membranes

Article
Fabrication, Optimization, and Performance of a TiO2 Coated
Bentonite Membrane for Produced Water Treatment: Effect of
Grafting Time
Mohamad Izrin Mohamad Esham 1 , Abdul Latif Ahmad 1, * and Mohd Hafiz Dzarfan Othman 2

1 School of Chemical Engineering, Engineering Campus, Universiti Sains Malaysia (USM),

Nibong Tebal 14300, Pulau Pinang, Malaysia; mohamadizrin@student.usm.my
2 Advanced Membrane Technology Research Centre (AMTEC), School of Chemical and Energy Engineering,
Universiti Teknologi Malaysia (UTM), Johor Bahru 81310, Johor, Malaysia; hafiz@petroleum.utm.my
* Correspondence: chlatif@usm.my; Tel.: +60-45996499; Fax: +60-45996908

Abstract: The main problem usually faced by commercial ceramic membranes in the treatment of
produced water (PW) is low water flux even though ceramic membrane was well-known with their
excellent mechanical, thermal, and chemical properties. In the process of minimizing the problem
faced by commercial ceramic membranes, titanium dioxide (TiO2 ) nanocomposites, which synthe-
sized via a sol-gel method, were deposited on the active layer of the hydrolysed bentonite membrane.
This paper studied the influence of grafting time of TiO2 nanocomposite on the properties and per-
formance of the coated bentonite membranes. Several characterizations, which are Fourier transform



infrared (FTIR), scanning electron microscopy (SEM), energy-dispersive X-ray Spectroscopy (EDX),


contact angle, porosity, and average pore size, were applied to both pristine and coated bentonite
Citation: Mohamad Esham, M.I.;
membranes to compare the properties of the membranes. The deposition of TiO2 nanoparticles on
Ahmad, A.L.; Othman, M.H.D.
the surface of the coated bentonite membranes was successfully confirmed by the characterization re-
Fabrication, Optimization, and
sults. The pure water flux performance showed an increment from 262.29 L h−1 m− 2 bar−1 (pristine
Performance of a TiO2 Coated
bentonite membrane) to 337.05 L h−1 m− 2 bar−1 (Ti-Ben 30) and 438.33 L h−1 m− 2 bar−1 (Ti-Ben 60)
Bentonite Membrane for Produced
Water Treatment: Effect of Grafting
as the grafting time increase but when the grafting time reached 90 min (Ti-Ben 90), the pure water
Time. Membranes 2021, 11, 739. flux was decreased to 214.22 L h−1 m− 2 bar−1 which is lower than the pristine membrane. The oil
https://doi.org/10.3390/ rejection performance also revealed an increase in the oil rejection performance from 95 to 99%. These
membranes11100739 findings can be a good example to further studies and exploit the advantages of modified ceramic
membranes in PW treatment.
Academic Editor: Harsh Vardhan
Keywords: bentonite membrane; titanium dioxide nanoparticle; produced water
Received: 4 August 2021
Accepted: 2 September 2021
Published: 28 September 2021

1. Introduction
Publisher’s Note: MDPI stays neutral
Produced water (PW), a troublesome by-product, can be defined as formation water
with regard to jurisdictional claims in
located in the reservoirs below the hydrocarbon layer. Once the reservoirs become mature
published maps and institutional affil-
and experience water cuts, the reservoirs will produce PW during oil and gas extraction.
iations.
PW contain various complex compounds. However, it can be grouped into inorganic
and organic compounds such as dispersed and dissolved oils, grease, chemicals, salts,
dissolved gases, anions, and cations [1]. PW also contain hazardous and radioactive heavy
metals, such as chromium, lead, barium, uranium, and cadmium [2]. Other than that,
Copyright: © 2021 by the authors.
some microorganisms may also live in the PW since PW has nutrients in it [3]. Hence, the
Licensee MDPI, Basel, Switzerland.
treatment of the PW must meet the offshore and onshore discharge rules and regulations
This article is an open access article
before being released into the environment. If not, PW may result in severe conflicting
distributed under the terms and
consequence on our flora and fauna in the marine and land environment [4,5]. What make
conditions of the Creative Commons
Attribution (CC BY) license (https://
more concern, when some of the previous studies have declared that PW have become
creativecommons.org/licenses/by/
one of the largest waste stream in oil and gas industry [6,7]. One of the previous studies
4.0/). reported that the volume of the PW is 10 times larger than the extracted hydrocarbon. The

Membranes 2021, 11, 739. https://doi.org/10.3390/membranes11100739 https://www.mdpi.com/journal/membranes

Membranes 2021, 11, 739 2 of 18

other study also reported that matured oil fields such as oil fields in Danish North Sea may
be one of the causes of increase production of PW, other than the demand of oil and gas.
Various standalone or combined physical, biological and chemical technique were proposed
for the treatment of PW either by removing the suspended solids, de-oiling, removal of
iron, desalting the organic compounds and softening the PW. The example of physical
treatment are activated carbon [8] which used as adsorbent media to remove organic
compounds like BTEX (Benzene, Toluene, Ethylbenzene and Xylene), hydrocyclone [9] for
the separation of solids in PW based on their density, electrodialysis(ED) [10] to separate
either anions or cations. Other than that, physical treatment also includes precipitation [11],
ion-exchange [12], American Petroleum Institute “API” separator [13], deep bed filter [14],
aeration [15], sedimentation [16] and also thermal desalination [17]. As for the chemical
treatment, the example are coagulation and flocculation [18–20], and lastly biological
treatment [21,22] which is superior in removing anaerobic and aerobic microorganisms.
In the beginning of the 20th century, membrane technology had become one of the
most effective methods introduced in the oil and gas industry [23–25]. Membranes usually
will be divided into two groups, which are inorganic and organic membranes, based on their
starting materials. Polymeric membranes can be classified in the organic membranes group,
while ceramic, metal oxides, and metallic membranes are inorganic membranes. Previous
studies have shown some interest in utilizing the ceramic membranes in the PW treatment
without pre-treatment process [26,27], since the ceramic membranes have great advantages
in their mechanical, chemical and thermal characteristics. Furthermore, ceramic membranes
also have the advantages in harsh environment, such as backwashing, autoclaving and
multiple cleaning process due to their high chemical and thermal stability. The lifetime
of the ceramic membrane is also longer and better than the polymeric membrane. In
the production of ceramic membranes in the industry and in research studies, there are
numerous techniques that can be applied. Flat sheet ceramic membranes are usually
fabricated via the tape casting and pressing technique, while for the fabrication of hollow
fiber ceramic membrane the technique that will be applied to produce it are the phase
inversion and slip casting technique [28]. There are numerous studies which have focused
on the application of ceramic hollow fiber membrane on the PW treatment. For example,
there was a study using a low-cost alumina-spinel composite hollow fiber microfiltration
membrane for the pretreatment of synthetic oily saline produced water [29]. The results
illustrated that the prepared membrane efficiently remove the oil droplet in PW by 92.41%.
The turbidity was also found to be reduced around 92%. Another study used a low cost
superhydrophobic-superoleophilic kaolin-based hollow fiber ceramic membrane for the
recovery of oil from synthetic produced water [30]. The results showed an average flux
of 80 L m−2 h−1 and 90% of oil recovery. Meanwhile, a study conducted using mullite–
kaolinite powder was used to treat PW from an oil refinery through which >97% TOC
removal was reported [31].
In the past few years, the interest in fabrication of ceramic membranes had turned to
the surface modification of the membrane, since the ceramic membranes itself have superior
in most of the physical and chemical of properties. Hence, new studies were more focused
in increasing the flux and rejection performance. A previous study successfully achieved
better oil rejection and flux by implementing a graphene oxide (GO) nanoparticle, modi-
fying the commercial ceramic alumina (Al2 O3 ) membrane by the vacuum technique [32].
Next, kaolin hollow fiber membranes were dip coated with nano-Al2 O3 for oil and water
separation [33]. As expected, the coated membranes demonstrated an improvement in
hydrophilicity, resulting in good oil removal efficiency up to 98%. A study also reported
that polyamide/titanium dioxide (TiO2 ) composite hollow fiber nanofiltration (NF) mem-
branes were successfully fabricated via the sol-gel method [34]. The coated NF membrane
was prepared by immersing it vertically in the prepared TiO2 sol. The resulting coated
NF membrane outperformed the pristine membranes by showing higher fluxes as high
as 105.5 L m−2 h−1 at four bar and higher oil rejection greater than 95%. In the next study,
an omniphobic mullite hollow fiber membrane with flower-like TiO2 was successfully
Membranes 2021, 11, 739 3 of 18

fabricated, and the membrane surface wettability were improved [35]. In another study, α-
Al2 O3 membranes were coated with TiO2 using a simple dip-coating process with tetraethyl
orthosilicate (TEOS) as a binder to increase the hydrophilic characteristic of the membrane
surface [36]. After testing, the methylene blue (MetB) removal efficiency (up to 91%), as
well as the flux of the modified membrane, exceeded the uncoated membrane. In this
study, the authors were inspired by the sol dip-coating and sol-gel method reported in
the literature and aimed to coat the bentonite membranes with TiO2 nanoparticles on the
membranes’ surface via dip coating and heat for calcination. Hence, the aim of this study is
to modify the surface of the bentonite membranes with nanoparticles via the combination
of sol-gel dip coating and calcination techniques, as inspired by the method used in the
reported previous study. The concept of sol gel dip-coating bentonite membranes using
nanoparticles can be the pioneer of study regarding to the surface modification of bentonite
membranes for the treatment of PW since there was not reported before. Thus, this study
will be more focused on the effect of grafting time (30, 60, and 90 min) of TiO2 nanoparticles
on bentonite membranes to find the optimum grafting time via the characterization and
the performance of the coated membrane toward oil rejection and pure water flux, using
synthetic PW.

2. Materials and Methods

2.1. Materials
The bentonite membranes with an average pore size of 1.75 µm were from previous
study [37]. Titanium butoxide (Ti(OC4 H9 )4 ), absolute ethanol (EtOH, 99.5%, A.R.), Hy-
drochloric acid (HCl, 37%), glacial acetic acid, and distilled water (H2 O) were used during
this study without any further purification.

2.2. Preparation of TiO2 Sol

In the preparation of TiO2 sol solution, 10 mL of titanium butoxide and 35 mL of
ethanol were mixed together using a magnetic stirrer for 10 min in room temperature to
obtain the TiO2 precursor solution. Concurrently, in another magnetic stirrer, 2 mL of
glacial acetic acid, 10 mL of distilled water, and another 35 mL of ethanol were stirred
for 10 min, then few drops of hydrochloric acid were used to adjust the pH to pH = 3.
After that, the prepared TiO2 precursor was slowly added to the second solution, then the
stirring continued for another 1 h to obtain the TiO2 sol. The prepared TiO2 sol then was
aged for three days at room temperature prior to the usage.

2.3. Deposition of TiO2 Nanoparticle on Bentonite Membrane

Before the process of dip-coating, the bentonite membranes were potted at both ends
using polytetrafluoroethylene (PTFE) film tape to avoid the TiO2 sol solution from entering
the lumen of the membranes. The potted membranes were dipped into ethanol for 30 min
to enhance the OH bond on the surface of the membranes. The process of dip-coating into
the TiO2 sol solution begun with various grafting times (30, 60, and 90 min). Then, the
coated bentonite membranes were dried for 1 h at room temperature before the repeated
cycle of dip-coating process. In this study, the cycle of dip-coating process was fixed to three
times for each membrane after the drying process, as suggested in a previous study [38].
After that, the membranes were dried in an oven for 30 min to remove the moisture content.
Lastly, the membranes were calcined with a fixed calcination temperature of 400 ◦ C, as
suggested by previous study which is around 400–450 ◦ C [39–41], for 1 h to remove all of
the organic chemicals and crystallize the anatase-TiO2 . Table 1 shows the prepared coated
bentonite membranes with different grafting times.
Membranes 2021, 11, 739 4 of 18

Table 1. Coated bentonite membranes with different grafting time.

Calcination
Membrane Grafting Time (min) Grafting Cycle (times)
Temperature (◦ C)
Pristine Ben 0 3 400
Ti-Ben 30 30 3 400
Ti-Ben 60 60 3 400
Ti-Ben 90 90 3 400

2.4. Membrane Characterization

Fourier transform infrared spectroscopy (FTIR, Thermo Scientific Nicolet Nexus 670,
Waltham, MA, USA) in the range of 4000–500 cm−1 using the attenuated total reflectance
technique (ATR) was used to check the functional groups related to TiO2 coated bentonite
membranes. Moreover, scanning electron microscopy (SEM, HITACHI S-3000 N, Hitachi
Ltd., Tokyo, Japan) was employed to observe membranes’ surfaces and cross-sections.
Moreover, energy-dispersive X-ray Spectroscopy (EDX) also was utilized at 10 KeV to
examine the TiO2 nanoparticle distribution and density on the coated membranes. Zeta
Potential (Malvern Ver. 2.3, Malvern Instruments Ltd., 2008, Malvern, UK) were performed
to study the charges of the TiO2 nanoparticle and oil particle, also the size of oil particle.
The pore size of membranes was analyzed using Capillary Flow Porometer (Porolux 1000,
Nazareth, Belgium) by following “dry up-wet up” method. The pressure of gas was
gradually increased from 1 to 5 bar and the gas flow rate was recorded. Bubble point was
recorded when the pressure was high enough to remove the liquid out from the largest
pores. As pressure gradually increased, smaller pores become unblocked by liquid and
the gas flow rate increased until the whole sample was completely dry. The cumulative
pressure was used to calculate pore size distribution and average pore size. The pore sizes
were estimated using perfluorether (porefil) solution. As for the porosity, the prepared
pristine and coated membranes were immersed in the distilled water for 12 h and then
weighed immediately after blotting. After that, the membranes were dried in oven for 12 h
before dry weighed recorded. The porosity of the membranes was calculated by using
Equation (1):
mw
ρdw
ε= mw md × 100% (1)
ρdw + ρb

where ε indicates the porosity of membrane (%), md and mw are the mass of the dry and
wetted membranes (g), respectively, and ρb and ρdw are the density of the bentonite and
density of distilled water (g cm−3 ), respectively. The calculation of porosity was taken with
five measurements and averaged to minimize error.
The contact angles of the membranes were estimated using the goniometer (Rame-Hart
250F-1, Succasunna, NJ, USA), to study the wettability of pristine and coated membranes.
The device was used to perform a static contact angle test for the membranes. The contact
angle test was run in two conditions, which are in air condition and underwater condition.
Distilled water and oil droplet was used as the droplet dropped on the membranes’ surface
for each contact angle test. The contact angle device was equipped with a camera, which
recorded every image/video of the droplet, starting from before the water droplet drop to
the membranes’ surface until there is no movement of the droplet. For each membrane, the
droplets were dropped at 10 difference parts of the membrane surface. Then, an average
value of contact angle was recorded and analyzed. Prior to the contact angle test for
underwater conditions, the prepared pristine and coated bentonite membranes will be
placed in a black box setup, as shown in Figure 1, for 1 h. A 36W UV lamp (Brand: Philips,
Tokyo, Japan) is used as the light source and the illumination distance is 35 cm.
 Membranes 2021,11,
Membranes2021, 11,739
x FOR PEER REVIEW 5 of 18
5 of 19

Figure 1. Black box setup for illumination of UV light.

Figure 1. Black box setup for illumination of UV light.

2.5. Water
WaterFlux
FluxPermeation
Permeationand
andOil
OilRejection
RejectionPerformance
Performance
AA crossflow
crossflowfiltration
filtrationsystem
system asas
illustrated in in
illustrated Figure 2 was
Figure used
2 was usedto perform a water
to perform a water
flux and
and oil
oilrejection
rejectionperformance
performanceofof pristine bentonite
pristine bentonite andand
coated bentonite
coated withwith
bentonite a length
a length
of 10 cm
cm at
at aapressure
pressureofof33barbarwith
withconstant
constantflow
flow rate ofof
rate 400 mLmL
400 min −1 and
−1 and
min constant roomroom
constant
temperature.
temperature.The Thewater
waterflux
flux(J(J
W) was
W ) wasobtained and
obtained calculated
and by
calculated using
by Equation
using (2):
Equation (2):
𝑉𝑤
JW = Vw (2)
JW = 𝐴×𝑡×𝑃 (2)
A×t×P
Membranes 2021, 11, x FOR PEER REVIEW
where VW is the permeate of water permeate through the membrane (L), A is the mem- 6 of 19
where VW is the
brane surface permeate
area (m2), t isofthe
water permeate
sampling timethrough
(h), andthe
P ismembrane (L), A is the
the transmembrane membrane
pressure
surface area (m2 ), t is the sampling time (h), and P is the transmembrane pressure (bar).
(bar).
The rejection (R) of the oil was measured under the same condition as the water per-
meation test to evaluate the benchmarking of oil rejection. The rejection rates were ob-
tained and calculated according to Equation (3).
(𝐶𝑓 − 𝐶𝑝 )
𝑅= × 100% (3)
𝐶𝑓
where R is the rejection rate of oil (%), Cf is the oil concentration of feed (mg/L) and Cp is
the oil concentration of permeate (mg/L). The concentration of oil in the produced water
before and after ultrafiltration using UV-visible spectrophotometer (Thermo Scientific,
Genesys 10S, Waltham, MA, USA) at a wavelength of 254 nm. The synthetic produced
water was produced by mixing the heavy crude oil obtained from PETRONAS Melaka
Refinery Complex and distilled water with the concentration of 1000 ppm. The concentra-
tion of surfactant (SDS) used in the mixing process was based on the ratio of oil and sur-
factant which is 9:1. The oil content was tested and analyzed prior to the oil rejection per-
formance test.

Figure 2. Crossflow
Figure 2. Crossflowfiltration
filtration setup
setup for for produced
produced water
water treatment.
treatment.

The rejection
3. Results (R) of the oil was measured under the same condition as the water
and Discussion
permeation test to evaluate the benchmarking of oil rejection. The rejection rates were
3.1. Characterization
obtained of Pristine
and calculated Bentonite
according and Coated
to Equation (3). Bentonite Membranes
3.1.1. FTIR Analysis 
C
Figure 3 shows the FTIR analysis − C p
f for the pristine bentonite membranes and coated
R= × 100% (3)
C f 60 and Ti-Ben 90). It can be clearly seen that Ti-
bentonite membranes (Ti-Ben 30, Ti-Ben
Ben 30, Ti-Ben 60 and Ti-Ben 90 had the peak of Ti-O at ~1000 cm−1 and peak of Ti–OH at
~1620 cm−1. Other than that, the peak at ~2950 which also can be seen on all of the TiO2
coated membrane are the characteristic of –CH2. All the peaks for coated bentonite mem-
branes, however, was not witnessed on the pristine bentonite membrane FTIR spectra.
 Membranes 2021, 11, 739 6 of 18

where R is the rejection rate of oil (%), Cf is the oil concentration of feed (mg/L) and Cp is
the oil concentration of permeate (mg/L). The concentration of oil in the produced water
before and after ultrafiltration using UV-visible spectrophotometer (Thermo Scientific,
Genesys 10S, Waltham, MA, USA) at a wavelength of 254 nm. The synthetic produced
water was produced by mixing the heavy crude oil obtained from PETRONAS Melaka
Refinery Complex and distilled water with the concentration of 1000 ppm. The concen-
tration of surfactant (SDS) used in the mixing process was based on the ratio of oil and
surfactant which is 9:1. The oil content was tested and analyzed prior to the oil rejection
performance test.

3. Results and Discussion

3.1. Characterization of Pristine Bentonite and Coated Bentonite Membranes
3.1.1. FTIR Analysis
Figure 3 shows the FTIR analysis for the pristine bentonite membranes and coated
bentonite membranes (Ti-Ben 30, Ti-Ben 60 and Ti-Ben 90). It can be clearly seen that
Ti-Ben 30, Ti-Ben 60 and Ti-Ben 90 had the peak of Ti-O at ~1000 cm−1 and peak of Ti–OH
at ~1620 cm−1 . Other than that, the peak at ~2950 which also can be seen on all of the
TiO2 coated membrane are the characteristic of –CH2 . All the peaks for coated bentonite
membranes, however, was not witnessed on the pristine bentonite membrane FTIR spectra.
This proves that the membrane itself has no peak at that wavenumber and that those peaks
mbranes 2021, 11, x FOR PEER REVIEW observed are due to the TiO2 deposition solely. Furthermore, by comparing the coated 7 of 19
membranes, we can find that the peak of Ti–O, Ti–OH and –CH2 peaks increased as the
grafting time increased. The results obtained are similar to the previous study. The peak
intensity increased as the concentration of the TiO2 on the membrane increased [42].

Figure 3. FTIR spectra for pristine bentonite membrane, Ti-Ben 30, Ti-Ben 60, and Ti-Ben 90.

Figure 3. FTIR spectra for pristine

3.1.2. Surface bentonite membrane,
and Cross-Sectional Ti-Ben 30, Ti-Ben
Morphology 60, and Ti-Ben 90.
Analysis
The SEM images were used to investigate the morphology of the surfaces and cross-
3.1.2. Surface
sectionsand Cross-Sectional
of the Morphology
membranes. Figure 4 shows the Analysis
SEM images of surfaces and the cross-section
The SEM images were used to investigate the morphology of the surfaces and cross-
sections of the membranes. Figure 4 shows the SEM images of surfaces and the cross-
section images of Ti-Ben 30, Ti-Ben 60, Ti-Ben 90 membranes. The surface SEM images
were all taken at 2000× magnification whereas the cross-section images were taken at 400×
 Membranes 2021, 11, 739 7 of 18

images of Ti-Ben 30, Ti-Ben 60, Ti-Ben 90 membranes. The surface SEM images were all
taken at 2000× magnification whereas the cross-section images were taken at 400× and
2000× magnification. The white particles observed on the coated membranes surfaces were
Membranes 2021, 11, x FOR PEER REVIEW 8 of 19
identified as TiO2 nanoparticles. This was deduced as they showed most on membrane
Ti-Ben 90 and the least on membrane Ti-Ben 30.

Figure
Figure 4. SEM
4. SEM images
images of of
(a)(a) Ti-Ben30,
Ti-Ben 30,(b)
(b)Ti-Ben
Ti-Ben 60,
60, and
and (c)
(c) Ti-Ben
Ti-Ben90
90on
on(1(1and
and2)2)cross-section of the
cross-section membranes
of the at 2000
membranes ×
at 2000×
and 500 × magnification, respectively, and (3) surface of the membrane at 2000 × magnification.
and 500× magnification, respectively, and (3) surface of the membrane at 2000× magnification.
The main observation which can be concluded from the surface SEM images is the
The main observation which can be concluded from the surface SEM images is the
different deposition densities of TiO2 on the membranes according to the grafting time. At
different deposition
the grafting time of densities
30 min, the ofsurface
TiO2 onSEM
the image
membranes according
of Ti-Ben 30 showed to the
somegrafting time.
parts of the At
the grafting time of 30 min, the surface SEM image of Ti-Ben 30 showed
membrane surface were started to be covered by the TiO2 nanoparticles. However, as the some parts of the
membrane
grafting time increased to 60 min (Ti-Ben 60) and 90 min (Ti-Ben 90), the TiO2 nanoparticles the
surface were started to be covered by the TiO 2 nanoparticles. However, as
grafting
started time increased
to cover to 60 min
almost every (Ti-Ben
part of 60) and 90
the membrane min (Ti-Ben
surface. 90),see
As we can theinTiO 2 nanoparti-
Figure 4(c1),
cles started
some of thetoTiOcover
2 almost
nanoparticlesevery
have part of
enteredthe
themembrane
pore of the surface.
membrane As we
whichcan
ledsee
to in Figure
partial
blockage on some of the membrane pores. It is notable that
4(c1), some of the TiO2 nanoparticles have entered the pore of the membrane the TiO 2 nanoparticles only
which led to
covered
partial some part
blockage of theofcoated
on some bentonite pores.
the membrane membranes’ surfacethat
It is notable similar to the
the TiO previous
2 nanoparticles
study
only [41]. As
covered the study
some part ofreported,
the coatedthe partially
bentonitecoated TiO2 would
membranes’ helpsimilar
surface in increasing
to the the
previ-
flux of the coated membrane. To support the SEM results of Ti-Ben 90 partial
ous study [41]. As the study reported, the partially coated TiO2 would help in increasing blockage, the
cross sectional FESEM and EDX was analyzed, as shown in Figure 5. As can be seen in the
the flux of the coated membrane. To support the SEM results of Ti-Ben 90 partial blockage,
Figure 5b, some of TiO2 was detected inside the pore of the coated membrane, as the light
the cross sectional FESEM and EDX was analyzed, as shown in Figure 5. As can be seen
blue color is also shown on the membrane surface.
in the Figure 5b, some of TiO2 was detected inside the pore of the coated membrane, as
the light blue color is also shown on the membrane surface.
Membranes 2021, 11, x FOR PEER REVIEW 9 of 19

Membranes 2021, 11, x FOR PEER REVIEW 9 of 19

Membranes 2021, 11, 739 8 of 18

Figure 5. FESEM at magnification of 200× and Ti mapping images of Ti-Ben90, (a) cross sectional FESEM, (b) EDX Mapping
Figure
ofFigure
Ti, and 5. FESEM
5. (c) at magnification
EDXatresult
FESEM ofof200
200×
of Ti percentage.
magnification × and
and Ti
Ti mapping
mappingimages
imagesofofTi-Ben90,
Ti-Ben90,(a)(a)cross sectional
cross FESEM,
sectional (b)(b)
FESEM, EDX Mapping
EDX Mapping
of Ti, and (c) EDX result of Ti percentage.
of Ti, and (c) EDX result of Ti percentage.
3.1.3. Wettability Analysis
3.1.3. Wettability Analysis
3.1.3. The
Wettability Analysis
wettability properties of the pristine and coated membranes were analyzed by
The wettability properties of the pristine and coated membranes were analyzed by
measuring
The the water
wettability
measuring the water contact contact
properties angle
of the
angle (WCA)
pristine
(WCA) and
and and oil contact
coated
oil contact membranes
angle angle
(OCA),(OCA),were under
underanalyzed
both air both
by air
measuring
and
and underwaterthe water contact
conditions.
underwater conditions. angle
These
These (WCA)
analysis
analysis and oil
also also contact
werewere angle
also identified
also identified (OCA),
to helpto under
in help both air
in clarifying
clarifying
and
the underwater
thechanges
changes in inpureconditions.
pure water
water These
flux
flux and andanalysis
oil oil flux
flux also were also
performance
performance [41].identified
It[41].
should to noted
It should
be helpbe in clarifying
noted
here thathere that
the
thechanges
the wettability
wettabilityin pure
test water
testfor
forWCA flux
WCA forand
for oil and
air air
and flux performance
underwater
underwater [41].shown
was
was only It should
only shown be noted
for Ti-Benfor 30, here
sincethat
Ti-Ben 30, since
the
thewettability
results of test
images for WCA
for Ti-Benfor60air
andand underwater
Ti-Ben 90 was
displayed only
the shown
same
the results of images for Ti-Ben 60 and Ti-Ben 90 displayed the same results. As shown in for
results. Ti-Ben
As shown30, since
in
the results of images forcontact
Ti-Benangle
Figures
Figures 66 and
and 7,7,these
these contact60angle
and Ti-Ben
values 90 measured
were
values displayed at
were measuredthethe same
first
at theresults.
moment As(1shown
first moment s) the (1ins) the
Figures 6 and 7,drop
water droplets theseonto
contact angle values
the membrane weresince
surface, measured
the water at the first moment
droplets (1 s) the
were quickly
water droplets drop onto the membrane surface, since the water droplets were quickly
water droplets
dissolved drop
into the onto the membrane
membrane. surface,
In air condition, since the
the WCA water30
of Ti-Ben droplets
was 10.21°were quickly
± 0.54°
dissolved into the membrane. In air condition, the WCA of Ti-Ben 30 was ◦10.21° ±◦ 0.54°
lower than pristine bentonite membrane which was 30.54° ± 0.17°. Then, after several±
dissolved into the membrane. In air condition, the WCA of Ti-Ben 30 was 10.21 0.54
sec-
lower
lower than
onds, than
the WCA pristine
pristine bentonite
bentonite
of both membrane
pristinemembrane
and Ti-Benwhich which
30 turned was 30.54°
◦
0°. In ±
wasto30.54 ± 0.17°.◦
0.17 . Then,
underwater Then, after several
after several
condition, the sec-
onds,
water the
seconds, WCA
the
dropletsWCA ofof
were both
bothpristine
quicklypristine and
dissolved Ti-Ben
andintoTi-Ben 30
30 turned
turned
the Ti-Ben to 00°.
to
30 membrane In underwater
◦ . In underwater condition, the
condition,
at the first moment
water
the droplets
thewater
waterdroplets were
dropletswere quickly
touchquickly dissolved
the membrane
dissolved into
surface, the
while
into the Ti-Ben 30 membrane
there30ismembrane
Ti-Ben still an angle aton at
thethethe first moment
pristine
first moment
the
the waterdroplets
bentonite
water dropletstouch
membrane touch
which the
the membrane
ismembrane
22.59° ± 0.71°. surface,
surface, whilewhile
therethere
is stillis still an angle
an angle on the onpristine
the pristine
bentonitemembrane
bentonite membranewhich which is is ◦ ±±
22.59°
22.59 ◦.
0.71°.
0.71

Figure 6. Images of water droplet on, (a) pristine bentonite membrane and (b) Ti-Ben 30, for air
condition.

Figure 6. Images of water droplet on, (a) pristine bentonite membrane and (b) Ti-Ben 30, for
Figure 6. Images of water droplet on, (a) pristine bentonite membrane and (b) Ti-Ben 30, for air
air condition.
condition.
Membranes
Membranes 2021,
2021, 11, 739
x FOR PEER REVIEW 9 of 1810 of 19

Membranes 2021, 11, x FOR PEER REVIEW 10 of 19

Figure 7. Images of water droplet on, (a) pristine bentonite membrane, and (b) Ti-Ben 30, for
Figure 7. Images of water
Figure droplet
7. Images on,droplet
of water (a) pristine
on, (a)bentonite membrane,
pristine bentonite and (b)
membrane, andTi-Ben 30, 30,
(b) Ti-Ben forfor
under-
under-
underwater condition.
water condition.
water condition.
Figure 8 displays
Figurethe8 OCA
displaysof pristine
OCAand
thepristine coatedand
of pristine membranes
coated for air and
membranes forunderwater
air and underwater
Figure
conditions. 8 displays
Surprisingly, the
no OCAOCA
static of can be and coated
measured formembranes
the pristine for air
membrane and underwater
for both
conditions. Surprisingly, no static OCA can be measured for the pristine membrane for
conditions.As
conditions. Surprisingly,
for conditions. no static OCA
coated membranes, can of
the value bethe
measured
OCA didfor
not the pristineoleophobic
reach membrane for
both As for coated membranes, the value of the OCAthedid not reach the oleo-
both conditions.
range when tested
phobic As
on for coated
air condition,
range membranes,
when tested which
on airare the
56.51value
condition, of the
◦ ± 0.27
which OCA 30),
◦ (Ti-Ben
are 56.51° did not(Ti-Ben
84.13
± 0.27° reach
◦ ± 0.20the
◦ oleo-
30), 84.13° ±
phobic 60)range when ◦
tested on◦
(Ti-Ben and 90.30
0.20° (Ti-Ben ± 60) andair
0.10 condition,
(Ti-Ben
90.30° ±90), which
0.10°but as for
(Ti-Ben are 56.51°
as for± underwater
underwater
90), but 0.27° (Ti-Ben
condition 30),UV
with
condition 84.13° ±
with UV
0.20°illumination,
light (Ti-Benlight
60) and
the 90.30°
OCA ± 0.10°
values (Ti-Ben
for all 90),
coated but as for
membranes underwater
increased condition
drastically
illumination, the OCA values for all coated membranes increased drastically to with
to UV
light ◦illumination,
130.24 ◦ (Ti-Ben 30), 147.00◦ ± 0.05◦ (Ti-Ben 60) and 155.50◦ ± 0.10◦ , as the grafting
± 0.10130.24° ±the OCA
0.10° (Ti-Benvalues for all± 0.05°
30), 147.00° coated membranes
(Ti-Ben increased
60) and 155.50° drastically
± 0.10°, to
as the grafting
time increased.
130.24° time
± 0.10° increased.
(Ti-Ben 30), 147.00° ± 0.05° (Ti-Ben 60) and 155.50° ± 0.10°, as the grafting
time increased.

Figure 8. Oil contact angle values for each membrane for air and underwater condition.
Figure 8. Oil contact angle values for each membrane for air and underwater condition.

Figure 8. Oil contact angle values for each membrane for air and underwater condition.
Membranes 2021, 11, x FOR PEER REVIEW 11 of 19
Membranes 2021, 11, 739 10 of 18

These results showed great enhancement in the wettability of the coated bentonite
membranes.These The most
results interesting
showed part is that in
great enhancement thethe
coated bentonite
wettability of themembranes tend to
coated bentonite
membranes.
switch The most
the wettability interesting part
of membrane fromishydrophilic
that the coated bentonite
membrane to membranes
underwatertend to
superhy-
switch the wettability of membrane from hydrophilic membrane
drophilic and superoleophobic after being illuminated with UV light. A similar trend of to underwater super-
hydrophilic
findings and superoleophobic
was reported on photoinduced after being illuminated
switchable with UV superoleophobicity-su-
underwater light. A similar trend
of findings was reported on photoinduced switchable underwater superoleophobicity-
peroleophilicity on laser modified titanium surfaces [43]. Similarly, as the TiO2 surface
superoleophilicity on laser modified titanium surfaces [43]. Similarly, as the TiO2 surface
radiated with UV while immersed in water, the water contact angle was found to be
radiated with UV while immersed in water, the water contact angle was found to be around
around 2° which also can be considered as super-hydrophilic surface. Another study also
2◦ which also can be considered as super-hydrophilic surface. Another study also showed
showed the membrane
the membrane ability ability
to switchto the
switch the wettability
wettability to underwater
to underwater superoleophobic
superoleophobic as the UV as
thelight
UV light illuminated the TiO 2 /TCNC
illuminated the TiO2 /TCNC membrane [44]. membrane [44].
This wettability
This wettabilityenhancement
enhancement can canbe beexplained
explained bybytwotwo phenomenon
phenomenon ofreactivity
of the the reactivity
of
of TiO
TiO22 toward
toward the theUV
UVlight
lightasas shown
shown in in Figure
Figure 9. The
9. The firstfirst phenomenon,
phenomenon, as theas the
TiO 2 TiO
surface
2 sur-
facebeing
being illuminate
illuminate with
with UV UVlight
lightin in
thethe
water, there
water, therewill be be
will thethe
appearance
appearance of holes andand
of holes
electrons.
electrons. The The electronsthat
electrons thathave
havebeenbeen photogenerated
photogenerated will willreact
reactwith
with oxygen
oxygen molecules
molecules
in the
in the surrounding,
surrounding, thenproduce
then producesuperoxide
superoxide radical
radicalanions,
anions,while
whilethe holes
the will
holes be be
will filled
filled
up by water molecules to produce OH radicals. These generated
up by water molecules to produce OH radicals. These generated groups are very strong groups are very strong
oxidizing agents which help in repelling the oil particles [45].
oxidizing agents which help in repelling the oil particles [45].

Figure 9. Photocatalytic
Figure (phenomenon
9. Photocatalytic (phenomenon1)1)and
andhydrophilicity (phenomenon2)2)mechanism.
hydrophilicity (phenomenon mechanism.

The other
The otherphenomenon
phenomenon is the theformation
formation of of water
water filmfilm onmembrane
on the the membrane surface.
surface. In this In
thisparticular
particularcase, thethe
case, electrons andand
electrons holesholes
werewere
also generated. However,
also generated. they will
However, undergo
they will un-
different
dergo reaction.
different The photogenerated
reaction. The photogeneratedelectrons tend totend
electrons reduced Ti(IV) cations
to reduced Ti(IV)tocations
Ti(III) to
andandthe holes will be oxidize −
into Ointo
Ti(III) the holes will be oxidize 2 anions. During
O2− anions. this mechanism,
During the oxygen
this mechanism, will
the oxygen
be thrown out and a group of oxygen vacancies are produced on the surface.
will be thrown out and a group of oxygen vacancies are produced on the surface. Finally, Finally, the
water molecules will occupy the empty sites and adsorbed OH radicals to create the water
the water molecules will occupy the empty sites and adsorbed OH radicals to create the
film, thus increase the hydrophilicity of the membrane [46,47].
water film, thus increase the hydrophilicity of the membrane [46,47].
3.2. Performance Tests of Pristine Bentonite Membrane and TiO2 Coated Bentonite Membranes
3.2. Performance
Figure 10Tests
showsof Pristine
the pureBentonite Membrane
water flux andbentonite
of pristine TiO2 Coated Bentoniteand
membranes Membranes
coated
Figure membranes
bentonite 10 shows the pure30,water
(Ti-Ben Ti-Benflux of pristine
60 and Ti-Ben 90)bentonite membranes
while Table 2 shows theand coated
average
pore size,
bentonite porosity and
membranes coating
(Ti-Ben 30,thickness
Ti-Ben 60ofand
the Ti-Ben
prepared90)membranes.
while TableAs shown the
2 shows in the
aver-
figure, two observations can be seen, where the flux decreased with time for
age pore size, porosity and coating thickness of the prepared membranes. As shown in all membranes
theand the changes
figure, of flux as the
two observations grafting
can time
be seen, was increased.
where In the beginning
the flux decreased of thefor
with time filtration,
all mem-
Ti-Ben 30 and Ti-Ben 60 have the highest pure water flux of about ~2700 L h − 1 m − 2 bar−1
branes and the changes of flux as the grafting time was increased. In the beginning of the
filtration, Ti-Ben 30 and Ti-Ben 60 have the highest pure water flux of about ~2700 L h−1
m−²bar−1 followed by pristine bentonite membranes and Ti-Ben 90, which are about ~1800
Membranes 2021, 11, 739 11 of 18

followed by pristine bentonite membranes and Ti-Ben 90, which are about ~1800 and
~1300 L h−1 m− 2 bar−1 , respectively. However, as the time increased, different steady-
state flux time were obtained for each membrane, for which the fastest time are pristine
bentonite membrane (t = 50 min), followed by Ti-Ben 90 and Ti-Ben 30 (t = 70 min), and
lastly Ti-Ben 60 (t = 100 min). It should be noted here that the pure water flux and oil
rejection performance were tested starting on the first minute of the filtration without the
flux stabilization process. This is why the fluxes were decreased as the time increased,
so that we can observed the steady-state flux time for each tested membrane. Results
also showed that the pure water flux at the end of the filtration for Ti-Ben 30 and Ti-Ben
60 were increased when compared with the pristine membranes, which were around
28% (337.05 L h−1 m− 2 bar−1 ) and around 67% (438.33 L h−1 m− 2 bar−1 ), respectively.
However, the increment trend of pure water flux did not continue as the grafting time
increased to 90 min (Ti-Ben 90) when the value decreased around 214.22 L h−1 m− 2 bar− 1 ,
which was also even lower than the pristine bentonite membranes. When comparing
the Ti-Ben 30 and Ti-Ben 60, the pure water results showed Ti-Ben 60 was performed
more effectively than Ti-Ben 30. This can be explained by the coating thickness of TiO2
nanoparticles on the membrane’s surface. Clearly, we can see from SEM images that the
coating thickness of TiO2 on the Ti-Ben 60 membranes’ surface much better in quantity
than Ti-Ben 30 (Figure 4). It was found that the increase of the coating thickness between
the Ti-Ben 30 and Ti-Ben 60 is very marginal (10 and 12 µm). Thus, it is expected that the
diffusion resistance between these two membranes is very little. However, Ti-Ben 60 would
have higher concentration of TiO2 which contributed to the higher pure water flux. The
sufficient coating thickness of TiO2 nanoparticles on the membranes’ surface help in better
acceptance of water droplet passing through the coated membranes [48]. In contrast, there
are two explanations that can be derived from the declination of pure water flux for Ti-Ben
90. The first explanation is that the membranes’ pore blockage resulting from the high
concentration of TiO2 nanoparticles on the membranes’ surface. As compared to other
coated membranes, Ti-Ben 90 had the highest coating thickness of TiO2 nanoparticles on
the membranes’ surface. The coating thickness is almost double (23 µm), which contribute
to the higher diffusion resistance. As shown in Table 2, even though the thickness of TiO2
coating increased the hydrophilicity of the membrane. However, the average pore size
(0.63 µm) and porosity (45%) was expected to be drastically decline due to the attachment
of nanoparticle of the membranes’ pores. The SEM and FESEM results on the cross-section
(Figure 4(c1) and Figure 5b) also showed that some of the TiO2 nanoparticle had entered
the membrane pores which resulting of the pore blockage. It can be concluded that even
though the coated membranes have higher hydrophilicity than the pristine membranes,
they would not give a higher advantage in flux to the coated membranes if there were a
drastically decreased in membrane porosity. These results are also similar to the previous
study, wherein they reported that the flux of the coated membrane did not only affect by the
membrane hydrophilicity and they argue that even though the hydrophilicity increased,
the improved membranes can be in a disadvantage situation if the membranes’ pores
were blocked by the coated hydrophilic nanoparticles [49]. The second explanation that
also affecting the declination of the flux for Ti-Ben 90 is the interaction force between the
hydrophilic coated membrane and the feed water. This interaction force restricting the flow
of the feed water to pass through the membranes’ surface [50].

Table 2. Average pore size, porosity and coating thickness of the prepared membranes.

Membrane Average Pore Size (µm) Porosity (%) Coating Thickness (µm)
Pristine Ben 1.75 62 -
Ti-Ben 30 0.92 61 10
Ti-Ben 60 0.89 58 12
Ti-Ben 90 0.63 45 23
MembranesMembranes
2021, 11, 2021,
x FOR PEER REVIEW
11, 739 12 of 18
13 of 1

Figure 10. Pure water flux performance of pristine bentonite membrane, Ti-Ben 30, Ti-Ben 60, and Ti-Ben 90.
Figure 10. Pure water flux performance of pristine bentonite membrane, Ti-Ben 30, Ti-Ben 60, and Ti-Ben 90.
The oil rejection test was conducted to evaluate the oil removal capacity of the mem-
Tablebranes.
2. Average pore
Figure size, porosity
11 shows and coating
the permeate thickness
flux and of the
oil removal preparedobtained
efficiency membranes.
for the
pristine bentonite membranes and coated bentonite membranes (Ti-Ben 30, Ti-Ben 60 and
Ti-Ben Averageflux
90). The permeate Pore
andSize
oil rejection through pristine bentonite
Membrane − 1 − 2 − 1
Porosity (%) Coating membranes
Thickness (µm)
was 183.15 h m bar (µm) and 95%, respectively. The high oil rejection in the pristine
bentoniteBen
Pristine membrane was further1.75 enhanced in all the62 coated membranes reporting -97, 98.5
and 99% for Ti-Ben 30, Ti-Ben 60 and Ti-Ben 90; respectively. As expected, the oil rejection
Ti-Ben 30 0.92 61
efficiencies were increased as grafting time of TiO2 on the membrane surface increased.
10
Ti-Ben
This60can be explained 0.89 58 point is the pore size of the
by several points. The first 12coated
Ti-Ben 90 whereas the oil
membranes, 0.63
particle size (1 µm) is 45
more than the pore size of the 23coated
membranes (0.95, 0.89 and 0.63 µm), as shown in Tables 2 and 3. Thus, this will give
an advantage for the coated membrane surface to not allow the oil particle to enter the
The oil rejection test was conducted to evaluate the oil removal capacity of the mem
membrane. Secondly, the zeta potential of the TiO2 particles was measured between −33 to
branes.
−37.8Figure
mV as 11
shownshows the permeate
in Figure 12. Since TiOflux and oil removal efficiency obtained for th
2 nanoparticles is a negatively-charge particles,
pristine
hence bentonite membranes
TiO2 nanoparticles on theand coated bentonite
membranes’ membranes
surface plays (Ti-Ben
a role in repelling 30, Ti-Ben 60 an
the negatively-
Ti-Ben 90). The
charged permeate
oil droplets which flux and oilaround
measured −5 tothrough
rejection −6 mV aspristine
shown in bentonite membranes
Table 3, which also wa
resulting in preventing the oil particle to enter the pores of the membranes
183.15 h m ² bar and 95%, respectively. The high oil rejection in the pristine bentonit
−1 − −1 [51], thereby
increasing the oil rejection performance.
membrane was further enhanced in all the coated membranes reporting 97, 98.5 and 99%
for Ti-Ben 30, Ti-Ben 60 and Ti-Ben 90; respectively. As expected, the oil rejection efficien
cies were increased as grafting time of TiO2 on the membrane surface increased.
 Membranes
Membranes 11, 11,
2021,
2021, 739 PEER REVIEW
x FOR 13 of 1814 of 19

Figure 11. Permeate flux and oil removal performance of pristine bentonite membrane, Ti-Ben 30, Ti-Ben 60, and Ti-Ben 90.
Figure 11. Permeate flux and oil removal performance of pristine bentonite membrane, Ti-Ben 30, Ti-Ben 60, and Ti-Ben
90. Table 3. Properties of the synthetic PW.

Parameter Value Unit

This can be explained by several points. The first point is the pore size of the coated
Size of oil droplets 1.0 µm
membranes, whereas pH the oil particle size (16 µ m) is more than the pore - size of the coated
membranes Zeta(0.95, 0.89 and 0.63 µ m), as−5shown
potential to −6 in Tables 2 and 3. Thus, mV this will give an
advantage for the coated membrane surface to not allow the oil particle to enter the mem-
brane.Figure
Secondly,
13 showsthe the
zetalong-term
potential of the TiO
permeate particles of
flux2evaluation wasthemeasured
TiO2 coatedbetween
bentonite−33 to
−37.8
membrane with optimum grafting time which is Ti-Ben 30. It can be seen that Ti-Benparticles,
mV as shown in Figure 12. Since TiO 2 nanoparticles is a negatively-charge 30
hence
able toTiO 2 nanoparticles
maintain on the
the flux around L h−1 m− 2 bar
300membranes’ −1 withplays
surface good a role inbetween
rejection repelling the nega-
98–99%
tively-charged oil droplets
of oil as the permeate which
sample weremeasured around
analyzed every −5 toas−6shown
60 min, mV as in shown
Table 4.inThis
Table 3,
indicate that TiO
which also resulting 2 coated membrane with optimum grafting time have the potential
in preventing the oil particle to enter the pores of the membranes for
a long-term
[51], thereby PW treatment.
increasing theTable 5 showsperformance.
oil rejection the properties of Ti-Ben 30 before filtration, in
first 3 h and at the end of filtration. The result showed the differences results of OCA
and zeta potential, which indicate that the TiO2 nanoparticle might leaching out a bit
during the long-term filtration. This particular study about the long-term evaluation can
be studied and discussed in the future on the durability of the TiO2 nanoparticle of the
membrane surface.

Table 4. Permeate sample taken for every 60 min.

Time (min) Rejection of Oil (%)

60 98
120 98
180 98.5
240 98.5
300 99
360 99
 Membranes 2021, 11, 739 14 of 18

Table 5. Properties of Ti-Ben 30 in first 3 h and at the of filtration.

Membranes 2021, 11, x FOR PEER REVIEW
Properties Before Filtration In First 3 h At the End of Filtration
OCA 147◦ 147◦ 130◦
Zeta potential −33 mV −33 mV −30 mV

Figure 12. Zeta potential distribution of (a)Ti-Ben 30, (b) Ti-Ben 60, and (c) Ti-Ben 90.
Figure 12. Zeta potential distribution of (a)Ti-Ben 30, (b) Ti-Ben 60, and (c) Ti-B

Table 3. Properties of the synthetic PW.

Parameter Value
Size of oil droplets 1.0
Membranes 2021, 11, x FOR PEER REVIEW 16 of 19

Membranes 2021, 11, 739 15 of 18

and discussed in the future on the durability of the TiO2 nanoparticle of the membrane
surface.

Figure 13. Permeate flux of Ti-Ben 60 for long-term evaluation.

Figure 13. Permeate flux of Ti-Ben 60 for long-term evaluation.

4. Conclusions
Table 4. Permeate sample taken for every 60 min.
TiO2 coated bentonite membranes were successfully prepared via the sol-gel dip
coating technique Time
using(min)
TiO2 nanoparticles. The technique Rejection
of the of Oil (%) of TiO2
deposition
nanoparticles was confirmed 60 via the characterization results. With the 98 increasing of graft-
ing time of TiO2 nanoparticles,
120 the concentration of TiO2 nanoparticles 98on the membrane’s
surface were increased 180 as results shown in SEM results. However,98.5when the grafting
time increased to 90 240 min, the TiO2 started to disturb the membranes’ 98.5pores. The results
also reported that the TiO2 distribution on the membranes’ surface was not fully-coated,
300 99
however, it did not affect the performance of the membrane since the results displayed
360 99
higher oil rejection and pure water flux than the pristine bentonite membranes. The Ti-Ben
60 with grafting time of 60 min displayed an outstanding pure water flux performance with
Table 5. Properties
67% increment of Ti-Ben
from 30 in first
the pristine 3 h andmembrane.
bentonite at the of filtration.
Ti-Ben 60 also was selected as the
best performance
Properties compared to pristine
Before Filtration and other TiO coated
In First 23 h membranes
At the End as of
it had higher
Filtration
oil rejection performance of 98.5%. Meanwhile, as for the contact angle results, the deposi-
OCA 147° 147° 130°
tion of TiO2 nanoparticles on the membranes’ surface allowing the TiO2 coated bentonite
Zeta potential −33 mV −33 mV
membranes to switch the wettability from hydrophilic to underwater superhydrophilic −30 mV
and superoleophobic properties in underwater condition. It could be concluded that the
4. Conclusions
surface coating of bentonite membranes with TiO2 nanoparticles has great potential for the
removal
TiOof oil frombentonite
2 coated industrialmembranes
wastewaters,were especially PW.
successfully prepared via the sol-gel dip
coating technique using TiO2 nanoparticles. The technique of the deposition of TiO2 nano-
Author
particlesContributions:
was confirmed Conceptualization, M.I.M.E. and A.L.A.;
via the characterization results.methodology, M.I.M.E. andofA.L.A.;
With the increasing grafting
validation, M.I.M.E. and A.L.A.; formal analysis, M.I.M.E. and A.L.A.; investigation, M.I.M.E., A.L.A.;
time of TiO2 nanoparticles, the concentration of TiO2 nanoparticles on the membrane’s
resources, A.L.A.; data curation, M.I.M.E.; writing—original draft preparation, M.I.M.E.; writing—
surface were increased as results shown in SEM results. However, when the grafting time
review and editing, M.I.M.E. and A.L.A.; visualization, M.I.M.E.; supervision, A.L.A. and M.H.D.O.;
increased to 90 min,A.L.A.;
project administration, the TiO 2 started
funding to disturb
acquisition, the
A.L.A. Allmembranes’ pores.
authors have read andThe results
agreed to thealso
reported that the TiO distribution
published version of the manuscript.
2 on the membranes’ surface was not fully-coated, how-
ever, it did not affect the performance of the membrane since the results displayed higher
Funding: This research was funded by the Long-Term Research Grant Scheme (LRGS/1/2018/USM/
oil rejection
01/1/4) and pure water flux
(203/PJKIMIA/67215002), than
which the pristine
is supported by thebentonite
Ministry ofmembranes.
Higher EducationTheMalaysia.
Ti-Ben 60
with grafting time of 60 min displayed
The APC was funded by Universiti Sains Malaysia. an outstanding pure water flux performance with
67% increment from the pristine bentonite membrane. Ti-Ben 60 also was selected as the
Institutional Review Board Statement: Not applicable.
best performance compared to pristine and other TiO2 coated membranes as it had higher
Informed
oil Consent
rejection Statement:ofNot
performance applicable.
98.5%. Meanwhile, as for the contact angle results, the depo-
sition of TiO2 nanoparticles
Data Availability Statement: Not onapplicable.
the membranes’ surface allowing the TiO2 coated bentonite
membranes to switch the wettability from hydrophilic to underwater superhydrophilic
and superoleophobic properties in underwater condition. It could be concluded that the
Membranes 2021, 11, 739 16 of 18

Acknowledgments: Authors would like to thank School of Chemical Engineering, Universiti Sains
Malaysia for providing lab space and facilities to conduct the research.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Ahmad, N.A.; Goh, P.S.; Yogarathinam, L.T.; Zulhairun, A.K.; Ismail, A.F. Current advances in membrane technologies for
produced water desalination. Desalination 2020, 493, 114643. [CrossRef]
2. Al-Ghouti, M.A.; Al-Kaabi, M.A.; Ashfaq, M.Y.; Da’na, D.A. Produced water characteristics, treatment and reuse: A review. J.
Water Process Eng. 2019, 28, 222–239. [CrossRef]
3. Rahman, A.; Agrawal, S.; Nawaz, T.; Pan, S.; Selvaratnam, T. A review of algae-based produced water treatment for biomass and
biofuel production. Water 2020, 12, 2351. [CrossRef]
4. Rosell-Melé, A.; Moraleda-Cibrián, N.; Cartró-Sabaté, M.; Colomer-Ventura, F.; Mayor, P.; Orta-Martínez, M. Oil pollution in soils
and sediments from the Northern Peruvian Amazon. Sci. Total Environ. 2018, 610–611, 1010–1019. [CrossRef] [PubMed]
5. Ossai, I.C.; Ahmed, A.; Hassan, A.; Hamid, F.S. Remediation of soil and water contaminated with petroleum hydrocarbon: A
review. Environ. Technol. Innov. 2020, 17, 100526. [CrossRef]
6. Nasiri, M.; Jafari, I.; Parniankhoy, B. Oil and gas produced water management: A review of treatment technologies, challenges,
and opportunities. Chem. Eng. Commun. 2017, 204, 990–1005. [CrossRef]
7. Nasiri, M.; Jafari, I. Produced water from oil-gas plants: A short review on challenges and opportunities. Period. Polytech. Chem.
Eng. 2017, 61, 73–81. [CrossRef]
8. Kusworo, T.D.; Aryanti, N.; Utomo, D.P. Oilfield produced water treatment to clean water using integrated activated carbon-
bentonite adsorbent and double stages membrane process. Chem. Eng. J. 2018, 347, 462–471. [CrossRef]
9. Ishak, K.E.H.K.; Ayoub, M.A. Predicting the efficiency of the oil removal from surfactant and polymer produced water by using
liquid–liquid hydrocyclone: Comparison of prediction abilities between response surface methodology and adaptive neuro-fuzzy
inference system. IEEE Access 2019, 7, 179605–179619. [CrossRef]
10. Sosa-Fernandez, P.; Post, J.; Leermakers, F.; Rijnaarts, H.; Bruning, H. Removal of divalent ions from viscous polymer-flooding
produced water and seawater via electrodialysis. J. Membr. Sci. 2019, 589, 117251. [CrossRef]
11. Zhang, Z.; Du, X.; Carlson, K.H.; Robbins, C.A.; Tong, T. Effective treatment of shale oil and gas produced water by membrane
distillation coupled with precipitative softening and walnut shell filtration. Desalination 2019, 454, 82–90. [CrossRef]
12. Fard, A.K.; Mckay, G.; Chamoun, R.; Rhadfi, T.; Preud’Homme, H.; Atieh, M.A. Barium removal from synthetic natural and
produced water using MXene as two dimensional (2-D) nanosheet adsorbent. Chem. Eng. J. 2017, 317, 331–342. [CrossRef]
13. Hedar, Y. Pollution impact and alternative treatment for produced water. In Proceedings of the E3S Web of Conferences, EDP
Sciences, Semarang, Indonesia, 15−16 August 2017; p. 03004.
14. Rawlins, C.H.; Sadeghi, F. Experimental study on oil removal in nutshell filters for produced-water treatment. SPE Prod. Oper.
2018, 33, 145–153.
15. Khalifa, O.; Banat, F.; Srinivasakannan, C.; Radjenovic, J.; Hasan, S.W. Performance tests and removal mechanisms of aerated
electrocoagulation in the treatment of oily wastewater. J. Water Process Eng. 2020, 36, 101290. [CrossRef]
16. Jiménez, S.; Micó, M.M.; Arnaldos, M.; Ferrero, E.; Malfeito, J.J.; Medina, F.; Contreras, S. Integrated processes for produced water
polishing: Enhanced flotation/sedimentation combined with advanced oxidation processes. Chemosphere 2017, 168, 309–317.
[CrossRef] [PubMed]
17. Bagheri, M.; Roshandel, R.; Shayegan, J. Optimal selection of an integrated produced water treatment system in the upstream of
oil industry. Process Saf. Environ. Prot. 2018, 117, 67–81. [CrossRef]
18. Shamaei, L.; Khorshidi, B.; Perdicakis, B.; Sadrzadeh, M. Treatment of oil sands produced water using combined electrocoagulation
and chemical coagulation techniques. Sci. Total Environ. 2018, 645, 560–572. [CrossRef]
19. Al Hawli, B.; Benamor, A.; Hawari, A.A. A hybrid electro-coagulation/forward osmosis system for treatment of produced water.
Chem. Eng. Process. Process Intensif. 2019, 143, 107621. [CrossRef]
20. Santos, A.S.; Oliveira, L.F.S.; Marques, A.M.; Silva, D.C.; Mansur, C.R. Evaluation of the efficiency of polyethylenimine as
flocculants in the removal of oil present in produced water. Colloids Surf. A Physicochem. Eng. Asp. 2018, 558, 200–210. [CrossRef]
21. Jain, P.; Sharma, M.; Dureja, P.; Sarma, P.M.; Lal, B. Bioelectrochemical approaches for removal of sulfate, hydrocarbon and
salinity from produced water. Chemosphere 2017, 166, 96–108. [CrossRef]
22. Lusinier, N.; Seyssiecq, I.; Sambusiti, C.; Jacob, M.; Lesage, N.; Roche, N. A comparative study of conventional activated sludge
and fixed bed hybrid biological reactor for oilfield produced water treatment: Influence of hydraulic retention time. Chem. Eng. J.
2021, 420, 127611. [CrossRef]
23. Chang, H.; Li, T.; Liu, B.; Vidic, R.D.; Elimelech, M.; Crittenden, J.C. Potential and implemented membrane-based technologies
for the treatment and reuse of flowback and produced water from shale gas and oil plays: A review. Desalination 2019, 455, 34–57.
[CrossRef]
24. Zoubeik, M.; Ismail, M.; Salama, A.; Henni, A. New developments in membrane technologies used in the treatment of produced
water: A review. Arab. J. Sci. Eng. 2018, 43, 2093–2118. [CrossRef]
25. Jepsen, K.L.; Bram, M.V.; Pedersen, S.; Yang, Z. Membrane fouling for produced water treatment: A review study from a process
control perspective. Water 2018, 10, 847. [CrossRef]
Membranes 2021, 11, 739 17 of 18

26. Atallah, C.; Tremblay, A.Y.; Mortazavi, S. Silane surface modified ceramic membranes for the treatment and recycling of SAGD
produced water. J. Pet. Sci. Eng. 2017, 157, 349–358. [CrossRef]
27. Weschenfelder, S.E.; Fonseca, M.J.C.; Costa, B.R.S.; Borges, C.P. Influence of the use of surfactants in the treatment of produced
water by ceramic membranes. J. Water Process Eng. 2019, 32, 100955. [CrossRef]
28. Kayvani Fard, A.; McKay, G.; Buekenhoudt, A.; Al Sulaiti, H.; Motmans, F.; Khraisheh, M.; Atieh, M. Inorganic membranes:
Preparation and application for water treatment and desalination. Materials 2018, 11, 74. [CrossRef] [PubMed]
29. Aziz, M.H.A.; Othman, M.H.D.; Hashim, N.A.; Rahman, M.A.; Jaafar, J.; Hubadillah, S.K.; Tai, Z.S. Pretreated aluminium dross
waste as a source of inexpensive alumina-spinel composite ceramic hollow fibre membrane for pretreatment of oily saline
produced water. Ceram. Int. 2019, 45, 2069–2078. [CrossRef]
30. Usman, J.; Othman, M.H.; Ismail, A.F.; Rahman, M.A.; Jaafar, J.; Raji, Y.O.; El Badawy, T.H.; Gbadamosi, A.O.; Kurniawan, T.A.
Impact of organosilanes modified superhydrophobic-superoleophilic kaolin ceramic membrane on efficiency of oil recovery from
produced water. J. Chem. Technol. Biotechnol. 2020, 95, 3300–3315. [CrossRef]
31. Twibi, M.F.; Othman, M.H.D.; Hubadillah, S.K.; Alftessi, S.A.; Kurniawan, T.A.; Ismail, A.F.; Rahman, M.A.; Jaafar, J.; Raji, Y.O.
Development of high strength, porous mullite ceramic hollow fiber membrane for treatment of oily wastewater. Ceram. Int. 2021,
47, 15367–15382. [CrossRef]
32. Hu, X.; Yu, Y.; Zhou, J.; Wang, Y.; Liang, J.; Zhang, X.; Chang, Q.; Song, L. The improved oil/water separation performance of
graphene oxide modified Al2 O3 microfiltration membrane. J. Membr. Sci. 2015, 476, 200–204. [CrossRef]
33. Raji, Y.O.; Othman, M.H.D.; Nordin, N.A.H.S.M.; Kurniawan, T.A.; Ismail, A.F.; Rahman, M.A.; Jaafar, J.; Adam, M.R.B.; Alftessi,
S.A.; El-badawy, T.; et al. Wettability improvement of ceramic membrane by intercalating nano-Al2 O3 for oil and water separation.
Surf. Interfaces 2021, 25, 101178. [CrossRef]
34. Li, Y.-X.; Cao, Y.; Wang, M.; Xu, Z.-L.; Zhang, H.-Z.; Liu, X.-W.; Li, Z. Novel high-flux polyamide/TiO2 composite nanofiltration
membranes on ceramic hollow fibre substrates. J. Membr. Sci. 2018, 565, 322–330. [CrossRef]
35. Abd Aziz, M.H.; Dzarfan Othman, M.H.; Alias, N.H.; Nakayama, T.; Shingaya, Y.; Hashim, N.A.; Kurniawan, T.A.; Matsuura,
T.; Rahman, M.A.; Jaafar, J. Enhanced omniphobicity of mullite hollow fiber membrane with organosilane-functionalized TiO2
micro-flowers and nanorods layer deposition for desalination using direct contact membrane distillation. J. Membr. Sci. 2020, 607,
118137. [CrossRef]
36. Magnone, E.; Kim, M.K.; Lee, H.J.; Park, J.H. Facile synthesis of TiO2 -supported Al2 O3 ceramic hollow fiber substrates with
extremely high photocatalytic activity and reusability. Ceram. Int. 2021, 47, 7764–7775. [CrossRef]
37. Raji, Y.O.; Othman, M.H.D.; Md Nordin, N.A.H.; Ismail, A.; Rahman, M.A.; Jaafar, J.; Usman, J.; Momah, S.C. Preparation and
Characterization of Bentonite-Based Ceramic Hollow Fiber Membrane. J. Membr. Sci. Res. 2021, 7, 95–101.
38. Jamalludin, M.R.; Hubadillah, S.K.; Harun, Z.; Othman, M.H.D.; Yunos, M.Z.; Ismail, A.F.; Salleh, W.N.W. Facile fabrication of
superhydrophobic and superoleophilic green ceramic hollow fiber membrane derived from waste sugarcane bagasse ash for
oil/water separation. Arab. J. Chem. 2020, 13, 3558–3570. [CrossRef]
39. Guo, H.; Zhao, S.; Wu, X.; Qi, H. Fabrication and characterization of TiO2 /ZrO2 ceramic membranes for nanofiltration. Microporous
Mesoporous Mater. 2018, 260, 125–131. [CrossRef]
40. Zhang, Q.; Wang, H.; Fan, X.; Lv, F.; Chen, S.; Quan, X. Fabrication of TiO2 nanofiber membranes by a simple dip-coating
technique for water treatment. Surf. Coat. Technol. 2016, 298, 45–52. [CrossRef]
41. Ahmad, R.; Lee, C.S.; Kim, J.H.; Kim, J. Partially coated TiO2 on Al2 O3 membrane for high water flux and photodegradation by
novel filtration strategy in photocatalytic membrane reactors. Chem. Eng. Res. Des. 2020, 163, 138–148. [CrossRef]
42. Li, X.; Chen, Y.; Hu, X.; Zhang, Y.; Hu, L. Desalination of dye solution utilizing PVA/PVDF hollow fiber composite membrane
modified with TiO2 nanoparticles. J. Membr. Sci. 2014, 471, 118–129. [CrossRef]
43. Yong, J.; Chen, F.; Yang, Q.; Farooq, U.; Hou, X. Photoinduced switchable underwater superoleophobicity–superoleophilicity on
laser modified titanium surfaces. J. Mater. Chem. A 2015, 3, 10703–10709. [CrossRef]
44. Zhan, H.; Peng, N.; Lei, X.; Huang, Y.; Li, D.; Tao, R.; Chang, C. UV-induced self-cleanable TiO2 /nanocellulose membrane for
selective separation of oil/water emulsion. Carbohydr. Polym. 2018, 201, 464–470. [CrossRef] [PubMed]
45. Sun, T.; Liu, Y.; Shen, L.; Xu, Y.; Li, R.; Huang, L.; Lin, H. Magnetic field assisted arrangement of photocatalytic TiO2 particles on
membrane surface to enhance membrane antifouling performance for water treatment. J. Colloid Interface Sci. 2020, 570, 273–285.
[CrossRef] [PubMed]
46. Li, R.; Ren, Y.; Zhao, P.; Wang, J.; Liu, J.; Zhang, Y. Graphitic carbon nitride (g-C3N4) nanosheets functionalized composite
membrane with self-cleaning and antibacterial performance. J. Hazard. Mater. 2019, 365, 606–614. [CrossRef] [PubMed]
47. Li, N.; Chen, G.; Zhao, J.; Yan, B.; Cheng, Z.; Meng, L.; Chen, V. Self-cleaning PDA/ZIF-67@PP membrane for dye wastewater
remediation with peroxymonosulfate and visible light activation. J. Membr. Sci. 2019, 591, 117341. [CrossRef]
48. Chen, X.; Huang, G.; An, C.; Feng, R.; Yao, Y.; Zhao, S.; Huang, C.; Wu, Y. Plasma-induced poly(acrylic acid)-TiO2 coated
polyvinylidene fluoride membrane for produced water treatment: Synchrotron X-Ray, optimization, and insight studies. J. Clean.
Prod. 2019, 227, 772–783. [CrossRef]
49. Li, C.; Sun, W.; Lu, Z.; Ao, X.; Yang, C.; Li, S. Systematic evaluation of TiO2 -GO-modified ceramic membranes for water treatment:
Retention properties and fouling mechanisms. Chem. Eng. J. 2019, 378, 122138. [CrossRef]
Membranes 2021, 11, 739 18 of 18

50. Marzouk, S.S.; Naddeo, V.; Banat, F.; Hasan, S.W. Preparation of TiO2 /SiO2 ceramic membranes via dip coating for the treatment
of produced water. Chemosphere 2021, 273, 129684. [CrossRef]
51. Shah, A.H.; Rather, M.A. Effect of calcination temperature on the crystallite size, particle size and zeta potential of TiO2
nanoparticles synthesized via polyol-mediated method. Mater. Today Proc. 2021, 44, 482–488. [CrossRef]