Journal of Membrane Science 546 (2018) 61–69

Contents lists available at ScienceDirect

Journal of Membrane Science

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

The comparative study for scale inhibition on surface of RO membranes in T

wastewater reclamation: CO2 purging versus three different antiscalants
⁎
Muhammad Kashif Shahida, Young-Gyun Choib,
a
Department of Environmental & Chemical Convergence Engineering, Daegu University, (712-714), Daegudae-ro 201, Jillyang, Gyeongsan, Gyeongbuk, Republic of Korea
b
Department of Environmental Engineering, Chungnam National University, Daejeon, Republic of Korea

A R T I C L E I N F O A B S T R A C T

Keywords: The application of reverse osmosis (RO) membrane in wastewater reclamation is emerged as a potential tech-
Antiscalant nology. As far as the operational conditions are concerned, inorganic fouling is a major challenge for membrane
CO2 treatment systems. At present various antiscalants and chemicals are commercially available for scale inhibition
Flux on the surface of membrane. In an earlier study we found that CO2 can effectively inhibit scale growth on the
Reverse osmosis
surface of RO membrane in wastewater reclamation. On the basis of previous study, the scale inhibiting effi-
Scale
ciency of CO2 was compared with three commercially available antiscalants. The RO system was operated at
constant applied pressure with four different scale inhibition methods including CO2, Flocon 260, Flocon 300
and Kuriverter N-500. The permeate flux decline was considered as an indication of scale growth on the
membrane surface. The percent salt rejection and ionic mass balance was used to determine the scaling behavior
of the RO modules. Membrane autopsy was done to determine the effect of CO2 and antiscalants on the mem-
brane structure. The experimental data revealed that CO2 can effectively inhibit scale growth as compared with
all of three antiscalants. However, the Flocon 260 was found better than other two antiscalants.

1. Introduction the loss in percent salt rejection [16].

Water has always been an utmost concern due to its utilization,
demand and insufficiency. Almost 1.2 billion people have no access to
fresh and clean drinking water and the annual death rate for humans is
in millions due to the consumption of contaminated water [1]. To re-
duce the water scarcity, different alternative water sources are being
favored i.e., decontamination, disinfection, desalination and reclama-
tion techniques. For last few decades, membrane technology has ap-
peared as one of the most remarkable technologies for sea water de-
salination and treatment of brackish water [2,3]. The reclamation of
domestic and industrial wastewater with membrane technology is also
encouraging [4–6]. The selective separation technology is found
equally successful for removal of inorganic salts and pathogens [7,8].
Water carries different inorganic and organic particles or species which
may cause the fouling, and hence, results in low productivity and high
operational cost [9–11]. As the membrane permits only specific ions or
species to pass through itself, the concentration of rejected ions can
increase at the surface of membrane which may contributes in scale
growth [9–14]. The precipitation of inorganic salts on the membrane
surface can cause the serious decline in permeate flux or an increase in
transmembrane pressure (TMP) [15]. An increase in fouling also cause
In membrane treatment systems, pre-treatment, clean-in-place (CIP)
and the antiscalant or acid dosage in influent stream are some common
practices for scale inhibition [2,17–23]. Although the precipitation of
sparingly soluble salts in the system can be controlled kinetically with
the acidification, antiscalant and acid can itself oppose their role above
the threshold limit. Following the same concept of lowering pH, issue of
lime-scale is successfully resolved in heat exchangers. CO2 was used for
lowering the pH of mill water from 8.5 to 7.5 [24]. In another study on
bench-scale RO system, the fouled membrane was effectively washed
with dissolved CO2 solution [25]. However, excluding our previous
work, there is no study reported on the scale inhibition in wastewater
reclamation by use of CO2. In earlier study we found that membrane
fouling can be avoided by use of CO2 [26,27]. Moreover, the combined
effect of CO2 and antiscalant was also discussed. Our earlier study was
limited to the scale inhibition on the surface of membrane operated at
constant flux mode while in current study we focused on constant
pressure mode. In constant flux mode of operation, the particle move-
ment towards the membrane can be controlled, and hence, TMP be-
comes the dependent variable. Most of the available literature on
membrane separation is based on constant flux mode. Alternatively, in
the constant TMP mode of operation, TMP remains constant during the

⁎
Corresponding author.
E-mail address: youngchoi@cnu.ac.kr (Y.-G. Choi).

http://dx.doi.org/10.1016/j.memsci.2017.09.087
Received 5 June 2017; Received in revised form 23 September 2017; Accepted 26 September 2017
Available online 07 October 2017
0376-7388/ © 2017 Elsevier B.V. All rights reserved.
M.K. Shahid, Y.-G. Choi Journal of Membrane Science 546 (2018) 61–69

entire filtration cycle and the flux decreases with the scale formation. Table 2
Due to the changing conditions based on variable flux with time, lim- Chemical components of the synthetic wastewater.
ited information exists in literature for the operation of RO membrane
Chemical Molarity (mM) Volume (L) Amount (g)
at constant TMP. Herein, we provided the deep study of the operational
behavior of RO membranes at constant TMP and presented the appro- NaCl 20 200 235
priateness of CO2 purging at constant TMP mode operation. Moreover, CaCl2 5 200 110
MgCl2 3.4 200 65
in earlier study we compared the effectiveness of CO2 purging with
KCl 1.14 200 17
Hypersperse MDC-220, a widely used antiscalant in wastewater re- KNO3 0.30 200 6
clamation plants of Korea, while in this study the effectiveness of CO2 MgSO4 0.08 200 2
was compared with three other worldwide used antiscalants. The KH2PO4 0.07 200 2
chemistry of antiscalants at their threshold limit in water is also dis-
cussed.
RO system was operated with CO2 in comparison with three anti- Table 3
Specifications of RO membrane.
scalants including Flocon 260, Flocon 300 and Kuriverter N-500. The
operational performance was evaluated on the basis of permeate flux, Parameter Properties
salt rejection, mass balance of principal ions and membrane autopsy.
Based on the results of comparative study, CO2 purging (in feed water) Manufacturer Hyundai Wacortec
Membrane Type Spiral wound Polyamide TFC
is proposed as a novel method for scale control during wastewater re- Membrane Area 0.4 m2
clamation. MWCO ~100 amu (Dalton)
Operating Temperature (maximum) 45 °C
Operating Pressure (maximum) 125 psi
2. Materials and methods Feed flow rate (maximum) 2 GPM
pH range (continuous operation) 3.0 – 10
2.1. Experimental conditions Maximum Feed Silt Density Index <5
Free chlorine tolerance < 0.1 ppm
The RO system was operated with synthetic wastewater. The in-
fluent of RO system was synthesized by taking into account the quality
RO1 module. Table 5 shows the pH measurement on the feed, permeate
of the advanced treated domestic wastewater of the “P” wastewater
and concentrate streams of all RO modules. The system was restored
reclamation plant in Korea (Table 1). The chemical components of the
with CIP when 10% loss in salt rejection was observed. The CIP was
synthetic wastewater are presented in Table 2. Around 100,000 m3/d of
done with 1% EDTA, 0.1% caustic and 0.2% HCl. All the chemicals
the tertiary treated wastewater is reclaimed by further treatment with
were injected into feed side with 50 mL/min flow rate and 1-h contact
RO membrane for the purpose of industrial water supply at the plant.
time for each chemical. CIP was conducted at 25 °C temperature and
All the chemicals and reagents were of analytical grade and purchased
1 bar applied pressure. All the operation data including flux decline,
from Fisher Scientific. CO2 with over 99% purity was purchased from
salt rejection, ion mass balance was recorded throughout the study.
domestic market. Commercially available antiscalants (Flocon 260;
Membrane autopsy results were also considered for the establishment of
Flocon 300; Kuriverter N-500) were purchased from Hansu Ltd., Korea.
conclusion.
Flocon antiscalants are composed of phosphonic acid derivative and
polycarboxylic acid while Kuriverter N-500 constitutes PBTC (2-phos-
2.3. Analysis of water quality constituents
phonobutane-1,2,4- tricarboxylic acid) and polyacrylic acid. Polyamide
thin film composite membranes (0.4 m2 area) were purchased from
The pH and conductivity were analysed by 96pH-L2 (samsan) and
Hyundai Wacortec. Membranes were designed to operate successfully
EC96 (M. Cubic) respectively. Other water quality analysis were per-
on maximum operating pressure 8.6 bar, maximum operating tem-
formed according to Standard Methods (APHA) [28]. The flux decline
perature 45 °C, maximum feed flow rate 2 GPM and pH range 3.0–10
was monitored manually with the graduated cylinder and digital stop
(Table 3).
watch. The concentration of principal ions was analysed with Ion
Chromatograph (ICS-1000 and ICS-5000 for cations and anions re-
2.2. Operation of the RO system spectively). Scanning Electron Microscopy coupled with Energy Dis-
persive Spectroscopy (SEM-EDS, S-4300) and FTIR (Vertex 80 & Hy-
The RO system was operated for 45 days with synthetic wastewater. perion 3000) were used during membrane autopsy tests.
Four single pass RO modules were operated with CO2, Flocon 260,
Flocon 300 and Kuriverter N-500 respectively (Fig. 1, Table 4). RO 3. Results and discussion
modules were initially operated at 4 bar constant applied pressure,
11.32 LMH permeate flux, 40% recovery and 99% salt rejection. The 3.1. Operational behavior (flux decline)
RO 1 feed water was conditioned in a 200 L tank with CO2. The CO2
injection rate was controlled at 300 ± 2 mL/min and about The RO system was operated for continuous 45 days with two fil-
1.5 ± 0.1 min injection time was estimated for lowering the pH of feed tration cycles. CIP was conducted on 18th day i.e., at the end of first
water from 7.18 ± 0.07 to 6.05. About 450 ± 30 mL CO2/200 Linfluent is filtration cycle while following filtration cycle was terminated on 27th
assessed to be required for operation of RO1 at pH 6. Following the pH day (45th as a whole) without any CIP step. To determine the opera-
adjustment in 200 L tank, the synthetic wastewater was supplied to tional behavior in terms of flux decline during long run, CIP was not
conducted in 2nd filtration cycle. At the termination of entire filtration
Table 1
period, about 12% flux decline was observed for the RO 1 operated with
Influent water quality of the RO system.
CO2 (Fig. 2). However, 30% decline in initial flux was observed in RO 2
Parameter Analytical result Parameter Analytical result and 4. The RO 3 showed 45% decline in initial flux at the end of op-
eration. On the basis of flux variation, it can be stated that CO2 was
pH 7.18 ± 0.07 Conductivity 2150 ± 300 µS
more effective in suppressing scale growth on the membrane surface
Ca2+ 200 ± 10 mg/L Mg2+ 95 ± 10 mg/L
Na+ 480 ± 10 mg/L K+ 50 ± 5 mg/L while the membranes operated with antiscalant were remained in-
effective for scale inhibition. Moreover, the RO module 3 indicated the

62
M.K. Shahid, Y.-G. Choi Journal of Membrane Science 546 (2018) 61–69

Fig. 1. P & ID diagram of the RO system.

Table 4 osmotic pressure build-up at the surface of membrane followed by flux

Characterization of the RO modules on the basis of scale inhibition methods. decline at constant pressure. Comparatively better flux curve was ob-
served during the operation of RO 1 at low pH. It seems that by use of
Module Scale control method Amount of injection
CO2 the deposition of inorganic salts can be controlled kinetically. This
RO 1 CO2 purging 300 ± 2 mL/min finding is similar to previously reported studies on application of CO2 in
RO 2 Flocon 260 3 ppm water treatment systems [24–26,31].
RO 3 Flocon 300 3 ppm
RO 4 Kuriverter N-500 3 ppm
3.2. Salt rejection

Table 5
pH values of influent, permeate and concentrate streams of the RO modules.
Parameter Influent Permeate
RO 1 6.05 ± 0.05 5.12 ± 0.13
RO 2 7.18 ± 0.07 5.53 ± 0.12
RO 3 7.18 ± 0.07 5.48 ± 0.12
RO 4 7.18 ± 0.07 5.54 ± 0.13
sharp decline in initial permeate flux which is supposed to happen due
to the higher rate of crystallization on the membrane surface. Initially
all RO modules indicated the uniform permeate flux but as the filtration
time increased the more significant difference in permeate flux was
observed for all RO modules. Near 1–2% loss in initial permeate flux
was observed after the restoration of RO module 2, 3 and 4 with CIP
while near 0.5% loss in initial flux was observed for RO 1. After CIP,
this reduction in initial permeate flux indicates an increase in total
resistance followed by filtration period. The intensely ingrained re-
sidual deposits can also be a reason of this reduction in initial flux [29].
The flux decline curve for RO 2, 3 and 4 indicated the ineffective-
ness of antiscalants for scale reduction. As the filtration period increases
the membrane resistance also increases due to the gathering of fouling
particles on the membrane surface. Subsequently, this decreases the
total permeability of the membrane followed by flux decline at a con-
stant applied pressure as showed by RO 2, 3 and 4. The concentration
polarization also affects the membrane permeability. An increase in the
concentration polarization results in the higher solute concentration at
the surface of membrane which may leads to deposition of sparingly
soluble salts on the membrane surface [11,15,30]. It results in high
All RO modules were operated at 99% initial salt rejection. The salt
rejection was determined on the basis of influent and effluent con-
Concentrate ductivities. In first filtration cycle the RO module operated with CO2
indicated the minimum drop in salt rejection (4%) as compared with
6.37 ± 0.15
other modules. The percent salt rejection for RO 2 was dropped at 90%
7.22 ± 0.19
7.16 ± 0.27 while for RO 3 and 4 it was dropped at 83% and 86% respectively after
7.40 ± 0.21 17 days (408 h) of operation (Fig. 3a). The increasing deposition of
foulants on the membrane surface developed the cake layer resulting in
decrease of percent salt rejection [16]. The existence of divalent cations
affected both the permeate flux and salt rejection performance of
membranes. The RO module operated with CO2 indicates that at low pH
the deposition rate for foulants can be avoided successfully. However,
the antiscalants particularly Flocon 300 and Kuriverter N-500 were
found ineffective to maintain percent salt rejection throughout the fil-
tration cycle. The embedded foulants on the membrane surface are the
major reason of loss in the membrane permeability followed by de-
crease in percent salt rejection. As the fouling progresses, two factors
affect salt rejection and permeate flux simultaneously. The back diffu-
sion of rejected ions hindered due to the cake layer at the membrane
surface and the transportation of ions from bulk solution towards the
membrane surface. Hence an increase in concentration polarization can
cause an increase in osmotic pressure and decrease in percent salt re-
jection [32,33]. An increase in salt concentration on the membrane
surface causes higher salt flux followed by loss in salt rejection. Parti-
cularly, solubility limits can be increased for multivalent ions and re-
sulted deposits can cause adverse effects on mass transport. The con-
ceptual illustration of mass transfer and accumulation of particles near
the membrane surface is presented in Fig. 3(b). The transportation of
particles within the concentration polarization layer can be described

63
M.K. Shahid, Y.-G. Choi
by the convective diffusion equation [34]. The fluid velocity leads feed
particles to move in transverse and longitudinal directions. The particle
movement in longitudinal direction and transverse direction is followed
by crossflow velocity and permeate velocity respectively. During the
operation at constant applied pressure, the crossflow velocity of fluid
(parallel to membrane surface) decreases with increasing osmotic
pressure build-up by increased concentration polarization. Later it
causes an increase in salt permeation and flux decline. Among the an-
tiscalants, Flocon 260 indicated the better performance for percent salt
rejection but overall it was found behind CO2 in terms of effectiveness.
3.3. Mass balance of the principal ions
The mass balance of Na+, Mg2+, K+, Ca2+, Cl-, NO32-, PO43- and
SO42- was used for the determination of real amount of monovalent and
divalent ions entering and leaving the system. The concentration of ions
in influent, effluent and concentrate stream was used for the calculation
of ionic mass balance, i.e.:
ΔMAcc = Qinf • Cinf –Qper • Cper –Qcon • Ccon
where, ΔMAcc = ion mass accumulated inside of the membrane module
Fig. 3. (a). Salt rejection (%) of the RO membranes during 1st filtration cycle for 408 h (1–17 days). (b). The conceptual illustration of particle transportation and concentration
polarization; an increase in cake-enhanced concentration polarization directly effects the cross flow velocity, permeate flux and salt passage.
64
Journal of Membrane Science 546 (2018) 61–69
Fig. 2. Comparison of the flux decline.
Qinf = influent flow rate of the membrane module
Qper = permeate flow rate of the membrane module
Qcon = concentrate flow rate of the membrane module
Cinf = ion concentration of the influent
Cper = ion concentration of the permeate
Ccon = ion concentration of the concentrate
The accumulated fraction of principal ions in the RO modules was
calculated with the ion concentrations measured in the input and
output streams, thus not considering the concentration polarization on
the membrane which accumulates the high share of ionic mass and is
not a part of scale deposition or cake layer. Therefore, the mass fraction
of ions presented in Fig. 4 is hypothetically not only limited to the scale
layer but also shares the major contribution from polarization layer. An
increased mass fraction for monovalent ions most probably is due to the
higher ionic concentration accumulated in polarization layer [11,35].
Shawky et al., while discussing the intrinsic properties of the salts in
water and behavior of ionic movement through the RO membrane,
denied the theoretical concept of higher rejection for Mg2+ and Ca2+
than that of Na+ in RO membrane [36,37]. It was stated that the hy-
drated ions have the different radii as compared with non-hydrated
M.K. Shahid, Y.-G. Choi
ions, and hence, the cationic diffusion through the membrane does not
depend on its size and charge. Fig. 4 presented the higher ionic dis-
charge for all monovalent and divalent ions in RO 1. The accumulated
ionic mass fraction was less than 5% for RO 1 which seems to have a
major share of polarization layer because the membrane autopsy did
not indicate the significant existence of scale deposits particularly salts
of divalent cations. About 13% of ionic mass was accumulated inside
the RO 2. RO 3 and 4 indicated extensively low level of ionic discharge
which shows the accumulation of sparingly soluble salts inside the
system. Over 30% of divalent cationic mass was accumulated in RO 3
while over 20% ionic mass was accumulated inside RO 4. The high
precipitation of calcium and magnesium ions inside the RO 3 and 4
shows the ineffectiveness of Flocon 300 and Kuriverter N-500 for scale
inhibition. The dissociation of inorganic salts and successful discharge
of respective ions from system shows the effectiveness of CO2 for scale
control. It can be stated that the operation of RO membrane with CO2 is
more effective than that of antiscalants. Another study also supported
this finding [38].
3.4. Membrane autopsy
The membrane autopsy was conducted to determine the effect of
operation and scale control methods on the membrane. At the end of
operation, the membrane elements were unrolled and the membrane
sheet was cut in small pieces followed by rinsing with deionized water.
The membrane samples were dried at room temperature and then
analysed using SEM, EDS and FTIR. SEM was used to analyse the
morphology and structure of the surface of membrane and the EDS
analysis was conducted for the identification of elemental composition
of fouled membrane. FTIR analysis was conducted for fouled mem-
branes in comparison with virgin membrane.
The visual inspection of the membranes did not identify any damage
to the membrane (Supporting information, Fig. S1). However, a sticky
layer was found at the feed spacer channel of the RO 4 membrane
which was apparently due to the biofouling as discussed in other studies
[39,40]. The SEM images of RO 1 membrane displayed the negligible
existence of deposition without any distinct crystallization (Fig. 5) The
EDS spectra indicated the monovalent ions like Na, Cl and K are
dominant on the membrane surface of RO 1. Hence it can be stated that
no carbonate or sulphate scale was accumulated in RO 1 during whole
filtration cycle. As shown in Fig. 5, the sphere-shaped agglomerated
crystal structures were observed at different positions of membrane
65
Journal of Membrane Science 546 (2018) 61–69
Fig. 4. The accumulated fraction (%) of the principal ions
(ΔMAcc/Qinf•Cinf). The error bars denote the standard deviation of
the three sampled days (days 43–45).
surface in RO 2. It was happened due to the gathering of bulk calcite
crystals and other inorganic salts. In earlier studies this phenomenon
was explained as a result of accumulation of calcite and gypsum salts
[41] or due to the effect of antiscalant on calcite crystals [42]. The
sludge like fine patches were also found at the membrane surface which
indicate the bulk deposition of sodium and potassium salts [43]. The
aragonite crystals were also observed on the membrane surface and
found similar as presented in another studies [42,44,45]. Needle like
gypsum crystals were also recognized on the membrane surface which
are supposed to be calcium sulphate dehydrate [46,47]. Although the
major area of membrane was covered with inorganic scale, the basic
polymeric structure of membrane can also be identified at some por-
tions of membrane surface. It shows that although the membrane was
spoiled with inorganic scale at the termination of filtration cycle,
Flocon 260 (RO 2) was still functioning to reduce scale growth on the
membrane surface. While the entire surface of membranes in RO 3 and
4 was covered by inorganic deposits and the polymeric membrane
structure cannot be identified. The well-developed rhombohedral
structure of calcite [48,49], patches like refined structure of sodium salt
and needle like somehow plate-shaped gypsum structure was observed
on the membrane surface of RO 3 and 4. RO 3 and 4 indicated the
similar gypsum structure on the membrane surface but the bulk gath-
ering of gypsum crystals having needle-like structure was found to be
transformed into flower like appearance [50]. The flowery structure of
crystals might be due to the combined effect of lateral scale growth and
the precipitation of bulk formed crystals on the surface of membrane
(Fig. 5). The crystal structure is highly dependent on the concentration
of divalent ions in the bulk. The presence of antiscalant (F-300) is also
assumed to be a major reason of change in prismatic structure of
gypsum [51,46]. The adsorption of antiscalants at the crystallization
site may result in alteration of crystal morphology. The acrylate func-
tional group (-COO-) has the specific affinity for calcium ion in the bulk
solution. As the calcium ion remains in the vicinity of acrylate ion the
probability of joining sulphate ion remains low. But if the calcium ions
will once approach sulphate ions, it will initiate nucleation and the
acrylate ion of antiscalant will not able to control further growth of
scale [51]. Another study discussed the threshold limit of PBTC anti-
scalant for scale inhibition but at the same time no significant mor-
phological change in gypsum crystals was identified [52]. Along with
other inorganic salts, the gypsum is also one of the major causes of flux
decline in polyamide membranes. P was also identified in the EDS
spectra of membranes from RO 2, 3 and 4 which can be consider as an
M.K. Shahid, Y.-G. Choi Journal of Membrane Science 546 (2018) 61–69

Fig. 5. SEM images and EDS spectra of the RO membranes.

66
M.K. Shahid, Y.-G. Choi
Fig. 6. Weight % of the elements on the surface of RO membranes.
existence of calcium phosphate deposition [53,54].
Due to the high level of solute rejection, the concentration of
sparingly soluble salts can certainly exceed the saturation state at the
membrane surface resulting in precipitation. The bulk phase also car-
ries the divalent ions which further contribute in the secondary crys-
tallization on membrane surface. The lateral growth of fouling on
membrane surface caused the blockage of pores resulted in perme-
ability loss. As the fouling progressed the sharper flux drop was ob-
served in RO 3 and 4. It is noteworthy that no significant fouling was
found on the membrane surface when RO module was operated at low
pH with CO2. It can be stated that the saturation of sparingly soluble
salts can be controlled kinetically at low pH, and hence, the deposition
can be avoided.
The EDS analysis identified O, S, C, Na, Mg, Si, P, Cl, Ca, Al, K, Fe,
Cu, Pt and Ti on the surface of membranes. In RO 3 and 4, the EDS
spectra unexpectedly indicated that the fouling layer had very low level
of Fe, Cu, Pt and Ti which were not identified during feed water ana-
lysis. The presence of Pt was apparently due to a mistake in ion de-
tection by the EDS, because it did not appear in the reanalysis for three
times (Supporting information, Fig. S2). On the other hand, the pre-
sence of Fe, Cu and Ti was probably due to a leaching from the system
parts particularly pressure regulator which was made up of Inconel (an
alloy constitutes Fe, Cu, Ti and some other elements). In addition to
these elements, carbon, oxygen and sulphur, the main constituents of
polyamide membrane were also ignored and rest of elements were
supposed to be deposited on the surface of membranes. The RO module
operated with CO2 indicated the high share of Na (72%) followed by Cl
(17%), Al (7%) and K (3%) approximately (Fig. 6). However, Ca, Mg, P
67
Journal of Membrane Science 546 (2018) 61–69
and other main constituents of inorganic scale were not identified in RO
1. Although Fig. 4 indicated Ca, Mg, P and K in RO1, it seems that this
finding is due to the accumulation of most of the ions in polarization
layer and not in scale deposits. Along with other elements the higher
share of Ca, Mg and P was identified on the membranes operated with
antiscalants. However, the share of deposited elements was quite well
for RO 2 as compared with RO 3 and 4. If this finding will correlate with
percent salt rejection and the SEM images of the membranes, it can be
concluded that Flocon 260 is more effective than that of Flocon 300 and
Kuriverter N-500. The existence of the main constituents of inorganic
salts on membrane surface shows that after a certain time of run the
threshold limit of antiscalants approached. In result no further scale
inhibition was observed. It was discussed in earlier studies that anti-
scalant can itself cause fouling in the system [55,56].
The FTIR spectra of the virgin membrane and the used membranes
are presented in Fig. 7. As expected the specific pattern of polyamide
membranes was reflected by virgin membrane [57,58]. The bands in
the region 1540–1640 cm−1 signified the existence of pure polyamide
layer. And the polysulfone layer was signified by bands at different
wavelengths like 1146, 1180, 1240, 1280, 1486 and 1585 cm−1 etc.
[59]. The same pattern can be seen in case of RO 1 membrane operated
with CO2. However, the membranes operated with antiscalants reflects
the either eliminated or rigorously tempered bands in range
1550–1650 cm−1 and 1010–1120 cm−1. This alteration in band pattern
signifies the influence of fouling on the membrane. It was noticed that
RO 2 membrane reflects the little change in band pattern as than that of
RO 3 and 4. Finally it can be stated on the basis of membrane autopsy
results that the use of CO2 in RO system can achieve better performance
as that of antiscalant. In case of antiscalants, Flocon 260 was found
better as compared with Flocon 300 and Kuriverter N-500. On the basis
of entire study, CO2 was found more effective as compared with anti-
scalants for scale inhibition on the surface of RO membrane in waste-
water reclamation.
4. Conclusions
In this study the application of CO2 for scale inhibition in RO system
was discussed in comparison with three different commercial anti-
scalants. The RO system was operated for 45 days at 4 bar constant
applied pressure mode with four different lines operated with CO2,
Flocon 260, Flocon 300 and Kuriverter N-500. The flux decline was
considered as a representation of fouling on the surface of membrane.
The percent salt rejection, ion mass balance and the results from
membrane autopsy indicated the successful inhibition of scale on
membrane surface when CO2 was applied. However serious scale
Fig. 7. Comparison of FTIR spectra of RO membranes.
M.K. Shahid, Y.-G. Choi
growth was identified on the membranes operated with antiscalants.
SEM images indicated the bulk accumulation of calcite, gypsum and
aragonite crystals on the surfaces of membranes operated with anti-
scalants. The well-developed crystals of inorganic salts were found on
the membrane surfaces. Though all the antiscalants showed limited
effectiveness for scale inhibition, Flocon 260 was found better in per-
formance than other antiscalants i.e., Flocon 300 and Kuriverter N-500.
Although the major area of membrane (operated with Flocon 260) was
covered with inorganic scale, the basic polymeric structure of mem-
brane was identifiable. However, the entire surface of RO 3 and 4
(operated with Flocon 300 and Kuriverter N-500, respectively) was
covered with deposits and the basic polymeric structure of RO mem-
branes was not visible in SEM images. The fine polymeric structure was
visible in the SEM images of membrane operated with CO2. Finally, it
can be stated that CO2 can effectually inhibit the scale growth on the
membrane surface. The CO2 purging in feed water has the potential to
become an important operational step of RO in future. CO2 purging can
reduce or replace the chemical consumption in RO systems, and hence,
makes the process eco-friendly.
Acknowledgments
This research was supported by 2015 Joint Lab supporting Program
of MOTIE (Ministry of Trade, Industry and Energy), Korea (Project #:
N0001672).
Appendix A. Supplementary material
Supplementary data associated with this article can be found in the
online version at http://dx.doi.org/10.1016/j.memsci.2017.09.087.
References
[1] M. a. Montgomery, M. Elimelech, Water and sanitation in developing countries:
including health in the equation, Environ. Sci. Technol. 41 (2007) 17–24, http://dx.
doi.org/10.1021/es072435t.
[2] C. Fritzmann, J. Löwenberg, T. Wintgens, T. Melin, State-of-the-art of reverse os-
mosis desalination, Desalination 216 (2007) 1–76, http://dx.doi.org/10.1016/j.
desal.2006.12.009.
[3] L.F. Greenlee, D.F. Lawler, B.D. Freeman, B. Marrot, P. Moulin, Reverse osmosis
desalination: water sources, technology, and today's challenges, Water Res. 43
(2009) 2317–2348, http://dx.doi.org/10.1016/j.watres.2009.03.010.
[4] T. Coskun, E. Debik, N.M. Demir, Treatment of olive mill wastewaters by nanofil-
tration and reverse osmosis membranes, Desalination 259 (2010) 65–70, http://dx.
doi.org/10.1016/j.desal.2010.04.034.
[5] C.R. Bartels, M. Wilf, K. Andes, J. Iong, Design considerations for wastewater
treatment by reverse osmosis, Water Sci. Technol. 51 (2005) 473–482.
[6] S.M.R. Razavi, T. Miri, A. Barati, M. Nazemian, M. Sepasi, Industrial wastewater
treatment by using of membrane, Membr. Water Treat. 6 (2015) 489–499, http://
dx.doi.org/10.12989/mwt.2015.6.6.489.
[7] M.L. Pype, M.G. Lawrence, J. Keller, W. Gernjak, Reverse osmosis integrity mon-
itoring in water reuse: the challenge to verify virus removal - a review, Water Res.
98 (2016) 384–395, http://dx.doi.org/10.1016/j.watres.2016.04.040.
[8] K.V. Plakas, A.J. Karabelas, Removal of pesticides from water by NF and RO
membranes - a review, Desalination 287 (2012) 255–265, http://dx.doi.org/10.
1016/j.desal.2011.08.003.
[9] M. Pontié, S. Rapenne, A. Thekkedath, J. Duchesne, V. Jacquemet, J. Leparc,
H. Suty, Tools for membrane autopsies and antifouling strategies in seawater feeds:
a review, Desalination 181 (2005) 75–90, http://dx.doi.org/10.1016/j.desal.2005.
01.013.
[10] A. Antony, J.H. Low, S. Gray, A.E. Childress, P. Le-Clech, G. Leslie, Scale formation
and control in high pressure membrane water treatment systems: a review, J.
Membr. Sci. 383 (2011) 1–16, http://dx.doi.org/10.1016/j.memsci.2011.08.054.
[11] S. Shirazi, C.J. Lin, D. Chen, Inorganic fouling of pressure-driven membrane pro-
cesses - a critical review, Desalination 250 (2010) 236–248, http://dx.doi.org/10.
1016/j.desal.2009.02.056.
[12] S. Lee, J. Kim, C.H. Lee, Analysis of CaSO4 scale formation mechanism in various
nanofiltration modules, J. Membr. Sci. 163 (1999) 63–74, http://dx.doi.org/10.
1016/S0376-7388(99)00156-8.
[13] M. Okazaki, S. Kimura, Scale formation on reverse osmosis membranes, J. Chem.
Eng. Jpn. 17 (1984) 145–151, http://dx.doi.org/10.1252/jcej.17.145.
[14] E.H.K. Zeiher, B. Ho, K.D. Williams, Novel antiscalant dosing control, Desalination
157 (2003) 209–216, http://dx.doi.org/10.1016/S0011-9164(03)00400-4.
[15] D.J. Miller, S. Kasemset, D.R. Paul, B.D. Freeman, Comparison of membrane fouling
at constant flux and constant transmembrane pressure conditions, J. Membr. Sci.
68
Journal of Membrane Science 546 (2018) 61–69
454 (2014) 505–515, http://dx.doi.org/10.1016/j.memsci.2013.12.027.
[16] S. Lee, J. Cho, M. Elimelech, Influence of colloidal fouling and feed water recovery
on salt rejection of RO and NF membranes, Desalination 160 (2004) 1–12, http://
dx.doi.org/10.1016/S0011-9164(04)90013-6.
[17] L.F. Greenlee, F. Testa, D.F. Lawler, B.D. Freeman, P. Moulin, The effect of anti-
scalant addition on calcium carbonate precipitation for a simplified synthetic
brackish water reverse osmosis concentrate, Water Res. 44 (2010) 2957–2969,
http://dx.doi.org/10.1016/j.watres.2010.02.024.
[18] A. Al-Amoudi, R.W. Lovitt, Fouling strategies and the cleaning system of NF
membranes and factors affecting cleaning efficiency, J. Membr. Sci. 303 (2007)
4–28, http://dx.doi.org/10.1016/j.memsci.2007.06.002.
[19] N. Ghaffour, T.M. Missimer, G.L. Amy, Technical review and evaluation of the
economics of water desalination: current and future challenges for better water
supply sustainability, Desalination 309 (2013) 197–207, http://dx.doi.org/10.
1016/j.desal.2012.10.015.
[20] N. Prihasto, Q.F. Liu, S.H. Kim, Pre-treatment strategies for seawater desalination
by reverse osmosis system, Desalination 249 (2009) 308–316, http://dx.doi.org/10.
1016/j.desal.2008.09.010.
[21] W. Ma, Y. Zhao, L. Wang, The pretreatment with enhanced coagulation and a UF
membrane for seawater desalination with reverse osmosis, Desalination 203 (2007)
256–259, http://dx.doi.org/10.1016/j.desal.2006.02.020.
[22] E. Filloux, J. Wang, M. Pidou, W. Gernjak, Z. Yuan, Biofouling and scaling control of
reverse osmosis membrane using one-step cleaning-potential of acidified nitrite
solution as an agent, J. Membr. Sci. 495 (2015) 276–283, http://dx.doi.org/10.
1016/j.memsci.2015.08.034.
[23] A.G. Pervov, Scale formation prognosis and cleaning procedure schedules in reverse
osmosis systems operation, Desalination 83 (1991) 77–118, http://dx.doi.org/10.
1016/0011-9164(91)85087-B.
[24] P.W. Hart, G.W. Colson, J. Burris, Application of carbon dioxide to reduce water-
side lime scale in heat exchangers, J. Sci. Technol. Prod. Process. 1 (2011) 67–70.
[25] E. Partlan, Dissolved Carbon Dioxide for Scale Removal in Reverse Osmosis,
Clemson University, 2013, 〈http://tigerprints.clemson.edu/all_theses/1788〉.
[26] M.K. Shahid, M. Pyo, Y. Choi, Carbonate scale reduction in reverse osmosis mem-
brane by CO2 in wastewater reclamation, Membr. Water Treat. 8 (2017) 125–136,
http://dx.doi.org/10.12989/mwt.2017.8.2.125.
[27] M.K. Shahid, M. Pyo, Y.-G. Choi, Inorganic fouling control in reverse osmosis
wastewater reclamation by purging carbon dioxide, Environ. Sci. Pollut. Res.
(2017), http://dx.doi.org/10.1007/s11356-017-9008-3.
[28] APHA/AWWA/WEF, Standard Methods for the Examination of Water and
Wastewater, 2012.
[29] W. Song, V. Ravindran, B.E. Koel, M. Pirbazari, Nanofiltration of natural organic
matter with H2O2/UV pretreatment: fouling mitigation and membrane surface
characterization, J. Membr. Sci. 241 (2004) 143–160, http://dx.doi.org/10.1016/j.
memsci.2004.04.034.
[30] J. Kucera, Desalination: Water from Water., 2014 〈http://dx.doi.org/10.1002/
9781118904855〉.
[31] I.S. Ngene, R.G.H. Lammertink, A.J.B. Kemperman, W.J.C. Van De Ven,
L.P. Wessels, M. Wessling, W.G.J. Van Der Meer, CO2 nucleation in membrane
spacer channels remove biofilms and fouling deposits, Ind. Eng. Chem. Res. 49
(2010) 10034–10039, http://dx.doi.org/10.1021/ie1011245.
[32] A. Seidel, J.J. Waypa, M. Elimelech, Role of charge (Donnan) exclusion in removal
of arsenic from water by a negatively charged porous nanofiltration membrane,
Environ. Eng. Sci. 18 (2001) 105–113, http://dx.doi.org/10.1089/
10928750151132311.
[33] E.M.V. Hoek, A.S. Kim, M. Elimelech, Influence of crossflow membrane filter geo-
metry and shear rate on colloidal fouling in reverse osmosis and nanofiltration
separations, Environ. Eng. Sci. 19 (2002) 357–372, http://dx.doi.org/10.1089/
109287502320963364.
[34] S. Hong, R.S. Faibish, M. Elimelech, Kinetics of permeate flux decline in crossflow
membrane filtration of colloidal suspensions, J. Colloid Interface Sci. 196 (1997)
267–277.
[35] C. Bartels, R. Franks, S. Rybar, M. Schierach, M. Wilf, The effect of feed ionic
strength on salt passage through reverse osmosis membranes, Desalination 184
(2005) 185–195, http://dx.doi.org/10.1016/j.desal.2005.04.032.
[36] H.A. Shawky, M.E.A. Ali, A.A.S. Gallab, M.S.A. Abdel-mottaleb, Influence of the Salt
Properties on their Transport through thin Film Reverse Osmosis Membranes, in:
BUE ACE1 SVT2016: 2016: pp. 1–14.
[37] G.M. Geise, D.R. Paul, B.D. Freeman, Fundamental water and salt transport prop-
erties of polymeric materials, Prog. Polym. Sci. 39 (2014) 1–42, http://dx.doi.org/
10.1016/j.progpolymsci.2013.07.001.
[38] A. Joss, C. Baenninger, P. Foa, S. Koepke, M. Krauss, C.S. McArdell, K. Rottermann,
Y. Wei, A. Zapata, H. Siegrist, Water reuse: > 90% water yield in MBR/RO through
concentrate recycling and CO2 addition as scaling control, Water Res. 45 (2011)
6141–6151, http://dx.doi.org/10.1016/j.watres.2011.09.011.
[39] S.S. Bucs, A.I. Radu, V. Lavric, J.S. Vrouwenvelder, C. Picioreanu, Effect of different
commercial feed spacers on biofouling of reverse osmosis membrane systems: a
numerical study, Desalination 343 (2014) 26–37, http://dx.doi.org/10.1016/j.
desal.2013.11.007.
[40] J.S. Vrouwenvelder, D.A. Graf von der Schulenburg, J.C. Kruithof, M.L. Johns,
M.C.M. van Loosdrecht, Biofouling of spiral-wound nanofiltration and reverse os-
mosis membranes: a feed spacer problem, Water Res. 43 (2009) 583–594, http://
dx.doi.org/10.1016/j.watres.2008.11.019.
[41] R. Sheikholeslami, H.W. Ong, Kinetics and thermodynamics of calcium carbonate
and calcium sulfate at salinities up to 1.5 M, Desalination 157 (2003) 217–234,
http://dx.doi.org/10.1016/S0011-9164(03)00401-6.
[42] C. Tzotzi, T. Pahiadaki, S.G. Yiantsios, A.J. Karabelas, N. Andritsos, A study of
M.K. Shahid, Y.-G. Choi
CaCO3 scale formation and inhibition in RO and NF membrane processes, J.
Membr. Sci. 296 (2007) 171–184, http://dx.doi.org/10.1016/j.memsci.2007.03.
031.
[43] E. Guillen-Burrieza, R. Thomas, B. Mansoor, D. Johnson, N. Hilal, H. Arafat, Effect
of dry-out on the fouling of PVDF and PTFE membranes under conditions simu-
lating intermittent seawater membrane distillation (SWMD), J. Membr. Sci. 438
(2013) 126–139, http://dx.doi.org/10.1016/j.memsci.2013.03.014.
[44] M. Brusilovsky, J. Borden, D. Hasson, Flux decline due to gypsum precipitation on
RO membranes, Desalination 86 (1992) 187–222, http://dx.doi.org/10.1016/0011-
9164(92)80033-6.
[45] C.L. Yao, C.X. Qi, J.M. Zhu, W.H. Xu, Unusual morphology of calcium carbonate
controlled by amino acids in agarose gel, J. Chil. Chem. Soc. 55 (2010) 270–273,
http://dx.doi.org/10.4067/S0717-97072010000200028.
[46] F. Rahman, Calcium sulfate precipitation studies with scale inhibitors for reverse
osmosis desalination, Desalination 319 (2013) 79–84, http://dx.doi.org/10.1016/j.
desal.2013.03.027.
[47] F. Alimi, A. Gadri, Kinetics and morphology of formed gypsum, Desalination 166
(2004) 427–434, http://dx.doi.org/10.1016/j.desal.2004.06.097.
[48] N. Andritsos, a.J. Karabelas, P.G. Koutsoukos, Morphology and structure of CaCO3
scale layers formed under isothermal flow conditions, Langmuir 7463 (1997)
2873–2879.
[49] M. Kitamura, Crystallization and transformation mechanism of calcium carbonate
polymorphs and the effect of magnesium ion, J. Colloid Interface Sci. 236 (2001)
2318–2327, http://dx.doi.org/10.1006/jcis.2000.7398.
[50] W.Y. Shih, A. Rahardianto, R.W. Lee, Y. Cohen, Morphometric characterization of
calcium sulfate dihydrate (gypsum) scale on reverse osmosis membranes, J. Membr.
Sci. 252 (2005) 253–263, http://dx.doi.org/10.1016/j.memsci.2004.12.023.
[51] S. Ben Ahmed, M.M. Tlili, M. Ben Amor, Influence of a polyacrylate antiscalant on
69
Journal of Membrane Science 546 (2018) 61–69
gypsum nucleation and growth, Cryst. Res. Technol. 43 (2008) 935–942, http://dx.
doi.org/10.1002/crat.200800066.
[52] M. Prisciandaro, E. Olivieri, A. Lancia, D. Musmarra, PBTC as an antiscalant for
gypsum precipitation: interfacial tension and activation energy estimation, Ind.
Eng. Chem. Res. 51 (2012) 12844–12851, http://dx.doi.org/10.1021/ie302060t.
[53] M. Raffin, E. Germain, S. Judd, Assessment of fouling of an RO process dedicated to
indirect potable reuse, Desalin, Water Treat. 40 (2012) 302–308, http://dx.doi.org/
10.1080/19443994.2012.671171.
[54] F. Tang, H.Y. Hu, L.J. Sun, Y.X. Sun, N. Shi, J.C. Crittenden, Fouling characteristics
of reverse osmosis membranes at different positions of a full-scale plant for muni-
cipal wastewater reclamation, Water Res. 90 (2016) 329–336, http://dx.doi.org/
10.1016/j.watres.2015.12.028.
[55] G. Greenberg, D. Hasson, R. Semiat, Limits of RO recovery imposed by calcium
phosphate precipitation, Desalination 183 (2005) 273–288, http://dx.doi.org/10.
1016/j.desal.2005.04.026.
[56] J.J. Qin, M.H. Oo, F.S. Wong, Effects of pH and antiscalant on fouling of RO
membrane for reclamation of spent rinse water from metal plating, Sep. Purif.
Technol. 46 (2005) 46–50, http://dx.doi.org/10.1016/j.seppur.2005.04.016.
[57] V. Freger, J. Gilron, S. Belfer, TFC polyamide membranes modified by grafting of
hydrophilic polymers: an FT-IR/AFM/TEM study, J. Membr. Sci. 209 (2002)
283–292, http://dx.doi.org/10.1016/S0376-7388(02)00356-3.
[58] E. Idil Mouhoumed, A. Szymczyk, A. Schäfer, L. Paugam, Y.H. La, Physico-chemical
characterization of polyamide NF/RO membranes: insight from streaming current
measurements, J. Membr. Sci. 461 (2014) 130–138, http://dx.doi.org/10.1016/j.
memsci.2014.03.025.
[59] F. a. Miller, C.H. Wilkins, Infrared spectra and characteristic frequencies of in-
organic ions, Anal. Chem. 24 (1952) 1253–1294, http://dx.doi.org/10.1021/
ac60068a007.