert particles, especially
PAC. Several trialswere done to obtain themaximum recovery
of the target materials. To extract the EPS, the samples (fou-
lants on themembrane surfaces, bulk samples in reactors and
influents to the reactors) were passed through 0.45 mm pore
size PVDF membrane (Millipore, USA) separately and then,
10 g of retained sediment on this membrane were placed in
a centrifuge tube and suspended in 25 ml of Milli-Q water,
shaken and then centrifuged at 3500 rpm for 10 min. Thesupernatant was decanted and set aside. The pellet remaining
in the centrifuge tube was re-suspended in 25 ml of 8.5% NaCl
and 0.22% formaldehyde. The mixture was vortexed at high
speed for 1 min to recover the capsule-bound material. The
previously retained supernatant was then added, and 15ml of
this mixture (now 50 ml plus original sediment sample) was
centrifuged at 12,000 rpm for 30 min. The supernatant was
collected and filtered through a 0.2 mm cellulose acetate filter
(Millipore, USA). Activated sludge (2 g) sample from the Shi-
nagawa Wastewater Treatment Plant (Japan) was used as
a positive control and EPS from this sludge sample were
extracted using the above protocol.
2.3.3. Target lectins
The lectins used in this study were: (1) D(þ) glucose and D(þ)
mannose (fluorescein isothiocyanate (FITC) fluorochrome); (2)
D(þ) galactose (FITC fluorochrome); (3) N-acetyl-D-galactos-
amine and D-galactose (FITC fluorochrome); (4) L() fucose
ays of filtration experiment. The membrane filtration flux
f filtration
washing
(min)
Aeration
(L/m3/min)
HRT
(h)
Operational
objective
nd 2 1000 2.4 Control (no PAC)
nd 2 1000 2.4 Standard operation
with PAC and
river water
nd 2 2000 1.2 Effect of HRT on
standard operation
nd 2 1000 2.4 Effect of pretreatment
of river water on
standard operation
nd 2 2000 1.2 Effect of HRT and
pretreatment of river
water on standard
operation(tetramethylrhodamine isothiocyanate (TRITC) fluoro-
chrome); and (5) oligosaccharides (TRITC fluorochrome) (Sig-
maeAldrich Co., USA). Each lectin corresponds to a specific
carbohydrate. The lectins used in this study were selected/
considered after detailed investigations (data not shown)
based on the abundance and availability of the corresponding
carbohydrates in the surface water sources and wastewater
systems. The lectin powders were diluted to 0.2% with 1 N
polyphosphate buffer (PBS) solution (8 g/L NaCl, 1.1 g/L
Na2HPO4 e anhydrous, 0.2 g/L KCl, and 0.2 g/L KH2PO4). The
dilute lectins were stored at 20 C and covered with
aluminum foil to protect them from light exposure.
2.3.4. Labeling of different lectins with extracted EPS samples
5-ml of the extracted EPS sample was placed drop-wise on
a glass slide and oven-dried at 90e95 C for 5 min. A labeled
lectin solution (10 ml) was then placed on top of the dried
droplet at room temperature. The labeled lectin was allowed
to react with the corresponding carbohydrate for 10min in the
dark, after which the slide was washed very gently with Milli-
Q water and then shaken until air-dried. One drop of SLOW
FADE (Molecular Probe, USA) was placed on top of the dried
samples at 490 nm was compared to the standard curve. The
performance. The addition of PAC reduced the membrane
resistance to filtration, which was reduced further in the
presence of a biofilter as a pretreatment (Khan et al., 2011).
However, in present experiment, a thick cake layer was not
formed on the outer surface of any of the fouled membranes.
During physical cleaning, very thin (0.4e0.9 mm thick) foulant
layerswere formed on themembrane surfaceswhere aeration
could reach and thicker cake (w3.5e5mm thick) was observed
to have accumulated primarily between membrane fibers at
the ends of the fibers, where aeration could not reach. Due to
the turbulence produced by aeration, the attached foulants to
the membrane surfaces were scoured continuously back into
the bulk solution of the reactors and the mass transport
through themembrane surfaceswere also enhanced (Chu and
Li, 2005; Sun et al., 2008). It is likely, then, that aeration
significantly contributed to the removal of cake fouling in this
system.
The influents to reactors R3 and R5 were different;
however, the HRT, filtration-backwashing cycle and PAC
loading conditions were similar in both reactors (Table 1).
Therefore, the reduction of HRT affected the membrane
fouling rate even though the aeration rate was doubled in
these two reactors compared to other reactors (R1, R2 and R4).
We did not operate a control reactor (rawwater and no PAC) at
lower HRT in this study; therefore, it is not possible to
compare the effect of aeration onmembrane fouling at higher
and lowerHRT. However, in previous reports (Ueda et al., 1997;
Khan et al., 2009), it was observed that aeration significantlyspectrophotometer U-2010 (Hitachi Co., Japan) was used to
measure the absorbance of the samples.
3. Results and discussion
3.1. TMP of membrane modules and effects of HRT on
membrane fouling
The transmembrane pressure (TMP) of membrane modules
inside all reactors increased with time (Fig. 2). In the case of
reactors operated at higher HRT (2.4 h), the system treating
raw water without PAC (reactor R1) accumulated TMP more
rapidly than the other two systems (reactors R2 and R4). The
number of cleaning cycles associated with membrane fouling
during the filtration period (45 days) was the highest (4
cleaning cycles) in reactor R1; however, that was lowered to
thrice in the presence of PAC in reactor R2, which was further
reduced to twice in the presence of PAC and biofilter-treated
raw water in reactor R4. However, the other reactors (reactors
R3 and R5) operated at lower (1.2 h) HRT and in the presence of
PAC had rapid TMP increase and required more cleaning
cycles (total of 4).
Interestingly, reactors R3 and R5 were operated in the
presence of PAC and the influents to the reactors were
different (settled riverwaterwithout andwith biofilter-treated
water, respectively). The decrease of HRT increased the
frequency of membrane cleaning in both reactors.
Prior work with PAC and microfiltration (Khan et al., 2011)
showed that cake and gel layer formation are the primarysample, and a cover slip was placed on top of it. Each lectin
labeled sample was observed under an epifluorescence
microscope (Nikon, Eclipse E 800, Japan) using a 100 objec-
tive. Several images (8e10 images) were captured for each
sample labeled with different lectins. FITC was excited at
494 nm, and TRITC was excited at 550 nm. The gain and offset
for each photomultiplier were adjusted to optimize lectin
detection. Stored images were analyzed in Photoshop (version
5.0) software to compute the light intensity (green from FITC
fluorochrome and red from TRITC fluorochrome), and a mean
value of light intensity for each lectin was estimated and
averaged.
2.3.5. Protein and polysaccharides measurements
The extracted EPS from different samples (foulants on the
membrane surfaces, bulk samples in reactors and influents to
the reactors) were used in the protein and polysaccharide
assays. The total protein was further extracted from EPS
samples using the Bicinchoninic Acid (BCA) protein assay
reagent kit (Pierce, BCA protein assay reagent kit, 23225).
Protein standards were prepared in the range of 0e50 mg/L
with bovine serum albumin (BSA) from 2 mg/ml of albumin
supplied with this kit. The enhanced protocol (Pierce,
instructions) was followed. The absorbance of the cooled
samples at 562 nm was compared to the standard curve. The
polysaccharide in the extracted EPS was measured using the
phenolesulfuric acid method. The absorbance of cooledmechanisms ofmembrane fouling: the adsorbed and attached
materials on the PAC were responsible for the loss in0
10
20
30
40
50
0 5 10 15 20 25 30 35 40 45 50
Filtration period (days)
T
r
a
n
s
m
e
m
b
r
a
n
e
 
p
r
e
s
s
u
r
e
 
(
k
P
a
)
R1
R2
R3
R4
R5
R1
R2
R3
R4
R5
I II III IV
I II III
I II III IV
I II
I II III IV
Fig. 2 e Transmembrane pressure (TMP) record for the MF
membrane modules in the reactors. One kPa [ 0.01 bar.
The Roman numerals at the top of the figure indicate the
number of cleanings for each module. The membranes
inside R1, R3 and R5 fouled 4 times; however, those inside
R2 and R4 fouled thrice and twice, respectively. The HRT of
reactors R1, R2 and R4 was kept at 2.4 h and that of reactors
R3 and R5 was kept at 1.2 h.reduced the fouling frequency of the membrane modules
when they were operated at similar HRT.
Based on our pervious study (Khan et al., 2009) and
microscopic observation, we noticed that during filtration,
suspended particles (including PAC in reactors R2 through R5)
inside the reactors accumulated on the membrane surfaces
and then were dispersed inside the reactors during back-
washing; i.e. no irreversible fouling effect. However, in this
study, we noticed significant irreversible fouling effects on the
membranes operated at lower HRT (reactors R3 and R5). After
physical cleaning of the membrane modules inside reactors,
the TMP started from higher values in the filtration cycle than
that in the previous cycle of operation especially R3 and R5.
This observation suggests that HRT is one of the most
important parameters during system (PAC-MF MBR) design.
This irreversible fouling effect also correlates with the
backwashing pressure (BWP) record of the membrane
modules (Fig. 3). The reactors were operated at a constant flux
(20.8 L/m2/h). During suction, the foulants were attached on
the surface of themembrane and backwashing removed some
of the foulants from the membrane pores and their
surrounding areas. However, the amounts of the foulants
removed during backwashing were less than the accumula-
tion during suction; therefore, TMP increased (Fig. 2) during
each filtration cycle and finally cleaning of membranes was
required. The BWP of membrane modules in reactors R3 and
R5 operated at lower HRT was the highest (>40 kPa) compared
to others at the end of each filtration cycle. In this study, the
backwashing flux in all reactors was kept similar to the
Interestingly, in our previous study (Khan et al., 2011), we
doubled the backwashing flux of the modules and irreversible
fouling effects were negligible.
3.2. The effect of HRT on organic carbon removal
The average TOC concentrations were similar to those in the
influents of raw and biofiltered water (2.35  0.25 mg/L in and
2.10  0.38 mg/L, respectively). During this study the bio-
filtration unit (used for the pretreatment of raw water)
removed only w10% of the organic carbon. In our previous
study (Khan et al., 2011) we demonstrated that the biofiltration
unit removed a significant amount (w80%) of biomass and
inert solids from the raw water and the organic carbon was
adsorbed by the biomass in the biofilter and finally bio-
degraded. Furthermore, the organic matter sloughed from the
biofiltration unit could potentially increase the organic carbon
concentration in its effluent.
R1 R2 R1 R4 R5filtration flux (20.8 L/m2/h). Due to reduced HRT, the BWP of
themodules inside these two reactors was higher than that in
other reactors operated at higher HRT, even though the
aeration rate was doubled inside the reactors R3 and R5.
0
10
20
30
40
50
0 5 10 15 20 25 30 35 40 45 50
Filtration period (days)
B
a
c
k
w
a
s
h
i
n
g
 
p
r
e
s
s
u
r
e
 
(
k
P
a
) R1
R2
R3
R4
R5
R1
R2
R3
R4
R5
I II III IV
I II III
I II III IV
I II
I II III IV
Fig. 3 e Backwashing pressure (BWP) record for the MF
membrane modules in the reactors. One kPa [ 0.01 bar.
The Roman numerals at the top of the figure indicate the
number of cleanings for each module. The membranes
inside R1, R3 and R5 fouled 4 times; however, those inside
R2 and R4 fouled thrice and twice, respectively. The HRT ofreactors R1, R2 and R4 was kept at 2.4 h and that of reactors
R3 and R5 was kept at 1.2 h.Reactor effluents
Fig. 4 e Removal efficiency of total organic carbon (TOC) by
the membrane reactors operated with and without PAC atThe removal efficiency of TOC in reactor R1 (system oper-
ated at higher HRT and treated raw water without PAC) was
the lowest (w30%) (Fig. 4). However, other reactors operated at
either higher or lower HRT that received different influents (R2
and R3 received raw water and R4 and R5 received biofiltered
water) with the addition of PAC showed >66% TOC removal.
No significant difference in TOC removal between these
reactors (R2eR5) was observed. It appears that TOC removal
efficiency is related to the presence of PAC and this efficiency
was not affected by the HRT and type of source water (influ-
ents to the reactors). According to the TOC mass balance, the
total amounts of organic carbon that entered into the reactors
receiving raw water (reactors R1eR3) and biofiltered water
(reactors R4 and R5) were 2.64 g and 2.36 g, respectively, during
the filtration period.
The reactors operated with PAC could remove or retain
>66e71% of the TOC that entered in the system; however,
<30% of the TOCwas removed by the reactor without PAC (R1)
(Fig. 4). The remaining TOC in R1 was either degraded inside
the reactor (probably biodegraded by the suspended biomass
and/or attached biofilm on the membrane surfaces) and/or
0
10
20
30
40
50
60
70
80
A
v
e
r
a
g
e
 
T
O
C
 
r
e
m
o
v
a
l
 
e
f
f
i
c
i
e
n
c
y
 
(
%
)different HRT. The HRT of reactors R1, R2 and R4 was kept
at 2.4 h and that of reactors R3 and R5 was kept at 1.2 h.
remained as adsorbed phases and therefore could be accu-
mulated over time. However, the remaining >66% of TOC
inside the reactors operated with PAC (R2eR5) was mostly
adsorbed on the PAC and then remained in as the attached
phase or was degraded by the microorganisms on the PAC or
in the bulk phase (Vigneswaran et al., 2006). The DOC levels
inside all reactors containing PAC were similar and did not
vary much during the filtration period (see Supplementary
Fig. 1). However, in the reactor without PAC (R1), the DOC
level increasedwith time of filtration and this level was higher
(wthrice) than that in other reactors. We observed significant
differences in DOC level between the reactors operated with
and without PAC. This clearly indicates that PAC not only
adsorbed but also maintained the level of TOC inside the
system; however, the PAC has finite adsorption capacity
(Khan et al., 2009).
3.3. Influence of carbohydrates on membrane
performance
Similar images were obtained for the other lectins and
their corresponding mean values of light intensity were
calculated. Because each fluorescent lectin has different light
scattering and absorption properties, it was not possible to
compare intensities among lectins in amanner that quantifies
the carbohydrate content. Comparisons were performed only
among samples labeled with the same lectin. Fig. 6 illustrates
the average light intensity of fluorescent N-acetyl-D-galac-
tosamine and D-galactose in the foulants on the all
membranes surfaces at the end of each fouling cycle and
activated sludge (positive control sample). The level of this
lectin increased with time inside reactor R5; however, this
level did not increase in reactor R4. Both were operated with
biofiltered water in the presence of PAC, but the HRT was
different. Therefore, the shorter HRT (reactor R5) enhanced
the accumulation of carbohydrates detected by this specific
lectin and potentially contributed to more fouling of the
membrane module. In membrane foulant samples at the end
of each fouling cycle for other four lectins, the highest light
) D(Lectins are proteins/glycoproteins with at least one non-
catalytic domain binding reversibly to specific mono-
saccharides or oligosaccharides without having any
enzymatic activity toward their carbohydrate ligands
(Goldstein et al., 1980; Kocourek and Horejsı´, 1981). By binding
to carbohydrate moieties on the cell surface, lectins partici-
pate in a range of cellular processes without changing the
properties of the carbohydrates involved (Lam and Ng, 2011;
da No´brega et al., 2012). The lectins used in this study are from
plant proteins. Fig. 5 shows the representative abundance of
different fluorescent lectins binding with the sample from
reactor R1. The abundance ofN-acetyl-D-galactosamine and D-
galactose was the highest and that of D(þ) galactose was the
lowest. In general, the abundance of other fluorescent lectins
is also the highest in the reactor operated without PAC, (R1,
data not shown) and this may indicate that the presence of
PAC changes the carbohydrate composition due to different
biochemical reactions/interaction in the presence of micro-
organisms (Jang et al., 2007).
Fig. 5 e Abundance of (A) D(D) glucose and D(D) mannose; (B(D) L(L) fucose; and (E) Oligosaccharides in the bulk phase of R1
membrane module in reactor R1.intensities were also observed in reactor R1 (raw water
without PAC) (Supplementary Table 1).
Except for L() fucose, the fluorescent intensities among
the four lectins in the membrane foulants of reactor R1 were
equal to or greater than those of the activated sludge (positive
control) sample. This observation corroborates with the TOC
adsorption data. The abundance of D(þ) glucose and D(þ)
mannose was higher in the reactors that received biofiltered
water (R4 and R5). In the reactors (R2 and R4) operated at
higher HRT and with PAC, the levels of lectins (fluorescent
intensities) were lower, but still higher than those in the
reactors were operated at lower HRT (R3 and R5) even in the
presence of PAC.
The average mean values of fluorescent light intensities of
lectins that detect all carbohydrates in both raw and bio-
filtered water were less than 1.0; however, those in the fou-
lants on membranes were 40e60 times higher than the
samples in bulk phase (Supplementary Table 2) of the reac-
tors, which supports the idea that the accumulation of
carbohydrates on membrane surfaces contributed to flux
D) galactose; (C) N-acetyl-D-galactosamine and D-galactose;(raw without PAC) on Day-10, the first fouling day of MF
3.4. Effects of polysaccharides and proteins on
membrane performance
The biofilter pretreated surface water contained more poly-
saccharides and proteins than the raw water (Fig. 7), which
indicates that the captured suspended solids inside the bio-
filter were biodegraded due to the microbial activities inside
the media and more polysaccharides and proteins were
produced. The average concentration of polysaccharides in
the foulants on the membrane surfaces in reactor R1 was 4
times higher than that in the bulk phase. Furthermore, the
polysaccharide concentration in the raw water was less than
2.0 mg/L.
During the filtration period,w410 mg/L of polysaccharides
entered into the reactor R1 andw7% of those polysaccharides
were present inside this reactor and in the foulants of
membrane surfaces on the last membrane cleaning day. This
indicates thatw93% of the polysaccharideswere biodegraded.
However, the presence of PAC in other reactors (R2eR5)
enhanced this degradation further by up to 97%. This degra-
dation of polysaccharides was not affected by the HRT. Poly-
saccharide is responsible for the evolution of irreversible
fouling (Kimura et al., 2004). The reactors operated at lowerdecline and TMP increase. Lee et al. (2001) reported that
despite the similar characteristics of soluble fractions in the
reactors, the membrane fouling rate due to attached growth
system was about 7 times higher than that of the suspended
0
10
20
30
40
50
60
70
1st 2nd 3rd 4th
Number of fouling in each series
t
h
g
i
l
f
o
e
u
l
a
v
n
a
e
m
e
g
a
r
e
v
A
i
n
t
e
n
s
i
t
y
 
on R1 on R2 on R3 on R4 on R5
Activated sludge 
sample
Fig. 6 e Average mean values of light intensity of N-acetyl-
D-galactosamine and D-galactose in the foulants on
membrane surfaces at the end of each filtration cycle. Error
bars show the ±standard error of the mean (SEM) was
determined in triplicate. The membranes inside R1, R3 and
R5 fouled 4 times; however, those inside R2 and R4 fouled
thrice and twice, respectively. The dotted line represents
the average mean values of light intensity of N-acetyl-D-
galactosamine and D-galactose in activated sludge sample
(35.13 ± 3.65) as positive control.growth system. This corroborates the effect of carbohydrate
accumulation on membrane performance.
The fluorescent light intensity levels of all lectins in the
bulk phases increased with time. However, the differences of
the fluorescent light intensity levels of these lectins in the
bulk phase and in the foulants of membranes indicate that
due to the biological activities inside the bioreactor, carbo-
hydrates not only accumulated, but also were produced inside
the systems. Doume`che et al. (2007) observed that fluorescent
lectins were conjugated or located in both microcolonies and
microbial cells, which indicates a high degree of spatial
organization and heterogeneity of carbohydrates within the
foulants. The TMP records (Fig. 2) indicate that the presence of
PAC and BF pretreatment of the surface water decreased TMP
during filtration at higher HRT. Therefore, we hypothesize
that the long chain carbohydrates act as bonding agents
among the particles and other foreign dissolved and sus-
pended matters, which causes stronger bridges among the
available membrane surfaces and also the particles, resulting
in increased TMP (Flemming, 1993). The amount of PAC in the
bulk phase was much higher than that in the foulants on the
membrane (Khan et al., 2011) and 40e60 times higher fluo-
rescent light intensities of lectins in the foulants on
membrane surfaces than that in bulk phase indicates that PAC
is not a good adsorbent for carbohydrate. Once the PAC and
other particles were attached to the membrane surface
forming foulant layers, higher level of carbohydrate was
observed because of their bonding properties to other
particles.HRT (reactors R3 and R5) suffered from the irreversible fouling
effects, which indicate that polysaccharide did not cause any
significant effects on the irreversible fouling in this
submerged PAC-MF MBR once a major portion of them were
biodegraded.
In our previous study (Khan et al., 2011), we observed that
the bacterial concentration was 2e4 times higher in the
reactors with PAC than those without PAC and most of
bacteriawere attached to PAC and other particles. Similarly, in
this study the higher number of bacteria attached to PAC and
other particles inside the reactors R2eR5 was probably
responsible for higher biodegradation rate of polysaccharides
0
5
10
15
20
25
30
35
40
45
R1 R2 R3 R4 R5 Raw
water
BF water
Reactors and influents to the reactors
s
n
o
i
t
a
r
t
n
e
c
n
o
c
n
i
e
t
o
r
p
d
n
a
e
d
i
r
a
h
c
c
a
s
y
l
o
P
(
m
g
/
L
)
Polysaccharide inside reactors and influents
Polysaccharide inside foulants
Protein inside reactors and influents
Protein inside foulants
Fig. 7 e Concentrations of polysaccharides and total
protein in the influents to the reactors and in the foulants
on the MF membrane surfaces on the last membrane
cleaning day (Day-45 for R1, R4 and R5 and Day-42 for R2
and R3). Error bars show the ±standard error of the mean
(SEM) was determined in triplicate. The membranes insideR1, R3 and R5 fouled 4 times; however, those inside R2 and
R4 fouled thrice and twice, respectively.
lower HRT, was almost twicemore than that inside reactor R4,
membrane biofoulant samples was developed. The effect ofwhich was operated at higher HRT. Therefore, HRT played
a significant role in the accumulation of polysaccharide in the
foulants. A similar trend was observed in the case of R2
and R3.
Proteins in the bulk fluid of the reactors accumulated
relatively steadily, starting at a low initial value. During the 45
days operational period,w730mg/L of protein from rawwater
andw952 mg/L of protein from biofiltered water entered into
the respective reactors. Biofiltered water had more protein
than raw water, which indicates that biodegradation of
different compounds and SS took place inside this biofilter
and more protein was produced by this system.
The influents to the reactors carried low concentration of
protein, but inside the reactors and in the foulants of all
membrane modules, the amount of total protein was 5e10
times higher. This suggests that microbial degradation and
consecutive accumulation of proteins from the source waters
resulted in substantial loading of protein to the membrane
reactors. The amounts of protein degraded and/or utilized by
the microorganisms inside reactors R1eR3 were 89%, 93% and
91%, respectively and those inside reactors R4 and R5 were
95% and 94.5%, respectively. Reactors R1eR3 received raw
water and with the exception of R1, the other two reactors
contained PAC. The PAC adsorbs higher numbers of bacteria
(Khan et al., 2011); therefore, enhanced biodegradation is
likely to have taken place in reactors R2 and R3. Moreover,
reactor R3 was operated at lower HRT and had slightly lower
protein degradation. Even though biofiltered water had higher
concentration of protein, reactors R4 and R5 showed more
protein degradation than other reactors that were fed raw
water. The HRT of reactors R4 and R5 was different, but they
both degraded protein at the same level. Based on the above
discussion, it is clear that the effect of protein level on
membrane fouling rate was more prominent in the PAC-MF
hybrid MBR; however, the presence of PAC enhanced the
degradation of both proteins and polysaccharides. In both
cases, HRT did not play a major role in their biodegradation.
Furthermore, a higher proteins/polysaccharides (PN/PS)
ratio indicates a higher level of hydrophobicity of EPS (Lee
et al., 2003; Sponza, 2003; Sun et al., 2008). The PN/PS ratio of
the raw and biofiltered water was 1.8 and 1.9, respectively;
however, this ratio increased from 2 to 5 fold for the samples
inside the reactors and for the membrane foulants. The
affinity between proteins and foulants should generally be
greater than that between polysaccharides and flocs in rela-
tion to their hydrophobicity and surface charge (Renard et al.,
2002). Therefore, a greater amount of proteinwas found inside
the reactors and foulants.
4. Conclusions
PAC-MF MBR hybrid systems are becoming one of the mostin these reactors than that inside the reactor R1. Both reactors
R4 and R5 received biofiltered water and operated with PAC;
however, polysaccharide concentration in the foulants on the
membrane surfaces inside reactor R5, which was operated atsustainable technologies for drinking water treatment
because of the increased stringency of water qualityHRT in the presence and absence of PAC and biofilter
pretreatment of surface water on the MF membrane perfor-
mance, and kinetics of adsorption and biodegradation of EPS
components and their effects on MF membrane fouling at
different operational conditions of PAC-MF MBR were quan-
tified in this study. The following conclusions can be drawn:
(1) The presence of PAC inside the MF MBR system improved
filtration performance and lowered the impact of organic
carbon, carbohydrates, polysaccharides and proteins on
membrane fouling.
(2) The decrease of HRT increased the frequency of
membrane cleaning and irreversible fouling effects on the
membranes even though the aeration rate was doubled.
Aeration significantly contributed to the removal of cake
fouling in this system.
(3) Biofiltration lessens fouling considerably, more than the
addition of PAC alone. The organic matter sloughed from
the biofiltration unit potentially increased the organic
carbon concentration in its effluent. TOC removal effi-
ciency is dependent on the presence of PAC and this effi-
ciency was not affected by the HRT and type of source
water. Moreover, the decrease of HRT enhanced the
accumulation of carbohydrates and potentially contrib-
uted to more fouling of the membrane module. PAC is not
a potential adsorbent for carbohydrate. Once the PAC and
other particles were attached on the membrane surface
forming foulant layers, higher level of carbohydrate was
observed because of their bonding properties to other
particles.
(4) Major portions of the polysaccharides were biodegraded
inside the MBR. Protein effects on membrane fouling rate
were more prominent than polysaccharides; however, the
presence of PAC enhanced the biodegradation of both
proteins and polysaccharides. In both cases, HRT did not
play a major role in their biodegradation. Polysaccharide
did not cause any significant effects on the irreversible
fouling in this submerged PAC-MF MBR once a major
portion of them were biodegraded. The biodegradation of
EPS components was not equal inside the reactors, which
could be due to the presence of different types of micro-
organisms responsible for biodegradation of the target
components. The HRT is one of the most important
parameters during system (PAC-MF MBR) design.
The MF systems with PAC could be operated longer than
the study period, and the frequency of PAC replacement in the
system would depend on the concentrations of the target
materials in the PAC-MF MBR effluent and PAC particle sizes.
Acknowledgmentsstandards, increasingly restricted water sources and supplies,
and their more competitive cost.
A robust and new method of EPS extraction from theThe authors would like to thank Professor Mark M. Benjamin
of the University of Washington for his valuable suggestions
and redeposition of Pseudomonas aeruginosa to surfaces.
Microbiology 148, 1161e1169.
Characteristics of soluble microbial products and extracellular
polymeric substances in the membrane bioreactor for water
reuse. Desalination 202, 90e98.
Jarusutthirak, C., Amy, G., 2006. Role of soluble microbial
products (SMP) in membrane fouling and flux decline.Huang, Z., Ong, S.L., Ng, H.Y., 2011. Submerged anaerobic
membrane bioreactor for low-strength wastewater treatment:
effect of HRT and SRT on treatment performance and
membrane fouling. Water Res. 45, 705e713.
Jang, N., Ren, X., Kim, G., Ahn, C., Cho, J., Kim, I.S., 2007.and comments on this report. This studywas supported by the
Ministry of Education, Japan.
Appendix A. Supplementary data
Supplementary data related to this article can be found at
http://dx.doi.org/10.1016/j.watres.2012.10.048.
r e f e r e n c e s
Barker, D.J., Stuckey, D.C., 1999. A review of soluble microbial
products (SMP) in wastewater treatment systems. Water Res.
33, 3063e3082.
Barker, D.J., Salvi, S.M.L., Langenhoff, A.A.M., Stuckey, D.C., 2000.
Soluble microbial products in ABR treating low-strength
wastewater. J. Environ. Eng. 126, 239e249.
Basar, C.A., Karagunduz, A., Cakici, A., Keskinler, B., 2004.
Removal of surfactants by powdered activated carbon and
microfiltration. Water Res. 38, 2117e2124.
Chen, V.J., Ma, P.X., 2004. Nano-fibrous poly(L-lactic acid) scaffolds
with interconnected spherical macropores. Biomaterials 25,
2065e2073.
Chu, H.P., Li, X.Y., 2005. Membrane fouling in a membrane
bioreactor (MBR): sludge cake formation and fouling
characteristics. Biotechnol. Bioeng. 90, 323e331.
da No´brega, R.B., Rocha, B.A.M., Gadelha, C.A.A., Santi-
Gadelha, T., Pires, A.F., Assreuy, A.M.S., Nascimento, K.S.,
Nagano, C.S., Sampaio, A.H., Cavada, B.S., Delatorre, P., 2012.
Structure of Dioclea virgata lectin: relations between
carbohydrate binding site and nitric oxide production.
Biochimie 94, 900e906.
Doume`che, B., Galas, L., Vaudry, H., Martino, P.D., 2007.
Membrane foulants characterization in a drinking water
production unit. Food Bioproducts Process. 85, 42e48.
Fabris, R., Lee, E.K., Chow, C.W.K., Chen, V., Drikas, M., 2007. Pre-
treatments to reduce fouling of low pressure micro-filtration
(MF) membranes. J. Memb. Sci. 289, 231e240.
Flemming, H.C., 1993. Mechanistic aspects of reverse osmosis
membrane biofouling and prevention, pp. 163e209. In:
Amjad, Z. (Ed.), Water Chemistry, and Industrial Applications.
Van Nostrand Reinhold, New York.
Fonseca, A.C., Summers, R.S., Greenberg, A.R., Hernandez, M.T.,
2007. Extra-cellular polysaccharides, soluble microbial
products, and natural organic matter impact on nanofiltration
membranes flux decline. Environ. Sci. Technol. 41, 2491e2497.
Goldstein, I.J., Hughes, R.C., Monsigny, M., Osawa, T., Sharon, N.,
1980. What should be called a lectin. Nature 285, 66.
Go´mez-Sua´rez, C., Pasma, J., van der Borden, A.J., Wingender, J.,
Flemming, H.-C., Busscher, H.J., van der Mei, H.C., 2002.
Influence of extracellular polymeric substances on depositionEnviron. Sci. Technol. 40, 969e974.Jia, Y., Wang, R., Fane, A.G., 2009. Hybrid PAC-submerged
membrane system for trace organics removal II. System
simulation and application study. Chem. Eng. J. 149, 42e49.
Khan, M.M.T., Ohgaki, S., Takizawa, S., Katayama, H., 2001.
Development of powdered activated carbon and
microfiltration membrane system for water treatment. J. Jpn.
Soc. Civil Eng. 8, 369e372.
Khan, M.M.T., Lewandowski, Z., Takizawa, S., Yamada, K.,
Katayama, H., Yamamoto, K., Ohgaki, S., 2009. Continuous
and efficient removal of THMs from river water using MF
membrane combined with high dose of PAC. Desalination 249,
713e720.
Khan, M.M.T., Stewart, P.S., Moll, D.J., Mickols, W.E., Burr, M.D.,
Nelson, S.E., Camper, A.K., 2010. Assessing biofouling on
polyamide reverse osmosis (RO) membrane surfaces in
a laboratory system. J. Memb. Sci. 349, 429e437.
Khan, M.M.T., Takizawa, S., Lewandowski, Z., Jones, W.L.,
Camper, A.K., Katayama, H., Kurisu, F., Ohgaki, S., 2011.
Membrane fouling due to dynamic particle size changes in the
aerated hybrid PACeMF system. J. Memb. Sci. 371, 99e107.
Kim, H.S., Katayama, H., Takizawa, S., Ohgaki, S., 2005.
Development of a microfilter separation system coupled with
a high dose of powdered activated carbon for advanced water
treatment. Desalination 186, 215e226.
Kim, H.S., Takizawa, S., Ohgaki, S., 2006. Application of
microfiltration systems coupled with powdered activated
carbon to river water treatment. Desalination 202, 271e277.
Kimura, K., Hane, Y., Watanabe, Y., Amy, G., Ohkuma, N., 2004.
Irreversible membrane fouling during ultrafiltration of surface
water. Water Res. 38, 3431e3441.
Kocourek, J., Horejsı´, V., 1981. Defining a lectin. Nature 290, 188.
Lam, S.K., Ng, T.B., 2011. Lectins: production and practical
applications. Appl. Microbiol. Biotechnol. 89, 45e55.
Lebeau, T., Lelievre, C., Buisson, H., Cleret, D., Van de
Venter, L.W., Cote, P., 1998. Immersed membrane filtration for
the production of drinking water: combination with PAC for
NOM and SOCs removal. Desalination 117, 219e231.
Lee, J., Ahn, W.J., Lee, C.H., 2001. Comparison of the filtration
characteristics between attached and suspended growth
microorganisms in submerged membrane bioreactor. Water
Res. 35, 2435e2445.
Lee, W., Kang, S., Shin, H., 2003. Sludge characteristics and their
contribution to micro-filtration in submerged membrane
bioreactors. J. Memb. Sci. 216, 217e227.
Lee, S., Lee, J.W., Kim, S., Park, P.K., Kim, J.H., Lee, C.H., 2009.
Removal of 17 b-estradiol by powdered activated carbon-
microfiltration hybrid process: the effect of PAC deposition on
membrane surface. J. Memb. Sci. 326, 84e91.
Meng, F., Shi, B., Yang, F., Zhang, H., 2007. Effect of hydraulic
retention time on membrane fouling and biomass
characteristics in submerged membrane bioreactors.
Bioprocess Biosyst. Eng. 30, 359e367.
Mikkelsen, L.H., Keiding, K., 2002. Physico-chemical
characteristics of full scale sewage sludges with implications
to dewatering. Water Res. 36, 2451e2462.
Nagaoka, H., Kono, S., Yamanishi, S., Miya, A., 2000. Influence of
organic loading on membrane fouling in membrane
separation activated sludge process. Water Sci. Technol. 41,
355e362.
Ng, C.A., Sun, D., Fane, A.G., 2006. Operation of membrane
bioreactor with powdered activated carbon addition. Separ.
Sci. Technol. 41, 1447e1466.
Oh, H.K., Takizawa, S., Ohgaki, S., Katayama, H., Oguma, K.,
Yu, M.J., 2007. Removal of organics and viruses using hybrid
ceramic MF system without draining PAC. Desalination 202,
191e198.
Park, N., Kwon, B., Kim, I.S., Cho, J., 2005. Biofouling potential of
various NF membranes with respect to bacteria and their
soluble microbial products (SMP): characterizations, flux
decline, and transport parameters. J. Memb. Sci. 258, 43e54.
Renard, D., Lavenant, L., Sanchez, C., Hemar, Y., Horne, D., 2002.
Heat-induced flocculation of microparticulated whey proteins
(MWP); consequences for mixed gels made of MWP and b-
lactoglobulin. Colloid. Surf. B Biointerface 24, 73e85.
Sponza, D.T., 2003. Investigation of extracellular polymeric
substances (EPS) and physicochemical properties of different
activated sludge flocs under steady-state conditions. Enzyme
Microb. Technol. 32, 375e385.
Sun, F.-Y., Wang, X.-M., Li, X.-Y., 2008. Visualisation and
characterisation of biopolymer clusters in a submerged
membrane bioreactor. J. Memb. Sci. 325, 691e697.
Tansel, B., Sager, J., Garland, J., Xu, S., Levine, L., Bisbee, P., 2006.
Deposition of extracellular polymeric substances (EPS) and
microtopographical changes on membrane surfaces during
intermittent filtration conditions. J. Memb. Sci. 285, 225e231.
Thiruvenkatachari, R., Shim, W.G., Lee, J.W., Aim, R.B., Moon, H.,
2006. A novel method of powdered activated carbon (PAC) pre-
coated microfiltration (MF) hollow fiber hybrid membrane for
domestic wastewater treatment. Colloid. Surf. A Physicochem.
Eng. Aspect. 274, 24e33.
Tsuneda, S., Aikawa, H., Hayashi, H., Yuasa, A., Hirata, A., 2003.
Extracellular polymeric substances responsible for bacterial
adhesion onto solid surface. FEMS Microbiol. Lett. 223,
287e292.
Turakhia, M.H., Characklis, W.G., 1989. Activity of Pseudomonas
aeruginosa in biofilms: effect of calcium. Biotechnol. Bioeng. 33,
406e414.
Ueda, T., Hata, K., Kikuoka, Y., Seino, O., 1997. Effects of aeration
on suction pressure in a submerged membrane bioreactor.
Water Res. 31, 489e494.
Vigneswaran, S., Guo, W.S., Smith, P., Ngo, H.H., 2006. Submerged
membrane adsorption hybrid system (SMAHS): process
control and optimization of operating parameters.
Desalination 202, 392e399.
Wingender, J., Neu, T.R., Flemming, H.-C., 1999. What are
bacterial extracellular polymeric substances?, pp. 1e19 In:
Wingender, J., Neu, T.R., Flemming, H.-C. (Eds.), Microbial
Extracellular Polymeric Substances. Springer, New York.
Wotton, R., 2004. The ubiquity and many roles of exopolymers
(EPS) in aquatic systems. Scientia Marina 68 (Suppl. 1), 13e21.
Yang, W., Paetkau, M., Cicek, N., 2010. Effects of powdered
activated carbon dosing on sludge characteristics and
estrogen removal in membrane bioreactors. Water Sci.
Technol. 61, 2193e2198.
Ye, Y., Le Clech, P., Chen, V., Fane, A.G., 2005. Evolution of fouling
during crossflow filtration of model EPS solutions. J. Memb.
Sci. 264, 190e199.
Zhang, X., Bishop, P.L., Kinkle, B.K., 1999. Comparison of
extraction methods for quantifying extracellular polymers in
biofilms. Water Sci. Technol. 39, 211e218.