Desalination 354 (2014) 3038

Contents lists available at ScienceDirect

Desalination
journal homepage: www.elsevier.com/locate/desal

A short review on reverse osmosis pretreatment technologies

S. Jamaly 1, N.N. Darwish 1, I. Ahmed 1, S.W. Hasan
Institute Center for Water and Environment (iWATER), Department of Chemical and Environmental Engineering, Masdar Institute of Science and Technology, P.O. Box 54224,
Abu Dhabi, United Arab Emirates

H I G H L I G H T S

Reverse osmosis (RO) water desalination has been getting very popular worldwide.
Conventional pretreatment operational cost is lower than non-conventional systems.
Non-conventional (membrane) pretreatment systems produce better water quality.
Membrane pretreatment capital cost increased by 2040% upon feed water quality.
NF is a better pretreatment method when compared to conventional and UF.

a r t i c l e

i n f o

Article history:
Received 10 April 2014
Received in revised form 10 September 2014
Accepted 11 September 2014
Available online xxxx
Keywords:
Reverse osmosis
Pretreatment
Desalination
Fouling
Cost

a b s t r a c t
Recent research reports have underlined reverse osmosis (RO) as the most optimized technology for water desalination related applications. However, implementing this technology to seawater desalination is facing
challenges of membrane fouling. This includes membrane biofouling, organic and inorganic fouling which
adversely affect the process performance and overall treatment cost. To overcome these issues, pretreatment
units ahead of the RO system are necessary to reduce RO membrane fouling and enhance its operational efciency. This article aimed at reviewing the literature and summarizing relevant methods, mechanisms and novel
developments which improve the performance of the RO systems when coupled with either conventional or
non-conventional pretreatment units. Several studies suggested that the non-conventional pretreatment units
were more efcient than the conventional systems for producing better water quality and minimizing the overall
treatment cost. Ultraltration appeared to be a cost effective and efcient method of removing suspended solids
(SS) and bacteria. The advent of nanostructured membranes nanoltration has the potential of becoming preferred non-conventional desalination pretreatment over a wide range of salinity, total dissolved solids (TDS),
inorganics, viruses, etc.
2014 Elsevier B.V. All rights reserved.

Contents
1.
2.

3.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . .
Coupling RO with conventional technologies . . . . . . . . . . .
2.1.
Chlorination . . . . . . . . . . . . . . . . . . . . . .
2.2.
Clarication/dissolved air oatation (DAF) . . . . . . . .
2.3.
Ozonation . . . . . . . . . . . . . . . . . . . . . . .
2.4.
Coagulationocculation . . . . . . . . . . . . . . . .
2.5.
Scale inhibitors . . . . . . . . . . . . . . . . . . . .
Reverse osmosis with non-conventional pretreatment technologies .
3.1.
Ultraltration (UF) and microltration (MF) . . . . . . .
3.2.
Nanoltration (NF) and nanostructured membranes . . . .
3.2.
Potential use of MBRs as RO pretreatment . . . . . . . .

Corresponding author. Tel.: +971 2 810 9237; fax: +971 2 810 9901.
E-mail address: swajih@masdar.ac.ae (S.W. Hasan).
1
Tel.: +971 2 810 9237; fax: +971 2 810 9901.

http://dx.doi.org/10.1016/j.desal.2014.09.017
0011-9164/ 2014 Elsevier B.V. All rights reserved.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

31
32
32
32
32
33
33
33
33
34
35

S. Jamaly et al. / Desalination 354 (2014) 3038

4.
Economical aspects of RO with pretreatment technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.
Conclusions and recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction
There're almost 26 countries that lack access to pure water resources
to sustain agricultural and economic developments, and more or less
one billion people have been deprived of pure drinking water. Middle
East is among those regions where fresh water scarcity has severely
affected agricultural capability and public life. Not to mention that the
demand for pure water will dramatically increases according to the
world statistics forecasting 4050% population growth over the next
50 years. Thus, conservation and recycling of water for consumption
can minimize the problem to some extent [76]. More or less, 98% of
the available water in the world isn't available for direct consumption
due to salinity. As water scarcity in many regions of the world is becoming an undeniable fact, efforts have been made to develop technologies
for alternative water resources. Thermal desalination processes have
been a great option, yet require high capital and operating costs due
to installation, maintenance, and energy used. Toward minimizing the
overall desalination cost, reverse osmosis membrane ltration (RO)
has been widely used and is recently becoming an important alternative
source of clean water [63]. There're more than 15,000 desalination
plants around the world providing fresh water from saline water
through which this number will continue to rise as researchers work
to improve the process, both in terms of cost effectiveness and energy
efciency.
Desalination can be traced in history as back as in 1558. Giovani
Batista Della Porta (15351615) mentions three desalination systems
in his books, Magiae Naturalis subsequently translated into French, Italian and German languages. During 1589, in the second edition, he described seven methods of desalination, including a solar distillation
apparatus that converted brackish water into fresh water. He also explained a method to obtain fresh water from the air by dehumidication
later reported by Delyannis [27]. The phenomenon of osmotic pressure
was rst observed by the French Cleric, Abb6 Nollet in 1748 [70], while
the rst semi-permeable membrane was prepared by traube in 1867.
This gelatinous lm of copper ferrocyanide supported on a porous clay
frit displayed remarkable selectivity to dilute solutions of electrolytes
had in fact pioneered the ultraltration as a technology. Many of the
RO and nanoltration (NF) membranes used these days are primarily
condensation polymers whose origin began with the rst synthesis of
nylon.
Hassler [47] marked the beginning of membrane research at the university level. In his report, entitled The Sea as the Sea as a Source of Fresh
Water, Hassler mentioned the possibility of vapor transfer through
sheets of cellophane. In a subsequent report, he described salt repelling
osmotic membranes and permselective lms. It is believed that this
historic unpublished document, dated August, 1950, introduced the
rst concept of membrane desalination. By the mid-1960s, two major
chemical companies, Dow Chemical and DuPont, acknowledged the
scope of large-scale membrane desalination. Both rms initiated R&D
efforts which resulted in the development of hollow ber desalination
modules. The Dow concept involved cellulose acetate bers as reported
by Bray [16], while DuPont focused on polyamides. Due to the core
problems faced during the operational performance of desalination systems; it was important to understand the membrane foulingphenomenon, causes and mechanism.
Ning et al. [67] divided commonly occurring fouling scales into two
major classes, a) hard scales and b) soft amorphous complexes.
Among these typically brackish waters, scale foulants are calcium carbonate, calcium sulfate, strontium sulfate, barium sulfate, calcium

31

35
36
37

uoride and calcium phosphate. Whereas amorphous gels consist

of natural exocellular polysaccharides from microorganisms, hydrated silica, ferric and aluminum hydroxides, colloidal iron and
manganese compounds and pretreatment polymers. Apart from
these factors, silt particles, clay and lter mass such as activated
carbon and manganese green sand from lters are damaging the
membranes [67].
Scales result from super-saturation of the RO brine relative to the
low solubility salts. At high total dissolved solids (TDS), natural crystallization rates and in the presence of seed crystals, scales can grow along
and move forward from the back-end of the RO system (i.e. high concentration) toward the front (i.e. low concentration). With the advent
of effective modern anti-scalants, scaling is a much smaller problem
than colloidal fouling by coagulation of ne particles [69]. Natural waters contain suspended particles that are extremely small in size (b 0.1
micron, dened as colloidal). At such sizes, surface to mass ratio is so
much larger than visible particles causing them to agglomerate in indefinite patterns or undergo coagulation. The aggregation of particles and
deposition on membrane surfaces results in what we observe in membrane autopsies as amorphous gels. Such foulants are complex mixtures
and are difcult, sometimes impossible to clean.
Larger, visible particles if not removed from the RO feed water will
naturally plug the feed ow channels in the membrane elements. To
prevent such fouling, RO feed water needs to have turbidities of less
than 1 NTU, and Silt Density Index (SDI), a ow rate over time through
a 0.45 micron lter, of less than 4.0. Turbidity and SDI don't detect colloidal fouling potentials. To control scaling, anti-scalants are used to
bind to nascent seed crystals preventing them from growing into scales
and safely discharged with the reject water. For this reason, it is commonly called threshold inhibition mechanism. Antifoulants for controlling colloidal fouling work on the principle of keeping the colloidal
particles from coagulating once formed on the membrane [67]. Colloidal
iron and manganese compounds, due to their positively charged characteristics, are particularly sticky on the negatively charged membranes
[67]. Special measures are required particularly on the mechanism of
fouling by colloidal silica and silicates due to the spontaneous polymerization of monomeric silicic acid in all natural waters [69]. Silicic acid
[Si(OH)4] is the reactive silica species that can be detected by the molybdate colorimetric assay. It is spontaneously polymerized by elimination
of water during RO concentration, generating in the RO concentrate a
reaction mixture of oligomeric silica and silicates. The silica in the ultimate dehydrated state is found as SiO2 (e.g. sand, quartz). When hydroxides of iron, aluminum, magnesium and calcium are involved in
copolymerization with silicic acid, complex silicate oligomers are
formed in the RO concentrate, some of which depending on their size
may be deposited on the membrane surface [68].
Several studies [6,57,58,66] reported that biofouling has been one of
the severe forms of membrane fouling affecting the performance of the
RO membranes. It is caused due to the bacterial growth on the inner surfaces of the membrane pores forming a thick layer of biolm clogging
the membrane surface. Asif et al. [8] suggested that the biolm behaves
as a second membrane, thus promoting high concentration polarization,
high salt passage, and low permeate ux. However, the scientic reason
behind the rapid growth and accumulation of microbial communities
on the RO membrane surface is still not well dened. Many research
studies focused on improving the RO membrane desalination technology in conjunction with reducing membrane fouling. For instance,
Flemming and Wingender [36], and Flemming [37] investigated the
chemical composition of biolm layers in the RO membranes. Xavier

32

S. Jamaly et al. / Desalination 354 (2014) 3038

et al. [88] concluded that the use of a xed lm aerobic biolter could
reduce the natural organic matters (NOMs) in the bulk solution, therefore, enhancing the membrane performance. Furthermore, Choi et al.
[22] and Belfer et al. [11] worked on surface modications of RO membrane, while Murugan et al. [65] and Ebrahim et al. [34] focused on conventional pretreatment techniques.
One of the limitations of the RO membrane desalination is to treat
efuents with a very low concentration of suspended solids to minimize
the problems associated with membrane fouling [38]. As a result, the
efciency of the desalination process is reduced as increasing the
osmotic pressure may increase the overall energy consumption. Therefore, the need for an appropriate pretreatment method becomes inevitable to ensure the feasibility and efciency of RO systems. Several
research investigations revealed that pretreatment technologies ahead
of the RO membrane desalination could have positive effects on the
overall operational performance such as minimizing membrane fouling,
increasing the removal of suspended solids and dissolved organic matters from sea waters. These pretreatment technologies may be conventional such as coagulation, occulation and scale inhibition or nonconventional such as ultraltration (UF), microltration (MF) and
nanoltration (NF).
The selection of the pretreatment technology is of a great importance while treating various types of saline waters by using diverse pretreatment strategies. In this article, an overview of the performance of
different RO membrane desalination systems coupled with various pretreatment technologies is summarized in the subsequent sections.
2. Coupling RO with conventional technologies
Comprehensive understanding of the raw water quality and characteristics, and type of water resource (e.g. surface water, brackish water,
sea water and industrial saline water) is essential to select the appropriate pretreatment technology ahead of the RO system. For instance, surface waters have high turbidities, SDI, and NOMs as compared to water
from the well source due to adsorption and ltration effect on underground water reserves. Similarly, well waters contain high silica content
than surface waters. The initially large particles, which may be pumped
from the well, are removed from the feed water using mesh strainers or
traveling screens. Traveling screens are more useful for surface water
sources, which typically have large concentrations of biological debris.
The conventional pretreatment process may consist of all or some of
the following treatment steps:

Large particle removal by coarse strainer

Chlorination
Clarication with oatation or occulation
Hardness removal by lime treatment
Filtration
Alkalinity reduction by pH control
Scale inhibitor
Removal of free chlorine by sodium bisulte or activated carbon
UV radiation
Suspended particles removal by cartridge ltration

2.1. Chlorination
Pre-chlorination is the process of adding chlorine to the raw water
after screening and before ash mixing. The residual chlorine is useful
in several stages of the treatment processaiding in coagulation, controlling algae problems in basins, reducing odor problems, and controlling mud ball formation. In addition, chlorine has a much longer contact
time when added at the beginning of the treatment process, so prechlorination increases safety in disinfecting heavily contaminated
water. Until the middle of the 1970s, water treatment plants typically
used both pre-chlorination and post-chlorination. However, the longer
contact time provided by pre-chlorination allows chlorine to react

with organics in the water and produces carcinogenic substances

known as trihalomethanes. As a result of concerns over trihalomethanes, pre-chlorination has become much less common in the United
States. Currently, pre-chlorination is only used in plants where trihalomethane formation is not a problem. When added at the beginning of
the treatment process, pre-chlorination enhances disinfection of heavily
contaminated water [39]. Ferric chloride was tested as a coagulant;
dosed at 50 mg/L through which the UF ux decline rate has reduced
by 43%, mainly due to a 38% decrease of the organic load. After coagulation, larger particles with narrower size distribution were observed in
the feed (average size ~0.5 m) [39]. In pre-chlorination, sodium hypochlorite or chlorine dioxide is also used to control the growth of sea
organisms/microorganisms (algae, mussels, etc) growing inside the
pipes' and tanks' walls in the intake system. It is found that shock chlorination is better than continuous chlorination which promotes destabilization and more coagulation of the natural colloidal polymers as well
as irritates sea organisms in the intake system allowing their division
to add foulants [53].
2.2. Clarication/dissolved air oatation (DAF)
During DAF processing, coagulants supported on sand or other
media lters are used to remove suspended particles from sea water.
Dispersion of air bubbles is introduced extending contact time with coagulants that enhance the removal of algae and NOMs sometimes called
transparent exopolymeric particles (TEP), or extracellular polymeric
substances (EPS) [23]. Researchers have used DAF as UF pretreatment
for algal laden surface water. They found that the UF ux could be increased 70% following DAF pretreatment, reducing capital costs ([14,
15]). These are sticky gels consisting of primarily complex polysaccharides [13]. Recently, DAF has been extensively piloted in seawater applications. Extensive piloting conducted in El Coloso, Chile indicated that
3-stage occulation, DAF and 2-stage ltration was able to produce RO
feed water with SDI less than 4 (typically less than 3) over a widerange of operating parameters when treating seawater possessing
high-concentrations of algae and zoo-plankton with maximum turbidity of 2 NTU [75].
2.3. Ozonation
Studies showed that the RO desalination technology serves as the
most used technique for potable water production from seawater [18].
A major challenge to seawater reverse osmosis (SWRO) desalination is
membrane productivity decline due to fouling caused by colloidal, particulate, dissolved organic and inorganic cake formation matters in feed
water, as well as biological growth in the RO process. Research studies
have been carried out to investigate the formation of cake-causing matter by ozonation of seawater which could be attributed to the improvement in calcium complexation by ozone leading to the conversion of
dissolved organic matter into colloidal particles. The raw seawater
was pre-ltered through 100 kDa, 0.45 m and 1 m membranes to prepare the particle-free seawater. The raw pre-ltered seawater was then
ozonated with an ozone diffuser at an ozone dose rate of 1.2 mg/L min.
The residual ozone was measured with the Indigo method whereas
the total residual oxidant (TRO) was measured with the DPD method;
(N,N-diethyl-p-phenylenediamine) used to measure the chlorine residual in water by titrating or comparing a color with standards color
[50,71].
Excessive ozonation causes foam fractionators and several studies
have been conducted to evaluate their effect on solids removal
(suspended solids (SS), volatile suspended solids (VSS), dissolved
organic carbon (DOC)). For example, three separate but identical
recirculating systems (4.5 m3 system volume) with foam fractionators
(300 mm in diameter, 3 m in height) were used during a 44 day experimental period [71]. One system (control: CS) without ozone, while the
other two systems were ozonated at a rate of either 20 g ozone/d (T 20)

S. Jamaly et al. / Desalination 354 (2014) 3038

or 40 g ozone/d (T 40) per kg of feed applied, respectively. A total of

107 kg of black sea bream Acanthopagrus schlegeli (Bleeker) with an
average weight of 0.34 kg was stocked into each system. Daily feeding rate was 1% of total body weight. The overall mean particle diameter of solids in the foam decreased as ozonation increased. As the
number of the bacteria in the inlet decreased, the bacteria in the
foam and the removal rate greatly decreased with increasing ozonation [71].

2.4. Coagulationocculation
The conventional technologies (coagulationocculation, disinfection, pH adjustment, scale inhibition, and ltration with granular
media) can be applied as pretreatment technologies to RO membrane
systems. Coagulation has been shown to be a successful method for
improving the water quality not only in conventional pretreatment
technologies, but also in low pressure membrane pretreatment technologies [40]. A study conducted by Duan et al. [32] deduced that the use of
powdered activated carbon (PAC) offers better efciency for the humic
acid adsorption in saline water compared to fresh water as well as low
conductivity water. It was found that the removal of humic acids
depends mainly on the coagulant dose, pH, and the order of the PAC
and metal salt coagulant addition when either aluminum sulfate or
ferric chloride was added. Their results showed that the addition of
PAC before the coagulant gives better humic removal. Gabelich et al.
[40] discussed the negative effects of the coagulant residuals that result
from the pretreatment process on the performance of the RO
membrane such as aluminum sulfate, ferric chloride coagulants, and
chloramines. The experimental dose of 68 mg/L as alum was used
with 1.0l.5 mg/L polydimethyl diallylammonium chloride (polyDADMAC) cationic polymer (Ageoc WT-20; CPS Chemical Co., Old
Bridge, NJ). Ferric chloride dosed in the range of 45 mg/L (as FeCl3)
with 1.52.0 mg/L cationic polymer. A free-chlorine residual of
2.53.5 mg/L was maintained at the lter efuent while directltration modules have been used from ltration plant. Their results
showed that the use of aluminum sulfate with multiple RO elements resulted in a rapid decline in specic ux that reached up to 60% over
around 100 h of operation, as well as a clear decrease in salt rejection,
whereas an increase in the specic ux was noticed using ferric chloride
and chloramines over time. However, the salt rejection reduced signicantly during the experiment.
On the other hand, coagulation can be used efciently for arsenic
removal. Iron based coagulants found to be better compared with the
aluminum based coagulants. The iron coagulants are hydrolyzed once
they are added to water in order to form ferric hydroxide with a positive
charge. However, the net positive charge is a strong function of pH; thus
as the pH decreases, the net number of the positively charged sited on
the surface of the ferric hydroxide elements increases. Thus, the arsenic
removal is usually enhanced at pH values that are less than 7. Moreover,
for the aluminum coagulants the best results were attained at a pH
value of 5.5, and the arsenic removal was in the range of 5999% at
dosages of 0.8 to 1.9 mg/L as aluminum, respectively. Whereas, in ferric
coagulants, the study showed that best results were obtained at a pH
value of 5.5, and the arsenic removal was in the range of 70 to 99.6%
at dosages of 1.7 to 3.8 mg/L as iron, respectively [86]. Den and Wang
[28] investigated the feasibility of electrocoagulation, as a pretreatment,
to remove silica. They found that the bipolar conguration offered
better silica removal than monopolar conguration of electrochemical
coagulation setup. Moreover, membrane fouling of the pretreated
seawater, and modied fouling index (MFI) with ultraltration (UF)
was investigated in terms of molecular weight distribution (MWD)
and membrane characterization. MFI values after pretreatments of
FeCl3 occulation and PAC adsorption signicantly decreased to
6900 and 6700, and to 2300 and 2500 s/L2 for 30 and 100 kDa UF,
respectively.

33

2.5. Scale inhibitors

Scale inhibitors are useful conventional pretreatment methods that
can be used to control scaling, thus, improve the performance of RO
membrane. Various types of antiscalants exist such as polyacrylates,
organophosphonates, and sodium hexametaphosphate (SHMP).
There're certain merits as well as demerits for using phosphate based
anti-scalants. Sweity et al. [80] showed despite the fact that the scale
formation in the RO desalination systems can be reduced using the
antiscalants, they can enhance the formation of biolm on the surface
of the RO membranes. It has been demonstrated that the use of
polyacrylate-based anti-scalant enhanced the formation of biolm by
changing the physico-chemical properties of the RO membranes including the hydrophobicity as well as the surface charge that stimulated the
deposition and attachment of the bacterial cells. Sweity et al. [80] further emphasized that the polyphosphate-based anti-scalants enhance
the membrane biofouling by acting as a phosphorous source of nutrients. Thus, anti-scalants should be screened for their biofouling impact
and the associated enhancing steps. One study showed that although
the SHMP is cheap, it wasn't suitable when compared to the polymeric
organic antiscalants [31]. Slight amounts were adsorbed on the surface
of the micro crystals, inhibiting additional growth of the crystals. Hydrolysis of SHMP should be avoided in the dosing feed tank given that it
won't only reduce the efciency of the scale inhibition, but also will generate the calcium phosphate scaling risk. On the other hand, organophosphates act as anti-foulants for insoluble aluminum and iron
which make them more stable and effective when compared to SHMP.
In spite of the fact that coagulation can be used for reducing fouling
to UF and MF membranes as well as enhancing their operational stability, many issues should be considered such as proper selection of coagulant for certain water quality, optimal dosage, and overall treatment
cost. In other studies, for instance, Ma et al. [62] reported that some materials such as microbial contamination, silica, and iron can be reduced
in the feed water through the use of enhanced coagulation with Fe
(VI) and UF membrane treatment process considering the fact that
bio-mineralization is caused by the microorganisms' metabolization at
iron and silica. Results showed that the turbidity is lower than 0.5
NTU, while iron and silica concentrations were 0.2 and 0.1 mg/L, respectively. The rate of algae and microbial removal was more than 98% by
enhanced bio-mineralization with 5 quantitative unit doses.
3. Reverse osmosis with non-conventional
pretreatment technologies
Membranes with pore sizes of 5 mm or greater are particulate
lters. Microltration membranes have pore sizes in the range of
1005000 nm and are capable of removing suspended particles like
blood cells and latex emulsions. Ultraltration membranes have
pore sizes in the range of 10100 nm and can remove large
molecules like albumin or pepsin within this range [17,52,74].
Nanoltration membranes can separate small molecules like dissolved salts, dissociated acids and sugar, and have pore sizes in the
range of 12 nm. RO membranes separate ions like sodium and chloride on the molecular level and have pore sizes in the range of few
angstroms. Non-porous membranes are used for gas separation,
vapor permeation, and evaporation.
3.1. Ultraltration (UF) and microltration (MF)
Several studies investigated the efciency of applying UF and MF
pretreatments in RO membrane technology [5,43]. Conventional pretreatment technologies require higher chemical additive doses, larger
footprint, less quality of the produced efuent, larger manpower, and
higher chemical and operating costs [10,33]. SDI, turbidity, and MFI
are used to assess the quality of the feed water to the RO membranes.
Some investigations were done using UF pretreatment to RO in the

34

S. Jamaly et al. / Desalination 354 (2014) 3038

Red Sea in Saudi Arabia [72]. The inlet feed turbidity to UF was found to
be in average of 4.5 NTU, and the SDI is 2.2 in the UF permeate water.
The addition of FeCl3 coagulant in phase 1 by ferric dosing of 2 ppm
and 0.7 ppm at pH 67 and a minimum contact time of 20 min maintained stable membrane permeability, consistent permeate water, and
SDI less than 3 [43,72].
The dual system UFRO was used in seawater treatment using Wang
Tan Power Plant; the permeate water quality results after UF pretreatment were: NTU N 1, SDI b 3, and 95% recovery [17]. Also, the RO membrane cleaning frequency was reduced as the water ux was between
9.5 and 16.5 LMH. The average pressure drop maintained 1.7 bar
while 180 ppm of efuent TDS was reported. Van Hoof et al. [83] introduced UFRO system for wastewater reuse related applications. Their
results showed that SDI of UF membrane was 1.8; yet operated at
2225 LMH water ux while recovering 70% of produced water.
UF pretreatment technology can be improved by adding coagulants and adsorbents. Whereas a backwashing strategy has been
employed to control different types of membrane fouling arising
from these coagulants, such as chemical enhanced backwashing
(CEB) with acid or NaOCl used to control ferric membrane fouling
and caustic CEB to control fouling due to algae. Air enhanced
backwashing can help reduce the particulate accumulation in
ber lumen based membranes, if frequency increased from once a
week to once a day [72]. YoungHong et al. [90] tested the impact
of chemical coagulation on UF membrane. Kaolinite and humic
acid (HA) were used to simulate the particles and NOM present in
source water. The results revealed that the particles and NOM compounds present in source water may have different fouling behavior, and may mitigate the irreversible fouling caused by humic acid.
The addition of coagulant can lead to a higher rate of removal of
large-sized hydrophobic compounds. Dong et al. [30] investigated
the impact and mechanism of preventing membrane fouling, by coagulation pretreatment, in terms of fractional component and molecular weight of NOM. They concluded that the coagulated ocs
could absorb neutral hydrophilic compounds resulting in an
increase in the operating ux. In addition, the application of adsorption technology has shown signicant improvements on the
operation of RO systems. Several studies proved its importance by
reducing humic acids, fulvic acids, low molecular weights matters
(LMW), and other organic matters.
Chatkaew et al. [21] determined that coupling PAC with UF membrane is a relevant pretreatment to the RO system because it removed
70% of organic matters and reduced the ux decline. Other research
study reported that backwashing and the combined PAC/UF system
were very effective in reducing a ux decline and producing a good
quality efuent when the dosage of PAC was below 20 mg/L [81].
Comparing UF with conventional pretreatment techniques, less space
can bring 30% cost saving as well as the need for RO disinfection/cleaning
to avoid considerable production loss [72]. MF pretreatment offers more
signicant improvements on the performance of the RO desalination systems. For example, Herzberg et al. [49] analyzed the effect of MF pretreatment on biofouling of RO membrane by treating secondary wastewater
efuent. They found that the salt rejection increased to 98.298.8% comparing it to RO without pretreatment (94.397%). Moreover, with MF
pretreatment, the permeate ux decline and particulate and colloidal
matter were reduced in RO while the oxygen uptake rate has increased.
Chakravorty and Layson [20] concluded that the use of polypropylene
membrane having 0.2 m pore diameter in continuous mode under a
pressure gradient of 1 bar resulted in 40% increase in water ux over traditional pretreatment methods. The study carried by Cardona et al. [19]
deduced that the energy saving in the order of 1315% has been reported.
Ahmad and Mariadas [2] investigated the impact of using tubular single
channel ceramic membranes with a nominal pore size of 0.2 m. They
found that the insertion of helical bafe increased the permeate ux
(520.8 LMH) up to 104.9% as compared to a system without bafe
using feed of 1.0 g/L TiO2 at 1.4 bar TMP.

Dey et al. [29] described the operational process at the pilot scale on
the Arabian coastal sea water using tubular ceramic membrane (average pore diameter of 0.1 m) of 19-channel conguration as pretreatment for RO system. Their results showed an increase in the permeate
ux between 350 and 370 LMH, a decrease in turbidity and SDI to 1.0
and b3, respectively while operating at xed TMP of 1.2 bar with 40 h
continuous run, using ltrate water of pressure sand lter (PSF) as
feed. Corral et al. [25] carried out a comparative study between MF
and slow sand ltration (SSF) as pretreatments prior to RO system.
They concluded that MF steadily provided ltrate with SDI b 3 while improving the long-term RO performance. Despite the economic costs of
MF and SSF pretreatments are similar; MF is preferred based on the
quality of treated water and stability of downstream RO operation.
The potential benets offered by membrane pretreatment compared
to conventional pretreatments can be summarized below:
Lower suspended solids and less biological content, resulting in improved RO operation
RO membrane cleaning cost savings in cleaning chemicals
Lower RO pressure drops from fouling, resulting in lower energy costs
Longer RO membrane life
Increased ux rates in the RO system
Shorter plant footprint size resulting in reduced capital investment
Lower chemical and sludge handling costs
3.2. Nanoltration (NF) and nanostructured membranes
Several studies have shown the evolving sequence of nanoltration
(NF) as a pretreatment stage in desalination industry; moving from
pilot to commercial/industrial scale offering an appropriate operational
technology. An earlier study carried by Al-So et al. [7] explored the
brackish water softening NF technique as a permeation pretreatment
of feed to seawater reverse osmosis (SWRO) and multi stage ash
(MSF). They concluded that total hardness was reduced by 86.5%
while chloride (CI) ion falls down from 22,780 ppm in feed seawater
to 16,692 ppm in NF permeate. Also, 26.7% reduction of Na+ and K+
ions was reported. Moreover, researchers [45,46] have demonstrated
that NF can be successfully incorporated prior to SWRO resulting in an
improved desalination efciency, reduced fouling, less frequent
cleaning of RO membrane. 6 NF spiral wound membrane elements of
size 8 40 in series connected by a SWRO unit of 3 HFF SWRO
elements 8 40 or 9 40 all in series were used. This conguration
has doubled the SWRO product water output and recovery ratio, while
producing high purity permeate (TDS b 200 ppm) during a single
stage SWRO. Their results revealed that NF can be integrated not only
at a pilot level, but also at industrial scale operations. A group of scientists [4] studied a new conguration of a NF as a pretreatment to
SWRO system at industrial unit, where in single SWRO desalination
plant was converted to a dual NFSWRO desalination process by introducing NF membrane pretreatment ahead of the SWRO desalination.
Their results showed that the permeate ow has increased signicantly
from 91.8 to 130 m3/h in a dual NFSWRO process. Recent studies [91]
provided many operational improvements using NFSWRO system.
These included 65% water recovery at pH = 6 with a low feed
pressure of b 25 bar, and 42% increase in production rate. In an effort
to avoid cleaning, NF unit was operated in Gulf seawater at ux of 12
gfd along with occasional ushing using pretreated seawater on monthly intervals without chemical cleaning up to two years.
Furthermore, modern technology has enabled exceptional control
over the fabrication of nanostructured materials, particularly, the capability to create well-dened, size-selective, nanostructured ltration
membranes. By enabling the manipulation of matter and control of
transport at nanometer length scales, the emergence of nanotechnology
offers new opportunities to advance water desalination technologies.
Size-selective membrane with pore sizes in the sub-nanometer range
are expected to allow water molecules to pass through, while ceasing

S. Jamaly et al. / Desalination 354 (2014) 3038

the ions that have a larger effective diameter due to their hydration
shells [44]. For example, the pore could act as a molecular sieve if the
pore diameter is smaller than that of a solvated ion (diameter of a
hydrated sodium ion is 7.6 A), yet larger than a water molecule. As
considerably high energy barrier is to strip the ion of its solvation shell
( 1709 kJ/mol for Na+) [78]; the applied pressure should be greater
than the osmotic pressure on the feed water to force the water molecules through these pores while ceasing the passage of ions. In addition
to steric exclusion, electrostatic and van der Waals interactions may be
used to achieve the desired selectivity.
Nanostructured materials that are promising for desalination
include zeolites [60,61]. They're aluminosilicate minerals with a microstructure composed of 38 A pores. Li et al. [60] used hydrothermal
synthesis to develop 0.53 m thick membranes consisting of hydrophobic MFI (mordenite framework inverted) type zeolites with an
average pore diameter of 5.6 A on a porous -alumina support. Their
results showed that those membranes have rejected 76% of Na+ ions
while permitting a water ux of 0.112 kg/m2 h. Carbon nanotubes [51]
and graphene [9,56,77] can be synthesized to have non-twisting pores
of the order of 1 nm or less and can be fabricated into macroscopic arrays. In addition to its electronic properties, graphene exhibits a high
breaking strength and impermeability to molecules as small as helium
in its pristine state [52]. These properties suggest that graphene has a
great potential to create ultrathin high ux membranes with sizetunable pores that can act as molecular sieves.
The NF membrane was able to completely remove nitrate ions; thus,
being coupled with RO would increase the production rate at high quality [12]. The advantages of NF membrane pretreatment, and to the overall system design fall into two broad categories, including overall
treatment cost reduction while maintaining the integrity of water supplies. The benets can be summarized as follows [73,74]:

Higher RO design ux and recovery may be possible

RO membrane replacement reduced signicantly
Treatment of surface water, with poor and/or variable quality
Reduced requirement for RO disinfection and cleaning

The capital cost of the membrane pre-treatment normally exceeds

that of conventional pre-treatment by a signicant amount. The membranes racks or assemblies are normally about twice as expensive as
the dual media lters, and the overall membrane system can exceed
the conventional system by 2050% depending on ow rate, feed quality, and local factors. If two stages of media ltration are needed, the cost
can come closer, though lower membrane uxes may be used in this
case. Although membranes provided a barrier to particles, colloids,
and most microorganisms, dissolved organics can still pass through
the membrane. Organics may be a problem to the RO/NF, since they
can cause fouling due to surface adsorption, or they may become a
food source to microorganisms.
3.2. Potential use of MBRs as RO pretreatment
The experimental results showed that the MBR pretreatment resulted in less RO membrane fouling when compared with conventional
activated sludge with tertiary membrane ltration pretreatment (CASTMF) [79]. Dukes and von Gottberg [33] and Lerner et al. [59] evaluated
the performance of MBR in the removal of chemical oxygen demand
(COD) and biological oxygen demand (BOD) so as to minimize the impact of membrane biofouling on RO membranes. According to Dukes
and von Gottberg [33], low concentration of phosphate (0.1 ppm) was
obtained via MBR pretreatment, and therefore, limiting calcium phosphate scaling in RO systems.
Although the application of the MBR as a pretreatment method
could be efcient in reducing RO membrane fouling; yet the operating
conditions of the MBR would have signicant impacts on the fouling
mechanism and cleaning frequency in RO systems. For instance, Grelier
et al. [42] analyzed two parallel MBR pretreatments with two different

35

ratios of food to microorganisms (F/M): 0.5 and 0.17 g/g.d. Their results
showed that RO membrane fouling rate was 4 times higher in the MBR
pretreatment that was fed with 0.5 g/g.d. Sludge retention time (SRT)
and hydraulic retention time (HRT) would also affect the performance
of the RO membranes. Grelier et al. [42] and Van den Broecka et al.
[82] concluded that RO membranes were susceptible to fouling at SRT
of 8 days. Ahmed et al. [3] investigated the effect of SRT on membrane
biofouling and microbial community in MBRs. Four sequential anoxic/
anaerobic membrane bioreactors at SRT of 20, 40, 60, 80, and 100 days
were operated. Their results indicated that the bound extra polymeric
substances (bound EPSs) values increased at short SRT (20 d) compared
to long SRT (60100 d); thus, the increased biofouling in MBR subsequently affects the higher RO fouling rates. Furthermore, the ltration
mechanism within an MBR would contribute to the quality of water
produced by the RO membrane systems. The submerged hollow ber
membrane systems (SMBR) using MF or UF membranes inside the bioreactor have demonstrated great potential in producing good quality of
permeate water product [35,89]. A research study conducted by Ye et al.
[89] showed that the SMBR using UF membrane was able to remove 60%
of biopolymers while low removal efciency of low molecular organics
and humics was reported.
Other researchers considered the impact of adsorption and coagulation processes in conjunction with membrane ltration in SMBRs used
as pretreatment to RO membrane systems. Jeong et al. [55] compared
the performance of two SMBRs: 1) adding 1.5 g/L of PAC adsorbents,
and 2) adding 1.5 g/L of PAC adsorbents and 0.51.0 mg/L of FeCl3 as a
coagulation agent (0.5 to 1 mg/l of Fe3+) in the removal of microorganisms and organic foulants while treating raw seawater. Their results
revealed that SMBR with PAC adsorption has removed 76.6% of dissolved organic carbon (DOC), 92.3% of biopolymer, 70% of humics,
89.5% of building blocks, and 88.9% of neutrals, whereas the SMBR
with the addition of PAC and 0.5 mg/L of FeCl3 has removed 83.9% of
DOC, 100% of biopolymer, 89% of humics, 92.5% of building blocks, and
87.8% of neutrals. Biopolymers were the major foulants detected on
the RO membrane surface. Nevertheless, 16.7 and 14.8 g/cm2 were
reported on the RO membrane surface after the pretreatment with
PAC and PAC/FeCl3, respectively when compared to initial 23.5 g/cm2
before pretreatment.
4. Economical aspects of RO with pretreatment technologies
Extensive attempts were made to evaluate the costs of construction
and maintenance of RO desalination plants. These methods ranged from
the use of empirical estimation according to the experience and opinion
of the experts, to sophisticated predictions through process simulations
based on material balances. Moreover, the economics of desalinating
water using RO has been constantly improving, due to the enhancements of the RO membrane technology. In spite of the fact that the
prices of seawater and brackish water membrane elements are slightly
different, the desalting cost of seawater RO is signicantly higher than
the cost of brackish water RO. However, the production of drinking
water from seawater is considered an affordable alternative when
there is no reliable fresh water sources are available. The RO factor
that has the main impact on both investment and operating cost is the
recovery rate of permeate. The RO feed ow is inversely related to the
recovery rate of the design. Consequently, the recovery rate directly impacts the size as well as the cost of all the equipment and power consumption. Yet, in RO systems, the recovery rate cannot exceed certain
limit, due to the fact that higher recovery rates will lead to higher feed
salinity that will result in higher osmotic pressure and consequently increase the permeate salinity [87]. In California, many water agencies
have embarked on exploring seawater desalination because of the
diminishing capacities of fresh surface and ground water. Most of the
water utilities in Southern California currently purchase imported
water from the Bay Delta and Colorado River at a rate of US $2.30 to
$2.45/1,000 gallons ($750 to $800/AF), and the cost of these water

36

S. Jamaly et al. / Desalination 354 (2014) 3038

supplies is very likely to increase by 15% or more through 2015 due to

additional expenditures needed to comply with more stringent drinking
water quality regulatory requirements promulgated by the US EPA
(Fig. 1).
Studies were carried out to evaluate the total cost of UF as pretreatment to RO technology. Some of them found that an UFRO system is
economically unfeasible [24,26,64] while others weren't in line [48].
The operating cost of RO can be increased due to the frequent replacement rate of its membranes. However, pretreatment by low-pressure
membrane ltration was proven to improve the cost of the RO in the
long-term as it increases the lifespan of the RO membrane by 2030%.
Voutchkov [84] showed that the use of UF and MF membrane pretreatment for RO systems signicantly lowered the SDI values of the feed
signicantly, hence less RO membranes biofouling potential. Both the
RO and the low pressure membranes need occasional chemical
cleaning. Pearce [74] examined the effect of UF/MF pretreatment on
the chemical cleaning cost of a RO plant. It was concluded that, by the
use of UF/MF pretreatment, the basic cleaning frequency would be
reduced to two or one cleaning per year. Nevertheless, conventional
pretreatment systems might be very attractive in terms of low energy
cost than non-conventional pretreatment systems [54]. Gleuckstern
and Priel [41] compared the treatment cost of the 90,000 m3/d RO sea
water desalination plant via using: 1) conventional and 2) UF as pretreatment technologies. Table 1 demonstrates the comparative cost
with respect to capital, operational and maintenance, energy, chemical,
and the total variable operational costs.
The conventional pretreatment systems have lower total cost, labor
cost, and unit cost. However, the performance of RO in producing a good
quality of efuent with conventional pretreatment is still limited. To
summarize, Table 2 demonstrates the capital and efuent costs of the
pretreatment systems based on salinity ranges.

Table 1
Comparison of water cost of UF and conventional pretreatment for the 90,000 m3/d RO
desalination plants [41].

5. Conclusions and recommendations

This article summarized the performance of RO coupled to various

pretreatment technologies to develop efcient RO technologies. The
lower investment and lower unit water cost will favor the conventional
pretreatment option but for sites with limited in space and high salinity
seawater will pull the choice bar down to membrane technologies (UF/
NF).The performance of RO is summarized according to every pretreatment as follows:
The applied pretreatments can help in extending the lifetime of RO
membranes, increasing its performance, producing better product

Filtration method

Capital cost
Fixed O&M cost
Energy cost
Chemical cost
Total variable operational cost
Total cost

Conventional

UF

$/m3

$/m3

0.22
0.07
0.16
0.05
0.22
0.51

0.23
0.09
0.16
0.03
0.20
0.52

water quality, and minimizing chemical cleaning frequency. The

non-conventional pretreatment is going to improve the operational
cost because of low chemical additives to saline feed water and low
operational maintenance of the system.
To produce a good quality of product water over a wide range of salinity (N35,000 ppm), membrane pretreatments (UF/NF) can be more
efcient with RO systems than the conventional one. It can effectively
reduce organic/biofouling and unit water cost as low as (0.05 $/m3).
MF is a known pretreatment technique to be coupled with RO systems; however, the pore size ranging between 0.1 and 10 m minimizes its efciency in overcoming biofouling (microbes) and TDS
removal.
Nanostructured (NF) membranes may also be used as pretreatment
which offers unique pore size and chemical selectivity features.
MBR technology, although considered as main stream treatment process, has the potential to be used as RO pretreatment. It can efciently
reduce biofouling in RO system in low saline waters; however, it can
be also applied in high saline water in conjunction with other conventional or non-conventional pretreatment technologies.
Conventional pretreatment operational cost is lower than nonconventional systems for less saline feed water. The quality of the produced water by treating seawater in RO membrane with conventional
pretreatment is lower than membrane pretreatment systems.
The different pretreatment systems cost is relevant with the type and
quality of feed water. In general, treating high saline water is more expensive than low saline water using any pretreatment systems. The
major part of a SWRO plant cost is expended in infrastructure as
well as energy.
Although organics may cause fouling in NF due to surface adsorption
as nutrients, molecular sieves of sub nanometer size, non-twisting

Fig. 1. SWRO plant construction cost breakdown [85].

S. Jamaly et al. / Desalination 354 (2014) 3038

37

Table 2
Comparison of capital cost, produced water cost, energy consumption, and the salinity range of the pretreatment systems.
Pretreatment

Salinity range
(ppm)

Capital cost

Energy consumption

Produced water cost

Reference

MF and UF

1500
35,000

The higher xed cost, 1.92 Cent/m3, is

associated with the lower energy and
chemical costs (0.02 Cent/m3 and
2.12 Cent/m3).

Around $40/m2, with an

overall area requirement of
24,000 m2, and a 7 year
membrane life

[1,41,74]

Conventional
systems

1500
35,000

22.13 (Cent/m3)

The O&M unit costs of MF are lower than UF, probably

because of lower membrane replacement costs and lower
energy consumption. The approximate OSM unit costs for
MF and UF processes in plants with a capacity of 1 MGD are
about $0.20/1000 gal and $0.43/1000 gal, respectively.
16.06 (Cent/m3)

pores and size selective ion/TDS removal capability of NF has made it a

better prospective method of pretreatment than UF, as it allows water
molecules to pass through, while ceasing the hydrated ions of a larger
effective diameter. Its ion selective features are expected to dominate
the desalination membrane technology having the potential of becoming the most cost effective technique in near future.
Apart from the above extracted conclusions to design an efcient desalination process, some core and crucial recommendations are to be
considered. These include:
Researchers need to focus on the greener additives and sustainable
pretreatment technologies with least additives relevant to the quality
of feed water to improve the efciency of RO membranes.
Apart from search for nanostructured membrane systems, feed water
may be exploited as source for aerobic and anaerobic bacteria to grow
and spread quickly keeping in mind enhanced bioremediation without interrupting the biodiversity of the aquifer.
Development of novel nano-based/structured membranes which contain molecular sieves of sub nanometer size, non-twisting pores and
size selective ion (TDS) removal capability.
The working principle of system should preferably be based on 3R's
(Reduce, Reuse, Recycle) of waste management, particularly for brine
waste produced.

References
[1] S. Adham, J. Jacangelo, J. Laine, Characteristics and costs of MF and UF plants, J. Am.
Water Works Assoc. 88 (1996) 2231.
[2] A.L. Ahmad, A. Mariadas, Bafed microltration membrane and its fouling control
for feed water of desalination, Desalination 168 (2004) 223230.
[3] Z. Ahmed, J. Cho, B.R. Lim, K.G. Song, K.H. Ahn, Effects of sludge retention time on
membrane fouling and microbial community structure in a membrane bioreactor,
J. Membr. Sci. 287 (2007) 211218.
[4] A.S. Al-Amoudi, A.M. Farooque, Performance restoration and autopsy of NF membranes used in seawater pretreatment, Desalination 178 (2005) 261271.
[5] A. Alhadidi, A.J.B. Kemperman, R. Schurer, J.C. Schippers, M. Wessling, W.G.J. Van der
Meer, Using SDI, SDI + and MFI to evaluate fouling in a UF/RO desalination pilot
plant, Desalination 285 (2012) 153162.
[6] R.A. Al-Juboori, T. Yusaf, Biofouling in RO system: mechanisms, monitoring, and controlling, Desalination 302 (2012) 123.
[7] M.A.K. Al-So, A.M. Hassan, G.M. Mustafa, A.-G.I. Dalvi, M.N.M. Kither, Nanoltration
as a means of achieving higher TBT of 120 C in MSF, Desalination 118 (1998)
123129.
[8] M. Asif, Z. Khan, S.M.J. Zaidi, M.C. Boyce, Biofouling in reverse osmosis membranes
for seawater desalination: phenomena and prevention, Desalination 281 (2011)
116.
[9] J. Bai, Graphene nanomesh, Nat. Nanotechnol. 5 (2010) 190194.
[10] W.T. Bates, Capillary UF as RO pretreatment, International Water Conference, Pittsburgh, USA, 1999.
[11] S. Belfer, Y. Purinson, R. Fainshtein, Y. Radchenko, O. Kedem, Surface modication of
commercial composite polyamide reverse osmosis membranes, J. Membr. Sci. 139
(1998) 175181.
[12] J. Bohdziewicz, M. Bodzek, E. Wasik, The application of reverse osmosis and
nanoltration on the removal of nitrates from groundwater, Desalination 121
(1999) 139147.
[13] V. Bonnelye, M.A. Sanz, J.P. Durand, L. Plasse, F. Gueguen, P. Mazounie, Reverse osmosis on open intake seawater: pre-treatment strategy, Desalination 167 (2004)
191200.
[14] A. Braghetta, F.A. DiGiano, W.P. Ball, NOM accumulation at NF membrane surface:
impact of chemistry and shear, J. Environ. Eng. ASCE 124 (1998) 10871098.

[22]

[15] A. Braghetta, J. Jacangelo, S. Chellam, M. Hotaling, B. Utne, Impact of dissolved air otation pretreatment on the performance of hollow-ber direct ow microltration,
J. AWWA 89 (1997) 1180.
[16] D.T. Bray, Reverse osmosis purication apparatus, U.S. Patent 3,417,870, December
24, 1968.
[17] M. Busch, R. Chu, U. Kolbe, Q.Q. Meng, L. SiJi, Ultraltartionpretreatment to reverse
osmosis for seawater desalinationthree years eld experience in the Wangtan
Datang power plant, Desalin. Water Treat. 10 (2009) 120.
[18] O. ByungSoo, J. HaYounh, C. JaeWeon, L. Syngyun, L. Eunkyung, I.S. Kim, H. TaeMun,
K. JoonWun, Effect of ozone on microltration as a pretreatment of seawater reverse
osmosis, Desalination 238 (2009) 9097.
[19] E. Cardona, A. Piacentino, F. Marchese, Energy saving in two-stage reverse osmosis
systems coupled with ultraltration processes, Desalination 184 (2005) 125137.
[20] B. Chakravorty, A. Layson, Ideal feed pretreatment for reverse osmosis by continuous microltration, Desalination 110 (1997) 143150.
[21] T. Chatkaew, L. Stephanie, C. Corinne, Adsorption combined with ultraltration to
remove organic matter from seawater, Water Res. 45 (2011) 63626370.
[22] H. Choi, J. Park, T. Tak, Y.N. Kwon, Surface modication of seawater reverse osmosis
(SWRO) membrane methacrylate-hydroxy poly(oxyethylene) methacrylate (MMAHPOEM) comb-polymer and its performance, Desalination 291 (2012) 17.
[23] C. Cleveland, D. Hugaboom, B. Raczko, W. Moughamian, DAF pretreatment for ultraltration: cost and water quality implications, Proceedings of the 2002 Biannual
AMTA Conference, 2002.
[24] W. Clunie, L. VandeVenter, S. Williams, E. Garana, Comparison of pre-treatment alternatives for seawatr reverse osmosis, Proceedings AWWA Membrane Technology
Conference, Phoenix, USA, 2005.
[25] A.F. Corral, Y. Umur, S. Roy, Y. Dongxu, H. Eric, H. Chris, P.E. Wendell, G.A. Robert,
Comparison of slow sand ltration and microltration as pretreatments for inland
desalination via reverse osmosis, Desalination 334 (2014) 19.
[26] P. Cote, S. Siverns, S. Monti, Comparison of membrane-based solutions for water reclamation and desalination, Desalination 182 (2005) 245251.
[27] E. Delyannis, Book Solar Energy, 75, Elsevier Ltd, 2003. 357366.
[28] W. Den, C. Wang, Removal of silica from brackish water by electrocoagulation pretreatment to prevent fouling of reverse osmosis membranes, Sep. Purif. Technol. 59
(2008) 318325.
[29] T. Dey, G.C. Sahoo, N.R. Somendra, B. Sibdas, Studies on pre-treatment of seawater
using tubular ceramic MF membrane of 19-channel conguration, Int. J. Sci. Res.
Public 3 (2013) 22503153.
[30] B.Z. Dong, Y. Chen, N.Y. Gao, J.C. Fan, Effect of coagulation pretreatment on the fouling of ultraltration membrane, J. Environ. Sci. 19 (2007) 278283.
[31] Dow Liquid Separations FILMTEC, Reverse Osmosis Membranes Technical Manual,
2004.
[32] J. Duan, F. Wilson, N. Graham, J.H. Tay, Adsorption of humic acid by powdered activated carbon in saline water conditions, Desalination 151 (2003) 5366.
[33] S. Dukes, A. von Gottberg, Membrane Bioreactors for RO Pretreatment, Water Environment Foundation, Wilmington, 2006.
[34] S.H. Ebrahim, M.M. Abdel Jawad, M. Safar, Conventional pretreatment system for the
Doha Reverse Osmosis Plant: technical and economic assessment, Desalination 102
(1995) 179187.
[35] 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.
[36] H.C. Flemming, J. Wingender, The biolm matrix, Nat. Rev. Microbiol. 8 (2010)
623633.
[37] H.C. Flemming, Microbial biofouling: Unsolved problems, insufcient approaches,
and possible solutions, 5, Springer, 2011. 81109.
[38] C.K. Fraser, F. Khosrow, M. Basuvaraj, J. Magdalena, N.L. Steven, Z. Hongde,
Water reclamation using reverse osmosis: analysis of fouling propagation
given tertiary membrane ltration and MBR pretreatments, J. Membr. Sci. 382
(2011) 328338.
[39] E. Friedler, I. Katz, C.G. Dosoretz, Chlorination and coagulation as pretreatments for
grey water desalination, Desalination 222 (2008) 3849.
[40] C.J. Gabelich, T.L. Yun, B.M. Coffey, I.H.M. Sffet, Effects of aluminum sulfate and ferric
chloride coagulant residuals on polyamide membrane performance, Desalination
150 (2002) 1530.
[41] P. Gleuckstern, M. Priel, Comparitive cost of UF vs conventional pretreatment for
SWRO systems, 5th IDS Annual Conference, Haifa, 2002.
[42] P. Grelier, S. Rosenberger, A. Tazi-Pain, Inuence of sludge retention time on membrane bioreactor hydraulic performance, Desalination 192 (2006) 1017.
[43] D.F. Halpern, J. McArdle, B. Antrim, UF pretreatment for SWRO: pilot studies, Desalination 182 (2005) 323332.

38

S. Jamaly et al. / Desalination 354 (2014) 3038

[44] J. Han, J. Fu, R.B. Schoch, Molecular sieving using nanolters: past, present and future, Lab Chip 8 (2008) 2333.
[45] A.M. Hassan, A.M. Farooque, A.S. Al-Amodi, Autopsy and characterization of NF
membranes after long term operation in an NFSWRO Pilot plant A.S. Presented
at IDA World Congress on Desalination and Water Reuse, San Diego, CA, USA,
1999.
[46] A.M. Hassan, A.M. Farooque, A.T.M. Jamaluddin, A.S. Al-Amoudi, M.A.K. Al-So, A.F.
Al-Rubaian, N.M. Kither, I.A.R. Al-Tisan, A.A. Rowaili, Demonstration plant based
on the new NFSWRO process, Desalination 131 (2000) 157171.
[47] G.L. Hassler, Report: the sea as a source of fresh water, UCLA Dept. of Engineering
Research Summary1950. 1150.
[48] L. Henthorne, Economic evaluation of membrane and conventional SWRO pretreatmentresults from pilot study, Proceedings AWWA Membrane Technology Conference. Phoenix, USA, 2005.
[49] M. Herzberg, D. Berry, L. Raskin, Impact of microltration treatment of secondary
wastewater efuent on biogouling of a reverse osmosis membranes, Water Res.
44 (2010) 167176.
[50] E.M.V. Hoek, A.S. Kim, M. Elimelech, Inuence of cross-ow membrane lter geometry on colloidal fouling of reverse osmosis membrane, Environ. Eng. Sci. 19 (2002)
357.
[51] J.K. Holt, G.H. Park, Y. Wang, M. Stadermann, A.B. Artyukhin, C.P. Grigoropoulos, A.
Noy, O. Bakajin, Fast mass transport through sub-2 nm carbon nanotubes, Science
312 (2006) 10341037.
[52] T. Humplik, J. Lee, S.C. O'Hern, B.A. Fellman, M.A. Baig, S.F. Hassan, M.A. Atieh, F.
Rahman, T. Laoui, R. Karnik, E.N. Wang, Nanostructured materials for water desalination, Nanotechnology 22 (292001) (2011) 119.
[53] R.A. Ingham, L. Mansour, T. Qadan, M.H. Al Hindi, O. Attawneh, RO pretreatment design and performance under challenging conditions in the Arabian Gulf, IDA World
CongressAtlantis, the PalmDubai, UAE, 2009, pp. 712.
[54] C.K. Ioannis, G.S. Petros, Water desalination cost literature: review and assessment,
Desalination 223 (2008) 448456.
[55] S. Jeong, G. Naidu, S. Vigneswaran, Submerged membrane adsorption bioreactor as a
pretreatment in seawater desalination for biofouling control, Bioresour. Technol.
141 (2013) 5764.
[56] D.E. Jiang, V.R. Cooper, S. Dai, Porous graphene as the ultimate membrane for gas
separation, Nano Lett. 9 (2009) 40194024.
[57] M.T. Khan, C.L. Manes, C. Aubry, J.P. Croue, Source water quality shaping different
fouling scenarios in a full scale desalination plant at the Red Sea, Water Res. 47
(2013) 558568.
[58] T.H. Kima, Y.S. Kima, Y.H. Choia, J.H. Kweona, J.H. Songb, N.W. Ganga, Biolm formation and its effect on biofouling in RO membrane processes for wastewater reuse,
Desalin. Water Treat. 2 (2009) 7175.
[59] M. Lerner, N. Stahl, N.I. Galil, Comparative study of MBR and activated sludge in the
treatment of paper mill wastewater, Water Sci. Technol. 55 (2007) 2329.
[60] L.X. Li, J.H. Dong, T.M. Nenoff, R. Lee, Desalination by reverse osmosis using MFI zeolite membranes, J. Membr. Sci. 243 (2004) 401404.
[61] L.X. Li, N. Liu, B. McPherson, R. Lee, Inuence of counter ions on the reverse osmosis
through MFI zeolite membranes: implications for produced water desalination, Desalination 228 (2008) 217225.
[62] W. Ma, Y. Zhao, L. Wang, The pretreatment with enhanced coagulation and a UF
membrane for seawater desalination with reverse osmosis, Desalination 203
(2007) 256259.
[63] J.E. Miller, Review of water resources and desalination technologies, Sandia National
Laboratories Report, SAND-2003-08002003.
[64] A. Mody, J. Dietrich, R. Reiss, M. Coates, C. Owen, D. McIntyre, Alternative pretreatment considerations for the Tampa Bay Water Gulf Coast Desalination Project, Proceedings AWWA Membrane Technology Conference, Phoenix, USA, 2005.
[65] V. Murugan, R. Nagaraj, Y. Dangore, S. Prabhakar, P.K. Tewari, Experiences with the
conventional pretreatment system in the seawater reverse osmosis plant at
Kalpakkam, Int. J. Nucl. Desalination (2008) 2732.
[66] T. Nguyen, F.A. Roddick, L. Fan, Biofouling of water treatment membranes: a review
of the underlying causes, monitoring techniques and control measures, Membranes
2 (2012) 804840.

[67] B. Ning, N. Graham, Y. Zhang, M. Nakonechny, M.G. El-Din, Degradation of endocrine

disrupting chemicals by ozone/AOPs, Int. Ozone Assoc. 29 (2007) 153176.
[68] R.Y. Ning, Pretreatment for reverse osmosis systems, expanding issues in desalination, InTech, 2011. 624629 (ISBN 978-953-307).
[69] R.Y. Ning, A.J. Tarquin, Crystallization of salts from super-concentrate produced by
tandem RO process, Desalin. Water Treat. 16 (2010) 238242.
[70] J.A. Nollet, Lecons de physique experimentale, Hippolyte-Louis Guerin and LouisFrancios Delatour, Paris, 1748.
[71] J. Park, Y. Kim, P.K. Kim, H.V. Daniels, Effects of two different ozone doses on seawater recirculating systems for black sea bream Acanthopagrus schlegeli (Bleeker): removal of solids and bacteria by foam fractionation, Desalination 44 (2011) 1924.
[72] G. Pearce, S. Talo, K. Chida, A. Basha, A. Gulamhushusin, Pretreatment options for
large scale SWRO plants: case studies of UF trials at Kindasa, Saudi Arabia, and conventional pretreatment in Spain, Desalination 167 (2004) 175189.
[73] G.K. Pearce, The case for UF/MF pretreatment to RO in seawater applications, Desalination 203 (2006) 286295.
[74] G.K. Pearce, UF/MF pre-treatment to RO in seawater and wastewater reuse applications: a comparison of energy costs, Desalination 222 (2008) 6673.
[75] M.A. Sanz, D. Guevara, F. Beltrn, E. Trauman, 4 Stages pre-treatment reverse osmosis for South-Pacic seawater: El Coloso Plant (Chile), Proceedings of the 2005 International Desalination Association World Congress, Singapore, 2005.
[76] M.A. Shannon, P.W. Bohn, M. Elimelech, J.G. Georgiadis, B.J. Marinas, A.M. Mayes,
Science and technology for water purication in the coming decades, Nature 452
(2008) 301310.
[77] K. Sint, B. Wang, P. Kral, Selective ion passage through functionalized graphene
nanopores, J. Am. Chem. Soc. 130 (2008) 1644816449.
[78] C. Song, B. Corry, Intrinsic ion selectivity of narrow hydrophobic pores, J. Phys.
Chem. B 113 (2009) 76427649.
[79] P. Sutton, Membrane bioreactors for industrial wastewater treatment: applicability
and selection of optimal system conguration, Water Environment Foundation, Eneld, 2006.
[80] A. Sweity, Y. Oren, Z. Ronen, M. Herzberg, The inuence of antiscalants on biofouling
of RO membranes in seawater desalination, Water Res. 47 (2013) 33893398.
[81] M. Tomaszewska, S. Mozia, Removal of organic matter from water by PAC/UF system, Water Res. 36 (2002) 41374143.
[82] R. Van den Broecka, J. Van Dierdonck, P. Nijskens, C. Dotremont, P. Krzeminski, J.H.J.
M. van der Graaf, J.B. van Lier, J.F.M. Van Impe, I.Y. Smets, The inuence of solids retention time on activated sludge bioocculation and membrane fouling in a membrane bioreactor (MBR), J. Membr. Sci. (2012) 4855.
[83] S.C.J.M. Van Hoof, A. Hashim, A.J. Kordes, The effect of ultraltration as pretreatment
to reverse osmosis in wastewater reuse and seawater desalination applications, Desalination 124 (1999) 231242.
[84] N. Voutchkov, Considerations for selection of seawater ltration pretreatment system, Desalination 261 (2010) 354364.
[85] water Association, Water reusewater desalination white paper, Seawater Desalination Cost2012. 119.
[86] WaterWiki, Retrieved from IWA Water Wiki: www.iwawaterwiki.org2013.
[87] M. Wilf, C. Bartels, Optimization of seawater RO systems design, Desalination 173
(2005) 112.
[88] S.F. Xavier, R. Elisabet, L. Joan, B. Sylvie, Study on the removal of biodegradable NOM
from seawater using bioltration, Desalination 316 (2013) 816.
[89] Y. Ye, L.N. Sim, B. Herulah, V. Chen, A.G. Fane, Effects of operating conditions on submerged hollow ber membrane systems used as pre-treatment for seawater reverse
osmosis, J. Membr. Sci. 365 (2010) 7888.
[90] L.I. YoungHong, J. Wang, W. Zhang, X.J. Zhang, C. Chen, Effects of coagulation on submerged ultraltration membrane fouling caused by particles and natural organic
matter (NOM), Environ. Eng. 56 (2011) 584590.
[91] A.A. Al-Hajouri, A.S. Al-Amoudi, A.M. Farooque, Long term experience in the operation of nanoltration pretreatment unit for seawater desalination at SWCC SWRO
plant, Desalin. Water. Treat. 51 (2013) 18611873.