Journal of Hazardous Materials 284 (2015) 121129

Contents lists available at ScienceDirect

Journal of Hazardous Materials

journal homepage: www.elsevier.com/locate/jhazmat

Comparative study on the treatment of raw and biologically treated

textile efuents through submerged nanoltration
Qing Chen a,b , Ying Yang a,b , Mengsi Zhou a,b , Meihong Liu a,b , Sanchuan Yu a,b, ,
Congjie Gao c
a

Department of Chemistry, Zhejiang Sci-Tech University, Hangzhou 310018, Peoples Republic of China
Engineering Research Center for Eco-Dyeing & Finishing of Textiles, Ministry of Education of China, Zhejiang Sci-Tech University, Hangzhou 310018,
Peoples Republic of China
c
The Development Center of Water Treatment Technology, Hangzhou 310012, Peoples Republic of China
b

h i g h l i g h t s

Raw textile efuent was efciently treated through submerged nanoltration.

A comparison between raw and biologically treated textile efuents was conducted.
Increase of TMP resulted in decreased water permeability, COD and color removal.
Increase of VCF resulted in both increased COD reduction and color removal.
Higher COD reduction and color removal were obtained with the raw efuent.

a r t i c l e

i n f o

Article history:
Received 14 September 2014
Received in revised form 9 November 2014
Accepted 13 November 2014
Available online 15 November 2014
Keywords:
Textile efuent
Nanoltration
Submerged ltration
Color removal
COD reduction

a b s t r a c t
Raw and biologically treated textile efuents were submerged ltrated using lab-fabricated hollow ber
nanoltration membrane with a molecular weight cut-off of about 650 g/mol. Permeate ux, chemical
oxygen demand (COD) reduction, color removal, membrane fouling, and cleaning were investigated and
compared by varying the trans-membrane pressure (TMP) and volume concentrating factor (VCF). It
was found that both raw and biologically treated textile efuents could be efciently treated through
submerged nanoltration. The increase of TMP resulted in a decline in water permeability, COD reduction,
color removal, and ux recovery ratio, while the increase of VCF resulted in both increased COD reduction
and color removal. Under the TMP of 0.4 bar and VCF of 5.0, uxes of 1.96 and 2.59 l/m2 h, COD reductions
of 95.7 and 94.2%, color removals of 99.0, and 97.3% and ux recovery ratios of 91.1 and 92.9% could be
obtained in ltration of raw and biologically treated efuents, respectively. After ltration, the COD and
color contents of the raw efuent declined sharply from 1780 to 325 mg/l and 1.200 to 0.060 Abs/cm,
respectively, while for the biologically treated efuent, they decreased from 780 to 180 mg/l and 0.370
to 0.045 Abs/cm, respectively.
2014 Elsevier B.V. All rights reserved.

1. Introduction
Efuents from textile industry are generally loaded with spent
dyes and chemical additives, and high in both biochemical oxygen demand (BOD) and chemical oxygen demand (COD) [1,2].
The extremely polluted textile efuents must be properly treated

Corresponding author at: Zhejiang Sci-Tech University, Department of Chemistry, Hangzhou 310018, Peoples Republic of China. Tel.: +86 571 86843217;
fax: +86 571 86843217.
E-mail addresses: yuschn@163.com, yuschn@mail.hz.zj.cn (S. Yu).
http://dx.doi.org/10.1016/j.jhazmat.2014.11.009
0304-3894/ 2014 Elsevier B.V. All rights reserved.

before disposal or reuse, since the presence of dyes even at a very

low concentration is highly visible and toxic to aquatic life through
damaging the esthetic nature of water and reducing the photosynthetic activity of aquatic organisms [35].
Biological and physicalchemical processes with previous
screening and pH adjusting steps are commonly adopted to treat
the textile efuents. Biological process using activated sludge offers
high efciency in the reduction of chemical oxidation demand
(COD), but can not completely eliminate the color due to the
biodegradable difculty of the dyes presented in the efuents [6,7].
Physicalchemical process, which reduces dissolved, suspended,
colloidal, and non-settleable materials from wastewater through

122

Q. Chen et al. / Journal of Hazardous Materials 284 (2015) 121129

chemical coagulationocculation followed by gravity settling, can

completely eliminate the color, but has the main drawbacks of
high chemical cost and low removal efciency of soluble COD
[8]. Advanced oxidation processes including chemical oxidation,
photocatalytic degradation, and electrochemical treatment that
have been proven to be alternatives for better treatment of textile
efuents are still quite costly for practical application [914]. Therefore, more advanced treatment processes are required to meet the
increasing stringent legislations on wastewater discharge, the necessary of wastewater reclamation, the reduction of the disposal
cost, as well as the minimization of the volume of the waste discharged.
Membrane technology, as an advanced separation technology
for water desalination and purication, has attracted more attention in treatment and reclamation of textile efuents for its high
treatment efciency and quality of treated water [15,16]. Among
the pressure-driven membrane processes of reverse osmosis (RO),
nanoltration (NF), ultraltration (UF), and microltration (MF)
that have been adopted for the treatment of textile efuents,
nanoltration has been proven to be the most effective one for its
unique separation characteristics and lower energy consumption
compared with reverse osmosis [17]. For example, Amar et al. [18]
found that, as the tertiary treatment of textile efuent for water
reuse, the NF membrane exhibited lower total dissolved solid rejection and higher yield compared with the RO membrane. Bes-Pi
et al. [19] tested and compared the behaviors of six nanoltration membranes TFC-SR2, ESNA, NF270, DS-5 DK, DS-5 DL, and
Duraslick in reclamation of a secondary textile efuent. They concluded that all the tested membranes yielded permeate streams
of high quality meeting reuse criteria. Ellouze et al. [20] applied
nanoltration as the post treatment of coagulationocculation
process to enhance the treatment of textile efuent and reported
that color retention of 100% was achieved. More recently, Sun et al.
[21] developed a novel double-repulsive NF hollow ber membrane
by cross-linking hyperbranched polyethyleneimine on the out surface of a polyamide-imide substrate hollow ber. Ong et al. [22]
adopted the membrane to treat textile wastewater through pilotscale tests. It was reported that the rejections to various tested dyes
were higher than 90%, to NaCl and Na2 SO4 were lower than 20 and
10%, respectively. When this hollow ber nanoltration membrane
was combined with chemical/occulation process, the removal of
the tested dyes could reach almost 100% [23].
However, nanoltration process now used were generally operated under the model of tangential pressurized ltration, which
has the main drawbacks of high energy consumption and severe
membrane fouling even with the pretreatment of activated sludge
treatment or coagulationocculation process[24,25]. Therefore,
innovative research still needs to be conducted on improvement
of the treatment efciency of nanoltration process.
In consideration of the fact that submerged ltration technology
has the advantages of lower operation pressure, energy consumption and cleaning requirement compared to the pressurized
ltration model [26,27], submerged ultraltration (UF) or microltration (MF) systems have been widely adopted for the advanced
treatment of wastewater in membrane bioreactors [28,29] and the
pretreatment of seawater in seawater reverse osmosis (SWRO)
desalination plants [30,31]. In recent years, membrane scientists have also devoted their attentions to the application of this
technology in nanoltration process [3235]. Compared with the
conventional submerged membrane bioreactor (SMBR) system
using UF/MF membrane, the combination of biological treatment and submerged nanoltration will produce ltrated efuent
of better quality, but has relatively lower water permeability.
In our previously study, submerged nanoltration with a negatively charged NF hollow ber membrane has been successfully
applied to the treatment of a specic biologically treated textile

Table 1
Properties of the lab-fabricated thin-lm composite hollow ber nanoltration
membrane used in this study.
Parameter

Unit

Value

Pure water permeability (PWP)a

Molecular weight cut-off (MWCO)b
Surface zeta potential at pH 7.0c
Rejection to NaCld
Rejection to Na2 SO4 d
Inner diameter of hollow ber (din )
Outer diameter of hollow ber (dout )

l/m2 h bar
g/mol
mV
%
%
mm
mm

9.2
650
2.5
10.5
8.0
0.40
0.51

0.3
40
0.8
1.0
0.8

Tested with de-ionized water under TMP of 0.8 bar and 25.0 C.
Obtained from the retention cure of PEG with different molecular weights
according to the method described in [37].
c
Obtained from the measured streaming potential according to the
HelmholtzSmoluchowski equation with the Fairbrother and Mastin substitution
[38].
d
Tested with 500 mg/l salt aqueous solution under TMP of 0.8 bar, 25.0 C, and
pH 7.0.
a

wastewater [36]. The color and COD presented in the biologically

treated textile wastewater could be effectively removed under low
suction pressures ranging from 0.5 to 0.9 bar.
Accordingly, in this work, we tried to apply submerged nanoltration with a positively charged NF hollow ber membrane to
the treatment of raw textile efuent and make a comparison with
the treatment of biologically treated textile efuent. Submerged
ltration tests were conducted employing a tailor-fabricated hollow ber membrane module under different trans-membrane
pressures (TMPs) and volume concentrating factors (VCFs). Water
permeate ux, COD reduction, color removal, membrane fouling as
well as membrane cleaning were investigated and compared.
2. Experimental
2.1. Materials
The raw and biologically treated textile efuents used in this
study were taken from a textile industry located in Zhejiang
Province, China. The biologically treated efuent was the output
of the existing wastewater treatment plant composed of screening,
acidication, activated sludge process, and clarier, while the raw
textile was only pretreated with screening, acidication, and clarier.
The nanoltration membrane used in this study was thinlm composite hollow ber nanoltration membrane with a
selective layer of cross-linked polyvinyl alcohol (PVA) and
polyquaternium-10 (a reaction product of hydroxyethyl cellulose
with a trimethylammonium substituted epoxide) on the outside
of the polypropylene microporous support hollow ber. The membrane was prepared in our lab according to the method described
in our previous work [35] and its properties are summarized in
Table 1. The reason of choosing this NF membrane is its relatively
low rejections to NaCl and Na2 SO4 , which generally presented in
the textile efuent.
De-ionized water with a resistivity of 18.0 M was used as pure
water. Other chemicals involved were all analytical grade and used
without further purication.
2.2. Lab-scale submerged ltration setup and tests
The lab-scale experimental setup for submerged ltration was
nearly the same as that described in our previous study [36]. It
consists of a suction pump, a 10 l feed tank equipped with a temperature controller, an aerator device and a submerged hollow ber
membrane module. Aeration was adopted to reduce membrane
fouling through scouring. Each module contains fty hollow bers

Q. Chen et al. / Journal of Hazardous Materials 284 (2015) 121129

of 40.0 cm length and has an effective outer surface ltration area

of 320.0 cm2 . The module was performed in an outside-to-inside
model under the constant temperature of 25.0 C and a certain TMP
generated by the suction pump.
Submerged ltration tests were rstly carried out at different
TMPs of 0.4, 0.5, 0.6, 0.7, and 0.8 bar under recirculation model,
through which the permeate stream was recycled into the feed tank
to keep the feed concentration approximately constant. The volume
of the feed solution for each experiment was 5.0 l. Each ltration
test consists of four steps: (1) ltration of de-ionized water under
a certain TMP for 60 min to determine the pure water ux of the
fresh membrane module (Jwi ); (2) ltration of raw or biologically
treated textile efuent under the same TMP to investigate the timedependent ux, color removal, and COD reduction until a steady
ux (Jws ) is obtained; (3) washing the fouled membrane module
with owing de-ionized water for 60 min to remove the foulants
that loosely deposited on the membrane surface; and (4) ltration
of de-ionized water under the same TMP as step 1 to determine the
pure water ux of the cleaned membrane module (Jwc ).
Submerged ltration tests were also conducted at the constant TMP of 0.4 bar under concentrating model, through which
the permeate stream was withdrawn from the system. The volume
concentrating factor (VCF) was calculated by Eq. (1):
VCF =

Vf
Vf Vp

(1)

where Vf and Vp are the volumes of the initial feed solution and
total withdrawn permeate stream, respectively. The volume of the
initial feed solution (Vf ) for volume reduction experiment was 9.0 l.
The performance at certain VCF was studied by checking the permeate water ux, color removal as well as COD reduction. After
each run of certain VCF, the fouled membrane module was washed
with owing de-ionized water for 60 min and measured for the ux
of de-ionized water (Jwc ) under the same TMP of 0.4 bar.
2.3. Analytical methods
Permeate volume collected over a certain period was measured
to determine the permeate ux. The COD concentrations were
determined by a HACH DR1010 (USA) direct reading spectrophotometer using a HACH COD reactor, according to the standard
methods [39]. Conductivity of all the samples was measured by
a conductance meter (DDSJ-308A, Cany Precision Instruments Co.,
Ltd., China). Color of each sample was analyzed by spectrophotometer (UV759, Shanghai) and calculated using the Eq. (2) [20]:
Color =

A2530 + A2590 + A2617

A530 + A590 + A617

(2)

where A530 , A590 , and A617 are the absorbance values measured at
three different wave lengths of 530, 590, and 617 nm, respectively.
2.4. Calculations
The volumetric permeate water ux (Jw ) in terms of liter per
square meter per hour (l/m2 h) was calculated by the Eq. (3):
Jw =

V
A t

(3)

where A is the effective area of the membrane module for permeation, and V is the volume of permeate collected over a time interval
t.
Membrane water permeability (WP) was calculated from the
volumetric water ux (Jw ) measured under certain TMP through
the Eq. (4):
WP =

Jw
TMP

(4)

123

The ux recovery ratio (FRR) of the fouled membrane after

cleaning with de-ionized water was calculated using the Eq. (5):
FRR(%) =

Jwc
100
Jwi

(5)

where Jwi and Jwc are the pure water uxes of the fresh and watercleaned membrane measured under the same TMP, respectively.
The observed COD reduction and color removal were calculated
using the Eq. (6):
R(%) =

Cf Cp
100
Cf

(6)

where Cf and Cp represent the contents of COD or color in feed and

permeate streams, respectively.
The hydraulic resistance of the fresh membrane (Rm ) was calculated from the pure water ux (Jwi ) measured under certain TMP
through the Eq. (7) [40]:
Jwi =

TMP
w Rm

(7)

where w is the water viscosity.

The hydraulic resistance of the fouling layer (Rf ) of the fouled
membrane was calculated from the steady-state water ux (Jws )
measured under certain TMP by adopting the resistance-in-series
model through the Eq. (8) [41]:
Jws =

TMP 
w (Rm + Rf )

(8)

where  is the trans-membrane osmotic pressure difference.

3. Results and discussion
3.1. Characteristics of the textile efuents
The raw and biologically treated textile efuents were characterized in terms of pH, conductivity, COD as well as color. The results
listed in Table 2 clearly show that the COD and color presented
in the raw efuent are partially removed by the activated sludge
process, and both COD and color of the biologically treated textile
efuent are still too high to be discharge or reused.
3.2. Submerged ltration under recirculation model
3.2.1. Inuence of TMP on permeate ux
Fig. 1 presents the time-dependant uxes in ltration of the raw
and biologically treated textile efuents, respectively, under the
TMPs of 0.4, 0.5, 0.6, 0.7, and 0.8 bar. It can be seen from the gure
that, the water uxes decrease continuously with ltration time
and reach to steady-state values after about 20.0 h. The decline of
ux is mainly due to the deposition and absorption of foulants on
the membrane surface. Under the same TMP, the ux to the raw
efuent decreases more quickly than the biologically treated efuent, showing that the fouling of raw efuent to membrane is more
severe. On the other hand, to the same efuent, higher TMP usually leads to a quick decrease of ux and more severe membrane
fouling.
Table 2
Characteristics of the raw and biologically treated textile efuents used in this study.
Parameter

Unit

Raw textile efuent

Conductivity
COD
Color
pH

s/cm
mg/l
Abs/cm

2250
1780
1.200
6.8

40
30
0.001
0.2

Biologically treated
textile efuent
2690
780
0.370
7.0

40
30
0.0008
0.2

124

Q. Chen et al. / Journal of Hazardous Materials 284 (2015) 121129

Pure water

Pure water

Txtile effluent

Raw textile effluent

Biologically treated effluent

Cleaning

Water permeability ( l/m h bar)

Flux ( l/m2 h)

0.7bar
0.6bar

0.8bar
6

0.5bar
4

0.4bar
3

a
2

5
0

300

600

900

1200

1500

1800

0.4

0.5

0.6

0.7

0.8

TMP (bar)

Filtration time (min)

Fig. 2. Steady-state water permeability (WP) values in ltration of raw (

biologically treated (
TMPs.

Pure water

Pure water

Txtile effluent

Cleaning

0.8bar
Flux ( l/m2 h)

0.4 bar, the steady-state water permeability values are up to 7.02

and 7.87 l/m2 h bar to the raw and biologically treated efuents,
respectively.

0.7bar
0.6bar

0.5bar
4

0.4bar
3

b
2
0

300

600

900

1200

1500

1800

Filtration time (min)

Fig. 1. Time-dependent uxes in ltration of raw (a) and biologically treated (b)
textile efuents, respectively, under 25.0 C and different TMPs of 0.4 (), 0.5 ( ),
0.6 (

) and

) textile efuents, respectively, under 25.0 C and different

), 0.7 ( ), and 0.8 (

) bar.

The values of the pure water ux of the fresh membrane (Jwi ),

steady-state water ux of the fouled membrane (Jws ) and pure
water ux of the cleaned membrane (Jwc ) under each TMP are
summarized in Table 3. The steady-state water permeability (WP)
values calculated through Eq. (4) are presented in Fig. 2. It is
obviously that both the normalized water ux (Jws /Jwi ) and membrane water permeability to each efuent decrease with increasing
TMP. This is due to the enhancements of deposition and adsorption
of foulants on the membrane surface under higher TMP. Additionally, under each TMP, Jws /Jwi and water permeability to raw efuent
are lower than those to the biologically treated efuent. At TMP of

3.2.2. Inuence of TMP on COD reduction

Fig. 3 depicts the variations of observed COD reduction ratios
with ltration time under different TMPs. Under each TMP, the
COD reduction decreases sharply-to-gradually and approaches to
a steady-state value after certain ltration period. Similar results
were also obtained by Chakraborty et al. [42]. The change trend can
be explained in terms of concentration polarization, which leads to
a build up of solute concentration over the membrane surface and
thereby enhancing the solute permeation by convection through
the membrane. The concentration polarization at the initial stage
of ltration results in a quick increase of solute in permeate stream
and thereby a sharp decline of COD reduction. With the ltration prolongs, the permeate ux declines slowly and reaches to a
steady-state value, the concentration polarization becomes weak,
as a result, the COD reduction decreases gradually and reaches to a
steady-state value.
At a xed TMP, steady state is attained within a shorter ltration period to raw textile efuent due to its higher load of foulants
compared with the biologically treated efuent. Furthermore, to
the same efuent, the steady-state COD reduction is lower at higher
TMP. As the TMP increases from 0.4 to 0.8 bar, the averaged steadystate COD contents in ltrated raw and biologically treated efuents
ascended from about 155.0 to 245.0 mg/l and 130.0 to 205.0 mg/l,
respectively, during the experiments. The concentration polarization at higher TMP is more severe due to the enhanced convective
ux through the membrane, and thus the COD content in permeate
stream increases and the observed COD reduction decreases.

Table 3
Fluxes (Jwi , Jws , and Jwc ) to the raw and biologically treated textile efuents under 25.0 C and different TMPs.
TMP (bar)

Jwi (l/m2 h)

0.4
0.5
0.6
0.7
0.8

3.68
4.60
5.51
6.45
7.35

Jws (l/m2 h)

Jwc (l/m2 h)

Jws /Jwi

Raw efuent

Biologically treated efuent

Raw efuent

Biologically treated efuent

Raw efuent

Biologically treated efuent

2.81
3.26
3.73
4.27
4.72

3.11
3.73
4.31
4.94
5.49

3.54
4.31
4.99
5.78
6.44

3.56
4.34
5.06
5.85
6.50

0.764
0.709
0.677
0.662
0.642

0.845
0.811
0.782
0.766
0.747

Q. Chen et al. / Journal of Hazardous Materials 284 (2015) 121129

0.030

95
90
85

0.4 bar
0.5 bar
0.6 bar
0.7 bar
0.8 bar

80
75

Color in permeaate stream (Abs/cm)

100

COD reduction (%)

125

0.020

0.015

0.4 bar
0.5 bar
0.6 bar
0.7 bar
0.8 bar

0.010

0.005

a
0.000

70
0

0.025

300

600

900

1200

300

600

900

1200

Filtration time (min)

Filtration time (min)

0.4 bar
0.5 bar
0.6 bar
0.7 bar
0.8 bar

COD reduction (%)

95
90
85
80
75

Color in permeaate stream (Abs/cm)

0.030
100

0.025

0.020

0.015

0.4 bar
0.5 bar
0.6 bar
0.7 bar
0.8 bar

0.010

0.005

b
0.000

70
0

300

600

900

1200

300

600

900

1200

Filtration time (min)

Filtration time (min)

Fig. 3. Time-dependent COD reductions in ltration of raw (a) and biologically
treated (b) textile efuents, respectively, under 25.0 C and different TMPs of 0.4

Fig. 4. Time-dependent color contents of the ltrated raw (a) and biologically
treated (b) textile efuents under 25.0 C and different TMPs of 0.4 (), 0.5 ( ),

(), 0.5 ( ), 0.6 (

0.6 (

), 0.7 ( ), and 0.8 (

) bar.

It should be pointed out here that, although the COD content

of the raw textile efuent is much higher than that of the biologically treated textile efuent, the steady-state COD content of the
ltrated raw textile efuent is only a littler higher than that of the
ltrated biologically treated efuent, especially under the low TMP
of 0.4 bar.
3.2.3. Inuence of TMP on color removal
The changes of color in permeate stream with ltration time
under different TMPs are presented in Fig. 4, which clearly shows
that the submerged nanoltration membrane used in this study
can effectively remove the color from both the raw and biologically treated textile efuents. The slight decrease of color removal
with ltration time under each TMP is mainly due to the increase
of dye concentration on the membrane surface as the result of
concentration polarization.
To the same efuent, higher TMP usually leads to increased color
content in permeate stream. The color removals to both efuents
decrease with increasing TMP. With increasing TMP from 0.4 to
0.8 bar, the observed color removals to the raw and biologically
treated efuents decrease from around 98.6 to 97.8% and 95.9 to
94.2%, respectively. As explained earlier, the concentration polarization becomes more severe at higher TMP due to the enhanced
convective ux through the membrane. This leads to an increase

), 0.7 ( ), and 0.8 (

) bar.

of color content on membrane surface and thereby an augment of

color content in permeate stream.

3.2.4. Inuences of TMP on cleaning efciency and hydraulic

resistance
The ux recovery ratios of the fouled membranes after water
cleaning were calculated through Eq. (5) by using the ux data of
Jwi and Jwc shown in Fig. 1. The results presented in Fig. 5 clearly
indicate that, the FRR is lower in ltration of raw efuent compared with the biologically treated efuent under each TMP, and
higher TMP usually leads to a lower water cleaning efciency. With
increasing TMP from 0.4 to 0.8 bar, the values of FRR in ltration of
raw and biologically treated efuents decrease from around 96.2
to 87.6% and 96.7 to 88.4%, respectively. At low TMP of 0.4 bar, the
values of FRR to both efuents are all higher than 96.0%, indicating high efciency of physical water cleaning for the submerged
ltration membrane.
Cake formation and pore blocking are assumed to be the predominant mechanism of the fouling of nanoltration membranes,
which generally add additional resistances to the permeation, both
external (cake layer) and internal (adsorption and pore blocking).

126

Q. Chen et al. / Journal of Hazardous Materials 284 (2015) 121129

98

1.0

Raw textile effluent

Biologically treated effluent

96

0.8

Jws/Jwi

94

FRR( %)

Raw textile effluent

Biologically treated effluent

0.9

92

0.7

90

0.6

88

0.5

0.4

86
0.4

0.5

0.6

0.7

0.8

TMP (bar)
Fig. 5. FRR values in ltration of raw ( ) and biologically treated (
ents, respectively, under 25.0 C and different TMPs.

) textile efu-

Rf = Rfr + Rr

(9)

where Rfr is the hydraulic resistance of the reversible fouling layer,

which is mainly due to the deposition of cake layer on the membrane surface and can be removed through water washing. Rr is
the hydraulic resistance of the irreversible fouling layer, which is
due to the adsorption and pore blocking of materials onto the membrane surface and pores and can not be removed by water cleaning
[44]. Thus, Rr can be calculated from the pure water ux of the
water-cleaned membrane (Jwc ) measured under certain TMP using
the following Eq. (10):
TMP
w (Rm + Rr )

Fig. 6. Inuence of VCF on normalized ux Jws /Jwi in ltration of raw (

) and bio-

logically treated ( ) textile efuents, respectively, at TMP of 0.4 bar and 25.0 C,
under which Jwi = 3.70 l/m2 h.

Therefore, the hydraulic resistance of the fouling layer (Rf ) is composed of two resistances in series [43]:

Jwc =

VCF

(10)

The hydraulic resistances were calculated through Eqs. (7)(10)

by using the ux data obtained in the ltration experiments as
shown in Fig. 1 and ignoring the trans-membrane osmotic pressure difference based on the fact that nearly all the electrolytes
presented in the efuent permeated through the membrane freely
during experiments. It is obviously from Table 4 that the hydraulic
resistance Rf formed in ltration of each efuent increases with
increasing TMP and is relatively higher to raw efuent compared
with the biologically treated efuent. Therefore, as illustrated in
Fig. 2, the steady-state water permeability decreases with increasing TMP and is relatively lower to the raw efuent under each TMP
compared with the biologically treated efuent.
It can also be seen from the table that, Rfr is much higher than
Rr , indicating that the cake layer provides more contribution to the
fouling resistance than the adsorbed particles which remains inside
the membrane after the washing stage with pure water, and relatively high cleaning efciency can be obtained for the nanoltration
membrane used under submerged ltration model. The ratio of Rr

to Rm in ltration of raw efuent under the TMP of 0.4, 0.5, 0.6,

0.7, and 0.8 bar is about 3.87, 6.60, 10.5, 11.6, and 14.1%, respectively. While in ltration of biologically treated efuent, the ratios
are lower and only 3.42, 5.92, 8.86, 10.3, and 13.0%, respectively.
Therefore, as illustrated in Fig. 5, the ux recovery ratio decreases
with increasing TMP and is also relatively lower to the raw efuent
under each TMP compared with the biologically treated efuent.
3.3. Submerged ltration under concentrating model
3.3.1. Inuence of VCF on permeate ux
The inuence of VCF on normalized ux Jws /Jwi under TMP of
0.4 bar is shown in Fig. 6. As expected, the permeate ux decreases
with the increase of VCF as the result of the enhanced membrane
fouling, which results in an increased hydraulic resistance to water
permeation. As the VCF increases from 1.0 to 5.0, the value of Jws /Jwi
to raw efuent decreases dramatically from 0.76 to 0.53, while to
biologically treated efuent, it declines gradually from 0.85 to 0.70,
indicating more severe membrane fouling in ltration of raw efuent, especially at higher VCF. However, it must be pointed out here
that, even at VCF of 5.0, the values of membrane water permeability
in ltration of raw and biologically treated efuents are still up to
4.90 and 6.48 l/m2 h bar, respectively.
3.3.2. Inuence of VCF on COD reduction
Fig. 7 presents the observed COD reductions under different
VCFs. It is apparent from the gure that an increase in VCF results in
an augment in COD reduction. Under each VCF, the COD reduction
to the raw efuent is higher than that to the biologically treated
efuent. With increasing VCF from 1.0 to 5.0, the observed COD
reductions to raw and biologically treated efuents ascend from
91.3 to 95.7% and 88.2 to 94.2%, respectively. The increase of COD
reduction is attributed to the formation of denser fouling layer

Table 4
Hydraulic resistances (Rm , Rf , Rfr , and Rr ) in ltration of raw and biologically treated textile efuents, respectively, under 25.0 C and different TMPs.
TMP (bar)

Rm 1013 (m1 )

0.4
0.5
0.6
0.7
0.8

4.39
4.39
4.40
4.38
4.39

Rf 1013 (m1 )

Rfr 1013 (m1 )

Rr 1013 (m1 )

Raw efuent

Biologically treated efuent

Raw efuent

Biologically treated efuent

Raw efuent

Biologically treated efuent

1.36
1.80
2.10
2.24
2.45

0.80
1.02
1.23
1.34
1.49

1.19
1.51
1.64
1.73
1.83

0.65
0.76
0.84
0.89
0.92

0.17
0.29
0.46
0.51
0.62

0.15
0.26
0.39
0.45
0.57

Q. Chen et al. / Journal of Hazardous Materials 284 (2015) 121129

100

127

100

95

Raw textile effluent

Biologically treated effluent

96

FRR (%)

COD reduction (%)

98

90

94

92

Raw textile effluent

Biologically treated effluent

90

85
1

88

VCF
Fig. 7. Inuence of VCF on COD reduction of the ltrated raw (

) and biologically

) textile efuents under TMP of 0.4 and 25.0 C.

treated (

99

Color removal (%)

Fig. 9. Inuence of VCF on FRR in ltration of raw ( ) and biologically treated (

textile efuents, respectively, under TMP of 0.4 and 25.0 C.

biologically treated efuent. With the VCF increases from 1.0 to

5.0, the observed color removal to raw efuent ascends slightly
from about 98.6 to 98.9%, while to the biologically treated efuent
it increases appreciably from about 95.9 to 97.3%. During the
experiments, even at VCF of 5.0, the averaged color contents of
the ltrated raw and biologically treated efuents were only 0.060
and 0.045 Abs/cm, respectively.

100

98

97

96

Raw textile effluent

Biologically treated effluent

95

94
1

VCF
Fig. 8. Inuence of VCF on color removal of the ltrated raw (
treated (

VCF

) and biologically

) textile efuents under TMP of 0.4 and 25.0 C.

on the membrane surface as the concentrating process continues

[19]. During the experiments, even at VCF of 5.0, the averaged COD
contents of the ltrated raw and biologically treated efuents were
only 325 10.0 and 180 5.0 mg/l, respectively.
3.3.3. Inuence of VCF on color removal
The inuence of VCF on color removal is depicted in Fig. 8.
It is apparent that the color removal increases with VCF. The
possible reason is the formation of denser fouling layer, which
results in a reduced mean pore size of the membrane used and
thereby an enhancement of color removal [45]. Under each VCF,
the color removal to the raw efuent is higher than that to the

3.3.4. Inuences of VCF on cleaning efciency and hydraulic

resistance
The ux recovery ratios of the membrane under different VCFs
are shown in Fig. 9. It can be seen from the gure that the FRR in
ltration of each efuent decreases with increasing VCF. Under each
VCF, the FRR of membrane ltrated with raw efuent is relatively
lower than that with the biologically treated efuent. Higher VCF
results in a lower water cleaning efciency. With increasing VCF
from 1.0 to 5.0, the values of FRR of the membrane ltrated with
raw and biologically treated efuents decline from 95.9 to 91.1%
and 96.2 to 92.9%, respectively.
From Table 5, one can nd that the hydraulic resistance Rf
increases with VCF due to the enhancement of membrane fouling.
The ratio of Rf to total permeation resistance (Rm + Rf ) in ltration
of raw efuent under the VCF of 1.0, 2.0, 3.0, 4.0, and 5.0 is 24.0,
31.8, 38.0, 42.9, and 47.0%, respectively. While in ltration of biologically treated efuent, the values are lower and only 15.3, 20.9,
24.8, 28.0, and 29.6%, respectively. Therefore, as illustrated in Fig. 6,
the normalized water ux Jws /Jwi decreases with increasing VCF and
is lower to the raw efuent at each VCF compared with the biologically treated efuent. Similarly, the ratio of Rr to Rm in ltration
of raw efuent under the VCF of 1.0, 2.0, 3.0, 4.0, and 5.0 is 4.11,
5.94, 7.09, 8.45, and 9.38%, respectively. While in ltration of biologically treated efuent, the values are also lower and only 3.89,
5.02, 5.95, 6.85, and 7.55%, respectively. Therefore, as illustrated in
Fig. 9, the ux recovery ratio decreases with increasing VCF and

Table 5
Hydraulic resistances (Rm , Rf , Rfr , and Rr ) in ltration of raw and biologically treated textile efuents, respectively, under TMP of 0.4 bar, 25.0 C and different VCFs.
VCF

Rm 1013 (m1 )

1.0
2.0
3.0
4.0
5.0

4.37
4.38
4.37
4.38
4.37

Rf 1013 (m1 )

Rfr 1013 (m1 )

Rr 1013 (m1 )

Raw efuent

Biologically treated efuent

Raw efuent

Biologically treated efuent

Raw efuent

Biologically treated efuent

1.38
2.04
2.68
3.29
3.87

0.79
1.16
1.44
1.70
1.84

1.20
1.78
2.37
2.92
3.46

0.62
0.94
1.18
1.40
1.51

0.18
0.26
0.31
0.37
0.41

0.17
0.22
0.26
0.30
0.33

128

Q. Chen et al. / Journal of Hazardous Materials 284 (2015) 121129

Fig. 10. Photograph of the samples of (a) raw textile efuent, (b) ltrated raw textile efuent obtained at TMP of 0.4 bar and VCF of 5.0, (c) biologically treated textile efuent,
and (d) ltrated biologically treated textile efuent obtained at TMP of 0.4 bar and VCF of 5.0.

Table 6
Characteristics of the representative ltrated raw and biologically treated textile efuents under TMP of 0.4 bar and VCF of 1.0 and 5.0, respectively.
Parameter

Conductivity
COD
Color
pH

Unit

s/cm
mg/l
Abs/cm

VCF = 1.0

VCF = 5.0

Filtrated raw efuent

Filtrated biologically treated efuent

Filtrated raw efuent

Filtrated biologically treated efuent

2180
153.0
0.017
6.83

2610
132.0
0.015
7.02

2210
325
0.060
6.86

2630
182
0.045
6.95

is also lower to the raw efuent at each VCF compared with the
biologically treated efuent.

3.4. Characteristics of the ltrated raw and biologically treated

textile efuents
The photograph of the samples of raw and biologically treated
textile efuents before and after submerged nanoltration is shown
in Fig. 10. Table 6 lists the characteristics of the representative ltrated raw and biologically treated textile efuents obtained under
TMP of 0.4 bar and VCF of 1.0 and 5.0, respectively. It can be seen
from the table that the COD and color contents of the ltrated
efuents are low and the treatment efciencies are attractive for
practical application. Although the COD and color contents of the
raw efuent are much higher than the biologically treated efuent,
the quality of the ltrated raw efuent is only slightly lower than
the ltrated biologically treated efuent. Additionally, the conductivity of the ltrated efuent is nearly the same as that of the feed,
indicating that the electrolytes present in the efuent are nearly
not retained by the membrane, especially under higher VCF.

4. Conclusions
Raw and biologically treated textile efuents have been treated
through submerged ltration using lab-fabricated hollow ber
nanoltration membrane with a molecular weight cut-off of about
650 g/mol. The treatment efciencies to both the raw and biologically treated textile efuents were high and attractive for practical
application. Raw textile efuents containing high contents of dyes
of biodegradable difculty are more suitable for direct submerged
nanoltration.
The water permeate ux, COD reduction and color removal were
largely affected by TMP and VCF. The increase of TMP resulted in
a decline in water permeability, COD reduction and color removal,

while the increase of VCF resulted in both increased COD reduction

and color removal.
Higher TMP and/or VCF resulted in more sever membrane fouling, decreased cleaning efciency, and lowered ux recovery ratio.
Under the same condition, the membrane fouling in ltration of
raw efuent was more sever than biologically treated efuent. Both
water permeability and ux recovery ratio to the raw efuent were
lower than those to the biologically treated efuent.
Acknowledgments
The authors gratefully acknowledge the nancial support by the
National Nature Science Foundation of China (Grant nos. 21276242
and 21476213), the National High-tech R&D Program of China
(Grant no. 2012AA03A601) and the 521 personnel training plan
of Zhejiang Sci-Tech University.
References
[1] G. Tchobanoglous, F.L. Burton, H.D. Stensel, Wastewater Engineering:
Treatment and Reuse, fourth ed., McGraw Hill New York, New York, USA,
2003.
[2] W.-J. Lau, A.F. Ismail, Polymeric nanoltration membranes for textile dye
wastewater treatment: preparation, performance evaluation, transport
modeling, and fouling control a review, Desalination 245 (2009) 321348.
[3] L.A.V. de.Luna, T.H.G. da.Silva, R.F. Pupo Nogueirab, F. Kummrow, G.A.
Umbuzeiro, Aquatic toxicity of dyes before and after photo-Fenton treatment,
J. Hazard. Mater. 276 (2014) 332338.
[4] C. Kannan, K. Muthuraja, M.R. Devi, Hazardous dyes removal from aqueous
solution over mesoporous aluminophosphate with textural porosity by
adsorption, J. Hazard. Mater. 244245 (2013) 1020.
[5] M. Akgl, Enhancement of the anionic dye adsorption capacity of
clinoptilolite by Fe3+ -grafting, J. Hazard. Mater. 267 (2014) 18.
[6] H. Keharia, D. Madamwar, Textile and dye efuent, in: A. Pandey (Ed.),
Concise Encyclopedia of Bioresource Technology, The Haworth Press, New
York, 2004, pp. 167175.
[7] B. Guieysse, Z.N. Norvill, Sequential chemicalbiological processes for the
treatment of industrial wastewaters: review of recent progresses and critical
assessment, J. Hazard. Mater. 267 (2014) 142152.

Q. Chen et al. / Journal of Hazardous Materials 284 (2015) 121129

[8] B. Marrot, N. Roche, Wastewater treatment and reuse in textile industries, a
review, Res. Adv. Water Res. 3 (2002) 4153.
[9] C. Ahmed Basha, J. Sendhil, K.V. Selvakumar, P.K.A. Muniswaran, C.W. Lee,
Electrochemical degradation of textile dyeing industry efuent in batch and
ow reactor systems, Desalination 285 (2012) 188197.
[10] Y. Yao, C. Zhao, M. Zhao, X. Wang, Electrocatalytic degradation of methylene
blue on PbO2 ZrO2 nanocomposite electrodes prepared by pulse
electrodeposition, J. Hazard. Mater. 263 (2013) 726734.
[11] B. Haspulat, A. Glce, H. Glce, Efcient photocatalytic decolorization of some
textile dyes using Fe ions doped polyaniline lm on ITO coated glass
substrate, J. Hazard. Mater. 260 (2013) 518526.
[12] X.-B. Zhang, W.-Y. Dong, F.-Y. Sun, W. Yang, J. Dong, Degradation efciency
and mechanism of azo dye RR2 by a novel ozone aerated internal
micro-electrolysis lter, J. Hazard. Mater. 276 (2014) 7787.
[13] A. Vijayabalan, K. Selvam, B. Krishnakumar, M. Swaminathan, Photocatalytic
degradation of Reactive Orange 4 by surface uorinated TiO2 Wackherr under
UV-A light, Sep. Purif. Technol. 108 (2013) 5156.
[14] K. Turhan, I. Durukan, S. Arda Ozturkcan, Z. Turgut, Decolorization of textile
basic dye in aqueous solution by ozone, Dyes Pigm. 92 (2012) 897901.
[15] E. Alventosa-deLara, S. Barredo-Damas, M.I. Alcaina-Miranda, M.I. Iborra-Clar,
Ultraltration technology with a ceramic membrane for reactive dye
removal: optimization of membrane performance, J. Hazard. Mater. 209210
(2012) 492500.
[16] E. Kurt, D.Y. Koseoglu-Imer, N. Dizge, S. Chellam, I. Koyuncu, Pilot-scale
evaluation of nanoltration and reverse osmosis for process reuse of
segregated textile dyewash wastewater, Desalination 302 (2012)
2432.
[17] A.I. Schfer, A.G. Fane, T.D. Waite, Nanoltration-Principles and Applications,
Elsevier, New York, 2005.
[18] N.B. Amar, N. Kechaou, J. Palmeri, A. Deratani, A. Sghaier, Comparison of
tertiary treatment by nanoltration and reverse osmosis for water reuse in
denim textile industry, J. Hazard. Mater. 170 (2009) 111117.
[19] A. Bes-Pi, B. Cuartas-Uribe, J.-A. Mendoza-Roca, M.I. Alcaina-Miranda, Study
of the behaviour of different NF membranes for the reclamation of a secondary
textile efuent in rinsing processes, J. Hazard. Mater. 178 (2010) 341348.
[20] E. Ellouze, N. Tahri, R.B. Amar, Enhancement of textile wastewater treatment
process using nanoltration, Desalination 286 (2012) 1623.
[21] S.P. Sun, T. Alan Hattona, S.Y. Chan, T.-S. Chung, Novel thin-lm composite
nanoltration hollow ber membranes with double repulsion for effective
removal of emerging organic matters from water, J. Membr. Sci. 401402
(2012) 152162.
[22] Y.K. Ong, F.Y. Li, S.-P. Sun, B.-W. Zhao, C.-Z. Liang, T.-S. Chung, Nanoltration
hollow ber membranes for textile wastewater treatment: lab-scale and
pilot-scale studies, Chem. Eng. Sci. 114 (2014) 5157.
[23] C.-Z. Liang, S.-P. Sun, F.-Y. Li, Y.-K. Ong, T.-S. Chung, Treatment of highly
concentrated wastewater containing multiple synthetic dyes by a combined
process of coagulation/occulation and nanoltration, J. Membr. Sci. 469
(2014) 306315.
[24] B. Van der Bruggen, M. Mnttri, M. Nystrm, Drawbacks of applying
nanoltration and how to avoid them: a review, Sep. Purif. Technol. 63 (2008)
251263.
[25] T.M. Patel, K. Nath, Alleviation of ux decline in cross ow nanoltration of
two-component dye and salt mixture by low frequency ultrasonic irradiation,
Desalination 317 (2013) 132141.
[26] M. Gander, B. Jefferson, S. Judd, Aerobic MBRs for domestic wastewater
treatment: a review with cost considerations, Sep. Purif. Technol. 18 (2000)
119130.

129

[27] A.G. Fane, A. Yeo, A. Law, K. Parameshwaran, F. Wicaksana, V. Chen, Low

pressure membrane processes doing more with less energy, Desalination 185
(2005) 159165.
[28] A. Santos, W. Ma, S.J. Judd, Membrane bioreactors: two decades of research
and implementation, Desalination 273 (2011) 148154.
[29] N.S.A. Mutamim, Z.Z. Noor, M.A. Abu Hassan, G. Olsson, Application of
membrane bioreactor technology in treating high strength industrial
wastewater: a performance review, Desalination 305 (2012) 111.
[30] J.-B. Castaing, A. Mass, M. Ponti, V. Schet, J. Haure, P. Jaouen, Investigating
submerged ultraltration (UF) and microltration (MF) membranes for
seawater pre-treatment dedicated to total removal of undesirable
micro-algae, Desalination 253 (2010) 7177.
[31] G. Di Proo, X. Ji, E. Curcio, E. Drioli, Submerged hollowber ultraltration as
seawater pretreatment in the logic of integrated membrane desalination
systems, Desalination 269 (2011) 128135.
[32] J.-H. Choi, K. Fukushi, K. Yamamoto, A submerged nanoltration membrane
bioreactor for domestic wastewater treatment: the performance of cellulose
acetate nanoltration membranes for long-term operation, Sep. Purif.
Technol. 52 (2007) 470477.
[33] M. Rhricht, J. Krisam, U. Weise, U.R. Kraus, R.-A. Dring, Elimination of
pharmaceuticals from wastewater by submerged nanoltration plate
modules, Desalination 250 (2010) 10251026.
[34] S. Yu, Z. Chenb, Q. Cheng, Z. L, M. Liu, C. Gao, Application of thin-lm
composite hollow ber membrane to submerged nanoltration of anionic dye
aqueous solutions, Sep. Purif. Technol. 88 (2012) 121129.
[35] Y. Zheng, G. Yao, Q. Cheng, S. Yu, M. Liu, C. Gao, Positively charged thin-lm
composite hollowber nanoltration membrane for the removal of cationic
dyes through submerged ltration, Desalination 328 (2013) 4250.
[36] Y. Zheng, S. Yu, S. Shuai, Q. Zhou, Q. Cheng, M. Liu, C. Gao, Color removal and
COD reduction of biologically treated textile efuent through
submergedltration using hollow ber nanoltration membrane,
Desalination 314 (2013) 8995.
[37] N. Hilal, M. Al-Abria, H. Al-Hinaib, M. Abu-Arabic, Characterization and
retention of NF membranes using PEG, HS, and polyelectrolytes, Desalination
221 (2008) 284293.
[38] A.E. Childress, M. Elimelech, Effect of solution chemistry on the surface charge
of polymeric reverse osmosis and nanoltration membranes, J. Membr. Sci.
119 (1996) 253268.
[39] Standard Methods for the Examination of Water and Wastewater: Including
Bottom Sediments and Sludges, rst ed., American Public Health Association,
New York, 1995.
[40] R.W. Baker, Membrane Technology and Applications, second ed., John Wiley &
Sons Ltd., Chichester, 2004.
[41] S. Mattaraj, W. Phimpha, P. Hongthong, R. Jiraratananon, Effect of operating
conditions and solution chemistry on model parameters in crossow reverse
osmosis of natural organic matter, Desalination 253 (2010) 3845.
[42] S. Chakraborty, M.K. Purkait, S. DasGupta, S. De, J.K. Basu, Nanoltration of
textile plant efuent for color removal and reduction in COD, Sep. Purif.
Technol. 31 (2003) 141151.
[43] J.L. Acero, F.J. Benitez, Ana I. Leal, F.J. Real, F. Teva, Membrane ltration
technologies applied to municipal secondary efuents for potential reuse, J.
Hazard. Mater. 177 (2010) 390398.
[44] X. Wang, L. Wang, Y. Liu, W. Duan, Ozonation pre-treatment for ultraltration
of the secondary efuent, J. Membr. Sci. 287 (2007) 187191.
[45] E. Sahinkaya, N. Uzal, U. Yetis, F. Dilek, Biological treatment and nanoltration
of denim textile wastewater for reuse, J. Hazard. Mater. 153 (2008)
11421148.