Available online at www.sciencedirect.

com

ScienceDirect
Agriculture and Agricultural Science Procedia 2 (2014) 257 264

ST26943, 2nd International Conference on Agricultural and Food Engineering, CAFEi2014

Wastewater Treatment of Palm Oil Mill Effluent (POME) by

Ultrafiltration Membrane Separation Technique Coupled with
Adsorption Treatment as Pre-treatment
Nazatul Shima Azmi, Khairul Faezah Md. Yunos*
Department of Process and Food Engineering, Faculty of Engineering,University of Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia

Abstract
The application of membrane separation technology coupled with adsorption treatment (pre-treatment) can be effective means of
treating palm oil mill effluent (POME).The pre-treatment process was done by stirring the raw palm oil mill effluent (POME)
with 0.20 g/L of palm kernel shell-based activated carbon (PKS-AC) in 35.94 min with stirring speed 39.82 rpm and able to
reduce 71.26% of suspended solid. Ultrafiltration membrane separation was subsequently applied for further treatment of POME.
From a random observation on the result, it was found that at pressure 2 bar with 600 rpm speed and pH 8 obtained lower
dissolved solid (123.70 mg/L) and turbidity (4.50 NTU) concentration.

by Published
Elsevier B.V.
This is an
open access article under the CC BY-NC-ND license
2014
2014 Published
The Authors.
by Elsevier
B.V.
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the Scientific Committee of CAFEi2014.
Peer-review under responsibility of the Scientific Committee of CAFEi2014
Keywords:POME; PKS-AC; adsorption; ultrafiltration membrane sepeeration; treated POME

1. Introduction
Malaysian commonly used palm oil as the alternative for vegetable oil in their daily life. This palm oil was
extracted from oil palm fruits. Malaysia itself managed 39% of the worlds palm oil production and 44% of the
worlds exports (MPOB, 2012). Furthermore, the current market showed the high demand in palm oil and oleo
chemical industries which will lead in high production rate (Ahmad et al.,2011). Unfortunately, along with the
increased in the palm oil production, it will also result in concomitant the production of waste. Normally, there were

* Corresponding author. Tel.: +603-89466348; fax: 603-89464440.

E-mail address:kfaezah@upm.edu.my

2210-7843 2014 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the Scientific Committee of CAFEi2014
doi:10.1016/j.aaspro.2014.11.037

258

Nazatul Shima Azmi and Khairul Faezah Md. Yunos / Agriculture and Agricultural Science Procedia 2 (2014) 257 264

two types of wastes produced in this industry, which are solid and liquid wastes. The liquid waste was known as
palm oil mill effluent (POME).
POME is a thick brownish viscous liquid waste, slurry, high in colloidal suspension and has an unpleasant odor
(Ahmad et al., 2009). The higher portion in the raw POME was water with 95-96% and made up with 0.6 - 0.7% oil
and 4-5% total solids (Ma, 2000; Ahmad et al., 2003). Given the huge concentration in biological oxygen demand
(25,000 mg/L), chemical oxygen demand (53,630 mg/L), oil and grease (8,370 mg/L) and suspended solid (19,020
mg/L), its disposal without proper treatment in water bodies has become undesirable (Ma, 1995; Wu et al., 2007).
The high amount of this degradable organic matter in the raw POME or partially treated POME was due to the
presence of the unrecovered palm oil (Ahmad et al., 2003). The palm oil industry has a big responsibility to face it in
term of environmental protection, economic viability, and sustainable development.
Over the last few decades, there are several innovative methods that have been developed and applied by palm oil
mills for the treatment of POME. Palm oil mill management commonly applied convectional biological treatments
of anaerobic or facultative digestion (Ahmad et al.,2003; Quahet al., 1982). The anaerobic and facultative ponds rely
on bacteria to break down the organic matters into simple end products of methane, carbon dioxide, hydrogen
sulphide and water. This system consists of a series of ponds connected which each of pond has its own purposes.
However, this biological treatment system needs proper maintenance and monitoring; increase the labour
requirement and cost. This is due to the processes relying solely on microorganism to break down the pollutants.
The microorganisms are very sensitive to the surrounding temperature and pH and thus extra care has to taken in
order to ensure a conducive environment for the microorganism to develop well (Ahmad et al., 2003). This
treatment also required large treatment area with a long treatment periods (80 to 120 days) (Ahmad and Chan,
2009). Moreover, biological treatment also generates vast amounts of biogas, which is corrosive and odorous. This
biogas is contains of methane, carbon dioxide and trace amounts of hydrogen sulphide (Ahmad et al., 2003). These
gases are corrosive and dangerous. Moreover, methane gas is more potent and fire hazard.
To encounter the current problems and challenges in POME treatment, it is suggested to apply membrane
separation technology. This technology is in a state of rapid growth and innovation. There are several advantages to
use membrane separation technology. It can be applied across a wide range of industries; the quality of the treated
water is more consistent regardless of the influent variations; it can be used in a process to allow the recycling of
selected waste streams within a plant; highly skilled operators would not be required when the plant can be fully
automated (Cheryan and Rajagopalan, 1998) and the water reclaimed from this treatment could be reused in the
mill. Thus, the primary advantages lie in the reduction of the cost for the water supply and its further treatment as
well as in the effective elimination of the pollutant from the POME. However, the palm oil mill industry still do not
apply and used the membrane technology to treat POME due to the high cost for installation of the membrane set up
and maintenance. The cost for maintenance can be reduced by first pre-treated the POME to lower the concentration
of the colloidal particles which can cause damage and fouling before applying it to ultradiltration membrane.
Moreover, the operational time of membrane technology was lesser than other convectional treatment and the cost
of the treatment can be reduced simultaneously. Hence, high cost in set up installation was only at the beginning to
compared with the cost of treatment area and time consuming for conventional treatment which is more expensive
In this current research, the objective is to investigate the performance of ultrafiltration membrane treatment in
reclaiming water reuse towards the different operating conditions.
2. Materials and methods
2.1. Materials
Sample of palm oil mill wastewater was taken from a local palm oil mill in Labu, Negeri Sembilan. Raw palm
oil mill effluent (POME) was taken from a pipe before pumping into the pond area that had a temperature of around
80C to 90C. The sample was stored at 4C to avoid biodegradation due to microbial action. Palm kernel shell
activated carbon (PKS-AC) is in granular form with size range less than 2000 m and more than 500 m and (K. D.
technology). PKS-AC was in laboratory grade and used directly as received from the supplier was physically
activated. The membrane use in this study was flat sheet regenerated cellulose (RC) membrane (Merck

Nazatul Shima Azmi and Khairul Faezah Md. Yunos / Agriculture and Agricultural Science Procedia 2 (2014) 257 264

MiliporeUSA) with 28.7 cm2 effective membrane area and membrane diameter of 63.5 mm. The molecular weight
cut-off (MWCO) of the RC membrane used was 5 kDa.
2.2. Laboratary scale experimental set-up
The experiments were carried out in two stages which were pre-treatment stage and membrane treatment stage.
Adsorption treatment (pre-treatment) was initially applied before ultrafiltration of POME, to reduce the sludge and
particles in POME. This pre-treatment was performed prior to reduce and avoid fouling effect later on membrane
during ultrafiltration membrane process. An amount of PKS-AC (0.2 g/L 0.6 g/L) was stirred with raw POME in a
certain operating conditions [stirring time (20-50 min) and stirring speed (20-40 rpm)] using Flocculator SW 1
(Stuart Scientific). After that, the POME was left for one hour for sedimentation process. The supernatant was then
pipette out and analyzed. The optimization of the treatment conditions was obtained with the aid of Design Expert
Version 6.0.4 software (Stat-Ease, Inc, Minneapolis).
180 ml of pre-treated POME was prepared for each run in ultrafiltration treatment. The experiment was
performed in batch mode using stirred ultrafiltration cell (Amicon 8200, Milipore USA). There were two parameters
measured to see the effectiveness of membrane in the POME treatment. The parameters varied in this batch
experiments were stirring speed (0.200 and 600 rpm) and pH (5.85, 7, 8 and 9) of the feed solution (pre-treated
POME). The membrane unit was pressurized using five different compressed air pressure (1, 1.5, 2, 2.5 and 3 bar).
For each cycle, the experiment was run in 90 min. The permeate flux was being observed by collected the permeate
volume for every five minutes and the volume was recorded. The permeate was then further analyzed for dissolved
solid and turbidity.The details of all the analytical methods conducted were based on procedures given in the APHA,
Standard Method for the Examination of Water and Wastewater (Hammer et al., 2005). Each analysis was done
twice.
Ultrafiltration membrane separation was measured using permeate flux analysis. The permeate flux was being
observed by collected the permeate volume for every five minutes and the volume was recorded. In accordance with
another study, the permeate flux, J, was calculated by Eq. (1) (Tchobanoglous et al., 2003).
J = (1/Am) dV/dT

(1)

whereJ is the permeate flux of solution (L/m2.h), Am is the membrane area, dV is the amount of permeate collected,
and dT is the time taken for the permeate flow to be collected.
3. Results and discussion
3.1. Quality of pre-treated POME
Batch adsorption was performed in 20 sets of 1000 ml of raw POME without any pH adjustment. Analysis of
variance (ANOVA) was further carried out to justify the adequacy of the model. The results show that the model Fvalue of 19.52 and Prob> F of < 0.0001 be state that it was good to validate the fit. It is indicating that the
interaction of the model was significant at 90% of the model and about 10% of the model cannot be explained by the
model which might lead to larger variation in the suspended solid reduction predicted from the model (Hameed et
al., 2008). The value of the adjusted determination coefficient (adjusted R2 = 0.854) is also high to advocate a high
significance of the model (Khuri and Cornall, 1987). The result of the experiments for suspended solid analysis
showed that dosage (A), time (B), dosage-time (AB), dosage-speed (AC), time-speed (BC) and the second order
effect of time (B2) is significant model terms. In order to improve the model, the insignificant model were reduced
or removed and the ANOVA can be seen in Table 2. The following regression equation is the empirical models in
term of coded factors for suspended solid as in Eq. (2).
Suspended solid reduction = 68.88- 1.43A+1.09B - 4.30B2+ 2.05AB - 2.87AC +2.03 BC (2)
The optimum values of selected variables were obtained by solving the regression equations. The optimum

259

260

Nazatul Shima Azmi and Khairul Faezah Md. Yunos / Agriculture and Agricultural Science Procedia 2 (2014) 257 264

values of the test variables in actual were as follows; PKS-AC dosage= 0.20 g/L, treatment time=35.94 min, stirrer
speed=39.82 rpm with 73.16% prediction of suspended solid reduction. The average value for suspended solidwas
71.26%.
The result of the pollutant in the raw POME was significantly reduced after the pre-treatment stage.The pH of
pre-treated POME shows some shifted from 3.9 to 5.85. This might be due to the pH of PKS-AC itself, 8 which is
more basicity and have more OH-.
As an overall, PKS-AC can be used as adsorbent in adsorption treatment for pre-treatment of POME. This pretreatment successfully reduced total solid, dissolved solid, suspended solid, BOD 5, COD and turbidity up to 67.30%,
47.11%, 71.26%, 63.23%, 42.38%, and 63.31% respectively. Using this pre-treated POME it is believed can reduce
the fouling effect during ultrafiltration process later.
3.2. Ultrafiltration of POME
3.2.1. Effect of stirring speed and pressure on the membrane performance (permeate flux, dissolved solid and
turbidity)
As shown in Fig.1(a), the permeate flux was significantly different between with and without stirrer. The trend
showed that the permeate flux increased as the stirring speed increased. The same trend was also observed in
investigations carried out at different operating pressure.
The higher flux is noticeable with the presence of stirring effect which will reduce concentration polarization
within the solute and membrane surface. Moreover, as stirring speed increased the shear stress and hydrodynamic
effect on membrane surface will increase too. As explained in the previous paragraph, the accumulated compounds
on membrane surface return in to the bulk of the fluid and concentration polarization effect diminishes (Mohammadi
and Esmaeelifar, 2005). Thus, it causes the osmotic pressure to decrease and permeation flux to increase (Scott et
al., 2000; Wang et al., 2000).
However, the permeate flux at 200 and 600 rpm at 1.5 bar were 46.20L/m2h, and 46.41L/m2h shown that it was
not significantly increased. This phenomenon was also observed by other researchers which in finding that at a
definite threshold, increasing speed does not affect permeation flux (Wang et al., 2000; Brincket al., 2000).The
negligible flux change at high stirring speed was due to the portion of the cake layer resistance which is relatively
small compared to the other resistances caused by the membrane itself and the fouling layer. Hence, the dislodging
of the cake layer at higher shear rates may not show any obvious in flux changes. Moreover, the cake layer is also
too sticky to be removed by the shear applied (Kang and Choo, 2003).
By referring to Fig.1(b), as the pressure increased until 2 bar, dissolved solid was decreased with increasing
stirring speed. In the other hand, the trend was suddenly changed at pressure 2.5 bar and 3 bar, dissolved solid
decreased from 0 rpm to 200 rpm but increased again at 600 rpm.
Dissolved solid concentration in permeate was higher when the solution was unstirred because there was no
hydrodynamic effect which can lead in increasing the deposition of cake layer on the surface of the membrane. This
deposition of cake layer will increased the possibility of dissolved solid to pass through the membrane with forces
from pressure applied. In the other hand, unstirred conditions showed decreasing in dissolved solid concentration
when the operating pressure is increasing. This phenomenon could be due to the existence of fouling layer and pore
plugging which would exaggerated at higher pressure (Nakatsuka and Michaels, 1992; Wu et al., 2007). At higher
pressure, it is believed that the gel polarization layer formed at the higher pressure is stable to disruption by pressure
fluctuation (Nakatsuka and Michaels, 1992). This layer will act as filter on the top of membrane and reduced the
permeability of dissolved solid through the membrane. When the solution was stirred at 200 rpm and 600 rpm at
pressure 1.5 and 2 bar, the dissolved solid concentration was decreased and increased back from 2.5 to 3 bar. This is
because at low pressure, the pressure will not enough to pressurized some of the deposition of cake layer on the
surface to the permeate side but at high pressure, it is possible with stirred condition. Moreover, when there the
stirring speed is too high with high pressure, it will sweep away the deposition of the cake layer on the membrane
surface and the dissolved particle will easily pass through the membrane (Ahmad et al., 2005). Hence, it can be
hypothesized that dissolved solid was sensitive to the hydrodynamic effect and operating pressure.
From Fig.1(c), as the stirring speed increased, the turbidity concentration was decreased. This is because by

Nazatul Shima Azmi and Khairul Faezah Md. Yunos / Agriculture and Agricultural Science Procedia 2 (2014) 257 264

increasing speed, the hydrodynamic effect and shear stress is also increased which then returned back the
accumulated compounds on the membrane surface back to the bulk of fluid (Mohammadi and Esmaeelifar, 2005).
Thus, the concentration polarization effect diminishes. As reported by Benito (2001), it is believed by increasing in
hydrodynamic effect the possibility of gel layer formation is lower as a result of higher shear rates at the membrane
surfaces. Thus, it will lower the possibility of the colloidal to pass through the membrane.
At 600 rpm, the permeate output is higher with lower value of dissolved solid and turbidity. Higher value of
permeate flux of permeate will promises higher yield rate.
(b)

(a)

(c)

Fig.1.Effect of stirrer speed on (a) permeate flux, (b) dissolved solid and (c) turbidity concentration after ultrafiltration of POME.

3.2.2. Effect of pH and pressure on the membrane performance (permeate flux, dissolved solid and turbidity).
As presented in Fig.2(a), it was found that pH does give impact on flux changes at low pressure. Meanwhile at
high pressure, permeate flux does not show any significant changes on different pH (flux become insensitive to pH).
At lower pressure, the permeate flux decreased when pH of the sample increased from pH 5.85 to 7 and increased
back as the pH increased from pH 8 to 9. It explains that the feed chemistry is changed at acidic and basic solutions.
Advantageously, this will causes the fouling resistance on the membrane surface to reduce and the permeate flux of
the membrane to enhance (Ahmad et al., 2003; Salahi et al., 2010).
The higher permeate flux can be observed at pH 8 and 9 rather than pH 5.85. When pH was adjusted to alkaline,
the carboxylic group on the surface fully dissociates and at the same time membrane surface gains its strongest
negative charge (Manttariet al., 2006). The negative charge of the polymer chains in three-dimensional network of
the surface start to repel to each other and later will open up the pore of the membrane skin layers. Hence, the water
permeability to permeate side will be more efficiently. It was also found that, at alkaline state, the particles form a
stable suspensions and the fatty acid molecules (POME is an oily wastewater) are converted into ions and their
accumulation on the surfaces reduces (Mohammadi and Esmaeelifar, 2005). As a result, the flux will be increased.

261

262

Nazatul Shima Azmi and Khairul Faezah Md. Yunos / Agriculture and Agricultural Science Procedia 2 (2014) 257 264

Results in Fig.2(b) represent the dissolved solid after ultrafiltration of POME at various pH and pressure.
Dissolved solid seem to be decreased as pH increased from 5.85 to 7 and decreased back as the pH increased to 8
and 9. This similar trend was also observed at other operating pressure.
At lower pH, this natural organic matter is seen to agglomerate with the presence of high multivalent cations
concentrations (Zularisamet al., 2006). Moreover, the concentration of H+ ions and density for the negative ions on
the surface particles became low as in acidic conditions (Ahmad et al., 2002). In this situation, the dissolved solid
particles carry (positive) opposite charge from the membrane (negative charge) which will increase the attraction
forces. The dynamic membrane will be formed due to the adsorption of the dissolved solid particles (Ghosh,
2003). This dynamic membrane has the same charge as the other dissolved solid particles in the solution. As a result,
the dissolved solid particles in the solution that approached the membrane surface will repel (self rejection) and
consequently reduced the transmission of dissolved solid. High reduction of dissolved in the permeate was also
partly attribute by the restrict the entrance of the solute through the membrane due to the compaction of membrane
pore size at low pH and high ionic strength (Braghettaet al., 1997; Zularisam et al., 2006).
Dissolved solid concentration at solution pH 8 and 9 were lower than pH 7. When POME is adjusted to alkaline,
the particles in POME were in stable conditions (Ahmad et al., 2002). Moreover, unlike in acidic conditions, the
density of negative ions on the particles surface was higher in alkaline condition. Alkaline solution also carries more
OH- ions. The repulsion forces will repel the dissolved solid particles away from the membrane surface due to the
same ion charge carry by the membrane and dissolved particles. This phenomenon called intrinsic electrostatic
rejection (Ghosh, 2003). In the other hand, dissolved solid concentration at high pH is higher than at low pH. This
is because at high pH, the negative charge of the polymer chains in three-dimensional network of the surface start to
repel to each other and later will open up the pore of the membrane skin layers (Manttari et al., 2006). Hence, it will
reduce the rejection of dissolved solid but increase in permeability of water through the membrane. It was proved by
high dissolved solid concentration in permeate with high permeate flux.
(a)

(b)

(c)

Fig.2.Effect of pH solution on (a) permeate flux, (b) dissolved solid and (c) turbidity concentration after ultrafiltration of POME.

Nazatul Shima Azmi and Khairul Faezah Md. Yunos / Agriculture and Agricultural Science Procedia 2 (2014) 257 264

As presented in Fig.2(c), the results show that pH of the solution does affect on the turbidity rejection. Obviously,
at pH 8, the turbidity was constantly lower regardless of operating pressure. Meanwhile, at pH 9, the turbidity seems
to decrease as the pressure increased. For solution at pH 5.85 and 7, the turbidity did not decreased continuously as
the pressure increase, but increased at certain pressure. The pH adjustment gives the effect of the electric charges to
the particles which will influence on the stability of the particles (Rohrsetzeret al., 1998; Ahmad et al., 2002).
The pre-treated POME might contain residual oily bearing cellulosic material which consist of free oil and
encapsulated oil. The encapsulated oil is found within the plant cell inside the fiber which linked to protein and
carbohydrates (Wahabet al., 2013). At pH 7, this protein accumulates and forms a densely packed layer on the
surface of the membrane due to the concentration polarization effect (Das et al., 2009). At this stage, the layer seem
to be constituted of the small protein molecules which form a structure that is less open and act as the other filter
media on the surface of the membrane. Hence, lower value in turbidity was observed at pH 7 compared to pH 5.85.
From the random observations based on the high permeate flux, lower dissolved solid and turbidity concentration
was observed at pressure 2 bar using stirring speed 600 rpm and at pH 8. As can be seen in Table 1, there were
significant difference in each parameter between the raw POME, pre-treated POME and treated POME. This study
showed that the responses were improved at each stage of treatment. The pollutant elements in the raw POME were
reduced more than 90% after applying ultrafiltration membrane treatment coupled with adsorption treatment as the
pre-treatment.
Table 1. Quality of raw POME, pre-treated POME and treated POME.
Parameter

Raw POME

Pre-treated POME

Treated POME

pH
Total solid, mg/L
Dissolved solid, mg/L
Suspended solid, mg/L
BOD5, mg/L
COD, mg/L
Turbidity, NTU

3.9
75,200
43,015
32,185
31,000
73,306
17,000

5.85
32,000
22,750
9,250
11,400
42,240
6236.65

8
173.70
123.70
50
360
150.49
4.50

4. Conclusions
The optimum conditions from the pre-treatment process with high desirability that satisfied obtained at 0.20 g/L
PKS-AC dosage in 35.94 min stirring time and 39.82 rpm stirring speed. This pre-treatment successfully reduced
total solid, dissolved solid, suspended solid, BOD5, COD and turbidity up to 67.30%, 47.11%, 71.26%, 63.23%,
42.38%, and 63.31% respectively. Then, the pre-treated POME was subsequently further treated using ultrafiltration
membrane technique and a better quality of treated POME was obtained. From the random observations, the best
conditions was at pH 8, 600 rpm of stirring speed and by applying 2 bar of operating pressure for 90 min of
operating time. Applying these optimum conditions successfully reduced the pollutant elements up to 90% with high
permeate flux rate, 57.23 L/m2h. According to the results, it can be concluded that ultrafiltration treatment coupled
with adsorption treatment as pre-treatment is and advantageous method for the POME treatment. This treatment
offer the easier operating conditions, more ability to combine with two technique and easier control of the system.
Acknowledgment
The support of the Universiti Putra Malaysia (RUGS 9199806) is gratefully acknowledged.
References
Ahmad, A. L., Chan, C.Y., 2009. Sustainability of Palm Oil Industries: An Innovative Treatment via Membrane Technology. Journal of Applied
Sciences 9(17), 3074-3079.

263

264

Nazatul Shima Azmi and Khairul Faezah Md. Yunos / Agriculture and Agricultural Science Procedia 2 (2014) 257 264
Ahmad, A. L., Chong, M. F., Bhatia, S., Ismail, S., 2006.Drinking Water Reclamation from Palm Oil Mill Effluent (POME) using Membrane
Technology.Desalination191(1-3), 35-44.
Ahmad, A. L., Ibrahim, N., Ismail, S., Bhatia, S., 2002.Coagulation-Sedimentation-Extraction Pre-treatment Method for the Removal of
Suspended Solids and Residual Oil from Palm Oil Mill Effluent (POME). IIUM EngineeringJournal 3(1), 25-33.
Ahmad, A. L., Ismail, S., Bhatia, S., 2003.Water Recycling from Palm Oil Mill Effluent (POME) using Membrane Technology.Desalination 157
(1-3), 87-95.
Ahmad, A. L., Ismail,S.,Bahtia, S., 2005. Ultrafiltration Behaviorin the Treatment of Agro Industry Effluent: Pilot Scale Studies. Chemical
Engineering Science 60(19), 5385-5394.
Ahmad, M. N., Mokhtar, M. N., Baharuddin, A. S., Hock, L. S., Ali, S. R. A., Aziz, S. A., Rahman, N. A., Hassan, M. A,. 2011. Changes in
Physiochemical and Microbial Community during Co-Composting of Oil Palm Frond with Palm Oil Mill Effluent Anaerobic Sludge,
Bioresource Technology 6(4), 4762-4780.
Benito, J.M., Ebel, S., Guttierez, B., Pazos, C., Coca, J., 2001. Ultrafiltration of a Waste Emulsified Cutting Oil Using Organic Membranes.
Water, Air and Soil Pollution 128(1-2), 181-195.
Braghetta, A. H., DiGiano, F. A., Ball, W. P., 1997. Nanofiltration of Natural Organic Matters: pH and Ionic Strength Effects. Journal of
Environmental.Engineering123(7), 628-641.
Brinck, J., Jonsson, A.S., Jonsson, B., Lindau, J., 2000. Influence of pH on the Adsorptive Fouling of Ultrafiltration Membranes by Fatty Acid,
Journal Membrane Science 164(1-2), 187-194.
Cheryan, M.,Rajagopalan, M., 1998. Membrane Processing of Oily Streams.Wastewater Treatment and Waste Reduction.Journal of Membrane
Science 151(1), 13-28.
Das, R., Bhattacherjee, C., Ghosh, S., 2009. Effects of Operating Parameters and Nature of Fouling Behavior in Ultrafiltration of Sesame Protein
Hydrolysate. Journal of Desalination 237, 268-276.
Ghosh, R. Protein BioseperationUsing Ultrafiltration. 2003. London: Imperial College Press
Hameed, B.H., Tan, I. A. W., Ahmad, A. L., 2008. Optimization of Basic Dye Removal by Palm Oil Palm Fibre-Based Activated Carbon
UsingRespone Surface Methodology. Journal of Hazardous Materials 158(2-3), 324-332.
Hammer, M. J., Junior, M. J. H., Salim, M. R., 2005. Laboratory Chemical Analyses, in Water and Wastewater Technology, 5th Ed., Prentice
Hall, Singapore, pp. 28-44.
Kang, S.K.,Choo, K.H., 2005. Use of MF and UF Membranes for Reclamation of Glass Industry Wastewater Containing Collodial Clay and
Glass Particles. Journal of Membrane Science 223(1-2), 89-103.
Ma, A. N. 1995.A Novel Treatment for Palm Oil Mill Effluent, Palm Oil Research.Institute. Malaysia (PORIM) 29, 201-212.
Ma, A. N., 2000.Environmental Management for the Oil Palm Industry. Palm Oil Development 30, 1-10.
Mohammadi, T.Esmaeelifar, A., 2005 Wastewater Treatment of a Vegetable Oil Factory by a Hybrid Ultrafiltration-Activated Carbon Process.
Journal of Membrane Science 254(1-2), 129-137.
MPOB, 2014. Monthly Production of Oil Palm Products Summary for the Month of May 2014. Retrieved 03 July,
2014.http://bepi.mpob.gov.my/index.php/statistics/production/125-production-2014/660-production-of-oil-palm-products-2014.html
Nakatsuka, S, Michaels,A.S., 1992. Transport and Separation of Proteins by Ultrafiltration through Sorptiveand Non-Sorptive Membrane,
Journal Membrane Science 69(3), 189-211.
Quah, S. K., Lim, K. H., Gillies, D., Wood, B. J., Kanagaratnam, J., 1982. POME Treatment and Land Application System.Proc. Regional
Workshop on Palm Oil Mill Technology and Effluent Treatment. Kuala Lumpur: PORIM, pp. 193-200.
Salahi, A., Abbasi, M., Mohammadi, T., 2010. Permeate Flux Decline during UF Of Oily Wastewater: Experimental and Modelling. Desalination
251(1-3), 153-160.
Scott, K., Mahmood, A.J., Jachuck, R.J., Hu B.,2000. Intensified Membrane Filtration with Corrugated Membranes, Journal of Membrane
Science 173(1),1-16.
Tchobanoglous, G., Burton, F. L., Stensel, H. D., 2003. Wastewater Engineering Treatment and Reuse, Metcalf and Eddy, Inc., McGraw-Hill
Companies.
Wahab, N. A., Aziz, A. A., Berahim, N., Ali, S., Kamal, S. M. M., Jalani, N. F., 2005. Combined Effect of Operational Variables On Aqueous
Enzymatic Oil Extraction from Palm Pressed Fibre. Journal of Oil Palm Research 25(2), 197-207.
Wang, J., Zhou, X., Ma, J., 2012. Preparation and Characteristics of a Paper-Based Ultrafiltration Membrane.Bioresource Technology 7(1), 545553.
Wang, P., Xu, N., Shi, J., 2000. APilot Study of the Treatment of Waste Rollingemulsionusing Zirconia Microfiltration Membranes, Journal of
Membrane Science173(2) 159-166.
Wong P.W., Sulaiman N.M., Nachiappan M., Varadaraj B., 2002. Pre-treatment and Membrane Ultrafiltration using Treated Palm Oil Mill
Effluent (POME).Songklanakarin Journal of Science and Technology 24, 891-898.
Wu, T. Y., Mohammad, A. W., Md. Jahim, J., Anuar, N., 2007. Palm Oil Mill Effluent (POME) Treatment and Bioresources Recovery using
Ultrafiltration Membrane: Effect of Pressureo Membrane Fouling. Biochemical Engineering Journal 35(3), 309-317.
Zularisam, A. W., Ismail, A. F., Salim R., 2006. Behaviours of Natural Organic Matter in Membrane Filtration for Surface Water Treatment A
Review. Desalination194(1-3), 211-231.

Accepted for poster presentation in CAFEi2014 (December 1-3, 2014 Kuala Lumpur, Malaysia) as paper 149.