close

Вход

Забыли?

вход по аккаунту

?

Fouling mechanisms of model polymeric substances.

код для вставкиСкачать
ASIA-PACIFIC JOURNAL OF CHEMICAL ENGINEERING
Asia-Pac. J. Chem. Eng. 2007; 2: 394–399
Published online 10 August 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI:10.1002/apj.071
Research Article
Fouling mechanisms of model polymeric substances
E. Negaresh, P. Le-Clech and V. Chen*
UNESCO Centre for Membrane Science and Technology, School of Chemical Science and Engineering, University of New South Wales, 2052 Sydney,
NSW, Australia
Received 17 October 2006; Revised 27 February 2007; Accepted 28 February 2007
ABSTRACT: Extracellular polymeric substances (EPS) and soluble microbial products (SMP) have been identified as
the main foulants in membrane bioreactor (MBR) operation. For a better understanding of the fouling mechanisms of
these substances, filtration of model solutions with submerged membranes was carried out. Filtration was conducted
following the flux-stepping method. In this work alginate was used as a model for polysaccharide, bovine serum albumin
(BSA) as a model for protein, and yeast (washed and unwashed) and bentonite were used for suspended solid contents.
Fouling behaviours and rejection propensities were also assessed for one-component and two-component solutions and
provided explanations for fouling mechanisms of carbohydrate and its interaction with protein and membrane in MBR
systems. During the filtration of most foulants (except BSA), introducing the alginate increased the reversible fouling
more than the irreversible one. In addition, a different behaviour was observed for the mixture of BSA and alginate.
Fouling rate was also affected by the introduction of alginate.  2007 Curtin University of Technology and John Wiley
& Sons, Ltd.
KEYWORDS: extracellular polymeric substances; membrane bioreactors; fouling; microfiltration
INTRODUCTION
Membrane bioreactor (MBR) combines the activated
sludge process with a membrane separation system and
is generally used in domestic or industrial wastewater treatment. MBR provides many advantages over
conventional treatments, such as small footprint and
reactor requirements, superior effluent quality, good
disinfection capability, and higher volumetric loading.
However, membrane fouling is a common problem in
membrane processes and it is even more difficult in
terms of predicting and controlling in MBRs: owing to
highly heterogeneous nature of the mixed liquor in the
bioreactor and also the effect of micro-organisms.
Extracellular polymeric substances (EPS) and soluble
microbial products (SMP) have been identified as the
main foulants in MBR operations. Both EPS and
SMP can be characterized by their relative levels
of polysaccharides, proteins, and, more rarely, lipids
and nucleic acids (Bura et al ., 1998; Laspidou and
Rittmann, 2002). For membrane units filtering activated
sludge, biofouling remains a major issue, as organic
adsorption and deposition on the membrane surface
significantly reduce hydraulic performances, leading to
*Correspondence to: V. Chen, UNESCO Centre for Membrane Science and Technology, School of Chemical Science and Engineering,
University of New South Wales, 2052 Sydney, NSW, Australia.
E-mail: v.chen@unsw.edu.au
 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
a rise in operational and maintenance costs. Therefore,
recent research has focussed on the mechanisms and the
control strategies of membrane fouling.
EPS are mostly produced by bacteria that participate in the formation of microbial aggregates whether
the bacteria grow in suspended culture or in biofilms.
EPS are mainly responsible for the structural and functional integrity of biofilms and are considered as the
key components that determine the physicochemical and
biological properties of biofilms. In general, the proportion of EPS in biofilms can vary between roughly
50% and 90% of the total organic matter (Nielsen
et al ., 1997). Chang and Lee (1998) measured the
EPS content quantitatively by separating the activated
sludge broth into three portions, i.e. cell, bulk and
EPS fraction. It was found that EPS was the major
component that was contributing to the total fouling
resistance. It was also found that, in any physiological state of the activated sludge tested, the content
of EPS in the activated sludge had a direct relation
with how the membrane fouling proceeded. The EPS
content of activated sludge was suggested as a probable index for membrane fouling in an activated sludge
MBR system. Carbohydrate was identified as the predominant constituent in the EPS of many pure cultures
(Cescutti et al ., 1999), whereas protein was found in
substantial quantities in the sludge of many wastewater treatment reactors (Jia et al ., 1996; Ruiz et al .,
1997).
Asia-Pacific Journal of Chemical Engineering
FOULING MECHANISMS OF MODEL POLYMERIC SUBSTANCES
SMPs are defined as soluble cellular components that
are released during cell lysis, which diffuse through the
cell membrane and are lost during synthesis (Laspidou
and Rittmann, 2002). Fouling mechanism of sodium
alginate as a model solution for EPS was measured
by (Ye et al ., 2005b). In that work, dead-end, unstirred
filtration was used. It was found that the standard poreblocking model and cake model are more applicable to
micro-filtration membranes. Characterizations of SMP
have been performed in different studies in terms of
their size, biodegradability, and toxicity using different
methods. 14 C-labelled compounds were used by (Boero
et al ., 1996) to determine the molecular weight (MW)
distribution of SMP in batch ultra-filtration cells.
Detailed studies of EPS and SMP filtration under controlled operation may provide a better understanding of
fouling propensity and process mechanisms in MBR
operation. The aim of this study is therefore to filter
model solutions for EPS and SMP in order to characterize fouling mechanisms and contributions of the
various components.
Figure 1. Schematic of the submerged fed with model
solutions. This figure is available in colour online at
www.apjChemEng.com.
MATERIALS AND METHODS
80
Experimental
 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
60
J(l.m-2.h-1)
A hydrophobic poly(vinylidene difluoride) (PVDF) hollow fibers membrane (0.006 m2 ) with pore size of
0.2 µm, internal diameter 0.39 mm and outer diameter 0.65 mm (Memcor, Australia) was submerged in
a reactor. For each experiment, a new membrane was
used and air scouring (0.12 l min−1 ) was introduced at
the bottom of the 2 l tank to induce mixing and limit
fouling formation (Fig. 1). Alginate (Sigma – Low viscosity) was used as a model foulant for polysaccharides
and bovine serum albumin (BSA) for protein analogues,
while bentonite and yeast (washed and unwashed) represented suspended solids content. Colorimetric method
(Peterson, 1979) was used in three different fluxes (10,
30, 50 l m−2 h−1 ) to measure the concentration of carbohydrate in the feed and permeate solutions.
Yeast (Tandaco Dry yeast, Cerebos Gregg’s Ltd.) was
used in two forms, washed and unwashed, as a model
(biomass) particle in this work. The washing procedure
was as follows: 2 g yeast was dissolved in 200 g water,
and then centrifuged at 2000 rpm for 15 min. The
supernatant liquid was then drawn out using a syringe.
The above process was repeated three times. The mean
particle size for washed yeast is smaller than unwashed
yeast because of the broken cells resulting from mixing
with a magnetic stirrer. Bentonite (Aldrich Chemicals)
is another model particulate used in this study. The
mean size and zeta potential are reported in Table 1.
The mean size was measured by a Zetasizer for alginate
and by a particle MasterSizer/E (Malvern) for other
particles.
Washed Yeast
Unwashed Yeast
BSA
Alginate
Bentonite
New membrane
70
50
40
30
20
10
0
0
2
4
6
TMP(kPa)
8
10
12
Figure 2. Fouling behaviour during flux-stepping filtration
of one-component model solutions.
Table 1. Particles size and zeta potential for model
solutions.
pH value
Solution
Particle
size
2.6
3
4.8
8.0
Alginate
BSA
Bentonite
Washed yeast
Unwashed yeast
0.2 µm
67 kDa
4.6 µm
5.2 µm
6.9 µm
i.p.
+
–
–
–
–
+
–
–
–
–
i.p.
–
–
–
–
–
–
–
–
i.p. = isoelectric point.
Zeta potential was measured in the pH range 3–8;
Negative values were observed for alginate, washed
yeast, unwashed yeast and bentonite in the operating
Asia-Pac. J. Chem. Eng. 2007; 2: 394–399
DOI: 10.1002/apj
395
396
E. NEGARESH, P. LE-CLECH AND V. CHEN
Asia-Pacific Journal of Chemical Engineering
range. Isoelectric point of the alginate was found to
be around pH 2.6 (Table 1). At the pH value of 4.8,
isoelectric point was observed for BSA (Kelly and
Zydney, 1997).
Evaluation of fouling propensity
Flux-stepping filtrations (step duration: 15 min, step
height around 10 l m−2 h−1 ) were carried out to assess
the fouling propensity of each solution (Le-Clech et al .,
2003) and to allow comparison between the different
operating conditions tested. Equation (1) was used for
calculation of permeability (K ).
K =
J
TMPav
(1)
where J (l m−2 h−1 ) is the permeate flux and TMPav
(kPa) is the average of all transmembrane pressures
(TMP) recorded at the corresponding flux.
The other parameter used in this study is dp/dt,
indicating the fouling rate.
TMPf − TMPi
dp
=
dt
tf − ti
(2)
where TMPi and TMPf are the TMP values obtained at
the beginning and end of the step of flux, respectively.
For comparison of dp/dt in individual and mixture
solutions, the ratio of fouling rates was calculated:
(
dp
dp
)with ( )without
dt
dt
(3)
where (dp/dt) is the fouling rate, in the last flux step for
solution mixed with alginate (with) and without alginate
(without).
Darcy’s law was used to evaluate the filtration
resistance during the filtration:
J =
TMP
TMP
=
µRt
µ(Rm + Rf )
(4)
where µ is the viscosity of permeate (Pa s), Rt the total
filtration resistance, Rm the virgin membrane resistance
and Rf the fouled membrane resistance (m−1 ).
Rf = Rr + Rirr
(5)
where Rr is the reversible fouling and Rirr the irreversible fouling. After the filtration membrane was
backwashed by Milli-Q water (10 min, 120 l m−2 h−1 )
and based on clean water test in the constant flux, Rbw
(resistance for backwashed membrane) was measured
(Rbw = Rm + Rirr ).
 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
Table 2. Fouling resistance and recovery percent
after backwashing for individual and mixed model
solutions.
Recovery
after BW (%)
86
87
90
96
92
80
82
67
Rf (m−1 × 10−11 )
1.13
0.96
0.52
0.71
1.99
3.78
2.16
2.63
Solution
Unwashed yeast
Washed yeast
Bentonite
BSA
Al-uwy
Al-wy
Al-bentonite
Al-BSA
Recovery percentage of permeability was calculated
by using Rm and Rbw (Table 2).
RESULTS AND DISCUSSION
One-component solutions
For investigating the fouling propensity of each model
solution, filtration was carried out individually. For
these series of filtration, a concentration of 100 mg l−1
was used.
Results clearly revealed the highest fouling resistance
for alginate among the foulants (Figure 2). Significant
fouling was observed at fluxes as low as 10 l m−2 h−1 .
During the filtration of alginate, the permeability gradually declined from 10.5 to 4.9 l m−2 h−1 kPa−1 . On
the other hand, bentonite was found to be the least fouling component. Membrane permeability decreased from
11.4 (for new membrane) to 8.2 l m−2 h−1 kPa−1 as
the flux was increased from 10 to 80 l m−2 h−1 . These
results were expected since the creation of a highly fouling hydrogel has been previously reported during the
filtration of alginate (Ye et al ., 2005b). However, the
observed similar fouling behaviour between washed and
unwashed yeast was more surprising, since the soluble
materials (SMP) present in the unwashed yeast solution
were suspected to further contribute to the membrane
fouling.
Results revealed that not only Rf obtained with alginate is the highest (3.91 × 1011 m−1 ) but the recovery
of permeability after backwashing was also the lowest (74%) in these series of filtration. While most of
the BSA (96%) was removed by simple backwashing,
unwashed and washed yeast and bentonite irreversibly
fouled the membrane during these short-term experiments (Table 2).
Two-component solutions
For a better understanding of the fouling mechanisms
of EPS and SMP in MBR systems, evaluations of the
Asia-Pac. J. Chem. Eng. 2007; 2: 394–399
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
FOULING MECHANISMS OF MODEL POLYMERIC SUBSTANCES
fouling propensity for a mixture of alginate with washed
(Al-wy), unwashed yeast (Al-uwy), BSA (Al-BSA) and
bentonite (Al-bentonite) were carried out. In these series
of filtration, the concentration of each chemical was
50 mg l−1 . All solutions were well mixed by using a
magnetic stirrer. The hydraulic conditions were similar
to those for the one-component experiments.
Results reveal the highest fouling resistance of Alwy over the other foulants (Fig. 3). During the Alwy test, permeability gradually declined from 11.4 to
5.2 l m−2 h−1 kPa−1 . On the other hand, Al-uwy was
found to be the least fouling component. Membrane
permeability decreased from 10.6 (for a new membrane)
to 6.7 l m−2 h−1 kPa−1 as the flux was increased from
10 to 80 l m−2 h−1 (Table 2). Although the washing
procedure was supposed to remove most of the cell
debris and soluble material, the fouling propensity of
washed yeast mixed with alginate was higher than those
of the unwashed. It could be due to the degree of
mixing carried out during the washing procedure, which
could have fractured the yeast into smaller particles with
higher fouling potential.
However, fouling resistance of Al-bentonite was
similar to Al-uwy: especially in first three flux steps.
It was observed that Al-BSA has the same fouling
propensity as alginate (50 mg l−1 ).
Results showed that Rf obtained after filtration of Alwy was the highest (3.78 × 1011 m−1 ), and the recovery
of permeability after backwashing was 80% (Table 2).
Al-BSA had the lowest recovery of permeability after
backwashing (62%). Although Al-wy, Al-uwy, Albentonite were found to be easily removable by simple
backwashing, they irreversibly fouled the membrane
during these short-term experiments.
Fouling mechanisms
The effect of polysaccharide in fouling mechanisms,
and hydraulic resistance of model solutions were investigated individually and in mixture to allow easier comparison between the different solutions tested.
80
Alginate
Al-uwy
Al-wy
Al-bsa
Al-Bentonite
New membrane
70
60
50
100
80
40
%R
J(l.m-2.h-1)
Although the alginate has the highest fouling resistance (3.91 × 1011 m−1 ) among the chemicals, most of
the fouling is reversible (70%) (Fig. 4). Fouling of alginate is more in the form of pore blockage and cake
formation (hydrogel layer) rather than pore narrowing (internal fouling). Moreover, high-molecular-weight
polysaccharides present good adhesive and gelling properties, which play an important role in formation of
sticky hydrogels on the membrane surface (Frank and
Belfort, 2003).
Although the fouling resistance of the bentonite was
modest (0.52 × 1011 m−1 ), it has the highest irreversible
form of fouling 71% (Fig. 3). It is expected to have
more cake formation rather than internal fouling due to
the mean particle size. A similar result was reported
by Broussous et al . (2001). Introducing the alginate
enhanced the fouling resistance of bentonite to 2.16 ×
1011 m−1 , and a variation in fouling mechanism was
also observed. Al-bentonite became more reversible
(65%) than bentonite alone (29%). It could be due
to the aggregations between bentonite and alginate,
during which small particles adhered to the alginate
aggregation and did not enter the membrane pores.
Washed and unwashed yeast have similar fouling
resistance in one-component solutions. Nonetheless,
fouling of washed yeast is more reversible (77%) in
contrast to unwashed yeast (53%). This difference was
expected, since in commercially available baker’s yeast
some cell debris and solubles are present; however, they
were discarded with the supernatant during the washing
step.
The presence of alginate enhanced the fouling resistance of washed yeast from 0.96 × 1011 m−1 for individual to 3.78 × 1011 m−1 for mixture (Fig. 4). However, no significant changes were observed in terms of
fouling mechanisms, compared to the individual solution. For both cases fouling appeared to be reversible.
A constant value was observed for rejection at imposed
fluxes of 10, 30 and 50 l m−2 h−1 (88%), which might
be due to porous cake formation.
60
30
40
20
20
10
With alginate
Without alginate
0
0
0
2
4
6
TMP(kPa)
8
10
12
Figure 3. Fouling behaviour during flux-stepping filtration
of two-component model solutions.
 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
Unwashed Washed
yeast
yeast
%Rr
Bentonite
BSA
%Rirr
Figure 4. Different resistances for foulants individually and
mixed with alginate.
Asia-Pac. J. Chem. Eng. 2007; 2: 394–399
DOI: 10.1002/apj
397
398
E. NEGARESH, P. LE-CLECH AND V. CHEN
Alginate increased fouling resistance of unwashed
yeast (from 1.13 × 1011 for individual to 1.99 ×
1011 m−1 for mixture) as well as formation of reversible
fouling, (up to 81%). Formation of reversible fouling
could be due to aggregation of cell debris and big particles present in the unwashed yeast. The same behaviour
was observed in rejection of carbohydrate for both Aluwy and Al-wy (88%). This might be due to porous
cake formation.
According to Fig. 4, BSA (fouling resistance of
0.71 × 1011 m−1 ) has the highest percentage of
reversible form of fouling (87%). It is found that the
fouling of BSA during micro-filtration occurs by two
different mechanisms (Kelly and Zydney, 1995): physically, deposition of large protein aggregates on or inside
the membrane structure, and chemically, attachment
of native (non-aggregated) BSA to these previously
deposited aggregates. This chemical addition occurs
via the formation of an intermolecular disulfide linkage between the bulk BSA and the protein aggregates
on the membrane surface. With these reactions, blocking the free sulfhydryl group on the BSA molecules is
completely inhibited. As a result, easy backwashing of
protein from the membrane surface can be caused by
deposition of large protein aggregations.
In presence of BSA, the existence of alginate in
the filtration system increased the fouling resistance to
2.63 × 1011 m−1 in the mixture. Irreversible form of
fouling increased as a result of alginate (from 13% to
71%). However, the presence of alginate in the mixture
feed solution seems to slightly loosen the cake formed
by BSA, perhaps by disrupting binding between alginate
aggregates (Neiser et al ., 1999; Yamasaki et al ., 2005).
Similar results were reported by Ye and Chen (2005a)
in cross-flow filtration of alginate and BSA with flat
sheet PVDF membrane.
As shown in Figs 3 and 4, unexpectedly, the fouling
propensity of Al-wy (3.78 × 1011 m−1 ) is higher than
that of Al-uwy (1.99 × 1011 m−1 ). It is probably due
to the stronger interactions between the broken washed
yeast (as a result of mixing by stirrer) cells with alginate
compared to unwashed yeast. This connection can
make a strong network, which increases the aggregation
on the surface of the membrane. Fouling rate was
calculated for all individual and mixed model solutions.
By introducing the alginate, fouling rates was found to
decrease in all flux steps. It was possibly due to the
different types of aggregations. Furthermore, fouling
is more in the form of hydrogel formation rather than
cake formation. Ratio of fouling rate in last flux steps
was measured by Eqn (3) (Table 3). It was shown that
although alginate increases the fouling resistance (Rf ),
it decreases the fouling rate in flux steps.
 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
Asia-Pacific Journal of Chemical Engineering
Table 3. The proportion of differentiation of individual
solution to mixture model solution.
(dp/dt)f with
/(dp/dt)f without
Unwashed yeast
Washed yeast
Bentonite
BSA
0.6
0.7
0.4
0.8
CONCLUSIONS
In this work, filtrations of model EPS and SMP solutions were conducted with submerged membrane modules.
Although these experiments were carried out at subcritical fluxes, they remain useful tools for comparing
operating conditions and the fouling propensity of
different model solutions. Fouling resistance (Rf ) is
one of the main parameters for comparing different
systems. From the results, the fouling propensity of each
model solution can be classified as follows: Alginate >
unwashed yeast > washed yeast > BSA > bentonite
for one-component solution and Al-wy > Al-BSA >
Al-bentonite > Al-uwy for two-component solutions.
During the filtration of all the foulants (except BSA),
introducing the alginate increased the reversible fouling more than the irreversible one. It is possibly
due to the different forms of aggregation occurring
between species. The calculated fouling rate (dp/dt)
also clearly indicated the higher fouling propensity of
one-component solutions compared to two-component
mixtures. However, high-molecular-weight polysaccharides, such as alginates, present thickening and gelling
properties and may play a strong role in the formation of adhesive hydrogels on membrane surfaces rather
than formation of simple particulate cakes. Different
behaviour in the mixture of BSA and alginate was
observed, which is possibly due to movement of binding
between alginate aggregates.
REFERENCES
Boero VJ, Bowers AR, Eckenfelder WW Jr. Molecular weight
distribution of soluble microbial products in biological systems.
Water Sci. Technol. 1996; 34(5–6): 241–248.
Broussous L, Prouzet E, Beque L, Larbot A. An experimental study
of helically stamped ceramic microfiltration membranes using
bentonite suspensions. Separat. Purificat. Technol. 2001; 24(1–2):
205–221.
Bura R, Cheung M, Liao B, Finlayson J, Lee BC, Droppo IG,
Leppard GG, Liss SN. Composition of extracellular polymeric
substances in the activated sludge floc matrix. Water Sci. Technol.
1998; 37(4–5): 325–333.
Cescutti P, Toffanin R, Pollesello P, Sutherland IW. Structural
determination of the acidic exopolysaccharide produced by a
Pseudomonas sp. strain 1.15. Carbohydr. Res. 1999; 315(1–2):
159–168.
Asia-Pac. J. Chem. Eng. 2007; 2: 394–399
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering
FOULING MECHANISMS OF MODEL POLYMERIC SUBSTANCES
Chang IS, Lee CH. Membrane filtration characteristics in coupled
activated sludge system – the effect of physicological states of
activated sludge on membrane fouling. Desalination 1998; 120:
221–233.
Frank BP, Belfort G. Polysaccharides and sticky membrane
surfaces: critical ionic effects. J. Memb. Sci. 2003; 212(1–2):
205–212.
Jia XS, Furumai H, Fang HHP. Extracellular polymers of hydrogenutilizing methanogenic and sulfate-reducing sludges. Water Res.
1996; 30(6): 1439–1444.
Kelly ST, Zydney AL. Mechanisms for BSA fouling during
microfiltration. J. Memb. Sci. 1995; 107(1–2): 115–127.
Kelly ST, Zydney AL. Protein fouling during microfiltration:
Comparative behavior of different model proteins. Biotechnol.
Bioeng. 1997; 55(1): 91–100.
Laspidou CS, Rittmann BE. A unified theory for extracellular
polymeric substances, soluble microbial products, and active and
inert biomass. Water Res. 2002; 36(11): 2711–2720.
Le Clech P, Jefferson B, Chang IS, Judd SJ. Critical flux
determination by the flux-step method in a submerged membrane
bioreactor. J. Memb. Sci. 2003; 227(1–2): 81–93.
 2007 Curtin University of Technology and John Wiley & Sons, Ltd.
Neiser S, Draget KI, Smidsrod O. Interactions in bovine serum
albumin-calcium alginate gel systems. Food Hydrocoll. 1999;
13(6): 445–458.
Nielsen PH, Jahn A, Palmgren R. Conceptual model for production
and composition of exopolymers in biofilms. Water Sci. Technol.
1997; 36(1): 11–19.
Peterson GL. Review of the folin phenol protein quantitation method
of lowry, rosebrough, farr and randall. Anal. Biochem. 1979;
100(2): 201–220.
Ruiz I, Veiga MC, de Santiago P, Blazquez R. Treatment of
slaughterhouse wastewater in a UASB reactor and an anaerobic
filter. Bioresour. Technol. 1997; 60(3): 251–258.
Yamasaki M, Ogura K, Hashimoto W, Mikami B, Murata K. A
structural basis for depolymerization of alginate by polysaccharide
lyase Family-7. J. Mol. Biol. 2005; 352(1): 11–21.
Ye Y, Chen V. Reversibility of heterogeneous deposits formed from
yeast and proteins during microfiltration. J. Memb. Sci. 2005a;
265(1–2): 20–28.
Ye Y, Le-Clech P, Chen V, Fane AG. Evolution of fouling during
crossflow filtration of model EPS solutions. J. Memb. Sci. 2005b;
264(1–2): 190–199.
Asia-Pac. J. Chem. Eng. 2007; 2: 394–399
DOI: 10.1002/apj
399
Документ
Категория
Без категории
Просмотров
0
Размер файла
132 Кб
Теги
mode, substances, fouling, mechanism, polymeric
1/--страниц
Пожаловаться на содержимое документа