Relation between EPS adherence, viscoelastic properties, and MBR operation: Biofouling study with QCM-D.

Abstract

Membrane fouling is one of the main constraints of the wide use of membrane bioreactor (MBR) technology. The biomass in MBR systems includes extracellular polymeric substances (EPS), metabolic products of active microbial secretion that adversely affect the membrane performance. Solids retention time (SRT) in the MBR is one of the most important parameters affecting membrane fouling in MBR systems, where fouling is minimized at optimal SRT. Among the operating parameters in MBR systems, SRT is known to strongly influence the ratio of proteins to polysaccharides in the EPS matrix. In this study, we have direct evidence for changes in EPS adherence and viscoelastic properties due to changes in the sludge removal rate that strongly correlate with the membrane fouling rate and EPS composition. EPS were extracted from a UF membrane in a hybrid growth MBR operated at sludge removal rates of 59, 35.4, 17.7, and 5.9 L day(-1) (corresponding SRT of 3, 5, 10, and 30 days, respectively). The EPS adherence and adsorption kinetics were carried out in a quartz crystal microbalance with dissipation monitoring (QCM-D) technology in several adsorption measurements to a gold sensor coated with Polyvinylidene Fluoride (PVDF). EPS adsorption to the sensor surface is characterized by a decrease of the oscillation frequency and an increase in the dissipation energy of the sensor during parallel flow of aqueous media, supplemented with EPS, above the sensor surface. The results from these experiments were further modeled using the Voigt based model, in which the thickness, shear modulus, and shear viscosity values of the adsorbed EPS layers on the PVDF crystal were calculated. The observations in the QCM-D suggested that the elevated fouling of the UF membrane is due to higher adherence of the EPS as well as reduction in viscosity and elasticity of the EPS adsorbed layer and elevation of the EPS fluidity. These results corroborate with confocal laser scanning microscopy (CLSM) image analysis showing thicker EPS in close proximity to the membrane surface operated at reactor conditions which induced more fouling at elevated sludge removal rates.

Full text

Page 1

Relation between EPS adherence, viscoelastic properties,
and MBR operation: Biofouling study with QCM-D
Amer Sweitya, Wang Yinga, Mohammed S. Ali-Shtayehb, Fei Yanga, Amos Bickc,
Gideon Orona, Moshe Herzberga,*
aBen Gurion University of the Negev, Zuckerberg Institute for Water Research, Sede Boqer Campus, Midreshet Ben Gurion, 84990, Israel
bBiodiversity & Environmental Research Center (BERC), Til Village, P.O.BOX 696, Nablus, West Bank, Palestinian Authority
cDepartment of Industrial Engineering and Management, Jerusalem College of Technology, Jerusalem, Israel
a r t i c l e i n f o
Article history:
Received 23 June 2011
Received in revised form
16 September 2011
Accepted 19 September 2011
Available online 29 September 2011
Keywords:
Biofouling
MBR
QCM-D
EPS
Ultrafiltration
Wastewater
a b s t r a c t
Membrane fouling is one of the main constraints of the wide use of membrane bioreactor
(MBR) technology. The biomass in MBR systems includes extracellular polymeric
substances (EPS), metabolic products of active microbial secretion that adversely affect the
membrane performance. Solids retention time (SRT) in the MBR is one of the most
important parameters affecting membrane fouling in MBR systems, where fouling is
minimized at optimal SRT. Among the operating parameters in MBR systems, SRT is
known to strongly influence the ratio of proteins to polysaccharides in the EPS matrix. In
this study, we have direct evidence for changes in EPS adherence and viscoelastic prop-
erties due to changes in the sludge removal rate that strongly correlate with the membrane
fouling rate and EPS composition. EPS were extracted from a UF membrane in a hybrid
growth MBR operated at sludge removal rates of 59, 35.4, 17.7, and 5.9 L day-1(corre-
sponding SRT of 3, 5, 10, and 30 days, respectively). The EPS adherence and adsorption
kinetics were carried out in a quartz crystal microbalance with dissipation monitoring
(QCM-D) technology in several adsorption measurements to a gold sensor coated with
Polyvinylidene Fluoride (PVDF). EPS adsorption to the sensor surface is characterized by
a decrease of the oscillation frequency and an increase in the dissipation energy of the
sensor during parallel flow of aqueous media, supplemented with EPS, above the sensor
surface. The results from these experiments were further modeled using the Voigt based
model, in which the thickness, shear modulus, and shear viscosity values of the adsorbed
EPS layers on the PVDF crystal were calculated. The observations in the QCM-D suggested
that the elevated fouling of the UF membrane is due to higher adherence of the EPS as well
as reduction in viscosity and elasticity of the EPS adsorbed layer and elevation of the EPS
fluidity. These results corroborate with confocal laser scanning microscopy (CLSM) image
analysis showing thicker EPS in close proximity to the membrane surface operated at
reactor conditions which induced more fouling at elevated sludge removal rates.
ª 2011 Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: þ972 8 6563520; fax: þ972 8 6563503.
E-mail address: herzberg@bgu.ac.il (M. Herzberg).
0043-1354/$ e see front matter ª 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.watres.2011.09.038
journal homepage: www.elsevier.com/locate/watres
Water Research 45 (2011) 6430e6440

Page 2

1.Introduction
Membrane fouling is one of the main constraints of the wide
use of membrane bioreactor (MBR) technology (Judd, 2006)
causing an increase in the trans-membrane pressure (TMP) or
a decrease in the permeateflux. During the biofouling process,
membrane permeability decreases and energy consumption
increases (Yang et al., 2006). Membrane fouling in MBR
processes almost always consists of a combination of
colloidal, organic, and microbial deposits (biofouling) as well
as inorganic precipitates (scaling). These fouling factors
increase the membrane hydraulic resistance over time and
the permeate flux is consequently reduced. In most cases,
deposition of the foulants are found both on the external
membrane surface with some degree of foulant deposition
inside themicrofiltration(MF)
membrane pores (Chang et al., 2002). Membrane biofouling is
strongly related to membrane properties, operational condi-
tions and biomass characteristics that include extracellular
polymeric substances (EPS) properties. Hybrid growth MBR
(HG-MBR) system can be defined as the combination of
a membrane separation process and a hybrid growth
processes, in which both suspended and attached-growth
microorganisms are part of the MBR (Sombatsompop et al.,
2006; Yang et al., 2006). HG-MBR allows for upgrading the
treatment capacities of existing MBR treatment plants by
increasingbiomass level.
suspended-growth are involved, the HG-MBR can be operated
at lower mixed liquor suspended solids (MLSS) concentra-
tions. Membrane fouling is minimized without loss of the
treatment efficiency due to biological activity of the microor-
ganisms that are attached to the support carriers.
EPS, metabolic products of active bacterial secretion
(Comte et al., 2006; Nuengjamnong et al., 2005), can be found
either in a soluble form (also termed as soluble microbial
products e SMP) or bound to cells or flocs in the reactor
formingthecohesivematrixof thebiofilms.BoundEPSconsist
of proteins, polysaccharides, nucleic acids and lipids accu-
mulating on the bacterial cell surface (Morgan et al., 1990). The
EPS strongly affect the microbial microenvironment hetero-
geneity including changes in porosity, density, water content,
sorption properties, charge, hydrophobicity, and mechanical
stability (Flemming and Wingender, 2001). One of the most
effective MBR operating parameters with an impact on fouling
propensity is solids retention time (SRT) or sludge age. SRT
affects various sludge properties such as floc size, bound and
soluble EPS content, and settling characteristics (Le-Clech
et al., 2006).
Contradictory reports regarding a relationship between
SRT and membrane biofouling show that even though higher
SRT leads inevitably to increase of MLSS concentration, this in
itself may not necessary lead to greater fouling. In general,
optimalSRT,reportedin plethora ofstudies between20 and 50
days, is required to achieve a minimal fouling tendency (Meng
et al., 2009; Drews, 2010; Kraume and Drews, 2010). Improved
membrane permeability was observed at longer SRT of 10 and
20 days in comparison to SRT of 3 and 5 days. The results were
attributed to elevated concentrations of SMP and EPS
concentrations that were observed to induce membrane
andultrafiltration(UF)
Sincebothattached-and
fouling rate when SRT was decreased (Ng et al., 2006). Cho
et al. (2005a) showed that as SRT decreased, the amount of
bound EPS in the sludge flocs increased (Cho et al., 2005b). Han
et al. (2005) has reported that membrane fouling rate
increasedwithincreasing SRTof30,50, 70,and 100daysdueto
a large amount of foulants and high sludge viscosity (Han
et al., 2005). In contrast, Lee et al. (2003) tested three lab-
scale submerged MBRs at SRT of 20, 40, and 60 days with
a constant permeate flux and no major change in EPS
concentration was observed as SRT increased (Lee et al., 2003).
In another study, at elevated MLSS concentrations from 7 to
18 g/l corresponding to an increase in SRT from 30 to 100 days,
fouling rate was twice for the extended SRT (Al-Amoudi and
Farooque, 2005). This increase was probably due to the
raised viscosity at the high MLSS concentration that attenu-
ates the effect of bubbling and scouring of the membrane
surface. Not surprisingly, fouling rate increased nearly 10
times when SRT was lowered from 10 to 2 days, probably due
to the increased levels of EPS production (Trussell et al.,
2006).Chang and Lee (1998) found that when the SRT was
increased from 3 to 8 and to 33 days, a significant increase in
sustainable flux was observed (Chang and Lee, 1998). The
reduced fouling rates associated with a decrease in sludge
production rates at longer sludge ages, is usually attributed to
lowerEPSconcentrationsin
increasing SRT could enhance the development of slow
growing microorganisms that are able to consume poly-
saccharides and proteins as substrates and produce less
biopolymers (Masse et al., 2006). Overall, it is likely that there
is an optimal SRT, between the high fouling tendency at very
low SRT and the high viscosity of mixed liquor at very long
SRT.
EPS play a major role in the cohesion of the sludge flocs in
the MBR as well as the cohesion of the biofilm layers located
on carriers in the HG-MBR systems. EPS are also in charge of
biofilms viscoelastic properties which in turn, can strongly
affect the microbial flocs and biofouling layer resistance to
shear. Eventually, EPS are recognized as the most direct and
significant factor affecting biofouling in MBRs (Laspidou and
Rittmann, 2002; Le-Clech et al., 2006). Soluble EPS in the
MLSS was reported as an important factor influencing
membrane fouling. A high concentration of soluble EPS was
shown to boost membrane fouling tendency (Kimura et al.,
2005). Ouyang and Liu (2009) showed that soluble EPS
concentration increased at shorter SRT, in which total protein
concentrations was higher than polysaccharides in the MLSS
supernatant, whereas the total polysaccharide content was
higher than the protein in the flocs attached to the membrane
surface causing a significant fouling. By increasing the SRT,
soluble EPS content was decreased on the membrane surface
and membrane filtration resistance was reduced (Ouyang and
Liu, 2009). EPS production and accumulation on the UF
membranes in MBR systems is a complex process influenced
by several factors like the substrate composition, mechanical
stress, organic loading rate, MLSS concentration, presence of
soluble EPS compounds and membrane properties (Chang and
Lee, 1998; Rojas et al. 2005; Rosenberger and Kraume, 2003).
Since it would be hard to point out how a combination of so
many parameters may influence the properties of the
thereactor.Inaddition,
Water Research 45 (2011) 6430e6440
6431

Page 3

accumulated on the membrane, direct membrane autopsy
and analysis of the accumulated EPS can help to relate
between EPS properties and membrane fouling.
In this study, we hypothesized that membrane filterability
is strongly influenced by EPS cohesion and viscoelastic prop-
erties, important properties of the EPS produced at different
sludge removal rates in the MBR (Ng et al., 2006; Ying et al.,
2009). Different EPS originated from the membrane at
different sludge removal rates showed different adherence
and viscoelastic properties that were correlated to the EPS
composition and to the fouling rate of the membrane. It was
intriguing to see how EPS adherence and viscoelasticity
change in correlation to different conditions that promote
biofouling to different degree. We also suggest a novel
parameter in fouling phenomena of membranes in general,
first to be applied to UF membranes, in this study e the fluidity
of the adsorbed EPS layer. This parameteris frequentlyused to
describe biopolymer layers in order to estimate their viscoe-
lasticy (deKerchove and Elimelech, 2006; Feiler et al., 2007).
The working objective of this study focused on defining if the
EPS accumulated on the UF membrane is more fluidic or more
rigid under conditions that promote biofouling.
To study the adherence and viscoelastic properties of the
EPS, we utilized a quartz crystal microbalance with dissipa-
tion monitoring (QCM-D) technology. QCM-D provides real-
time, label free measurements of molecular adsorption and/
or interactions taking place on various surfaces (Eydelnant
and Tufenkji, 2008; Wang et al., 2007). In addition to assess-
ing adsorbed mass (ng/cm2sensitivity), measured as changes
in oscillating frequency (F) of the quartz crystal, the energy
dissipation (D), which is the reduced energy per oscillation
cycle provides novel insights regarding structural properties
of adsorbed layers (Nguyen and Elimelech, 2007; Voinova
et al., 1999). EPS originated from the UF membrane at
different sludge removal rates during the MBR operation was
extracted and analyzed. Furthermore, confocal laser scanning
microscopy (CLSM) of the biofilm on the UF membrane and
EPS composition results were correlated to fouling rate of the
UF membrane and to the EPS cohesion and viscoelastic
properties.
2.Materials and methods
2.1.HG-MBR system and operating conditions
The HG-MBR was equipped with an immersed UF membrane
module of ZeeWeed?(ZW-10) (Zenon Environmental Inc,
Canada). The membrane module was made of hollow fibers of
polyvinylidene fluoride (PVDF) with a mean pore size of
0.04 mm and a total effective filtering surface area of 0.93 m2
allowing the removal of pathogens and organic matter. The
volume of the bioreactor process tank was 190 L and included
activated sludge, AqWise carriers (AqWise, Israel), and the
membrane module. AqWise carriers were filled as biofilm
support with a filling ratio (carrier volume/reactor volume) of
50% (13.64 kg). The carriers are made from high-density
(0.96 g/cm3) polyethylene with diameter and height of
13 mm and a specific surface area of 600 m2/m3. The carriers’
circulation was driven by an air diffuser. The membrane
module was surrounded by an 8 mm mesh for avoiding
damage from the moving carriers. The system operated under
constant-flux mode with a mode of 5 min filtration and 15 s
backwash. A feed domestic sewage mixed with chickens’
manure was injected into the bioreactor that was operated
under desert ambient conditions. Membrane cleaning was
maintained by soaking the membrane module in 750 mg/L
sodium hypochlorite supplemented with 250 mg/L sodium
dodecyl sulfate (SDS) solution for 16 h, repeatedly for 4 times
after each experiment, until the membrane permeability was
recovered. Aeration was done through an air diffuser installed
directly beneath the membrane module for supplementing
oxygen to microorganisms, mixing the liquor and cleaning the
membrane with aeration rate of 2.3 m3/h. Airflow rate was
controlled by a rotameter, filtration flux of permeate was
monitored volumetrically and TMP was monitored by a digital
pressure indicator. The mixed liquor temperature was moni-
tored by a temperature indicator located in the reactor MLSS.
The dissolved oxygen (DO) concentration is the mean of the
upper, middle and bottom locations in the bioreactor vessel
(Model 550, YSI, USA). The bioreactor was employed with
a water level sensor was used to keep a constant liquid level in
the bioreactor. The HG-MBR was operated over a period of two
months at sludge removal rate values of 2, 5, 10, and 30 days.
The hydraulic retention times (HRT) of this HG-MBR was 5.5 h
for all the experiments. Operating conditions of the HG-MBR
at different sludge removal rates are listed in Table 1. The
influent and effluent characteristics of the HG-MBR operated
at different sludge removal rates are listed in Table 2. As ex-
pected, at different sludge removal rates, biomass concen-
tration varies.The suspended
concentrations versus time at different sludge removal rates
arepresentedin thesupporting
(Figure S1). By reducing the suspended sludge age, an
increased washout of the suspended biomass was observed:
The MLSS concentration was lower at shorter sludge removal
rates. The mean MLSS concentrations were 4055, 2686, 1678
and 1392 mg/L for the sludge removal rates of 30, 10, 5 and 3
days, respectively (Table 1). Interestingly, the attached biofilm
concentration was also reduced. In a similar trend of the
decline in MLSS concentration, the decline of the attached
biofilm concentration was also observed (Figure S1). The
sludge removal rate calculation is taking into account that
there was no biomass lost in the effluent of the MBR during its
entire operational period. Therefore, the biomass concentra-
tion in both the reactor and in the removed sludge stream is
the same. SRT, was calculated following Li et al. (1984), SRT ¼
V$XV=VWS$DXV The reactor volume was 190 L multiplied by
the volumetric fraction occupied by the biofilm’s carriers
(50%). V is the reactor volume occupied by MLSS (L), VWSis the
flow rate of the removed sludge per day (L day?1), XV is the
MLSS concentration and DXV is the MLSS concentration in the
removed sludge per day. Since in this work, XV ¼ DXV, the
suspended sludge retention time is the reactor volume (L)
divided by sludge removal rate (L day?1), V/VWS.
andattached biomass
information section
2.2.EPS extraction and analysis
EPS extraction was performed from a single hollow fiber that
was cut from the ZW-10 module at the end of every
Water Research 45 (2011) 6430e6440
6432

Page 4

experiment. The EPSextractionstep was carriedout according
to Liu and Fang (Liu and Fang, 2002). A 10 cm piece of the fiber
was cut and the ends of the fibersattached to the modulewere
sealed. Briefly, the fiber was suspended into 10 mL of 0.1 M
NaCl solution in a 50 mL polypropylene tube, and vortexed for
45 min to make sure that the biofilm is totally suspended.
Then, 60 mL of 35% formaldehyde (SigmaeAldrich, Israel) were
added to the solution and incubated 1 h in a Vortex Genie 2?
(Scientific Industries, USA) at a minimum mixing setting and
4?C, followed by the addition of 4 mL 1 M sodium hydroxide at
4?C for 3 h incubation period in order to facilitate dissociation
of the acidic groups from the EPS to the solution. Thereafter,
the suspension was centrifuged (35,000 rpm, 30 min, 4?C), the
supernatant was filtered through a 0.2 mm hydrophilic nylon
filter(MilliporeCo.),and dialyzed
membrane of 3500 Da (Spectra/Por) for a few days until salts
were completely removed. Then the extracted EPS was
a lyophilized (FreeZone 2.5 plus) at ?80?C and 0.01 mbar for
48 h. The frozen and dried EPS samples were re-dissolved in
10 mL of double distilled water (DDW) for the determination of
dissolved organic carbon (DOC), proteins, and polysaccharides
concentrations.
througha dialysis
Extracellular protein of the extracted EPS was analyzed
using the colorimetric quantitative protein determination
with the Bio-RadªProtein Assay according to Bradford
(Bradford, 1977). Polysaccharides contents were determined
according to Dubois et al. (DuBois et al., 1956), using glucose
and alginic acid as standards. EPS extracted was expressed as
DOC concentration measured by using an Apollo 9000 TOC
Analyzer (Teledyne Tekmar, United States).
2.3.
QCM-D
Adherence and viscoelastic properties analysis with
EPS was extracted from the UF membrane surface after
operating the HG-MBR under different conditions, i.e., at
different sludge removal rates of 59, 35.4, 17.7, and 5.9 L day?1
(correspond to calculated SRT of 3, 5, 10, and 30 days). The
adherence and adsorption kinetics of the EPS was carried out
in a QCM-D (Q-Sense AB, Gothenburg, SWEDEN). The QCM-D
measurements were performed with AT-cut quartz crystals
mounted in an E1 system (Q-sense AB, Gothenburg, SWEDEN).
The gold coated crystals with a fundamental resonant
frequency of around 5 MHz were coated with Polyvinylidene
Fluoride (PVDF) batch number (QSX999, Q-sense). Before each
measurement, the crystals were soaked in a 5 mM ethyl-
enediaminetetraacetic acid (EDTA) solution for 30 min, rinsed
thoroughly with DDWand driedwith pureN2gas. The EPSwas
used in several adsorption measurements to the QCM-D PVDF
coated gold sensor. EPS adsorption to the sensor surface is
characterized by the change of the oscillation frequency of the
PVDF coated gold sensor during parallel flow of aqueous
media with flow rate of 150 ml/min above the sensor surface.
The variations of frequency, f (Hz) and dissipation factor, D
were measured for the three overtones (n ¼ , 5, 7, and 9). The
working stages for applying aqueous media to the QCM-D flow
cell include 5 stages of 20 min each at constant temperature
(22?C). The stages include the following fluids being injected
to the QCM-D flow cell: DDW, 10 mM NaCl aqueous solution,
20mg/Lof EPSas DOC(frommembraneafter MBRoperationat
different sludge removal rates) dissolved in 10 mM NaCl,
10 mM NaCl aqueous solution, and DDW. The QCM-D results
from these experiments were further modeled in which the
thickness, shear modulus, and shear viscosity values of the
adsorbed EPS layers on the PVDF crystal were calculated. The
viscoelastic properties of the EPS layers were calculated based
on the Voigt model according to Voinova et al. (Voinova et al.,
1999). The density and viscosity of the solution used in this
model were 1 g/cm3and 10?3Pa s, respectively. The density of
the adsorbed layer was fixed at 1.030 g/cm3, following the
recommendations of Gurdak et al. (2005). The best fitting
values of the shear viscosity (h), shear modulus (m), and
thickness of the adsorbed layer were obtained by modeling
the experimental data of f and D for three overtones using the
program Q-Tools provided by Q-Sense AB.
2.4. CLSM analysis
At the end of every experiment in which different sludge
removal rates were applied in the HG-MBR operation,
membrane autopsies were carefully cut to pieces of around
Table 1 e Operating conditions of the HG-MBR at different removal rates (L dayL1) of MLSS.
Parameter5.9?L day?1
17.7?L day?1
35.5?L day?1
59?L day?1
Estimated SRT (days)
Temperature
30
15.5 w 29.4?C
(mean 22.3?C)
631.1
10
22.2 w 28.9?C
(mean 26.3?C)
465.5
5
20.0 w 30.6?C
(mean 28.5?C)
449.8
3
25.5 w 29.4?C
(mean 28.0?C)
527.7
Initial membrane permeability
(L/(m2.hr.bar)) at 20?C
Initial membrane resistance (m?1)
Filtrate flux (L/(m2h))
Aeration rate (m3/hr)
Hydraulic retention time (hours)
pH in the reactor
Dissolved oxygen, mg O2/L
0.56 ? 1012
44.5 to 38.9
2.3
5.1 w 5.8
6.6 w 7.8
0.31 w 6.5
(mean 2.6)
2440 to 4580
(mean 4055)
13.64 kg carriers in the reactor with a bulk filling ratio of 50%
3426
0.76 ? 1012
44.8 to 40.6
2.3
5.0 w 5.5
6.2 w 6.9
0.7 w 5.6
(mean 2.8)
1000 to 3520
(mean 2686)
0.79 ? 1012
45.2 to 38.58
2.3
5.0 w 5.9
6.6 w 6.9
0.25 w 3.66
(mean 0.89)
1225 to 2110
(mean 1678)
0.67 ? 1012
45.2 to 37.7
2.3
5.0 w 6.0
6.7 w 7.1
2.4 w 6.2
(mean 4.3)
1170 to 1575
(mean 1392)
MLSS range (mg/L)
AqWise carriers in the reactor
Operating time (days)169
Water Research 45 (2011) 6430e6440
6433

Page 5

1 cm length from the fiber that was cut for the EPS extraction.
The membrane pieces were double stained with concanavalin
A (ConA) conjugated to Alexa fluor 633, and SYTO9 for probing
EPS or microorganisms, respectively. Microscopic observation
and image acquisition were performed using Zeiss-Meta 510,
a CLSM equipped with Zeiss dry objective LCI Plan-NeoFluar
(25 ? magnification and numerical aperture of 0.8). The
CLSM was equipped with detectors and filter sets for moni-
toring SYTO9 stained cells and Alexa fluor 633 dye (excitation
wavelengths of 488 and 633 nm, respectively). CLSM images
were generatedusing the Zeiss LSM ImageBrowser. Gray scale
images were analyzed, and the specific biovolume (mm3/mm2)
in the biofouling layer was determined by COMSTAT image-
processing software (Heydorn et al., 2000b). For every sample
between 4 and 6 positions on the membrane were chosen and
microscopically observed, acquired, and analyzed. The ConA,
conjugated to Alexa fluor 633 (Invitrogen Co.), was used as
a probe to determine the presence of EPS.Briefly, frozen
(?20?C) 100 mL aliquots of 1 mg/mL labeled ConA stock solu-
tion were thawed and diluted in 10 mM phosphate buffer (pH
7.5) to 100 mg/mL prior to use in 10 mM phosphate buffer (pH
7.5). An excess electrolyte solution was carefully drawn off
from the fouled membrane by gently touching the edge of the
specimens with an adsorbing paper (Kimwipes). Then, 100 mL
of ConA staining solution were added to cover the samples,
which were then incubated in the dark at room temperature
for 20 min. Unbound ConA was drawnoff the specimens using
a three-step wash of 10 mM phosphate buffer. The unbound
ConA solution and the washing solutions were carefully
removed by gently touching the edge of the specimen with an
adsorbing paper. CYTO9 was used for probing the microor-
ganisms in the fouling layer. Excess electrolyte solution was
carefully drawn off from a piece of a fouled membrane in the
same manner used for ConA staining. Then, 5 mM SYTO9
solution (prepared in 10 mM phosphate buffer, pH 7.5) was
added to cover the samples, which were then incubated in the
dark at room temperature for 20 min. Excess SYTO9 solution
was carefully drawn off with an adsorbing paper. The excess
SYTO9 nucleic acid stain that did not bind to the samples was
then removed by rinsing three times with a 10 mM phosphate
buffer at pH 7.5.
0510 15
Time, Days
202530
35
40
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Transmembrane Pressure, Bar
TMP, 5.9 L·day
TMP, 17.7 L·day
TMP, 35.4 L·day
TMP, 59 L·day
-1
-1
-1
-1
Fig. 1 e The effect of different sludge removal rates
(L dayL1) on the UF membrane TMP (Bar).
Table 2 e The influent and effluent characteristics of the HG-MBR operated at different removal rates (L dayL1) of MLSS.
Parameter
5.9 L day?1(estimated SRT ¼ 30 days)
17.7 L day?1(estimated SRT ¼ 10 days)
59 L day?1(estimated SRT ¼ 3 days)
35.5 L day?1(estimated SRT ¼ 5 days)
Influent
Effluent
Percent
removal
Influent
Effluent
Percent
Removal
Influent
Effluent
Percent
Removal
Influent
Effluent
Percent
removal
COD, mg/L
418 ? 123
36 ? 13
91.4
529 ? 205
37 ? 18
93.1
483 ? 140
42 ? 25
91.3
495 ? 50
51 ? 18
89.8
BOD, mg/L
171 ? 45
1.3 ? 0.4
99.2
231 ? 60
1.1 ? 0.5
99.5
229 ? 63
1.2 ? 0.4
99.5
317 ? 67
1.5 ? 0.3
99.5
NHþ
4eN, mg/L
30 ? 6.8
1.4 ? 2.8
95.4
38 ? 16
0.1 ? 0.2
99.8
38 ? 5.7
0.3 ? 0.2
99.1
38 ? 4.6
0.9 ? 1.1
97.7
TN, mg/L
35 ? 8.1
24 ? 7.4
31.2
49 ? 12
37 ? 6.3
24.4
46 ? 6.3
31 ? 3.3
33.0
41.9 ? 15
34.6 ? 4.8
17.4
PO?3
4eP, mg/L
10 ? 4.2
7.6 ? 3.3
25.0
14 ? 6.8
12 ? 4.4
15.2
12 ? 3.6
8.4 ? 2.2
29.9
18.6 ? 11.4
16.2 ? 11.3
12.9
TSS, mg/L
171 ? 63
0.2 ? 0.5
99.9
176 ? 97
0.8 ? 1.0
99.5
176 ? 70
0.8 ? 1.1
99.5
269 ? 105
1.0 ? 1.2
99.6
Turbidity, NTU
231 ? 79
0.2 ? 0.1
99.9
245 ? 186
0.3 ? 0.1
99.9
201 ? 85
0.3 ? 0.3
99.8
330 ? 186
0.3 ? 0.1
100.0
EC, mS/cm
1.3 ? 0.2
1.2 ? 0.2
e
1.2 ? 0.1
1.1 ? 0.1
e
1.3 ? 0.2
1.1 ? 0.1
e
1.3 ? 0.1
1.1 ? 0.1
e
pH
7.5 ? 0.3
7.5 ? 0.4
e
7.3 ? 0.2
6.7 ? 0.4
e
7.5 ? 0.2
7.2 ? 0.3
e
7.0 ? 0.2
7.4 ? 0.1
e
Water Research 45 (2011) 6430e6440
6434

Page 6

3.Results and discussion
3.1.
performance of the MBR
The effect of sludge removal rate on the filtration
Fig. 1 shows the variations of TMP over time at the various
sludge removal rates. At sludge removal rate of 5.9 and
17.7 L day?1(estimated SRT of 10 and 30 days), the TMP
increased slowly, displaying a linear tendency with the
increasing TMP rate of 0.0055 and 0.0064 bar per day, respec-
tively. It seems that at sludge removal rate of 17.7 L day?1
(estimated SRT of 10 days), the system has the lowest fouling
rate. When sludge removal rate was changed to 59 L$day?1
(estimated SRT of 3 days), a very sharp increase of TMP from
0.08 to 0.55 bar was observed after 13 days, with an increase
rate of 0.027 bar/day, while at sludge removal rate of
35.5 L day?1(estimated SRT of 5 days), an increase rate of
0.016 bar/day was observed. In other words, the fouling rate at
sludge removal rate of 59 L day?1(estimated SRT of 3 days) is
nearly 5 times higher than that of sludge removal rate of
17.7 L day?1(estimatedSRT of 10 days). The extent of fouling is
likely to vary according to the MLSS composition including
EPS and SMP in the bioreactor that interact with the
membrane surface and pores (Chang et al., 2002; Drews, 2010).
Therefore, we decided to analyze and compare the adherence
andviscoelastic properties
membrane. EPS extracted from the membrane operated in the
MBR at different SRTs was used for adsorption experiments to
a PVDF coated sensors in the QCM-D as well as fouling
experiments of single fiber UF membrane unit.
of EPSdepositedon the
3.2.
and viscoelastic properties
The effect of sludge removal rate on EPS adherence
In this part of the study, EPS adherence and viscoelastic
properties were analyzed by conducting EPS adsorption
experiments to PVDF coated sensors in a QCM-D flow cell
(Kwon et al., 2006; Li and Wang, 2006; Voinova et al., 1999). As
a proof of concept, we used PVDF coated crystals as a model
that mimics membrane surface as a substratum for EPS to
delineate their adherence and viscoelastic characteristics.
EPS were extracted from the membrane surface at the end of
each of the fouling experiments (estimated SRT of 3, 5, 10 and
30 days). The final EPS solution was set to 20 mg DOC per
liter. Fig. 2(AeB) describes the decrease in frequency and
increase in dissipation energy of the PVDF crystal due to
adsorption of EPS originated from the membrane taken from
the MBR operated at different sludge removal rates. It should
be mentioned that EPS measurements with QCM-D are from
EPS that was reconstituted on the QCM-D sensor and due to
the methodology, physical characteristics of the EPS might be
different compare to the EPS on the membrane. Interestingly,
the results were very consistent with the effect of sludge
removal rate on membrane performance (Fig. 1). The highest
EPS adsorption rate expressed as a decrease in the crystal
frequency was observed for the EPS extracted from the
membrane fiber surface at sludge removal rate of 59 L day?1
(estimated SRT of 3 days) while the lowest EPS adsorption
rate was observed for the EPS originated from MBR operation
at sludge removal rate of 17.7 L day?1(estimated SRT of 10
days).
Thesimultaneousmeasurements
frequency Df are associated with changes in adsorbed mass
per area according to the Sauerbrey relation: Dm ¼ ?C/nDf,
where Dm is the mass adsorbed to the sensor, n is the over-
tone mumber (n ¼ 1, 3,.), and C is the mass sensitivity
constant of the crystal (C ¼ 17.7 ng Hz?1cm?2for a 5 MHz
quartz crystal). This relation holds for sufficiently thin, rigid,
and non-dissipative film with very limited viscoelastic
behavior. Biofilm in general, and EPS layers in particular, are
not rigid and they undergo deformation under shear oscilla-
tory motion. In this case, the fluidity of the film can be inferred
from the dissipation of the crystal oscillation. The dissipation
factor, D, is defined as the ratio of the dissipated and stored
energies according to the following: D ¼ Edissipated/2p Estored.
of thechangein
020406080100
0.0
0.5
1.0
1.5
r
o t c
a
F
n
o i t
a
p i s
s i
D
Time, Minutes
59 L·day-1
35.4 L·day-1
17.7 L·day-1
5.9 L·day-1
020406080100
-8
-6
-4
-2
0
Frequency Shift, Hz
Time, Minutes
59 L·day-1
35.4 L·day-1
17.7 L·day-1
5.9 L·day-1
DDW
10 mM
NaCl
10 mM
NaCl
10 mM
NaCl
NaCl
10 mM
NaCl
NaCl
DDW
DDW
DDW
EPS
EPS
AB
Fig. 2 e EPS adherence properties, extracted from the UF membrane, after runs operated at different sludge removal
rates (L dayL1): Frequency shifts (A) and dissipation factors (B) during EPS adsorption to PVDF coated QCM-D sensors.
A background solution of 10 mM NaCl and ambient pH of 6.2 supplemented with EPS at 20 mg DOC/L was injected to an
E1 QCM-D parallel flow cell (Q-Sense, SWEDEN) at a flow rate of 150 mL/min.
Water Research 45 (2011) 6430e6440
6435

Page 7

The shifts in dissipation (D) associated with the decreased
frequency (F ) during EPS adsorption to the PVDF coated
sensor are presented in Fig. 2B. Fig. 2B shows representative
dissipation shifts obtained during adsorption of EPS extracted
from the membrane originated from the MBR operated at
different sludge removal rate conditions. As previously re-
ported, the slope of DD over DF gives the magnitude of the
variations in the adsorbed layer fluidity, the main factor in
charge of damping the quartz vibration (Notley et al., 2005;
Schofield et al., 2007). This DD over DF shows the induced
energy dissipation per coupled unit mass, eliminating time as
an explicit parameter, and making it possible to analyze the
effects of EPS adsorption on the damping of the crystals’
resonance frequency.
The fluidic properties of the EPS layer on the crystal are
determined by studying this relationship, between the shifts
in dissipation (D) and the shifts in frequency (F ) obtained by
the QCM-D (Fig. 2). Harmonic 7 (35 MHz) was used for this
relation. For each of the sludge removal rate conditions,
a linear relationship was observed between D and F during the
EPSadsorptionontothecrystalsurface.Eachlinear
correlation corresponds to the EPS adsorption stage after
acquiring a baseline with the background solution of 10 mM
NaCl. The slopes of the linear relationship between D and F for
each of the HG-MBR operational conditions are shown in
Fig. 3. The trends observed for the change in slopes show an
interesting behavior in which at higher sludge removal rate of
35.5 and 59 L day?1(estimated SRT of 5 and 3 days, respec-
tively), theextractedEPSlayersaremore fluidcompared to the
EPS layers extracted from the membrane exposed to slower
sludge removal rates of 5.9 and 17.7 L day?1(estimated SRT of
30 and 10 days, respectively). It seems that in addition to
a higher EPS adherence (Fig. 2A), fluidity of the fouling layer is
likely playing an important role in its accessibility to the
membrane pores that eventually are being accumulated more
rapidly by the EPS extracted at a faster sludge removal rate
(estimated SRT of 5 and 3 days). Recently, using similar UF
membrane, we have shown that EPS fluidity and swelling
induced at high pH, have major contribution to pore clogging
(Sweity et al., 2011).
The fitted values of the elastic shear modulus and shear
viscosity of the adsorbed EPS layers were calculated using the
Voigt model (Voinova et al., 1999, Q-Tools software of the
QCM-D). The fitted values further confirmed the results
showing higher fluidity of EPS extracted from the faster fouled
membrane. The variations in these two parameters are
calculated for each EPS obtained from the membrane under
different conditions of sludge removal rate and are shown in
Table 3. It is shown in Table 3 that for the slower sludge
removal rate of 5.9 and 17.7 L day?1(estimated SRT 30 and 10
days), the EPS is much more viscoelastic. An ambitious study
would be to find the operational conditions that induce such
characters of the EPS that eventually deposits on the UF
membranes. As already mentioned, the way soluble EPS is
produced and deposited on the UF membrane is a complex
process affected by many parameters. Possible reasons for the
differences in the adherence and viscoelasticity of the EPS
originated from the membrane can be differences in the
biomass concentration in the HG-MBR (Supporting informa-
tion e Figure S1), feed to biomass (F/M) ratio (Supporting
information e Figure S2) as well as different levels of proteins
and polysaccharides in EPS at different locations in the HG-
MBR (Supporting information e Figure S3). In conclusion,
EPS extracted from the membrane operated at lower sludge
removal rate (longer estimated SRT) was more viscoelastic
with morerigidconformation analyzedin theQCM-D,whilein
contrast, a higher fluidity was detected for EPS extracted from
the membrane operated at faster sludge removal rates
(shorter estimated SRT) (Table 1 and Fig. 3).
Table 3 e Thickness, shear modulus, and viscosity of the deposited EPS layers extracted from the HG-MBR at different
removal rates (L dayL1) of MLSS (presented in duplicate).
59 L day?1
(estimated SRT ¼ 3 days)
Thickness, nm2.8 2.43.5
Viscosity, kg m?1s?1
0.0010 0.00120.0011
Shear modulus, Pa2.6∙104
3.2∙104
6.4∙104
35.5 L day?1
(estimated SRT ¼ 5 days)
17.7 L day?1
(estimated SRT ¼ 10 days)
2.2
0.0015
1.5∙105
5.9 L day?1
(estimated SRT ¼ 30 days)
3.3
0.0018
2.1∙105
2.6
0.0011
4.5∙104
2.8
0.0018
1.5∙105
2.8
0.0018
4.5∙105
-1 -2-3 -4-5 -6-7-8 -9
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
Disspation Factor
Frequency Shift, Hz
59 L·day-1 S=0.23±0.011
35.4 L·day-1 S=0.15±0.0088
17.7 L·day-1 S=0.11±0.0095
5.9 L·day-1 S=0.088±0.013
Fig. 3 e Comparison of the fluidity of different EPS,
extracted from the UF membrane, after runs operated at
different sludge removal rates (L dayL1): Dissipation
factors versus frequency shifts during adsorption. Slope of
the linear regression for the different plots is presented as
S, indicating the relative fluidity/rigidity of the EPS layer at
each condition (smaller slope relates to higher rigidity of
the layer). An ionic strength of 10 mM was adjusted with
NaCl at an ambient pH of 6.2 ± 0.1.
Water Research 45 (2011) 6430e6440
6436

Page 8

3.3.
and membrane fouling rate
The relation between EPS composition, adherence,
To further study EPS adherence and accumulation on the
membrane, filtration of the extracted EPS from the membrane
of the MBR was performed through a single UF fiber under
representativeionicstrengthof10mMwithandwithout0.5mM
of calcium cations (Sweity et al., 2011). Hence, faster decline in
membrane permeability was observed for EPS originated at
higher sludge removal rate of 59 L day?1(estimated SRT of 2
days) under both conditions (Fig.4A and B). The slowest decline
in membrane permeability was observed for sludge removal
rates of 17.7 and 5.9 L day?1(estimated SRT of 10 and 30 days),
with and without calcium (Fig. 4A and B). EPS composition and
amountper membrane surface areawas quantifiedand related
to the membrane fouling rate in the HG-MBR and in the single
fiber filtration unit as well as to the QCM-D analysis. For EPS
extraction and analysis, one fiber was cut from the membrane
module at the end of each experiment operated at a constant
sludge removal rate. Fig. 5 presents the EPS (as DOC content),
proteins, and polysaccharides accumulation on the membrane
surface (mg/cm2) of the HG-MBR at the different sludge removal
rates. The results show that the highest EPS accumulation on
themembraneoftheHG-MBRoccurredatasludgeremovalrate
of59L day?1(estimatedSRT of3 days),followed, inturn,during
MBR operation at sludge removal rates of 35.5, 5.9 and
17.7 L day?1(estimated SRT of 5, 30 and 10 days, respectively).
The protein accumulation at the estimated SRT of 10, 5 and 3
days was very similar and at a relatively low level (Fig. 5).
However,the accumulation
membrane surface exhibited a different behavior in contrast to
theproteins,inwhichanextremelyhighlevelofpolysaccharide
accumulationwasobserved for thehighest sludge removalrate
of 59 L day?1(estimated SRT of 3 days), while at sludge removal
rate of 17.7 L day?1(estimated SRT of 10 days), the lowest
accumulation was obtained (Fig. 5). Combining the results so
far, at the highest removal rate of sludge, polysaccharides
content in EPS is elevated on the membrane (Fig. 5) and in
general, per biomass unit (VSS), also in other locations in the
HG-MBR (Figure S4). The increased polysaccharides content on
the membrane is proposed to be a result of stronger adherence
properties of the EPS. This stronger adhesion of EPS, eventually
reduce membrane permeability observed in Fig. 4 and most
likely increase the rate of TMP elevation (bar/day) in the
MBR (Fig. 1).
It is generally accepted that polysaccharides can mediate
cohesion of cells, and play an important part in maintaining
the structural integrity of biofilms (Christensen, 1989; Liu and
Tay,2001;Ross,1984).Polysaccharidescanmediatecell-to-cell
interaction in two ways: firstly, polysaccharides bridge cells to
form a three-dimensional structure, which may then interact
with more bacterial cells and particulate matter (Ross, 1984);
secondly, dispersed bacteria are negatively charged at usual
pH values, and electrostatic repulsion exists between cells. It
had been proposed that extracellular polymers could change
the surface negative charge of bacteria, and thereby bridge
two neighboring bacterial cells physically to each other as well
as other inert particulate matter (Schmidt and Ahring, 1994;
Shen et al., 1993). In this study, the higher polysaccharides
content in EPS extracted at the fastest sludge removal rate
(estimated SRT of 3 days) showed the strongest adherence as
of polysaccharideson the
0100200 300400500
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Normalized Permeability
Time, Minutes
5.9 L·day-1
17.7 L·day-1
35.4 L·day-1
59 L·day-1
A
0100 200
Time, Minutes
300400 500
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
5.9 L·day-1
17.7 L·day-1
35.4 L·day-1
59 L·day-1
Normalized Permeability
B
Fig. 4 e Normalized permeability during fouling of a single
fiber UF membrane with EPS extracted from the HG-MBR-
UF membrane at the end of runs operated at different
sludge removal rates (L dayL1). Fouling experiments were
carried out at ionic strength of 10 mM (adjusted with NaCl)
with (A) and without (B) 0.5 mM calcium cations. The
pressure was set at all experiments between 0.14 and
0.18 bar. Initial permeability of the UF PVDF fibers (Zenon,
GE) was 0.15 ± 0.02 cm∙minL1∙barL1.
Fig. 5 e The concentrations of accumulated components of
EPS presented as TOC, proteins, and polysaccharides on
the membrane surface. Inserted figure shows protein/
polysaccharide ratio of EPS components on the membrane
surface.
Water Research 45 (2011) 6430e6440
6437

Page 9

analyzed in the QCM-D and led to a greater loss of filterability
in the single UF fiber unit (Figs. 2 and 4). Other studies also
correlatedbetweenpolysaccharides
membrane fouling rate in MBR systems (Rosenberger et al.,
2006; Fan et al., 2006). Interestingly, a higher fluidity of the
EPS adsorbed layer was observed for the EPS extracted from
the membrane at the faster sludge removal rate of 59 and
35.5 L day?1(estimated SRT of 3 and 5 days). The higher
fluidity of EPS is likely a part of increased accessibility of the
EPS to the membrane pores that eventually increase its
concentration within the membrane (Fig. 5).
EPS adherence properties are primarily affected by poly-
saccharides (Herzberg et al., 2009a,b) and corroborated with
PN/PS ratios analyzed in the EPS extracted from the
membrane at different sludge removal rates (Fig. 5, inserted
graph): Lower PN/PS ratio correlates well with the higher
adherence of the EPS observed by the QCM-D (Fig. 2). With
regard to the fluidity of the EPS layer, previous results in our
lab also showed a decrease in elasticity (lower shear
modulus) of EPS due to over-expression of alginate (Results
not shown) as well as a relation between increased EPS
swelling, fluidity, and reduced UF membrane permeability
(Sweity et al., 2011).
concentration and
3.4.
different SRTs
Variations in biofilm formation on the membrane at
A correlation between a decrease in membrane performance
at different sludge removal rates and an increase in the EPS
content on the membrane is observed in Figs. 1 and 6. Varia-
tions in biofilm formation (amount of EPS and viable cells)
were observed on the membrane surface using CLSM imaging
(Fig. 6AeD) and analyzed using COMSTAT biofilm software
(Heydorn et al., 2002, 2000a). On the membrane surface, at the
highest sludge removal rate of 59 L day?1(estimated SRT of 3
days), the highest polysaccharides content was analyzed
using the labeled lectin, concanavalin A (Figs. 5 and 6A). The
lowest polysaccharides content was observed at sludge
removal rate of 17.7 L day?1(estimated SRT of 10 days) (Figs. 5
and 6C). CLSM results also corroborate with the measured
fouling rate, in which at a sludge removal rate of 17.7 L day?1
(estimated SRT of 10 days), the increase in TMP was the lowest
and at sludge removal rate of 59 L day?1(estimated SRT of 3
days), the increase in TMP was the highest.
4.Concluding remarks
In this work, a novel approach of EPS analysis was taken for
studying membrane biofouling in HG-MBR system. EPS, orig-
inating from a fouled UF membrane was extracted and its
adherence and viscoelasticity were determined using QCM-D.
EPS was collected from the membrane under different fouling
conditions affected by the sludge removal rate from the HG-
MBR. The different fouling conditions of the UF membrane
were correlated well to EPS adherence, where stronger adhe-
sion of the EPS was observed for EPS extracted from the
fouling experiments conducted under conditions of higher
sludge removal rate, in which the TMP elevation rate was
higher. EPS layer fluidity, a new parameter to be used in
membrane fouling phenomena, as well EPS viscoelastic
properties can also explain the stronger fouling propensity of
EPS extracted from membranes with lower permeability. We
propose that the more fluidic the EPS layers are, their acces-
sibility to the membrane pores is higher, where they can
penetrate and block the pores. Shear modulus of elasticity and
shear viscosity are critical parameters influencing biofilm and
EPS cohesion (Ahimou et al., 2007; deKerchove and Elimelech,
2006). These parameters correlate to an improved membrane
performance: As the EPS in the membrane is more elastic and
viscous, reduced fouling is observed and the ratio of proteins
to polysaccharides is higher.
Fig. 6 e CLSM analysis of the biofouling layer on the membrane surface at the end of runs operated at different sludge
removal rates (L dayL1). Sludge removal rates are (A) 59 L dayL1; (B) 35.4 L dayL1; (C) 17.7 L dayL1; and (D) 5.9 L dayL1. Blue
and green spots represent extracellular polysaccharides and microorganisms, respectively. Total biomass of EPS and cells is
expressed as specific biovolume (mm3/mm2) as analyzed with COMSTAT biofilm software: (A) 95.8 ± 25.2; (B) 45.4 ± 10.1; (C);
31.1 ± 13.2 and (D) 57.2 ± 22.2 for EPS and (A) 28.5 ± 5.1; (B) 24.3 ± 3.6; (C) 6 ± 3.5; and (D) 13.1 ± 2.71 for viable cells. (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Water Research 45 (2011) 6430e6440
6438

Page 10

Acknowledgement
This study was supported by USAID Middle East Regional
Cooperation (MERC) Program, project number: M29-048.
Appendix. Supplementary material
Supplementary data associated with this article can be found
in the online version, at doi:10.1016/j.watres.2011.09.038.
r e f e r e n c e s
Ahimou, F., Semmens, M.J., Novak, P.J., Haugstad, G., 2007.
Biofilm cohesiveness measurement using a novel atomic force
microscopy methodology. Applied and Environmental
Microbiology 73, 2897e2904.
Al-Amoudi, A.S., Farooque, A.M., 2005. Performance restoration
and autopsy of NF membranes used in seawater pretreatment.
Desalination 178, 261e271.
Bradford,M.,1977.Arapidandsensitivemethodforthequantitation
of microgram quantities of protein utilizing the principle of
protein dye-binding. Analytical Biochemistry 72, 248e254.
Chang, I.-S., Lee, C.-H., 1998. Membrane filtration characteristics
in membrane-coupled activated sludge system e the effect of
physiological states of activated sludge on membrane fouling.
Desalination 120, 221e233.
Chang, I.S., Le Clech, P., Jefferson, B., Judd, S., 2002. Membrane
fouling in membrane bioreactors for wastewater treatment.
Journal of Environmental EngineeringeASCE 128, 1018e1029.
Cho, J., Song, K.-G., Ahn, K.-H., 2005a. The activated sludge and
microbial substances influences on membrane fouling in
submerged membrane bioreactor: unstirred batch cell test.
Desalination 183, 425e429.
Cho, J., Song, K.G., Yun, H., Ahn, K.H., Kim, J.Y., Chung, T.H.,
2005b. Quantitative analysis of biological effect on membrane
fouling in submerged membrane bioreactor. Water Science
and Technology 51, 9e18.
Christensen, B.E., 1989. The role of extracellular polysaccharides
in biofilms. Journal of Biotechnology 10, 181e201.
Comte, S., Guibaud, G., Baudu, M., 2006. Relations between
extraction protocols for activated sludge extracellular
polymeric substances (EPS) and EPS complexation properties
part I. Comparison of the efficiency of eight EPS extraction
methods. Enzyme and Microbial Technology 38, 237e245.
deKerchove, A.J., Elimelech, M., 2006. Structural growth and
viscoelastic properties of adsorbed alginate layers in
monovalentand divalent salts. Macromolecules39, 6558e6564.
DuBois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F.,
1956. Colorimetric method for determination of sugars and
related substances. Analytical Chemistry 28, 350e356.
Drews, A., 2010. Membrane fouling in membrane bioreactors e
characterisation, contradictions, cause and cures. Journal of
Membrane Science 363, 1e28.
Eydelnant, I.A., Tufenkji, N., 2008. Cranberry derived
proanthocyanidins reduce bacterial adhesion to selected
biomaterials. Langmuir 24, 10273e10281.
Fan, F., Zhou, H., Husain, H., 2006. Identification of wastewater
sludge characteristics to predict critical flux for membrane
bioreactor processes. Water Research 40, 205e212.
Feiler, A.A., Sahlholm, A., Sandberg, T., Caldwell, K.D., 2007.
Adsorption and viscoelastic properties of fractionated mucin
(BSM) and bovine serum albumin (BSA) studied with quartz
crystal microbalance (QCM-D). Journal of Colloid Interface
Science 315, 475e481.
Flemming, H.C., Wingender, J., 2001. Relevance of microbial
extracellular polymeric substances (EPSs) e part I: structural
and ecological aspects. Water Science and Technology 43, 1e8.
Gurdak, E., Dupont-Gillain, C.C., Booth, J., Roberts, C.J., Rouxhet, P.
G., 2005. Resolution of the vertical and horizontal
heterogeneity of adsorbed collagen layers by combination of
QCM-D and AFM. Langmuir 21, 10684e10692.
Han, S.S., Bae, T.H., Jang, G.G., Tak, T.M., 2005. Influence of sludge
retention time on membrane fouling and bioactivities in
membrane bioreactor system. Process Biochemistry 40,
2393e2400.
Herzberg, M., Kang, S., Elimelech, M., 2009a. Role of extracellular
polymeric substances (EPS) in biofouling of reverse osmosis
membranes. Environmental Science and Technology 43,
4393e4398.
Herzberg, M., Rezene, T.Z., Ziemba, C., Gillor, O., Mathee, K.,
2009b. Impact of higher alginate expression on deposition of
Pseudomonas aeruginosa in radial stagnation point flow and
reverse osmosis systems. Environmental Science and
Technology 43, 7376e7383.
Heydorn, A., Ersboll, B., Kato, J., Hentzer, M., Parsek, M.R.,
Tolker-Nielsen, T., Givskov, M., Molin, S., 2002. Statistical
analysis of pseudomonas aeruginosa biofilm development:
impact of mutations in genes involved in twitching motility,
cell-to-cell signaling, and stationary-phase sigma factor
expression. Applied and Environmental Microbiology 68,
2008e2017.
Heydorn, A., Ersboll, B.K., Hentzer, M., Parsek, M.R., Givskov, M.,
Molin, S., 2000a. Experimental reproducibility in flow-chamber
biofilms. Microbiology-UK 146, 2409e2415.
Heydorn, A., Nielsen, A.T., Hentzer, M., Sternberg, C., Givskov, M.,
Ersboll, B.K., Molin, S., 2000b. Quantification of biofilm
structures by the novel computer program COMSTAT.
Microbiology-UK 146, 2395e2407. Uk.
Judd, S., 2006. The MBR Book: Principles and Applications of
Membrane Bioreactors in Water and Wastewater Treatment.
Elsevier Science, London.
Kimura, K., Yamato, N., Yamamura, H., Watanabe, Y., 2005.
Membrane fouling in pilot-scale membrane bioreactors
(MBRs) treating municipal wastewater. Environmental Science
and Technology 39, 6293e6299.
Kraume, M., Drews, A., 2010. Membrane bioreactors in waste
water treatment e status and trends. Chemical Engineering
and Technology 33, 1251e1259.
Kwon, K.D., Green, H., Bjoorn, P., Kubicki, J.D., 2006. Model
bacterial extracellular polysaccharide adsorption onto silica
and alumina: quartz crystal microbalance with dissipation
monitoring of dextran adsorption. Environmental Science and
Technology 40, 7739e7744.
Laspidou, C.S., Rittmann, B.E., 2002. A unified theory for
extracellular polymeric substances, soluble microbial
products, and active and inert biomass. Water Research 36,
2711e2720.
Le-Clech, P., Chen, V., Fane, T.A.G., 2006. Fouling in membrane
bioreactors used in wastewater treatment. Journal of
Membrane Science 284, 17e53.
Lee, W., Kang, S., Shin, H., 2003. Sludge characteristics and their
contribution to microfiltration in submerged membrane
bioreactors. Journal of Membrane Science 216, 217e227.
Li, A.Y., Kothari, D., Corrado, J.J., 1984. Application of membrane
anaerobic reactor system for the treatment of industrial
wastewater. In: Proc. 39th Industrial Waste Conf. Purdue Univ.,
West Lafayette, Indiana, pp. 627e636.
Li, X.Y., Wang, X.M., 2006. Modelling of membrane fouling in
a submerged membrane bioreactor. Journal of Membrane
Science 278, 151e161.
Water Research 45 (2011) 6430e6440
6439

Page 11

Liu, H., Fang, H.H.P., 2002. Extraction of extracellular polymeric
substances (EPS) of sludges. Journal of Biotechnology 95,
249e256.
Liu, Y., Tay, J.H., 2001. Detachment forces and their influence on
the structure and metabolic behaviour of biofilms. World
Journal of Microbiology & Biotechnology 17, 111e117.
Masse, A., Sperandio, M., Cabassud, C., 2006. Comparison of
sludge characteristics and performance of a submerged
membrane bioreactor and an activated sludge process at high
solids retention time. Water Research 40, 2405e2415.
Meng, F., Chae, S., Drews, A., Kraume, M., Shin, H., Yang, F., 2009.
Recent advances in membrane bioreactors (MBRs): membrane
foulingandmembranematerial.WaterResearch43,1489e1512.
Morgan, J.W., Forster, C.F., Evison, L., 1990. A comparative study
of the nature of biopolymers extracted from anaerobic and
activated sludges. Water Research 24, 743e750.
Ng, H.Y., Tan, T.W., Ong, S.L., 2006. Membrane fouling of
submerged membrane bioreactors: impact of mean cell
residence time and the contributing factors. Environmental
Science and Technology 40, 2706e2713.
Nguyen, T.H., Elimelech, M., 2007. Adsorption of plasmid DNA to
a natural organic matter-coated silica surface: kinetics,
conformation, and reversibility. Langmuir 23, 3273e3279.
Notley, S.M., Eriksson, M., Wagberg, L., 2005. Visco-elastic and
adhesive properties of adsorbed polyelectrolyte multilayers
determined in situ with QCM-D and AFM measurements.
Journal of Colloid and Interface Science 292, 29e37.
Nuengjamnong, C., Kweon, J.H., Cho, J., Ahn, K.H., Polprasert, C.,
2005. Influence of extracellular polymeric substances on
membrane fouling and cleaning in a submerged membrane
bioreactor. Colloid Journal 67, 351e356.
Ouyang, K., Liu, J., 2009. Effect of sludge retention time on sludge
characteristics and membrane fouling of membrane
bioreactor. Journal of Environmental Sciences 21, 1329e1335.
Rojas, M.E.H., Van Kaam, R., Schetrite, S., Albasi, C., 2005. Role
and variations of supernatant compounds in submerged
membrane bioreactor fouling. Desalination 179, 95e107.
Rosenberger, S., Kraume, M., 2003. Filterability of activated sludge
in membrane bioreactors. Desalination 151, 195e200.
Rosenberger, S., Laabs, C., Lesjean, B., Gnirss, R., Amy, G., Jekel, M.,
Schrotter, J.-C., 2006. Impact of colloidal and soluble organic
material on membrane performance in membrane bioreactors
for municipal wastewater treatment. Water Research 40,
710e720.
Ross, W.R., 1984. The phenomenon of sludge pelletization in the
anaerobic treatment of a maize processing waste. Water SA
10, 197e204.
Schmidt, J.E., Ahring, B.K., 1994. Extracellular polymers in
granular sludge from different upflow anaerobic sludge
blanket (UASB) reactors. Applied Microbiology and
Biotechnology 42, 457e462.
Schofield, A.L., Rudd, T.R., Martin, D.S., Fernig, D.G., Edwards, C.,
2007. Real-time monitoring of the development and stability
of biofilms of streptococcus mutans using the quartz crystal
microbalance with dissipation monitoring. Biosensors and
Bioelectronics 23, 407e413.
Shen, C.F., Kosaric, N., Blaszczyk, R., 1993. The effect of selected
heavy-metals (Ni, Co and Fe) on anaerobic granules and their
extracellular polymeric substance (EPS). Water Research 27,
25e33.
Sombatsompop, K., Visvanathan, C., Ben Aim, R., 2006.
Evaluation of biofouling phenomenon in suspended and
attached growth membrane bioreactor systems. Desalination
201, 138e149.
Sweity, A., Ying, W., Belfer, S., Oron, G., Herzberg, M., 2011.
pH effects on the adherence and fouling propensity of
extracellular polymeric substances in a membrane bioreactor.
Journal of Membrane Science 378, 186e193.
Trussell, R.S., Merlo, R.P., Hermanowicz, S.W., Jenkins, D., 2006.
The effect of organic loading on process performance and
membrane fouling in a submerged membrane bioreactor
treating municipal wastewater. Water Research 40, 2675e2683.
Voinova, M.V., Rodahl, M., Jonson, M., Kasemo, B., 1999.
Viscoelastic acoustic response of layered polymer films at
fluid-solid interfaces: continuum mechanics approach.
Physica Scripta 59, 391e396.
Wang, X., Ruengruglikit, C., Wang, Y.W., Huang, Q., 2007.
Interfacial interactions of pectin with bovine serum albumin
studied by quartz crystal microbalance with dissipation
monitoring: effect of ionic strength. Journal of Agricultural
and Food Chemistry 55, 10425e10431.
Yang, Q., Chen, J., Zhang, F., 2006. Membrane fouling control in
a submerged membrane bioreactor with porous, flexible
suspended carriers. Desalination 189, 292e302.
Ying, W., Herzberg, M., Yang, F., Bick, A., Oron, G., 2009. Hybrid
growth membrane bioreactor (HG-MBR): the indirect impact of
sludge retention time on membrane fouling. Desalination and
Water Treatment 10, 27e32.
Water Research 45 (2011) 6430e6440
6440