chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 284293

Contents lists available at ScienceDirect

Chemical Engineering Research and Design

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

The potential of direct contact membrane

distillation for industrial textile wastewater
treatment using PVDF-Cloisite 15A nanocomposite
membrane

N.M. Mokhtar a,b , W.J. Lau b, , A.F. Ismail b , S. Kartohardjono c , S.O. Lai d ,
H.C. Teoh d
a Faculty of Engineering Technology, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Kuantan, Pahang,
Malaysia
b Advanced Membrane Technology Research Centre (AMTEC), Universiti Teknologi Malaysia, 81310 Skudai, Johor,

Malaysia
c Process Intensication Laboratory, Department of Chemical Engineering, Universitas Indonesia, Depok 16424,

Indonesia
d Department of Chemical Engineering, Faculty of Engineering and Science, Universiti Tunku Abdul Rahman,

Jalan Genting Kelang, 53300 Kuala Lumpur, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: This work demonstrates the feasibility of employing direct contact membrane distillation
Received 20 May 2015 (DCMD) for treating industrial textile wastewater for clean water production. Experi-
Received in revised form 27 April mental results showed that the in-house fabricated polyvinylidene uoride-Cloisite 15A
2016 polymerinorganic nanocomposite membrane is robust and able to treat the industrial
Accepted 20 May 2016 efuent by reducing at least 89% of the initial values of the water quality parameters mea-
Available online 27 May 2016 sured. However, the membrane permeate ux was reported to decline almost 50% in the
rst few hours of the 40-h treatment process before reaching water ux of 1322 kg/m2 h. It
Keywords: is believed that the initial ux decline is mainly caused by the foulants accumulated on the
Membrane distillation membrane outer surface that increases mass transfer resistance of water molecules and
Polymerinorganic nanocomposite reduces water productivity. With respect to separation characteristics, the DCMD process
membrane has shown better performance for COD and color removal in comparison to the other com-
Textile wastewater monly used pressure-driven membrane processes. Further improvement on the membrane
Membrane fouling surface properties is necessary to reduce fouling propensity and pore wetting caused by the
Pore wetting surfactants and other foulants in the textile wastewater. This is of particular importance for
long-term operation of DCMD process.
2016 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

1. Introduction industry has accounted about 1.7% of the world exportation in

2007 which was equivalent to US$ 238.1 billions (Firmino et al.,
Textile industry has been playing an important role in the 2010). Rapid industrialization, however, requires adequate
global economic development, international business and care at the source itself to attain a balanced sustainability
trade. According to the World Trade Report 2008, the textile between economic growth and waste generation. On the input

Corresponding author. Tel.: +60 75535926.
E-mail addresses: lwoeijye@utm.my, lau woeijye@yahoo.com, lau.woeijye09@gmail.com (W.J. Lau).
http://dx.doi.org/10.1016/j.cherd.2016.05.018
0263-8762/ 2016 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 284293
Table 1 Typical characteristics of textile wastewater and standard for efuent discharge.
Parameter Unit
pH
Temperature C 3080
Color ADMIc
Biological oxygen demand (BOD5 at 20 C) mg/L
Chemical oxygen demand (COD) mg/L
Total suspended solids (TSS) mg/L
Total dissolved solid (TDS) mg/L
Organic nitrogen mg/L
Total phosphorus mg/L
Total chromium mg/L
a
The data was obtained from Marcucci et al. (2003).
b
Extracted from Environmental Quality (Industrial Efuents) Regulations 2009 (PU (A) 434). Standard A is applied to the efuent discharged to
the areas upstream of surface or above subsurface water supply intakes. Meanwhile, Standard B is applied to any other inland waters.
c
ADMI stands for American Dye Manufactures Institute.
side, the textile processing industry is regarded as a water
intensive sector as it uses a huge amount of fresh water in
the production process (Ntuli et al., 2009; Vajnhandl and Valh,
2014). Meanwhile, on the output side, the textile manufac-
turing process usually generates large quantity of wastewater
with a wide range of pollutant concentrations (Mokhtar et al.,
2013). Direct discharge of textile efuent into the water reser-
voirs without treatment leads to severe problems for the
ecosystem as the wastewater might comprise of process
water, cleaning water, non-contact cooling water, storm water,
etc. (Dasgupta et al., 2015). Moreover, the wastewater char-
acteristics from the textile factories are highly variable from
day to day, depending on the type of dye, material of fab-
ric, process step and concentration of the chemical agents
used (Brik et al., 2006; Lau and Ismail, 2009; Ismail and Lau,
2010).
Table 1 compares the typical textile efuent characteristics
and the discharge industrial efuent regulations in Malaysia.
Depending on the discharge location, the efuents from indus-
trial sectors must comply with either Standard A or B under
environmental regulation (Environmental Quality (Industrial
Efuent) Regulations, 2009). As can be seen, the quality of tex-
tile efuent is far below the standard of discharge and proper
treatment is therefore needed in order to reduce its environ-
mental impacts. Treating textile efuent is urgently needed
as textile industry is one of the biggest sources of industrial
water pollution in Malaysia (Mohammadi et al., 2010). Cur-
rently, the treatment systems based on biotechnology and
physico-chemical methods are widely used for dye removal
(Chen et al., 2007; Chhabra et al., 2015). Compared to the
chemical-based treatment method, the biological treatment
process gains remarkable development in textile industry due
to its higher efciencies for chemical oxygen demand (COD)
reduction and lower amount of sludge generated (Tfekci
et al., 2007). However, these processes are associated with
major drawback, i.e. inadequate to remove color and dissolved
ions effectively from textile efuents.
Although alternative treatment methods by membrane
technology such as nanoltration (NF) and reverse osmo-
sis (RO) have shown promising results in producing treated
efuent of good quality, the signicant amount of energy
consumption resulted from relatively high operating pressure
during ltration has set a limit for industrial implementation
(Gullinkala et al., 2010; Liu et al., 2011; Ong et al., 2012). To
date, membrane distillation (MD) has been seen as a potential
candidate in treating textile efuents as this thermally driven
285
Typical rangea Standard Ab Standard Bb
210 69 5.59
40 40
100 200
200300 20 40
505000 80 250
50500 50 100
15002200
1839
0.315
0.20.5 0.05 0.05
membrane process has unique advantages over pressure-
driven membrane processes (Mozia et al., 2010; Lin et al.,
2014a,b; Mokhtar et al., 2015a; Qu et al., 2014). In our previous
work (Mokhtar et al., 2015b), we have evaluated the potential
of in-house made polymerinorganic nanocomposite mem-
brane that composed of polyvinylidene uoride (PVDF) and
Cloisite 15A clay in treating synthetic feed solutions contain-
ing dyes and salts via direct contact membrane distillation
(DCMD) system. It can be deduced from the study that the
polymerinorganic nanocomposite membrane demonstrated
excellent results in eliminating almost all non-volatile compo-
nents present in the feed solutions with consistent permeate
ux irrespective of dye properties, feed concentrations and
feed ow rates. In view of this, the main objective of this work
is to further evaluate the feasibility of the polymerinorganic
nanocomposite hollow ber membrane for industrial tex-
tile wastewater treatment by analyzing the physicochemical
interactions between foulants and membrane material that
might lead to membrane fouling and pore wetting during
the DCMD process. To the best of our knowledge, this is the
rst reported work on the MD utilization for industrial textile
wastewater treatment using a polymerinorganic nanocom-
posite membrane.
2. Experimental
2.1. Textile efuent sample
The wastewater sample used in this work was collected from
a textile factory in Kulai district, Johor, Malaysia. Different
types of sewing threads and industrial yarns are processed
in the factory with a production rate of 1.6 ton/h. The factory
discharges around 120 m3 /day textile wastewater with tem-
perature vary in the range of 8090 C. Currently, the factory
employs biological process to treat the efuent produced. In
order to treat the wastewater using in-house made membrane
in laboratory, the wastewater sample collected from the equal-
ization tank was rst preserved at 4 C to prevent the sample
from undergoing biodegradation due to microbial action. No
pretreatment process was carried out and the wastewater
sample was used directly for DCMD process. Table 2 summa-
rizes the basic quality parameters of the wastewater sample
collected.
Color and COD of the sample solutions were detected by
a UV-vis spectrophotometer (DR5000, Hach). Biological oxy-
gen demand (BOD5 ) values of the samples were analyzed
286 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 284293
Table 2 Physicochemical analysis of the industrial
textile wastewater.
Parameter Unit Values
pH 10.511.8
BOD5 mg/L 1115
COD mg/L 405477
TDS mg/L 774996
Color Pt/Co 346526
Turbidity NTU 2225
Conductivity S/cm 12941673
NTU, nephalometer turbidity unit.
using a dissolved oxygen meter (ProODO, YSI) according to
standard methods prescribed in APHA (1992) methods. The
ionic conductivity and total dissolved solid (TDS) of the sam-
ple solutions were determined using a bench top conductivity
meter (4520, Jenway). The turbidity was measured by using
portable turbidimeter (2100Q, Hach) and pH of wastewater was
measured using a pH meter (EUTECH instrument).
2.2. Membrane fabrication
The nanocomposite hollow ber membrane used in this work
was prepared using dope solution consisted of 12 wt% PVDF,
80 wt% N-methyl-2-pyrrolidone (NMP), 8 wt% ethylene glycol
(EG) and 3 wt% Cloisite 15A (determined based on the total
weight of PVDF). This dope formulation was selected because
it produced the membrane with best results from our previous
work (Mokhtar et al., 2014a). To prepare the dope solution, pre-
weighted PVDF in pellet form and clay powder were dissolved
in the mixture of NMP and EG and stirred continuously at 60 C
until the solution became homogeneous. After that, the hol-
low ber membrane was fabricated using dry-jet wet spinning
technique in which the detailed fabrication procedure can be
found in our previously published work (Mokhtar et al., 2014a).
After air-drying process, the membrane was subject to several
characterizations to determine its structural properties.
2.3. Membrane characterizations
2.3.1. Porosity and membrane pore size measurements
The membrane porosity (%), , which is dened as the volume
of the pores per the total volume of the porous membrane was
calculated using the following equation (Chabot et al., 1997).
(wwet wdry )/w
(%) = 100
((wwet wdry )/w ) + (wdry /p )
where wwet is the weight of wet membrane (g), wdry is the
weight of dry membrane (g), p is the density of the polymer
(g/cm3 ) and w is the density of water (g/cm3 ).
The bubble point and pore size distribution of the pre-
pared membrane were investigated using a capillary ow
porometer (Porous Materials Inc.). The ber was fully wet-
ted with isopropanol (surface tension = 21.7 dynes/cm), and
then the measurements were carried out following the proce-
dure described elsewhere (Mokhtar et al., 2014a). The pore size
diameter, dp could be calculated by the following equation.
4 cos 
dp =
P set at 0.023 m/s and 0.002 m/s, respectively throughout the
Table 3 Details of membrane and membrane module.
Membrane PVDF-Cloisite 15A
Pore size (m) 0.088
Porosity (%) 83.70 0.67
Number of bers 10
Module inner dia. (mm) 10
Fiber outer dia. (m) 763 19
Fiber inner dia. (m) 511 15
Module length (mm) 220
Effective ber length (mm) 190
Effective membrane area (m2 ) 0.005
where  is the surface tension of liquid, is the contact angle
of liquid and P is the external pressure.
2.3.2. FESEM and EDX analysis
The membrane sample was immersed in liquid nitrogen and
fractured, followed by sputter-coated with platinum using a
sputtering device. The membrane cross-section, outer and
inner surface were examined using eld emission scanning
electron microscope (FESEM) (SU-8020, Hitachi). Meanwhile,
X-ray energy-dispersive (EDX) spectrometer (XFlash 430H,
Bruker) was used for elemental analysis in order to identify
elements present on the fouled membrane.
2.3.3. Contact angle measurement
Water contact angle of the membrane was measured using a
contact angle goniometer (OCA15plus, DataPhysics), equipped
with image-processing software via sessile drop technique.
The size of each droplet is 0.5 L and 15 spots for each sample
were randomly analyzed to yield an average value of contact
angle.
2.3.4. FTIR
Fourier-transform infrared (FTIR) spectra were obtained for
the PVDF-Cloisite 15A membrane before and after testing
with textile wastewater using Nicolet 5700 FTIR spectrome-
ter (Thermo Scientic Inc.). The measurements were recorded
using attenuated total reectance (ATR) technique with multi-
coated, conical shape germanium tip crystal and the incidence
angle was set at 45 . Each spectrum has been achieved in the
range of 4000650 cm1 at a spectral resolution of 1 cm1 and
with 32 scans.
2.4. DCMD experiments
A stainless steel module containing PVDF-Cloisite 15A
nanocomposite membrane was prepared prior to membrane
performance evaluation. Table 3 shows the details of the
(1)
membrane properties and its module. The DCMD system was
designed to have two circulating streams, i.e. hot stream also
known as feed stream (circulated through the membrane
shell-side) and cold stream (fed through the lumen-side of the
hollow ber membrane), as illustrated in Fig. 1. The purpose
of choosing outside-in ltration mode is mainly due to the
larger membrane surface area (for higher water permeabil-
ity) and less tendency of membrane surface fouling compared
to inside-out ltration mode. Besides, the membrane surface
analysis is much easier for process conducted in the mode of
outside-in ltration. The temperatures of both streams were
controlled at 90 1 C and 25 2 C using coiled heater (830,
PROTECH) and chiller (F26-ED, JULABO), respectively. Mean-
while, the cross-ow velocities of hot and cold stream were
(2)
 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 284293 287

Fig. 1 Schematic DCMD experimental setup.

experimental period. The change of feed properties (industrial

textile wastewater) in the feed tank during the experiment was
assumed to be negligible as large feed volume (5 L) was used.
Prior to the textile wastewater treatment process, the
permeate ux, Jv of membrane (kg/m2 h) was calculated as
follows.

W
Jv = (3)
At

where W (kg) is the weight of permeate collected over a pre-

determined time t (h) of process and A (m2 ) is the effective
membrane area. To determine the rejection, R (%) of the mem-
brane against each parameter, the following equation was
employed.

Fig. 2 Permeate ux of the PVDF-Cloisite 15A membrane

 
Cp during short-term study.
R (%) = 1 100 (4)
Cf
3. Results and discussion

where Cp and Cf stand for permeate and feed concentra-
tion (mg/L), respectively. The ux and rejection reported in
this work were the average of three measurements and the
standard deviations of the average measured values were indi-
cated using error bars.
In order to determine the membrane stability for the
industrial wastewater treatment process, prolonged DCMD
process (up to 40 h) was also performed to evaluate ux and
rejection (with respect to color and COD removal) of the
polymerinorganic nanocomposite membrane as a function
of time. To avoid signicant changes in feed solution proper-
ties (due to the high rejection capability of MD process), the
feed solution was replaced by a new batch of textile efu-
ent every 8 h. Detailed characterizations on membrane fouling
resulted by dyes and other components in the efuent were
also performed using different analytical instruments.
3.1. Comparison of MD performance with
pressure-driven membrane processes
The membrane was rst tested for 150-min using indus-
trial textile wastewater as feed to determine its permeate
ux as well as separation performance. As can be seen from
Fig. 2, the permeate ux varied in the range of 3238 kg/m2 .
This ux variation is reasonable for a MD process of real
industrial wastewater treatment as reported elsewhere (El-
Abbassi et al., 2013a; Mokhtar et al., 2015c; Li et al., 2016).
Industrial efuent normally contains many types of com-
ponents compared to the synthetic solution and this has
made the water ux to uctuate and rather difcult to pre-
dict. Overall, the membrane achieved a stable water ux at
about 37.8 kg/m2 after 70-min operation. In terms of ef-
ciency, the permeate quality of MD obtained in this work was
288 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 284293

Table 4 Comparison between the performance of MD process and other pressure-driven membrane processes in
treating textile wastewater.
Membrane process MF UF NF RO MD

Reference Fersi et al. (2009) Fersi et al. (2009) Fersi et al. (2009) Liu et al. (2011) In this work
Membrane T170-5 TA T170-50nTZ F 200 BW30a PVDF-Cloisite 15A
Pore size (m) 5 0.05 0.0005 N/A 0.088
Pressure (bar) 0.2 3 10 10 <0.1
Feed pre-treatment Pre-ltrated with 0.45 m lter
Color, R (%) 44.2 55.4 93.3 Colorlessb 95.3
Conductivity, R (%) 3.6 3.8 55.5 99.1b 93.7
Turbidity, R (%) 45.3 78.6 95.7 93.0
TDS, R (%) 4.6 7.9 65.2 92.0b 93.6
COD, R (%) 13.8 65.5 78.5 91.0b 89.6
BOD5 , R (%) 94.5b 90.8

a
The membrane purchased from Dow FilmTec (Minneapolis, MN).
b
The membrane rejection is obtained based on volume ratio of the initial efuent to the concentrated efuent, VCF = 1 as reported in the work.

Fig. 4 Removal efciency of the PVDF-Cloisite 15A

membrane and photos of efuent sample before and after
Fig. 3 Permeate ux of the PVDF-Cloisite 15A membrane
treatment process.
during prolonged study period.

that the initial permeate ux of MD which recorded at around

compared to other pressure-driven membrane processes and
the results are presented in Table 4. This comparison is of
importance to show that MD process could achieve compara-
ble or even better results than those of NF and RO process, but
at much lower operating pressure. Although RO could achieve
the highest separation efciency among all the membranes
studied owing to its smallest surface pore size, the main draw-
back of this process is its high energy consumption resulted
from high osmotic pressure during ltration (Gullinkala et al.,
2010). MD, however, does not have such problem as its driving
force is purely based on the thermal difference between two
streams. The hydrophobic nature of the membrane surface
is able to prevent water penetration during separation and
eliminate almost all of the dissolved solutes in the aqueous
solution (El-Abbassi et al., 2013b). However, it must be pointed
out that the incomplete removal of some pollutants in this
work could be related to partial wetting problem and/or foul-
ing issue as discussed in other works (Khaing et al., 2010; Lin
et al., 2014a,b).
3.2. Performance evaluation of DCMD under prolonged
study period
Fig. 3 shows the permeate ux of the PVDF-Cloisite 15A hol-
low ber nanocomposite membrane during MD process of
industrial textile wastewater for up to 40 h. It can be seen
38 kg/m2 h tended to decrease rapidly during the rst 10 h of
treatment process, showing approximately 50% ux reduc-
tion. From 11th h to 40th h, the ux of membrane continued
to decline, but at a slower rate. The membrane ux prole
obtained in this work is in agreement with the previous studies
in which MD process was employed for industrial wastewa-
ter treatment (Khaing et al., 2010; El-Abbassi et al., 2013b; Lin
et al., 2014a,b). The decreasing trend is likely caused by the
severe deposition of solutes on the outer surface of membrane
at the beginning of the separation process, leading to surface
pore blocking and increased water vapor transport resistance.
Although ux decline is mainly caused by membrane fouling,
for the MD process, partial pore wetting problem and/or ther-
mal polarization effect should also be taken into consideration
as both effects could inuence MD ux as well.
In this study, two important water quality parameters (i.e.
COD and color) were analyzed and the results are shown in
Fig. 4 together with photos of efuent sample before and
after treatment process. As can be seen, relatively high rejec-
tion rates for both COD (7280%) and color (7289%) could
be achieved throughout the 40-h operation. Overall, the MD
showed better results in removing efuents color in compar-
ison to COD value. Color is relatively easy to reject because of
the larger particle sizes (6001000 g/mol) of textile dyes, pig-
ments and other colored compounds present in the efuent
(Ismail and Lau, 2009). Unlike color, COD of the wastewater
 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 284293
is a complex mixture of textile dyes, suspended solids, min-
eral oils, electrolytes and surfactants (Ogunlaja and Aemere,
2009; Ong et al., 2014). It has been previously reported that the
organic contaminants in wastewater could fall in the range
of 0.001100 m (Shon et al., 2007). Because of this reason,
small solutes (e.g. ions) could pass through the larger pores
of membrane to the permeate side especially in the case of
partial wetting (Qu et al., 2009). El-Abbassi et al. (2013a,b) fur-
ther explained that some of components might be possibly
transported through the membrane pores as vapor. It must be
pointed out that although the removal efciencies of COD and
color were reported to be 7289% throughout the study period,
the treated permeate quality still adhere to the Standard B
discharge limit under the Malaysian Environmental Quality
Regulations.
3.3. Impact of membrane fouling on permeate ux
decline
Compared to the pressure-driven membrane processes, it has
been generally agreed that the fouling problem in MD process
is relatively less severe particularly in the cases of efuent that
contains high pollutant load (Gryta, 2008; Ramezanianpour
and Sivakumar, 2014; Warsinger et al., 2015). However, the
key factor affecting fouling in MD process is the hydropho-
bic nature of the membrane material itself. It is because the
membrane hydrophobic characteristics would have greater
interactions with aromatic rings of the dye molecules, caus-
ing fouling tendency to increase (Mokhtar et al., 2015d). In view
of this issue, visual inspection was performed on the fouled
membrane to examine any physical change/damage on the
membrane surface after treatment process.
Fig. 5 compares the FESEM image of the pristine membrane
with the fouled membrane after 40-h experimental study. A
thin and dense layer (around 1.0 m) was found on the outer
surface of the fouled membrane, which is likely caused by the
attachment of organophilic compounds present in the textile
efuent. This additional dense layer is possible to reduce the
open pore area available for mass and heat transfer across
the membrane, resulting in ux decline as experienced in this
work. Hausmann et al. (2013) also reported that dense cake
layer was gradually developed during MD process of dairy
streams and this has led to decreasing permeate ux follow-
ing the vapor pressure reduction caused by the higher solids
content of the fouling layer.
To further identify the elements caused by the pollutants
on the membrane surface, EDX analysis was performed. The
detections of magnesium, iron, calcium, sodium and sulfur on
the membrane surface as shown in Table 5 strongly indicated
the formation of additional layer on the membrane outer sur-
face. The presence of these elements is understandable as they
are commonly found in the industrial textile wastewater (Fersi
et al., 2009). Meanwhile, the detection of molybdenum could
be possibly attributed to the stainless steel pipe used for the
textile processing (Klueh, 1982; Raj et al., 2004). It is also impor-
tant to point out that the original content of the uorine in the
pristine PVDF-Cloisite 15A membrane was decreased by more
than 80% after MD process. This revealed that the membrane
outer surface was greatly covered by foulants. Meanwhile,
the higher concentration of the carbon and oxygen in the
fouled membrane is likely caused by organic and/or biological
foulants (Melin-Martel et al., 2012).
289
Table 5 EDX quantitative analysis of the outer surface
of prinstine and fouled membrane.
Surface element Pristine membrane Fouled membrane
(wt%)
Carbon 40.68 50.45
Oxygen 4.30 29.33
Fluorine 54.47 10.18
Silicon 0.38 0.39
Aluminum 0.17 0.37
Magnesium 4.17
Iron 2.34
Calcium 0.83
Molybdenum 0.82
Sodium 0.61
Sulfur 0.51
Total 100 100
3.4. Impact of surfactant on membrane wetting
As mentioned earlier, membrane wetting is one of the leading
causes for the reduction of COD and color removal ef-
ciencies during the long-term study. Camacho et al. (2013)
stated that membrane wetting can occur under three con-
ditions; (1) the hydraulic pressure exceeds the membrane
liquid entry pressure, (2) the fouling alters the membrane
surface hydrophobicity and (3) high organic content and sur-
factant in the feed solution that reduce membrane surface
tension. In this work, the membrane wetting is most likely
caused by the chemical and physical interactions between
the foulants and the polymeric membrane structure coupled
with the adsorption of surfactants into the membrane pores.
In order to testify that the membrane fouling had played a
signicant role in reducing the surface hydrophobicity, Fig. 6
compares the water contact angle of the pristine and fouled
membranes. The membrane hydrophobicity was found to be
greatly altered after 40-h distillation process, decreasing from
97.72 2.54 in the pristine membrane to 70.88 0.42 in the
fouled membrane. This was about 30% reduction in its surface
hydrophobicity.
It is postulated that the presence of surfactants in textile
efuent is the major factor affecting membrane hydrophobic-
ity as surfactants exhibit amphiphilic property that is likely
to reduce surface tension of membrane made of hydropho-
bic polymer (Lin et al., 2014a,b). The industry that provided
us the sample for this study has also conrmed the use of
surfactants in the production but details of the surfactant
properties could not be disclosed without authorization. In
this work, the surfactants that anchored on the membrane
surface would render the hydrophobic membrane surface into
more hydrophilic, causing the wetting problem and permeate
quality deterioration. The observation was in agreement with
the previous work conducted by Lin et al. (2014a,b). Another
possible factor contributing to low water contact angle is the
complex composition of the organic foulants adhered onto the
membrane surface. Goh et al. (2013) stated that the presence of
organic foulants could alter the membrane surface properties,
especially the hydrophobicity.
Fig. 7 shows the FTIR results of the surface of both pristine
and fouled membranes. It can be seen that the used mem-
brane sample showed very weak bands at peaks of around
840, 876, 1180 and 1400 cm1 . These peaks are originally
attributed by the C H2 rocking and C F2 asymmetric stretch-
ing of the PVDF-Cloisite 15A membrane (Mokhtar et al., 2014a).
The results suggested that the fouling layer had reduced
290 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 284293

Fig. 5 FESEM images of the (a) pristine and (b) fouled membrane, (i) cross section (magnication of 250), (ii) outer surface
(magnication of 20,000) and (iii) inner surface (magnication of 5000).

Fig. 6 Contact angle images of (a) pristine and (b) fouled membranes after 40-h treatment.

the hydrophobic functionality of the virgin membrane and
changed the chemical bonds and molecular structure of the
polymerinorganic nanocomposite membrane. In comparison
to the pristine membrane, the used membrane sample showed
various additional peaks, especially in the wavenumber range
between 2850 and 4000 cm1 . These peaks are likely caused by
the organic compounds in the textile efuent (Nordin et al.,
2013). The peak at 3447 cm1 indicates the N H stretching
of primary amine, while peaks at 2919 and 2851 cm1 denote
C H stretching of alkane (Nordin et al., 2013). In addition, peak
at 1540 cm1 represents the C N stretching of azo groups in
colored compounds (Mostaco-guidolin et al., 2010).
Although the DCMD process shows promising results in
the removal of non-volatile compounds, more research should
 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 284293 291

Table 6 Comparison of the results obtained in this study with the literature in the MD process for textile wastewater
treatment.
MD membrane MD conguration Cf a Jv (kg/m2 h) Tf(in) ( C) Tp(in) ( C) Rcolor (%) References

Commercial PP capillary Hybrid 0.03 g/L AR18 3.5 103 65 20 100 Mozia et al. (2010)
membrane from photocatalysis-
Membrana GmbH, DCMD
Germany
Commercial PP hollow ber SPMDR 0.4 g/L RB5 4.56 65 100 Qu et al. (2014)
membrane from
Hangzhou Kaijie
Membrane Company,
China
In-house made PVDF DCMD 0.05 g/L RB5 5.64 0.10 80 20 99.83 0.01 Mokhtar et al. (2014b)
hollow ber membrane
In-house made PVDF DCMD 0.5 g/L RB5 9.82 0.52 60 20 99.86 0.04 Mokhtar et al. (2015d)
hollow ber membrane
Commercial PTFE at sheet DCMD Industrial 22 60 21 Lin et al. (2014a,b)
membrane from textile
EF-Materials Industries wastewater
Inc., Taiwan
In-house made DCMD 0.05 g/L AR1 12.42 0.93 70 20 99.92 0.07 Mokhtar et al. (2015b)
PVDF-Cloisite 15A hollow and 1 M NaCl
ber membrane
In-house made DCMD Industrial 36.82 0.88b 90 25 95.33b This study
PVDF-Cloisite 15A hollow textile
ber membrane wastewater

PP, polypropylene; PVDF, polyvinylidene uoride; PTFE, polytetrauoroethylene; DCMD, direct contact membrane distillation; SPMDR, submerged
photocatalytic membrane distillation rector; Tf(in) , Feed inlet temperature, Tp(in) , Permeate inlet temperature, Rcolor , color rejection.
a
Cf , feed concentration, AR18 (Acid red 18), 375.4 g/mol; RB5 (reactive black), 991.82 g/mol; ARI (acid red 1), 509.42 g/mol; NaCl (sodium chloride),
58.44 g/mol.
b
Average result for the 150-min experimental work.

surface modication using advanced materials. For example,

Lin et al. (2014a,b) developed novel anti-wetting membrane
or omniphobic membrane by modifying a hydrophilic glass
ber membrane with silica nanoparticles followed by surface
uorination and polymer coating. It was found that the mem-
brane performance was very robust against the feed water
containing surfactants. Lu et al. (2016) on the other hand
explored an effective method to fabricate amphiphobic mem-
branes with anti-fouling and anti-wetting properties for MD
process.

3.5. Comparison with other MD membrane for textile

wastewater treatment

Table 6 compares the performance of the MD process studied

Fig. 7 FTIR-ATR analyses of the PVDF-Cloisite 15A
membranes before and after 40-h treatment with textile.
be carried out to further improve the membrane permeate
ux for possible industrial implementation. Moreover, mem-
brane fouling and pore wetting are critical issues in membrane
treatment for real wastewater, as they could compromise the
productivity and efciency of the treatment process in long
run. Future research is necessary to mitigate the fouling layer
caused by the attachment of organic compounds present in
the wastewater on the membrane surface. Membrane clean-
ing using water or mild chemical solution (acidic and/or basic
solution) could be employed to recover membrane surface
hydrophobicity after its use (Zhang et al., 2015; Peng et al.,
2015; Ding et al., 2016). Other possible solutions to reduce
membrane fouling and wetting phenomena is membrane
in this work with the recently published works for the treat-
ment of dyeing solutions. As can be seen, there is only one
study from open literature reported on the MD of industrial
textile wastewater treatment process in which a commer-
cial polytetrauoroethylene membrane modied with agarose
hydrogel was used (Lin et al., 2014a,b). Although the mem-
brane reported in this work was self-fabricated in laboratory,
its performance during MD process was comparable with the
commercial membrane with respect to permeate ux and
color removal. It is also worth to mention that only minor
reduction was observed on the color rejection compared with
our previous work that used synthetic wastewater as feed
(Mokhtar et al., 2015b). The slight reduction in color rejec-
tion might be attributed to the complex composition of the
industrial wastewater, in which some of the organic foulants
might cause the fouling and affect the permeate quality. Over-
all, the results of this work demonstrated that the MD process
is potential for industrial textile wastewater treatment if more
292 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 284293
research is conducted to address its fouling and pore wetting
problems.
4. Conclusion
The feasibility of applying MD process for industrial textile
wastewater treatment has been experimentally demonstrated
in this work. The ndings of this work revealed that the
MD is promising to overcome the limitations of pressure-
driven membrane processes such as UF, NF and RO in terms
of removal efciencies of pollutants and energy consump-
tion (low operating pressure). Furthermore, MD can exploit
the free energy given by hot efuent discharged by textile
industry and requires signicantly lower energy consump-
tion during operation. The in-house made PVDF-Cloisite 15A
nanocomposite membrane showed huge potential to produce
treated efuent of good quality and acceptable ux under low
hydraulic pressure. However, foulant depositions on mem-
brane outer surface that led to increase in mass resistance and
reduced permeate ux need further research to improve mem-
brane anti fouling resistance. In addition to membrane fouling,
wetting and thermal polarization effect cannot be ruled out
as the factors that might partly contribute to ux decrease.
With respect to separation characteristics, MD process could
achieve better COD (75 3%) and color removal (83 6%) dur-
ing the treatment process of industrial textile wastewater,
producing a permeate of high quality for possible reuse. In
summary, more research is still needed, especially on fouling
mechanism related to surfactants.
Acknowledgements
The authors gratefully acknowledge Universiti Teknologi
Malaysia (UTM) for funding this project with Research Uni-
versity Grant (Tier 1) (Vot No: Q.J130000.2509.05H48, Title:
Separation and Purication of Textile Wastewater using Low-
Energy Direct Contact Membrane Distillation (DCMD) Process).
The authors also gratefully thank Mrs. Aiza binti Mohd Zabidy
for her technical assistance on wastewater sampling.
References
Brik, M., Schoeberl, P., Chamam, B., Braun, R., Fuchs, W., 2006.
Advanced treatment of textile wastewater towards reuse
using a membrane bioreactor. Process Biochem. 41, 17511757.
Camacho, L., Dume, L., Zhang, J., Li, J., Duke, M., Gomez, J., Gray,
S., 2013. Advances in membrane distillation for water
desalination and purication applications. Water 5, 94196.
Chabot, S., Roy, C., Chowdhury, G., Matsuura, T., 1997.
Development of poly(vinylidene uoride) hollow-ber
membranes for the treatment of water/organic vapor
mixtures. J. Appl. Polym. Sci. 65, 12631270.
Chen, J., Wang, Q., Hua, Z., Du, G., 2007. Research and application
of biotechnology in textile industries in China. Enzyme
Microb. Technol. 40, 16511655.
Chhabra, M., Mishra, S., Sreekrishnan, T.R., 2015. Combination of
chemical and enzymatic treatment for efcient
decolorization/degradation of textile efuent: high
operational stability of the continuous process. Biochem. Eng.
J. 93, 1724.
Dasgupta, J., Sikder, J., Chakraborty, S., Curcio, S., Drioli, E., 2015.
Remediation of textile efuents by membrane based
treatment techniques: a state of the art review. J. Environ.
Manage. 147, 5572.
Ding, S., Yang, Y., Li, C., Huang, H., Hou, L., 2016. The effects of
organic fouling on the removal of radionuclides by reverse
osmosis membranes. Water Res. 95, 174184.
El-Abbassi, A., Khayet, M., Kiai, H., Hadi, A., Garca-Payo, M.C.,
2013a. Treatment of crude olive mill wastewaters by osmotic
distillation and osmotic membrane distillation. Sep. Purif.
Technol. 104, 327332.
El-Abbassi, A., Hadi, A., Khayet, M., Garca-Payo, M.C., 2013b.
Integrated direct contact membrane distillation for olive mill
wastewater treatment. Desalination 323, 3138.
Environmental Quality (Industrial Efuent) Regulations 2009,
http://www.doe.gov.my/portalv1/wp-content/uploads/2015/
01/Environmental Quality Industrial Efuent Regulations
2009 - P.U.A 434-2009.pdf.
Fersi, C., Gzara, L., Dhahbi, M., 2009. Flux decline study for textile
wastewater treatment by membrane processes. Desalination
244, 321332.
Firmino, P.I.M., da Silva, M.E.R., Cervantes, F.J., dos Santos, A.B.,
2010. Colour removal of dyes from synthetic and real textile
wastewaters in one- and two-stage anaerobic systems.
Bioresour. Technol. 101, 77737779.
Goh, S., Zhang, J., Liu, Y., Fane, A.G., 2013. Fouling and wetting in
membrane distillation (MD) and MD-bioreactor (MDBR) for
wastewater reclamation. Desalination 323, 3947.
Gryta, M., 2008. Fouling in direct contact membrane distillation
process. J. Membr. Sci. 325, 383394.
Gullinkala, T., Digman, B., Gorey, C., Hausman, R., Escobar, I.C.,
2010. Desalination: reverse osmosis and membrane
distillation. In: Escobar, I.C., Schafer, A.I. (Eds.), Sustainability
Science and Engineering. Elsevier, Ohio, pp. 6593.
Hausmann, A., Sanciolo, P., Vasiljevic, T., Weeks, M., Schron, K.,
Gray, S., Duke, M., 2013. Fouling mechanisms of dairy streams
during membrane distillation. J. Membr. Sci. 441,
102111.
Ismail, A.F., Lau, W.J., 2010. Application of response surface
methodology in PES/SPEEK blend NF membrane for dyeing
solution treatment. Membr. Water Treat. 1, 4960.
Ismail, A.F., Lau, W.J., 2009. Inuence of feed conditions on the
rejection of salt and dye in aqueous solution by different
characteristics of hollow ber nanoltration membranes.
Desalin. Water Treat. 6, 281288.
Khaing, T.H., Li, J., Li, Y., Wai, N., Wong, F., 2010. Feasibility study
on petrochemical wastewater treatment and reuse using a
novel submerged membrane distillation bioreactor. Sep. Purif.
Technol. 74, 138143.
Klueh, R.L., 1982. Chromium-molybdenum steels for fusion
reactor rst walls a review. Nucl. Eng. Des. 72,
329344.
Lau, W.J., Ismail, A.F., 2009. Polymeric nanoltration membranes
for textile dye wastewater treatment: preparation,
performance evaluation, transport modelling, and fouling
controla review. Desalination 245, 321348.
Li, J., Wu, J., Sun, H., Cheng, F., Liu, Y., 2016. Advanced treatment
of biologically treated coking wastewater by membrane
distillation coupled with pre-coagulation. Desalination 380,
4351.
Lin, S., Nejati, S., Boo, C., Hu, Y., Osuji, C.O., Elimelech, M., 2014a.
Omniphobic membrane for robust membrane distillation.
Environ. Sci. Technol. Lett. 1, 443447.
Lin, P.J., Yang, M.C., Li, Y.L., Chen, J.H., 2014b. Prevention of
surfactant wetting with agarose hydrogel layer for direct
contact membrane distillation used in dyeing wastewater
treatment. J. Membr. Sci. 475, 511520.
Liu, M., L, Z., Chen, Z., Yu, S., Gao, C., 2011. Comparison of
reverse osmosis and nanoltration membranes in the
treatment of biologically treated textile efuent for water
reuse. Desalination 281, 372378.
Lu, X., Peng, Y., Ge, L., Lin, R., Zhu, Z., Liu, S., 2016. Amphiphobic
PVDF composite membranes for anti-fouling direct contact
membrane distillation. J. Membr. Sci. 505, 6169.
Marcucci, M., Ciabatti, I., Matteucci, A., Vernaglione, G., 2003.
Membrane technologies applied to textile wastewater
treatment. Ann. N. Y. Acad. Sci. 984, 5364.
Melin-Martel, N., Sadhwani, J.J., Malamis, S.,
Ochsenkhn-Petropoulou, M., 2012. Structural and chemical
characterization of long-term reverse osmosis membrane
 chemical engineering research and design 1 1 1 ( 2 0 1 6 ) 284293
fouling in a full scale desalination plant. Desalination 305,
4453.
Mohammadi, M., Man, H.C., Hassan, M.A., Yee, P.L., 2010.
Treatment of wastewater from rubber industry in Malaysia.
Afr. J. Biotechnol. 9, 62336243.
Mokhtar, N.M., Lau, W.J., Goh, P.S., 2013. Effect of hydrophobicity
degree on PVDF hollow ber membranes for textile
wastewater treatment using direct contact membrane
distillation. J. Teknol. (Sci. Eng.) 65, 7781.
Mokhtar, N.M., Lau, W.J., Ismail, A.F., Ng, B.C., 2014a.
Physicochemical study of polyvinylidene
uoride-Cloisite15A composite membranes for membrane
distillation application. RSC Adv. 4, 6336763379.
Mokhtar, N.M., Lau, W.J., Ismail, A.F., 2014b. The potential of
membrane distillation in recovering water from hot dyeing
solution. J. Water Process Eng. 2, 7178.
Mokhtar, N.M., Lau, W.J., Ismail, A.F., 2015a. Dye wastewater
treatment by direct contact membrane distillation using
polyvinylidene uoride hollow ber membranes. J. Polym.
Eng. 35 (5), 471479.
Mokhtar, N.M., Lau, W.J., Ismail, A.F., Youravong, W.,
Khongnakorn, W., Lertwittayanon, K., 2015b. Performance
evaluation of novel PVDFCloisite 15A hollow ber composite
membranes for treatment of efuents containing dyes and
salts using membrane distillation. RSC Adv. 5, 3801138020.
Mokhtar, N.M., Lau, W.J., Ismail, A.F., Veerasamy, D., 2015c.
Membrane distillation technology for treatment of wastewater
from rubber industry in Malaysia. Procedia CIRP 26, 792796.
Mokhtar, N.M., Lau, W.J., Ng, B.C., Ismail, A.F., Veerasamy, D.,
2015d. Preparation and characterization of PVDF membranes
incorporated with different additives for dyeing solution
treatment using membrane distillation. Desalin. Water Treat.
56, 19992012.
Mostaco-guidolin, L.B., Murakami, L.S., Batistuti, M.R., Nomizo,
A., Bachmann, L., 2010. Molecular and chemical
characterization by Fourier transform infrared spectroscopy of
human breast cancer cells with estrogen receptor expressed
and not expressed. Spectroscopy 24, 501510.
Mozia, S., Morawski, A.W., Toyoda, M., Tsumura, T., 2010.
Integration of photocatalysis and membrane distillation for
removal of mono- and poly-azo dyes from water. Desalination
250, 666672.
Nordin, N., Amir, S.F.M., Riyanto, Othman, M.R., 2013. Textile
industries wastewater treatment by electrochemical
oxidation technique using metal plate. Int. J. Electrochem. Sci.
8, 1140311415.
Ntuli, F., Ikhu-omoregbe, D., Kuipa, P.K., Muzenda, E., Belaid, M.,
2009. Characterization of efuent from textile wet nishing
293
operations. In: Proceedings of the World Congress on
Engineering and Computer Science, 2022 October, San
Francisco, USA.
Ogunlaja, O.O., Aemere, O., 2009. Evaluating the efciency of a
textile wastewater treatment plant located in Oshodi, Lagos.
Afr. J. Pure Appl. Chem. 3, 189196.
Ong, C.S., Lau, W.J., Ismail, A.F., 2012. Treatment of dyeing
solution by NF membrane for decolorization and salt
reduction. Desalin. Water Treat. 50, 245253.
Ong, Y.K., Li, F.Y., Sun, S.P., Zhao, B.W., Liang, C.Z., Chung, T.S.,
2014. Nanoltration hollow ber membranes for textile
wastewater treatment: lab-scale and pilot-scale studies.
Chem. Eng. Sci. 114, 5157.
Peng, Y., Ge, J., Li, Z., Wang, S., 2015. Effects of anti-scaling and
cleaning chemicals on membrane scale in direct contact
membrane distillation process for RO brine concentrate. Sep.
Purif. Technol. 154, 2226.
Qu, D., Qiang, Z., Xiao, S., Liu, Q., Lei, Y., Zhou, T., 2014.
Degradation of Reactive Black 5 in a submerged photocatalytic
membrane distillation reactor with microwave electrodeless
lamps as light source. Sep. Purif. Technol. 122, 5459.
Qu, D., Wang, J., Hou, D., Luan, Z., Fan, B., Zhao, C., 2009.
Experimental study of arsenic removal by direct contact
membrane distillation. J. Hazard. Mater. 163, 874879.
Raj, B., Choudhary, B., Singh Raman, R., 2004. Mechanical
properties and non-destructive evaluation of
chromiummolybdenum ferritic steels for steam generator
application. Int. J. Press. Vessel. Pip. 81, 521534.
Ramezanianpour, M., Sivakumar, M., 2014. An analytical ux
decline model for membrane distillation. Desalination 345,
112.
Shon, H.K., Vigneswaran, S., Kandasamy, J., Cho, J., 2007.
Characteristics of Efuent Organic Matter in Wastewater.
Eolss Publishers, Oxford, UK.
Tfekci, N., Sivri, N., Toroz, I., 2007. Pollutants of textile industry
wastewater and assessment of its discharge limits by water
quality standards. Turk. J. Fish. Aquat. Sci. 7, 97103.
Vajnhandl, S., Valh, J.V., 2014. The status of water reuse in
European textile sector. J. Environ. Manage. 141,
2935.
Warsinger, D.M., Swaminathan, J., Guillen-Burrieza, E., Arafat,
H.A., Lienhard-V, J.H., 2015. Scaling and fouling in membrane
distillation for desalination applications: a review.
Desalination 356, 294313.
Zhang, P., Kntig, P., Gray, S., Duke, M., 2015. Scale reduction and
cleaning techniques during direct contact membrane
distillation of seawater reverse osmosis brine. Desalination
374, 2030.