Separation and Puri cation Technology 361 (2025) 131435

Contents lists available at ScienceDirect

Separation and Purification Technology

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

Enhanced performance of novel nano-enabled PVDF membranes with

amino and chloro-based clay for membrane distillation
Toheeb O. Obidara a , Mohammed Abdul Azeem b,c , Dahiru U. Lawal b,c, Konstantinos Spyrou d ,
Sakavitsi Viktoria d , Md Emdad Hossain a , Turki N. Baroud a,b,*
a
Department of Material Science and Engineering, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia
b
Interdisciplinary Research Center for Membranes & Water Security, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia
c
Department of Mechanical Engineering, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia
d
Department of Materials Science & Engineering, University of Ioannina, 45110 Ioannina, Greece

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

Editor: C. Marangoni Recent advancements in membrane distillation have prompted considerable interest in enhancing polymeric
membrane performance through various surface modification processes. In this study, we reported the synthesis
Keywords: of novel polyvinylidene fluoride (PVDF) composite membranes composed of amino (Am(2)-SCA) and chloro-
Composite membranes based (Ch-SCA) clay nanoparticles (CNPs). The sol–gel chemical method was used to synthesize the CNPs, and
Clay nanoparticles
the composite membranes were fabricated from amino-based CNPs (i.e., hydrophilic membranes) and chloro-
Hydrophilic
based CNPs (i.e., hydrophobic membranes), with the addition of various compositions of the CNPs in the
Hydrophobic
Membrane distillation PVDF polymer matrix. The performance of the developed membranes was evaluated by air gap membrane
Desalination distillation (AGMD) using a highly saline feed solution (70 gL− 1 NaCl). The inclusion of CNPs in the polymer
significantly improved the performance of PVDF membranes. Notably, the hydrophobic composite membranes
demonstrated excellent performance with high flux and stable rejection, while the hydrophilic composite
membranes exhibited moderate permeation flux at lower CNPs concentrations (0.50 wt%) and higher flux with a
corresponding decrease in rejection as the concentration increased. The composite membrane containing 1 wt%
hydrophobic CNPs displayed the best AGMD performance with a permeate flux of 12.80 kg m− 2 h− 1 and 99.99 %
salt rejection. The superior performance of the composite membranes was ascribed to the effective dispersion of
CNPs in the membrane matrix and improved membrane properties suitable for efficient desalination in MD.

1. Introduction materials for synthesizing MD membranes include polyethylene (PE),

polypropylene (PP) [11], PVDF [12,13], polyether sulfone [14] , and
The increasing population, industrialization, and climate change polyimides [15,16]. PVDF has been a polymer of choice and is widely
have all contributed to the issue of water scarcity, necessitating an ur­ used for membrane distillation because of its good thermal stability,
gent need for a sustainable clean water supply [1]. Seawater desalina­ high organic selectivity, good thermal resistance, and membrane-
tion and wastewater treatment are considered efficient purification forming ability with different solvents [17,18]. Membrane distillation
processes for producing freshwater using membrane-based desalination faces significant challenges such as, pore wetting, scaling, and fouling
methods [2]. These methods include reverse osmosis (RO), filtration, after continuous operation, which affects membrane efficiency [19,20].
pervaporation, and membrane distillation (MD) [3,4,5,6] Membrane Pore wetting affects membrane separation efficiency, which is a major
distillation is a highly efficient technology that integrates both thermal issue faced during MD test. Hence, there is need to enhance polymeric
and membrane processes and can be employed in food, bioprocessing, membrane properties to achieve efficient membrane distillation per­
highly saline water and wastewater treatment [7,8,9]. A microporous formance [21]. Optimization of MD membrane properties is needed to
hydrophobic membrane is usually synthesized for MD applications, prevent pore wetting by maintaining a balance between the liquid entry
allowing only vapor transport from the feed stream through a porous pressure and the pore structure [22]. Numerous efforts have been made
layer before being collected as permeate [10]. Frequently used polymer to enhance the properties of PVDF membranes by incorporating

* Corresponding author.
E-mail address: [email protected] (T.N. Baroud).

https://doi.org/10.1016/j.seppur.2025.131435
Received 1 July 2024; Received in revised form 28 December 2024; Accepted 2 January 2025
Available online 3 January 2025
1383-5866/© 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
T.O. Obidara et al. Separation and Puri cation Technology 361 (2025) 131435

nanoparticles and functionalizing their surface to reduce their suscep­ precursor to 50 mL of an ethanolic solution of MgCl2 (183 mM) at room
tibility to scaling and pore wetting [23]. The addition of nanoparticles temperature. Subsequently, to achieve sol–gel transition, the pH of the
has been used to enhance the surface features of the membranes for sol was changed to basic conditions (pH ~ 10.4), using an aqueous so­
better performance, such as activated carbon, MXenes, carbon nano­ lution of NaOH (1 M). It is worth mentioning that the gelation point was
tubes, TiO2, SiO2, and graphene oxide [24,25]. Nanomaterials enhance strongly influenced by the type of silane precursor used. The resulting
membrane performance by reducing pore wetting and increasing dispersion was stirred at room temperature to obtain a white slurry,
permeate flux [26,27]. Ramjou et al. [28] employed a low-temperature washed with ethanol, and dried overnight at 40 ◦ C. Poorly ordered
hydrothermal method to synthesize a superhydrophobic PVDF mem­ gelled aggregates of nanoclay sheets were obtained after drying. The
brane via TiO2 nanoparticle coating and fluorosalination. clay sheets were exfoliated in deionized water, followed by an equal
Clay nanoparticles (CNPs) are widely utilized to improve membrane volume of ethanol, then centrifuged and dried overnight at 40 ◦ C to
characteristics because of their abundant nature and inexpensive or improve their ordering.
facile synthesis methods for applications including gas separations [29],
membrane distillation [30], and ultrafiltration (UF) [31,32]. Pramit 2.2.2. Preparation of Chloro-Based CNPs Analogous (Ch-SCA)
et al. [33] synthesized a cost-effective UF poly composite membrane by Chloro-synthetic clay was synthesized using the same procedure for
incorporating sodium-based nanoclay particles through phase inversion Am(2)-SCA using sol–gel method. The pH of the sol was changed to a
to treat water pollutants. Hesavi et al. [34] synthesized composite basic condition (pH ~ 8.6). Additionally, for hydrophobic modification,
membranes from PVDF/Al2O3, TiO2, and clay nanoparticles for indus­ CPTMOS was used as a silane precursor.
trial wastewater removal of toxic metals. The composite membranes
demonstrated superior adsorption of lead and arsenic ions. Clay nano­ 2.3. Fabrication of PVDF- clay composite membranes
particles have also been used in membrane distillation. Khemakhem
et al.[35] synthesized clay ceramic composite membranes through the The pristine PVDF membrane, denoted P0 was prepared by adding
grafting of perfluorodecytriethoxysilane for synthetic and seawater PVDF polymer (15 wt%) into DMF under stirring for 12 h, followed by
desalination in AGMD. Lai et al. [36] synthesized composite PVDF the addition of PVP (5 wt%) and further stirring until homogeneity was
membranes with enhanced mechanical strength by incorporating cloi­ reached. Composite membranes were prepared by adding the CNPs to
site clay nanoparticles. Prince et al. [37] prepared an electrospun the solution. The hydrophobic (Ch-SCA) or hydrophilic clay particles
nanofiber PVDF composite membrane by blending clay nanoparticles for (Am(2)-SCA) were dissolved in DMF and mixed using a probe sonicator
membrane distillation. Monticelli et al. [38] reported the preparation of (Vibra-cell processors, obtained from Sonics & Materials, Inc.) for 15
nanocomposite membranes based on polysulfone and commercial cloi­ min to obtain a uniform dispersion of the particles. The dispersed clay
site clay particles via a phase inversion process. The composite mem­ solution was added to the polymer solution at different concentrations
branes are composed of clay particles homogeneously dispersed, (i.e., 0.5, 1, 3, and 5 wt%). The mixed polymer–clay solution was stirred
resulting in porous membranes. for 12 h and maintained at room temperature for 6 h without stirring to
Owing to the numerous benefits associated with clay nanoparticles, remove any bubbles before pouring the solution on a glass slide. A flat
this study focused on synthesizing novel PVDF composite membranes. thin film was made with a thickness of 250 μm using an adjustable
Two-layered synthetic clays with different functionalities on the surface casting blade and immediately immersed into a coagulation bath of DI
and in the interior of the lamellar structure were synthesized. The first water/isopropanol (80:20) at 30◦ C for 10 min to obtain a uniform solid
synthetic clay possesses amine functionalities, which gives the clay a phase membrane through the nonsolvent-solvent exchange [39]. This
hydrophilic character with flexible chains that may play a crucial role in was repeated for all the polymer-clay solutions and the resulting mem­
the dispersibility of the clay in the final membrane. Conversely, chloro- branes were stored in DI water to eliminate any residual solvent, after
based CNPs were synthesized with hydrophobic behavior and less flex­ which they were dried under filter paper for 24 h. The control membrane
ibility, enabling the hydrophobic PVDF to encapsulate it with its hy­ from PVDF polymer was termed as P0, whereas PVDF incorporated with
drophobic characteristics. Furthermore, the concentration of the CNPs amino-based CNPs at different compositions of 0.5,1,3, and 5 wt% were
was analyzed on the structural effects of the composite membranes for termed hydrophilic composite membranes designated as P1-P4 and the
optimal MD performance using various characterization techniques to chloro-based CNPs termed hydrophobic composite membranes equiva­
explain their morphology, porous nature, wettability, and surface lent to P5-P8 (Table 1).
roughness. Finally, the composite membranes were evaluated in a high
saline synthetic feed (NaCl) containing 70 g L− 1 via AGMD design.
2.4. Characterization of clay nanoparticles and PVDF-CNPs composite
2. Materials and methods membranes

2.1. Materials The chemical structure of the CNPs was evaluated by FTIR (Fourier
transform infrared, iS50 Nicolet Thermo Fisher Scientific, Massachus­
Polyvinylidene difluoride (PVDF), polyvinylpyrrolidone (PVP), N, N- sets) spectroscopy in ATR mode (Attenuated Total Reflectance), using a
Dimethylformamide (DMF, 99.9 %), and 2-propanol was purchased diamond crystal for 32 scans in the range of 400–––4000 cm− 1 with a
from Sigma Aldrich. Magnesium chloride hexahydrate (MgCl2⋅6H2O)
was obtained from Riedel-de Haen; N-[3-(trimethoxysilyl) propyl]eth­ Table 1
ylenediamine (EDAPTOS) from Sigma Aldrich, (3-chloropropyl) trime­ Membrane Designation of PVDF/CNPs Composition.
thoxysilane (CPTMOS, 98 %) from Acros Organics; sodium chloride Membrane ID PVDF (wt.%) PVP Clay Nanoparticles Clay (wt.%)
(NaCl, 99.5 %) from Fisher Chemical, sodium hydroxide (NaOH) and (wt. %) (CNPs)
ethanol from Merck. The chemicals were used in this study as received. P0 15 5 No 0.0
P1 15 5 Hydrophilic 0.5
P2 15 5 Hydrophilic 1
2.2. Synthesis of clay nanoparticles
P3 15 5 Hydrophilic 3
P4 15 5 Hydrophilic 5
2.2.1. Preparation of Amino-Based CNPs Analogous (Am(2)-SCA) P5 15 5 Hydrophobic 0.5
The sol–gel chemical method was used for the synthesis process, P6 15 5 Hydrophobic 1
using EDAPTEOS (diamine) as the silane precursor. The sol phase of the P7 15 5 Hydrophobic 3
P8 15 5 Hydrophobic 5
clay colloid was produced by dropwise addition of 10 mmol of the

2
T.O. Obidara et al. Separation and Puri cation Technology 361 (2025) 131435

resolution of 4 cm− 1. Furthermore, a powder X-ray diffraction (XRD) the vapor flow (permeate) that is regulated by resistance to mass transfer
(D8 advance Brucker diffractometer) with a monochromatic Cu kα across the membrane thickness. The membrane thickness was measured
source (wavelength 1.5418 Å) was used to obtain the diffraction spectra using a high-precision thickness measuring instrument (LITEMATIC VL-
of the clay particles. A field emission scanning electron microscope (FE- 50A) that uses a sensitive probe. Each membrane sample, 2.5 cm long
SEM), (FEI Quanta 250 FEG, USA) was used to examine the surface was placed flat on the glass surface. The thickness measurement was
morphology and cross-section of the membranes. The magnification taken from 15 different areas of the membrane sample and the average
used for the surface morphology was ’10 000 x’, while ‘1000 x’ was used thickness was recorded and reported with standard deviations.
for the cross-section, except for P2-P4, and P7- P8 with ‘1500 x’ to point The mean pore size and pore size distribution of the membranes were
out the clay particles. The working distance ranged between 9.6 to 10.2 measured using Quantachrome 3Gzh porometer. The porosity (“ԑ”) of
mm for surface morphology and 11.4 to 11.8 mm for the cross-section. the composite membranes was evaluated following Eq. (1). The tortu­
Atomic force microscopy (AFM, Oxford instruments MFD-3D Infinity) osity of the membrane’s pore was estimated by the correlation in Eq. (2)
analysis was used to evaluate the surface roughness of the membranes by Mackie and Meares [40]. Srisurichan et al. [41] described the cor­
using a Si3N4 cantilever beam in tapping mode at a scanning rate of 0.8 relation as the most effective method for estimating the tortuosity of
Hz, with a resolution of 5 nm in the scanning range of 10 x 10 µm2. phase inversion membranes.
Thermogravimetric analysis (TGA) of the membranes was analyzed by a (ww − wd )
thermogravimetric analyzer (SDT Q600 V 20.9 build 20). The temper­ ε (%) = (
ρe
) (1)
ature ramp rate used for the samples was 10◦ C /min, under a nitrogen ww − wd
+ wd
gas atmosphere. ρe ρm

The contact angle equipment was used to measure the membranes’

water contact angle (WCA) by placing a 5 μL droplet of DI water onto the (2 − ε)2
τ= (2)
membrane surface. The droplets were allowed to sit for 3–4 s, then ε
images were captured using a high-resolution camera coupled with
software that can measure WCA. This was done repeatedly in 5 different Where, ww, wd, ρe, ρm and τ represent the weight of wet and dry
locations on the sample and the average was recorded. membrane, density of ethanol (789 kg m− 3), density of PVDF membrane
The liquid entry pressure of water (LEPw) of the membranes was (1740 kg m− 3), and membrane tortuosity, respectively.
determined using a laboratory dead-end cell setup. Each of the mem­
brane samples was placed in the cell and topped up with DI water, after 2.5. Membrane distillation performance Setups
which it was pressurized until a droplet of water was observed on the
membrane surface. The process was repeated 3 times, and the mean The performance of all synthesized composite membranes (P0-P8)
LEPw was recorded. In membrane distillation, membrane thickness was tested in a laboratory-scale air gap membrane distillation (AGMD)
plays a crucial role in the performance of the membrane, particularly in process (see Fig. 1). The setup is made up of a water heater (Julabo,

Fig. 1. Schematic of Air gap membrane distillation (AGMD) setup.

3
T.O. Obidara et al. Separation and Puri cation Technology 361 (2025) 131435

CORIO CD), chiller (G 50 Thermo scientific), flow meter, valves, con­ nanoparticles altered the pore structure with large pore networks
ductivity meter and weighing balance. The feed water was pumped from created in the composite membranes (P1-P4) during phase inversion in
the water heater to the membrane-contained cell. The feed flow was water. The finger-like structure increased with an increasing hydrophilic
controlled and monitored using a flowmeter, the vapor from the feed CNP loading, indicating an accelerated demixing process during phase
side was transported across the microporous membrane and condensed inversion due to the high water affinity of clay nanoparticles. A similar
on a plate as permeate. The amount of permeate passing through the structure was reported in another study, in which the hydrophilic par­
membrane was collected and measured using a weighing balance ticles aided in the nucleation process of enlarging the pore structure
(RADWAG Analytical balance) connected to the computer and measured during phase inversion [46]. The addition of hydrophobic CNPs also
in real time. The salt concentration of the permeate water was also resulted in an asymmetric structure but with more suppressed fingers (as
measured using a conductivity meter (HANNA HI5522). Table 2 shows seen in the cross-section of the composite membranes) which can be
the test conditions of the membrane distillation process. The permeate attributed to slower demixing during the phase inversion process [47].
flux and salt rejection were evaluated during the experiment using Eqns. An increase in solution viscosity due to the addition of nanoparticles into
(3) and (4), respectively. the polymer solution affects membrane structure, suppressing finger-
Permeate flux (J) = t.A
Δm
E
(kg m− 2h− 1) (3). like structures and causing slower demixing in hydrophobic composite
Cf − Cp membranes [48]. Traces of hydrophobic clay particles can be noticed in
Salt Rejection (S.R) = Cf (%) (4).
the image, as shown in Fig. 3 (P6, P7 and P8) indicating the incorpo­
Where Δm, t, AE , Cf and Cp represent, mass of the permeate collected
ration of the particles within the membrane. However, no traces of hy­
in kg, time in h, effective membrane area in m2, feed, and permeate
drophilic nanoparticles were detected, because some of the particles
concentration, respectively.
may have leached out during the phase inversion process.
The obtained roughness parameters from AFM analysis provided
3. Results and Discussions quantitative insights into the membrane surface morphology. Fig. 4a, b,
and c show the surface topography of 3 different membrane samples
3.1. Characterization of clay nanoparticles (PDVF-base membrane, P0, PVDF hydrophilic composite membrane, P4
and hydrophobic membrane, P8). The composite membranes exhibited
Fig. 2 shows the characterization of amino-based and chloro-based a higher surface roughness than P0 due to the presence of clay particles
CNPs. The SEM micrographs of CNPs in Fig. 2a and b, show a non- which, demonstrated a heterogeneous exchange mechanism during the
uniform flaky structure with random sizes ranging from 200 nm to 50 phase inversion process, resulting in a non-uniform surface [49]. The
µm with more agglomerates and rough surfaces observed in chloro- increase in the surface roughness of the composite membranes can be
based CNPs. Fig. 2c and d represent the elemental analysis of the seen by the roughness parameters, as shown in Fig. 4d. Surface rough­
CNPs, with chlorine identified in the chloro-based CNPs. Fig. 2e shows ness enhances membrane hydrophobicity and reduces the tendency to­
the FTIR spectra of CNPs with absorption bands at 450–550 cm− 1 and ward membrane wetting during the MD process [50]. The roughness
1000–1186 cm− 1 corresponding to Mg-O and Si-O-Si stretching vibra­ parameters of the hydrophobic clay membrane increased by more than
tions. Amino-based CNPs show an absorption band at 1550–1650 cm− 1 2-fold compared to those of the base membrane.
and can be ascribed to the bending vibrations of the amino group, while The surface chemical structures of the membranes were analyzed by
the spectrum of Chloro-based CNPs has a vibration band at 696 cm− 1 FTIR spectroscopy, as displayed in Fig. 5a. The chemical structure of the
and can be ascribed to C-Cl [42,43]. The functional groups observed in PVDF base shows CH2 asymmetric and symmetric vibrations at 3022
the FTIR spectrum verify the hydrophilic and hydrophobic affinity of cm− 1 and 2980 cm− 1, respectively, which can also be observed in the
CNPs used in this study. XRD patterns of the CNPs are displayed in composite membrane, confirming that the PVDF structure was main­
Fig. 2f. Both samples exhibit a sharp peak of spacing d001 around 5◦ tained in the composite membrane.
along with broad in-plane diffraction peaks d020, 110 in the range of In addition, a C–C bond was observed at 1185 cm− 1 for the base and
15–31◦ and d130,200 in the range of 32–40◦ [44]. The characteristic d060 composite membranes. The spectra at 1403 cm− 1 present in all three
diffraction peak at approximately 60◦ corresponds to 2:1 trioctahedral membranes were ascribed to CH2 wagging vibrations [51]. The peak
Mg-phyllosilicate clay. The characteristic (d060) diffraction peak around observed in the hydrophobic composite membrane at 696 cm− 1 is due to
2θ = 60◦ corresponding to the 2:1 trioctahedral smectite clay is weak the C-Cl bond in the clay particles. Thermogravimetric curves of the
due to the staking disorder induced by the organofunctional groups PVDF base membrane and composite membranes incorporating 1 wt%
[45]. The pattern of the amino-based CNPs sample shows a peak at 5.3◦ , CNPs (P0, P2, and P6) are displayed in Fig. 5b. The membranes exhibit
corresponding to a basal spacing of d001 = 16.6 Å while chloro-based distinct thermal stability patterns due to the alteration in their chemical
CNPs exhibited a peak at 6.5◦ , corresponding to a basal spacing of composition resulting from the addition of clay particles. As evidenced
d001 = 13.6 Å. by the curves, the initial decomposition temperature of the membranes
noted after 5 % weight loss was found to be higher (approximately
3.2. Characterization of fabricated membranes 400◦ C) in the case of PVDF composite membranes with CNPs addition as
compared to pristine PVDF membrane (approximately 360◦ C) indi­
SEM images of the surfaces and cross-sections of the fabricated cating higher thermal stability in the PVDF membranes incorporated
membranes P0–P8 are shown in Fig. 3. All membranes developed in this with CNPs. The overall weight loss % of the pristine PVDF (P0), PVDF
study have asymmetric structures with spongy surface morphologies hydrophilic (P2), and PVDF hydrophobic (P6) composite membranes
and a finger-like cross-section. The addition of hydrophilic clay were estimated as 78 %, 69 %, and 68 %, respectively. The weight de­
creases in PVDF-hydrophilic and PVDF-hydrophobic clay composite
Table 2 membranes differ because of the different organic chains of the pre­
Membrane Distillation Test Conditions. cursors in the CNPs that are N-[3-(trimethoxysilyl)propyl]ethylenedi­
MD Test conditions Values amine (EDAPTEOS, 97 %), for PVDF-hydrophilic clay and (3-
chloropropyl) trimethoxysilane (CPTMOS, 98+%) for hydrophobic clay.
Feed Salinity 70 g L− 1
Feed flowrate 0.8 ± 0.1 L min− 1
The PVDF-hydrophobic clay has a lower amount of carbon on the carbon
Feed Temperature 70 ± 0.8 ◦ C chain than the hydrophilic clay. Furthermore, the weight loss ratio as
Coolant flow rate 2.0 ± 0.1 L min− 1 depicted in Fig. 5 from the TGA analysis indicated higher weight loss in
Coolant temperature 20 ± 0.2 ◦ C the pristine PVDF membrane as compared to the PVDF hydrophilic clay
Effective membrane area 7.316 × 10− 4 m2
composite membrane. This is attributed to the presence of organic and

4
T.O. Obidara et al. Separation and Puri cation Technology 361 (2025) 131435

Fig. 2. Characterization of clay particles: (a and b) SEM micrographs of amino-based (Am(2)-SCA and chloro-based (Ch-SCA) CNPs; (c and d) Elemental composition
spectrum of amino-based and chloro-based CNPs; (e) FTIR spectra of amino-based and chloro-based; (f) XRD patterns of amino-based and chloro-based CNPs.

Fig. 3. SEM micrographs showing membrane surface morphology and cross-sectional SEM images of PVDF-base and PVDF clay composite membranes.

inorganic chains in the clay material. The precursor in hydrophilic clay i. alkyl radicals that may have formed during the decomposition of PVDF-
e. N-[3-(trimethoxysilyl)propyl]ethylenediamine (EDAPTEOS, 97 %) clay composite contributes to weight loss % [52,53,54,55]. These radi­
utilizes carbon and amino groups. The relatively high amount of HF and cals form char at the end of the decomposition stage resulting in less

5
T.O. Obidara et al. Separation and Puri cation Technology 361 (2025) 131435

Fig. 4. Atomic Force Microscopy showing 3D surface topography of (a) PVDF base membrane (b) Hydrophilic composite membrane (c) Hydrophobic composite
membrane and (d) Roughness parameters. (z-axis scale is in µm).

Fig. 5. A) fourier transform infrared (ftir) b) thermogravimetric analysis c) and d) pore size distribution measured by capillary porometer for pvdf-base and pvdf
composite membranes.

weight loss as compared to pristine PVDF membrane. Fig. S1 displays approximately 5 % wt. Moreover. The organic part of the hydrophilic
the TGA of the CNPs, the hydrophilic clay incorporated in the PVDF clay is estimated at 41 % wt. while for the hydrophobic clay, the
polymer possesses a higher amount of water (25–150 ◦ C), with a weight decomposition of the organic moieties is estimated at approximately 34
loss of 12 % wt. compared to hydrophobic clay with a weight loss of % wt.

6
T.O. Obidara et al. Separation and Puri cation Technology 361 (2025) 131435

The pore sizes of the pristine PVDF and composite membranes were influence the wettability of the membrane. It is noted that membrane P1
compared (Fig. 5c and 5d). Hydrophilic composite membranes (P1-P4) exhibited an increase in surface porosity which is evident from the
have large pore sizes and wide pore distributions due to the hydrophilic porosity, pore size data, and surface texture from the SEM image (Fig. 3).
nature of clay particles, leading to the formation of a more porous The increase in WCA in this membrane can be attributed to the rough
network during phase inversion in water. Table 3 shows the mean pore texture which causes water droplets to lie on the surface where air
sizes (MPS) of PVDF hydrophilic composite membranes between 0.28 pockets are trapped between the liquid and the membrane surface as
and 0.33 μm. The hydrophobic CNPs incorporated membranes have a explained by the Cassie-Baxter model. However, with an increase in the
MPS between 0.12 and 0.21 μm. The hydrophobic CNPs within the PVDF concentration of hydrophilic clay particles, the influence of hydrophilic
matrix were increased from 3 wt% and 5 wt% which reduced the mean elements on the surface wettability is predominant which reduces the
pore size to 0.16 and 0.12 μm, respectively, resulting in a narrow pore contact angle as observed in the membranes from P1 to P4 (Fig. 6). This
size distribution. [56]. The porosity of the modified membranes was behavior can be explained through Wenzel’s model, wherein the
high (>75 %), as observed by the highly porous morphologies from the increased roughness coupled with the more hydrophilic surface due to
SEM analysis compared with the pristine membranes, as shown in the presence of hydrophilic clay nanoparticles, contributes to a reduc­
Table 3. The porosity of P1-P4 was higher (between 79 and 85 %) owing tion in the hydrophobicity of the membranes [59]. The low WCA of P3
to the hydrophilic nature of the CNPs. Hydrophobic clay nanoparticles and P4 results in poor wetting resistance, as confirmed by their low
incorporated with 0.5–1 wt% (P5 and P6) also showed a higher porosity LEPw.The bulk properties of all developed PVDF/composite membranes
than the base PVDF membrane. Whereas 3–5 wt% (P7 and P8) had are reported in Table 3. The LEPw values, which determine the mem­
lower porosity compared to the base PVDF due to possible agglomera­ brane’s resistance to long-term wetting, show a drop in LEPw for the
tion, thereby blocking the pores within the membrane. The high vis­ hydrophilic clay-incorporated membranes compared to base membrane
cosity of hydrophobic CNPs at high concentrations hinders solvent and P0 which had an LEP value of 0.8 bar. P1-P4 had a drop in LEP values
nonsolvent exchange during phase inversion. Adding CNPs to the from 0.6 to 0.4 bar as clay particles were increased in the membrane
polymeric solution increased the viscosity of the dope solution similar to matrix (i.e., from 0.5 % to 5 %). On the other hand, the hydrophobic
previous reports [48]. Typically, the pristine PVDF solution viscosity membranes (P5-P8) have shown an increase in LEP values between 1.2
was measured as 426 cP which increased to 520 cP and 579 cP with the and 1.8 bar as the clay particles increased in the membrane matrix from
addition of 5 wt% hydrophilic and hydrophobic CNPs respectively. 0.5 to 5 wt% which can be attributed to factors such as increase in hy­
Fig. 6a and 6b show the water contact angle values of the developed drophobicity due to the increase in the hydrophobic character from
PVDF/composite membranes. There was a decrease in WCA from P1-P4 higher amount of CNPs and decrease in pore size due to increase in dope
due to the incorporation of hydrophilic clay nanoparticles, while a solution viscosity from 426 cP for pristine PVDF membrane to 579 cP
progressive increase in contact angle was noticed for hydrophobic CNPs, with the addition of 5 wt% hydrophobic CNPs. It was noticed that an
as observed previously [57]. The WCA of the PVDF hydrophobic mem­ increase in hydrophobic clay nanoparticles above the optimum con­
branes increased from 87◦ to 109◦ (24 % increase) from 0-5 wt% hy­ centration results in constriction of the pores, enhancing the mem­
drophobic CNPs. The addition of a higher % of nanoparticles in the brane’s resistance to wetting, and thereby increasing the liquid entry
casting film hinders dispersibility and introduces particle agglomeration pressure. The tortuosity of a membrane is a critical factor in the mech­
during the phase inversion process. This is indicated in the cross-section anism of mass transfer and is typically linked to the pore structure and
images of the composite membranes in Fig. 3. Hence, it is implied that geometry of the membrane [60,61]. In order to achieve optimal mass
nanoparticle modifications require low concentrations to prevent ag­ transfer, membranes used in MD processes should be porous with suit­
glomerations as this can obstruct the formation of pores and lowers able thickness and low tortuosity.
membrane permeability [58]. As shown in Table 3, the tortuosity values of the membranes fall
Moreover, the WCA of the hydrophilic composite membranes was between the range of 1.7 and 2.14, which is considered optimal [62].
maximum at 0.5 wt% of hydrophilic CNPs, and decreased progressively The hydrophilic composite membranes P1-P4 have low tortuosity
to 88.5◦ from increasing the concentration to 5 wt%. The addition of values, indicating less constriction in the porous structure. On the other
hydrophilic material to a membrane would generally increase its hy­ hand, the hydrophobic CNPs composite membranes P5 and P6 have
drophilicity. However, there are other surface properties which average tortuosity values indicating a narrow size distribution, whereas
P7, P8, and PVDF base membrane P0 have values greater than or equal
to 2 which indicates higher tortuosity in these membrane structure as
Table 3 compared to other composite membranes.
Bulk Properties of PVDF Composite Membranes. The higher tortuosity also reflects a narrower pore size distribution
Membranes LEPw Mean Porosity Tortuosity Thickness in the membrane structure. Khayet et al. [63] noted that the existence of
(bar) Pore (%) (μm) tortuous paths within the membrane can reduce MD flux, as diffusing
Size
molecules move along the membrane pores. Membrane thickness plays a
(μm)
crucial role in the performance of the membrane [64,65]. The initial
P0 0.80 ± 0.18 ± 74.00 ± 2.14 ± 0.013 98.3 ± 5 casting thickness was maintained uniformly at 250 μm for both pristine
0.10 0.01 0.80
P1 0.60 ± 0.28 ± 79.70 ± 1.77 ± 0.080 109.7 ± 4
PVDF and composite membranes. However, the difference in viscosity of
0.10 0.02 1.40 the solution causes a differential in membrane creation on the glass
P2 0.55 ± 0.33 ± 81.60 ± 1.72 ± 0.010 110.8 ± 6 surface, resulting in a variable membrane volume after drying. The
0.05 0.01 0.90 membranes in our study were fabricated without support which rela­
P3 0.40 ± 0.32 ± 83.50 ± 1.63 ± 0.011 112.7 ± 5
tively shrink or expand with respect to the viscosity of the dope solution
0.10 0.02 0.60
P4 0.40 ± 0.39 ± 85.00 ± 1.55 ± 0.07 108.0 ± 5 in the phase inversion process.
0.10 0.04 1.50
P5 1.20 ± 0.23 ± 77.05 ± 1.79 ± 0.006 114.8 ± 3 3.3. MD performance of PVDF clay composite membranes
0.10 0.02 0.30
P6 1.50 ± 0.25 ± 75.80 ± 1.85 ± 0.008 115.3 ± 2
0.10 0.01 0.50 The PVDF-base membrane and PVDF composite membranes fabri­
P7 1.70 ± 0.16 ± 75.20 ± 1.99 ± 0.012 119.4 ± 3 cated in this work were tested for synthetic NaCl solution of 70 g/L (7.0
0.05 0.02 0.50 wt%) using air gap membrane distillation. All experiments were done
P8 1.80 ± 0.14 ± 73.50 ± 2.11 ± 0.013 127.6 ± 2 based on the AGMD process parameters mentioned in Table 2. The flux
0.05 0.02 0.60
and salt rejection of PVDF hydrophilic composite membranes P1-P4

7
T.O. Obidara et al. Separation and Puri cation Technology 361 (2025) 131435

Fig. 6. The water contact angle of (a) pristine PVDF membrane in relation to PVDF hydrophilic clay composite membrane, (b) PVDF hydrophobic clay compos­
ite membranes.

compared to PVDF-base membrane P0 are shown in Fig. 7a and 7b, separation process, resulting in the nominal pore size distribution in this
respectively. The PVDF-base membrane showed a mean flux of 7.30 kg sample. With a further increase in hydrophilic clay particle concentra­
m− 2h− 1 and a salt rejection of 98.64 % which dropped to 98 % after 24 h tion to 1.0 wt% improvement in the flux by 26.50 % as compared to the
because of gradual membrane wetting. The addition of 0.5 wt% to 5.0 wt 0.5 wt% was obtained. An increase in clay concentration to 3 wt%
% hydrophilic clay particles (P1-P4) has shown an increase in average demonstrated the highest flux among all samples, which was enabled by
flux from 10.81 to 13.80 kg m− 2h− 1. Specifically, with the addition of higher permeability of the membranes with lower rejection having an
0.50 wt% clay, the increase in flux with stable salt rejection was noted, average value of 98.5 %. The fluxes of P3 and P4 were comparable
which can be attributed to the number of clay particles that aided in because of the addition of more particles (higher concentration of CNPs)
enhancing the surface roughness and opening of pores during the phase and the possible effects of agglomeration in the P4 membrane.

Fig. 7. Performance Evaluation of PVDF-base and PVDF composite membranes a) and b) flux and salt rejection of PVDF-hydrophilic membranes; c) and d) Flux and
salt rejection of PVDF-hydrophobic membranes.

8
T.O. Obidara et al. Separation and Puri cation Technology 361 (2025) 131435

Incorporating hydrophilic CNPs within the PVDF matrix contributed to performance. We observed the flux to be around 18.30 kg m− 2h− 1 for P6
the formation of large pores during water immersion, resulting in an and P1 around 11.21 kg m− 2h− 1 with salt rejection of more than 99.9 %.
increased flux and a decrease in the salt rejection factor due to poor Also shown in Fig. 8 is the comparison of 1 wt% hydrophobic clay
wetting resistance. As a result, it was discovered that the rise in the flux composite membrane with the commercial PVDF membrane (TISCH
over time in PVDF hydrophilic composite membranes suffers a lower salt Scientific) tested in AGMD under 70,000 ppm feed solution for 60 h. The
rejection rate. P1-P4 exhibited good performance in the first 3–6 h of the results indicated the superior performance of composite membrane (P6)
MD, but subsequently experienced surface wetting and a decline in suggesting the importance of chloro-based CNPs in improving the per­
rejection. However, P1 membrane demonstrated better performance formance of PVDF membranes.
and maintained wetting resistance throughout the test due to the min­ The performance of modified PVDF composite membranes in this
imal amount (0.5 wt%) of CNPs) added to the PVDF membrane. It was study was compared with relevant composite membranes from the
observed that membranes with large pores contributed to their low LEP literature using the same AGMD process as shown in Table 4. The
and lower rejection. addition of clay particles performs excellently well than adding hexag­
Fig. 7c and 7d also display the flux and salt rejection of PVDF hy­ onal boron hydride, rGO, and GO to the PVDF matrix. The higher flux
drophobic composite membranes P5-P8 compared with the PVDF-base and rejection obtained by Mao et al. [68] by embedding silica nano­
membrane P0. The addition of hydrophobic clay particles enhances particles to improve the hydrophobicity of graphene oxide membranes
the hydrophobicity of the composite membranes as the concentration of for water desalination was due to the lower feed concentration of 35 g/L.
the clay particles increases [30]. Adding 0.5 wt% hydrophobic CNPs Triple-layer membrane was fabricated by Prince et al. [37] and reached
improved the vapor transport within the membrane, increasing water a flux of 15.2 kg m− 2h− 1 due to the porous electrospun layer above the
flux, as seen in Fig. 7c. The clay composition was further increased to 1 PVDF membrane. The nanofiber layer increases the hydrophobicity and
wt%, leading to a 75 % increase in flux compared to the base PVDF vapor transport across the membrane while the PVDF membrane
membrane, and excellent rejection > 99.99 % for 24 h test. Moreover, without electrospun fiber has a flux of 10 kg m− 2h− 1. Moreso, Navarro-
the excellent result can be attributed to the increase in hydrophobicity, Tovar et al. [69] recently synthesized PVDF-clay composite membranes
moderate pore size of the membrane, and optimum liquid entry pres­ using montmorillonite and cloisite 20A, showing the suitability of clay
sure, which made it resist wetting during testing for more than 24 h. materials for developing mixed-matrix membranes for desalination (35
However, increasing the clay particles to 3 and 5 wt% reduced the flux g/L). The performance of the membranes was evaluated by AGMD
due to lower porosity and pore sizes. The performance of composite achieving a flux of 8.6 kg m− 2h− 1 and salt rejection of 99.8 %. It can be
membranes in MD can be influenced by embedding nanoparticles in the seen from this comparison that the clay-modified membrane developed
membrane. However, there exists an amount of nanoparticles that can in this study performed excellently well in treating highly saline water in
be embedded in the membrane, diminishing the positive effects of the MD.
membrane’s performance. Dumee et al. [66] mentioned that a lower
WCA favors membrane wetting and reduction in permeation flux, while 4. Conclusion
a higher hydrophobic contact angle improves vapor transport. The
outstanding performance of hydrophobic composite membranes can be This study explored the synthesis of enhanced PVDF composite
attributed to the development of a thin air film on the membrane sur­ membranes composed of amino and chloro-based CNPs. A detailed
face, which is facilitated by hydrophobic forces [66,67]. The PVDF hy­ investigation of the structure and desalination performance of the
drophobic membranes of 1 wt% (P6) has an outstanding result reaching composite membranes was also investigated in comparison with the
12.80 kg m− 2h− 1 and salt rejection of 99.99 %. As mentioned earlier, the pristine membrane in AGMD. The addition of hydrophilic CNPs in
rough surface provided by the clay nanoparticles improves the hydro­ membrane fabrication imparted higher porosity in the PVDF membrane,
phobicity of membranes, leading to an increased resistance to water whereas hydrophobic CNP membranes exhibited a narrow distribution
permeation in MD and better salt rejection. The flux of the PVDF-base of pores with enhanced surface roughness and wetting resistance. The
membrane was enhanced by incorporating clay nanoparticles hydro­ hydrophilic composite membranes showed a general increase in flux as
philic while optimum flux and stable salt rejection were maintained by the clay concentration increased from 0.5 to 5 wt%. At lower concen­
the hydrophobic particles. Enhancement in flux is due to the presence of trations of CNPs, the membrane (P1) exhibited moderate flux, while at
more surface area for vapor transport within the membrane [64]. higher concentrations (P2-P4), the flux increased progressively
The stability of the composite membranes was further analyzed as compromising salt rejection due to poor wetting resistance. In contrast,
shown in Fig. 8 by subjecting the membranes to a seawater sample. The the hydrophobic composite membranes showed a nominal flux with
membrane P1 and P6 under the same test condition for long-term per­ stable salt rejection performance. The AGMD test results showed that the
formance was tested for 50 h using real seawater (35,403 ppm) obtained 1 wt% CNP hydrophobic composite membrane showed optimal perfor­
from Arabian gulf, to confirm the PVDF novel composite membrane’s mance with a permeate flux of 12.80 kg m− 2h− 1 and a stable salt

Fig. 8. Long term AGMD performance of PVDF clay composite membranes using A) Seawater as feed and B) 70,000 ppm feed comparison with commercial
PVDF membrane.

9
T.O. Obidara et al. Separation and Puri cation Technology 361 (2025) 131435

Table 4
Comparison of PVDF Composite Membranes developed in this study with previously reported membranes for AGMD.
Membranes Modification Porosity Thickness Feed Feed Flux Salt Ref.
Method (%) (μm) Conc. Temp. Rejection
(g/L) (◦ C) (kg m- (%)
2 ¡1
h )

Modified clay composite Silane grafting − − 58.4 95 5–7.2 ¡ [35]

PVDF/rGO Filler incorporation 70.1 ​ 35 − 7.0 99.99 [70]
Montmorillonite and Cloisite Clay composite Phase inversion 75 − 35 60 8.6 99.8 [69]
Membrane
Dual Layer PVDF Phase inversion − 150 35 80 10 99.99 [37]
PVDF/PMMA Phase inversion 63.4 ± 0.8 − 35 70 18.50 99.99 [71]
Triple-layer Composite membrane Phase inversion/ − 175 35 80 15.2 99.99 [37]
Electrospinning
Hyperbranched modified PVDF Solgel/Electrospinning − − 35 60 10.7 99.90 [72]
PVDF Carboxylated carbon nanotubes Phase inversion 42.53 ± 143.1 ± 0.5 35 − 6.38 99.90 [73]
0.86
PVDF/Graphene Quantum dots Electrospinning 90.66 195.3 ± 3.1 35 60 17.6 99.70 [74]
Hybrid PVDF/graphene Phase inversion 74.0 − 35 80 11.30 99.90 [75]
Hexagonal boron hydride (h-BN)/ PVDF Embedment of h-BN/Phase 65 ± 1.3 − 35 80 7.1 99 [76]
Inversion
Composite membrane Phase inversion 64.7 ± 6.3 − 30 80 7.8 − [77]
PVDF APTS-functionalized GO − 59.6 ± 2.1 19.6 ± 0.4 35 85 6.2 99.9 [78]
PVDF Chloro-Based Clay Composite Phase inversion 75.80 ± 115.3 ± 70 70 12.80 99.99 This
Membrane 0.50 0.2 Study

rejection (99.99 %) for more than 24 h. The result shows that above 0.5 [7] A. Alkhudhiri, N. Darwish, N. Hilal, Membrane distillation: A comprehensive
review, Desalination 287 (2012) 2–18, https://doi.org/10.1016/J.
wt%, the hydrophilic clay has more affinity towards water during phase
DESAL.2011.08.027.
inversion leading to poor wetting resistance. The chloro-based CNPs [8] M. Gryta, M. Tomaszewska, K. Karakulski, Wastewater treatment by membrane
have exhibited a high degree of compatibility with PVDF polymer, distillation, Desalination 198 (1–3) (2006) 67–73, https://doi.org/10.1016/J.
making them suitable for developing mixed-matrix membranes for MD DESAL.2006.09.010.
[9] L. Han, Y.Z. Tan, T. Netke, A.G. Fane, J.W. Chew, Understanding oily wastewater
applications. treatment via membrane distillation, J Memb Sci 539 (2017) 284–294, https://doi.
org/10.1016/J.MEMSCI.2017.06.012.
[10] S. S. Ray et al., “Recent Developments in Nanomaterials-Modified Membranes for
Declaration of competing interest Improved Membrane Distillation Performance,” Membranes 2020, Vol. 10, Page
140, vol. 10, no. 7, p. 140, Jul. 2020, doi: 10.3390/MEMBRANES10070140.
[11] M. Gryta, “The Influence of Talc Addition on the Performance of Polypropylene
The authors declare that they have no known competing financial Membranes Formed by TIPS Method,” Membranes 2019, Vol. 9, Page 63, vol. 9, no.
interests or personal relationships that could have appeared to influence 5, p. 63, May 2019, doi: 10.3390/MEMBRANES9050063.
the work reported in this paper. [12] A. Akbari, M. Hamadanian, V. Jabbari, A. Yunessnia Lehi, M. Bojaran, “Influence of
PVDF concentration on the morphology, surface roughness, crystalline structure,
and filtration separation properties of semicrystalline phase inversion polymeric
Acknowledgments membranes,” Desalination, Water Treat 46 (1–3) (2012) 96–106, https://doi.org/
10.1080/19443994.2012.677524.
[13] E. Yuliwati, A.F. Ismail, Effect of additives concentration on the surface properties
This research was funded by King Fahd University of Petroleum and and performance of PVDF ultrafiltration membranes for refinery produced
Minerals, KFUPM (INMW2421). wastewater treatment, Desalination 273 (1) (2011) 226–234, https://doi.org/
10.1016/J.DESAL.2010.11.023.
[14] M.A. Mousavi, H. Reza Mortaheb, M. Baghban Salehi, Hydrophobizing polyether
Appendix A. Supplementary material sulfone membrane by sol-gel for water desalination using air gap membrane
distillation, Polym.-Plast. Technol. Mater. (2020) 47–59, https://doi.org/10.1080/
Supplementary data to this article can be found online at https://doi. 25740881.2020.1784225.
[15] T.O. Obidara, M.A. Azeem, D.U. Lawal, M.A. Alfaraj, M.A. Abdulhamid, T.
org/10.1016/j.seppur.2025.131435. N. Baroud, Novel hexa-fluorinated intrinsically porous polyimide membranes for
the desalination of high saline water by air-gap membrane distillation, Desalination
Data availability 566 (Nov. 2023) 116948, https://doi.org/10.1016/j.desal.2023.116948.
[16] P. Vandezande, X. Li, L.E.M. Gevers, I.F.J. Vankelecom, High throughput study of
phase inversion parameters for polyimide-based SRNF membranes, J Memb Sci 330
Data will be made available on request. (1–2) (Mar. 2009) 307–318, https://doi.org/10.1016/j.memsci.2008.12.068.
[17] M. Pagliero, A. Bottino, A. Comite, C. Costa, Novel hydrophobic PVDF membranes
prepared by nonsolvent induced phase separation for membrane distillation, J
References Memb Sci 596 (Feb. 2020) 117575, https://doi.org/10.1016/J.
MEMSCI.2019.117575.
[1] R. Jain, Providing safe drinking water: A challenge for humanity, Clean Technol [18] M.O. Aijaz, M.R. Karim, M.H.D. Othman, U.A. Samad, Anti-fouling/wetting
Environ Policy 14 (1) (2012) 1–4, https://doi.org/10.1007/S10098-011-0446-1/ electrospun nanofibrous membranes for membrane distillation desalination: A
METRICS. comprehensive review, Desalination 553 (May 2023) 116475, https://doi.org/
[2] A.M. Alklaibi, N. Lior, Membrane-distillation desalination: Status and potential, 10.1016/J.DESAL.2023.116475.
Desalination 171 (2) (2005) 111–131, https://doi.org/10.1016/j. [19] M. Gryta, Fouling in direct contact membrane distillation process, J Memb Sci 325
desal.2004.03.024. (1) (Nov. 2008) 383–394, https://doi.org/10.1016/J.MEMSCI.2008.08.001.
[3] J. R. Werber, C. O. Osuji, and M. Elimelech, “Materials for next-generation [20] L.D. Tijing, Y.C. Woo, J.S. Choi, S. Lee, S.H. Kim, H.K. Shon, Fouling and its control
desalination and water purification membranes,” Nature Reviews Materials 2016 1: in membrane distillation—A review, J Memb Sci 475 (Feb. 2015) 215–244, https://
5, vol. 1, no. 5, pp. 1–15, Apr. 2016, doi: 10.1038/natrevmats.2016.18. doi.org/10.1016/J.MEMSCI.2014.09.042.
[4] N.C. Darre, G.S. Toor, Desalination of Water: a Review, Curr Pollut Rep 4 (2) (2018) [21] F.E. Ahmed, B.S. Lalia, R. Hashaikeh, Membrane-based detection of wetting
104–111, https://doi.org/10.1007/S40726-018-0085-9/FIGURES/3. phenomenon in direct contact membrane distillation, J Memb Sci 535 (Aug. 2017)
[5] L. Y. Jiang, Y. Wang, T. S. Chung, X. Y. Qiao, and J. Y. Lai, “Polyimides membranes 89–93, https://doi.org/10.1016/J.MEMSCI.2017.04.035.
for pervaporation and biofuels separation,” Nov. 2009. doi: 10.1016/j. [22] A. Criscuoli, M.C. Carnevale, E. Drioli, Evaluation of energy requirements in
progpolymsci.2009.06.001. membrane distillation, Chem. Eng. Process. 47 (2008) 1098–1105, https://doi.org/
[6] S. S. Ray et al., “Developments in forward osmosis and membrane distillation for 10.1016/j.cep.2007.03.006.
desalination of waters,” Dec. 15, 2018, Springer Verlag. doi: 10.1007/s10311-018-
0750-7.

10
T.O. Obidara et al. Separation and Puri cation Technology 361 (2025) 131435

[23] M. El Haj Assad et al., “Applications of nanotechnology in membrane distillation: A [46] M. Baghbanzadeh, D. Rana, C.Q. Lan, T. Matsuura, Effects of hydrophilic silica
review study,” Jul. 01, 2020, Desalination Publications. doi: 10.5004/ nanoparticles and backing material in improving the structure and performance of
dwt.2020.25821. VMD PVDF membranes, Sep Purif Technol 157 (2016) 60–71, https://doi.org/
[24] L. Zhao, C. Wu, X. Lu, D. Ng, Y.B. Truong, Z. Xie, Activated carbon enhanced 10.1016/j.seppur.2015.11.029.
hydrophobic/hydrophilic dual-layer nanofiber composite membranes for high- [47] J. Huang, K. Zhang, K. Wang, Z. Xie, B. Ladewig, H. Wang, Fabrication of
performance direct contact membrane distillation, Desalination 446 (Nov. 2018) polyethersulfone-mesoporous silica nanocomposite ultrafiltration membranes with
59–69, https://doi.org/10.1016/J.DESAL.2018.09.002. antifouling properties, J Memb Sci 423–424 (Dec. 2012) 362–370, https://doi.org/
[25] Y.C. Woo, et al., Graphene/PVDF flat-sheet membrane for the treatment of RO 10.1016/J.MEMSCI.2012.08.029.
brine from coal seam gas produced water by air gap membrane distillation, J Memb [48] C.S. Fernandes, et al., Explication of hydrophobic silica as effective pore former for
Sci 513 (Sep. 2016) 74–84, https://doi.org/10.1016/J.MEMSCI.2016.04.014. membrane fabrication, Appl. Surf. Sci. Adv. 3 (Mar. 2021), https://doi.org/
[26] L.Y. Ng, A.W. Mohammad, C.P. Leo, N. Hilal, Polymeric membranes incorporated 10.1016/j.apsadv.2020.100051.
with metal/metal oxide nanoparticles: A comprehensive review, Desalination 308 [49] B. Rahmaniyan, T. Mohammadi, M.A. Tofighy, Development of high flux PVDF/
(Jan. 2013) 15–33, https://doi.org/10.1016/J.DESAL.2010.11.033. modified TNTs membrane with improved properties for desalination by vacuum
[27] V. Kochkodan, N. Hilal, A comprehensive review on surface modified polymer membrane distillation, J Environ Chem Eng 9 (6) (2021) Dec, https://doi.org/
membranes for biofouling mitigation, Desalination 356 (Jan. 2015) 187–207, 10.1016/j.jece.2021.106730.
https://doi.org/10.1016/J.DESAL.2014.09.015. [50] P.S. Goh, R. Naim, M. Rahbari-Sisakht, A.F. Ismail, Modification of membrane
[28] A. Razmjou, E. Arifin, G. Dong, J. Mansouri, V. Chen, Superhydrophobic hydrophobicity in membrane contactors for environmental remediation, Sep Purif
modification of TiO 2 nanocomposite PVDF membranes for applications in Technol 227 (Nov. 2019) 115721, https://doi.org/10.1016/J.
membrane distillation, J Memb Sci 415–416 (Oct. 2012) 850–863, https://doi.org/ SEPPUR.2019.115721.
10.1016/J.MEMSCI.2012.06.004. [51] H. Bai, X. Wang, Y. Zhou, L. Zhang, Preparation and characterization of poly
[29] J.P.G. Villaluenga, M. Khayet, M.A. López-Manchado, J.L. Valentin, B. Seoane, J. (vinylidene fluoride) composite membranes blended with nano-crystalline
I. Mengual, Gas transport properties of polypropylene/clay composite membranes, cellulose, Prog. Nat. Sci.: Mater. Int. 22 (3) (Jun. 2012) 250–257, https://doi.org/
Eur Polym J 43 (4) (Apr. 2007) 1132–1143, https://doi.org/10.1016/j. 10.1016/j.pnsc.2012.04.011.
eurpolymj.2007.01.018. [52] G. Yi, J. Li, L.C. Henderson, W. Lei, L. Du, S. Zhao, Enhancing Thermal
[30] J.A. Prince, G. Singh, D. Rana, T. Matsuura, V. Anbharasi, T. Conductivity of Polyvinylidene Fluoride Composites by Carbon Fiber: Length Effect
S. Shanmugasundaram, Preparation and characterization of highly hydrophobic of the Filler, Polymers (basel) 14 (21) (2022) Nov, https://doi.org/10.3390/
poly(vinylidene fluoride) – Clay nanocomposite nanofiber membranes (PVDF–clay polym14214599.
NNMs) for desalination using direct contact membrane distillation, J Memb Sci [53] C. Van Goethem, M. Mertens, I.F.J. Vankelecom, Crosslinked PVDF membranes for
397–398 (Apr. 2012) 80–86, https://doi.org/10.1016/J.MEMSCI.2012.01.012. aqueous and extreme pH nanofiltration, J Memb Sci 572 (Feb. 2019) 489–495,
[31] J.C. Mierzwa, V. Arieta, M. Verlage, J. Carvalho, C.D. Vecitis, Effect of clay https://doi.org/10.1016/j.memsci.2018.11.036.
nanoparticles on the structure and performance of polyethersulfone ultrafiltration [54] M. Behera, S.K. Biswal, M.A. Ahemad, B.S. Panda, Demonstration of enhanced
membranes, Desalination 314 (Apr. 2013) 147–158, https://doi.org/10.1016/J. thermal stability, dielectric constant and low tangent loss by particle-reinforced
DESAL.2013.01.011. silver/poly (Vinylidene difluoride) polymer nanocomposites, Biointerface Res Appl
[32] H. Rajabi, N. Ghaemi, S.S. Madaeni, P. Daraei, M.A. Khadivi, M. Falsafi, Nanoclay Chem 11 (5) (Oct. 2021) 12584–12595, https://doi.org/10.33263/
embedded mixed matrix PVDF nanocomposite membrane: Preparation, BRIAC115.1258412595.
characterization and biofouling resistance, Appl Surf Sci 313 (Sep. 2014) 207–214, [55] W. Li, H. Li, Y.M. Zhang, Preparation and investigation of PVDF/PMMA/TiO2
https://doi.org/10.1016/j.apsusc.2014.05.185. composite film, J Mater Sci 44 (11) (Jun. 2009) 2977–2984, https://doi.org/
[33] P.R. Chowdhury, S.K. Ray, Synthesis and characterization of polymer-clay 10.1007/s10853-009-3395-x.
nanocomposite membranes for micellar enhanced ultrafiltration of synthetic dyes [56] D. Hou, G. Dai, H. Fan, J. Wang, C. Zhao, H. Huang, Effects of calcium carbonate
from water, J Taiwan Inst Chem Eng 165 (Dec. 2024), https://doi.org/10.1016/j. nano-particles on the properties of PVDF/nonwoven fabric flat-sheet composite
jtice.2024.105768. membranes for direct contact membrane distillation, Desalination 347 (Aug. 2014)
[34] M. Hesavi, E. Derikvand, M.S. Babarsad, M.S. Bejestan, M. Zendehdel, Preparation, 25–33, https://doi.org/10.1016/j.desal.2014.05.028.
characterization and performance of PVDF/Al2O3, TiO2 and clay membrane for [57] M.O. Mavukkandy, M.R. Bilad, J. Kujawa, S. Al-Gharabli, H.A. Arafat, On the effect
removal of toxic metals, Desalination Water Treat 313 (Nov. 2023) 82–91, https:// of fumed silica particles on the structure, properties and application of PVDF
doi.org/10.5004/dwt.2023.30070. membranes, Sep Purif Technol 187 (Oct. 2017) 365–373, https://doi.org/10.1016/
[35] S. Khemakhem and R. Ben Amar, “Modification of Tunisian clay membrane surface J.SEPPUR.2017.06.077.
by silane grafting: Application for desalination with Air Gap Membrane Distillation [58] J. Ravi et al., “Polymeric membranes for desalination using membrane distillation:
process,” Colloids Surf A Physicochem Eng Asp, vol. 387, no. 1–3, pp. 79–85, Aug. A review,” Sep. 15, 2020, Elsevier B.V. doi: 10.1016/j.desal.2020.114530.
2011, doi: 10.1016/j.colsurfa.2011.07.033. [59] R.N. Wenzel, Resistance of solid surfaces to wetting by water, Ind Eng Chem 28 (8)
[36] C.Y. Lai, A. Groth, S. Gray, M. Duke, Preparation and characterization of poly (Aug. 1936) 988–994, https://doi.org/10.1021/IE50320A024/ASSET/
(vinylidene fluoride)/nanoclay nanocomposite flat sheet membranes for abrasion IE50320A024.FP.PNG_V03.
resistance, Water Res 57 (Jun. 2014) 56–66, https://doi.org/10.1016/j. [60] J. Phattaranawik, R. Jiraratananon, A.G. Fane, Heat transport and membrane
watres.2014.03.005. distillation coefficients in direct contact membrane distillation, J Memb Sci 212
[37] J.A. Prince, V. Anbharasi, T.S. Shanmugasundaram, G. Singh, Preparation and (1–2) (Feb. 2003) 177–193, https://doi.org/10.1016/S0376-7388(02)00498-2.
characterization of novel triple layer hydrophilic–hydrophobic composite [61] S. B. Iversen, V. K. Bhatia, K. Dam-Johansen, and G. Jonsson B’, “Characterization
membrane for desalination using air gap membrane distillation, Sep Purif Technol of microporous membranes for use in membrane contactors,” 1997.
118 (Oct. 2013) 598–603, https://doi.org/10.1016/J.SEPPUR.2013.08.006. [62] S.S. Ibrahim, Q.F. Alsalhy, Modeling and simulation for direct contact membrane
[38] O. Monticelli, A. Bottino, I. Scandale, G. Capannelli, S. Russo, Preparation and distillation in hollow fiber modules, AIChE J 59 (2) (Feb. 2013) 589–603, https://
properties of polysulfone-clay composite membranes, J Appl Polym Sci 103 (6) doi.org/10.1002/aic.13845.
(Mar. 2007) 3637–3644, https://doi.org/10.1002/app.25511. [63] M. Khayet, “Membranes and theoretical modeling of membrane distillation: A
[39] M. A. Azeem, D. U. Lawal, H. Al Abdulgader, and T. N. Baroud, “Enhanced review,” May 11, 2011, Elsevier B.V. doi: 10.1016/j.cis.2010.09.005.
performance of superhydrophobic polyvinylidene fluoride membrane with [64] J. Xu, Y.B. Singh, G.L. Amy, N. Ghaffour, Effect of operating parameters and
sandpaper texture for highly saline water desalination in air-gap membrane membrane characteristics on air gap membrane distillation performance for the
distillation,” Desalination, vol. 528, p. 115603, Apr. 2022, doi: 10.1016/J. treatment of highly saline water, J Memb Sci 512 (Aug. 2016) 73–82, https://doi.
DESAL.2022.115603. org/10.1016/j.memsci.2016.04.010.
[40] “The diffusion of electrolytes in a cation-exchange resin membrane. II. [65] L. Eykens, et al., Influence of membrane thickness and process conditions on direct
Experimental,” Proc R Soc Lond A Math Phys Sci, vol. 232, no. 1191, pp. 510–518, contact membrane distillation at different salinities, J Memb Sci 498 (Jan. 2016)
Nov. 1955, doi: 10.1098/rspa.1955.0235. 353–364, https://doi.org/10.1016/j.memsci.2015.07.037.
[41] S. Srisurichan, R. Jiraratananon, A.G. Fane, Mass transfer mechanisms and [66] L.F. Dumée, S. Gray, M. Duke, K. Sears, J. Schütz, N. Finn, The role of membrane
transport resistances in direct contact membrane distillation process, J Memb Sci surface energy on direct contact membrane distillation performance, Desalination
277 (1–2) (Jun. 2006) 186–194, https://doi.org/10.1016/j.memsci.2005.10.028. 323 (Aug. 2013) 22–30, https://doi.org/10.1016/j.desal.2012.07.012.
[42] A. Bilgiç, A. Çimen, Removal of chromium(vi) from polluted wastewater by [67] H. Asanuma, H. Noguchi, K. Uosaki, H.Z. Yu, Water structure at superhydrophobic
chemical modification of silica gel with 4-acetyl-3-hydroxyaniline, RSC Adv 9 (64) quartz/water interfaces: A vibrational sum frequency generation spectroscopy
(2019) 37403–37414, https://doi.org/10.1039/c9ra05810a. study, J. Phys. Chem. C 113 (50) (2009) 21155–21161, https://doi.org/10.1021/
[43] F. Adam, H. Osman, K.M. Hello, The immobilization of 3-(chloropropyl) jp906607s.
triethoxysilane onto silica by a simple one-pot synthesis, J Colloid Interface Sci 331 [68] Y. Mao, et al., Roughness-enhanced hydrophobic graphene oxide membrane for
(1) (Mar. 2009) 143–147, https://doi.org/10.1016/j.jcis.2008.11.048. water desalination via membrane distillation, J Memb Sci 611 (Oct. 2020) 118364,
[44] B. Narayanamoorthy, K.K.R. Datta, M. Eswaramoorthy, S. Balaji, Improved oxygen https://doi.org/10.1016/J.MEMSCI.2020.118364.
reduction reaction catalyzed by Pt/Clay/nafion nanocomposite for PEM fuel cells, [69] R. Navarro-Tovar, P. Gorgojo, M. Jobson, P. Martin, M. Perez-Page, Innovations in
ACS Appl Mater Interfaces 4 (7) (Jul. 2012) 3620–3626, https://doi.org/10.1021/ water desalination: enhancing air gap membrane distillation performance by the
am300697q. incorporation of clay nanoparticles into PVDF matrix membranes, Environ Sci
[45] G. Johnsy, K.K.R. Datta, V.A. Sajeevkumar, S.N. Sabapathy, A.S. Bawa, (camb) (2024), https://doi.org/10.1039/d4ew00326h.
M. Eswaramoorthy, Aminoclay: A designer filler for the synthesis of highly ductile [70] A. Abdel-Karim, et al., PVDF membranes containing reduced graphene oxide:
polymer-nanocomposite film, ACS Appl Mater Interfaces 1 (12) (Dec. 2009) Effect of degree of reduction on membrane distillation performance, Desalination
2796–2803, https://doi.org/10.1021/am9005226. 452 (Feb. 2019) 196–207, https://doi.org/10.1016/J.DESAL.2018.11.014.

11
T.O. Obidara et al. Separation and Puri cation Technology 361 (2025) 131435

[71] Y.X. Lin, et al., Preparation of PVDF/PMMA composite membrane with green [75] H. R. Mortaheb, M. Baghban Salehi, and M. Rajabzadeh, “Optimized hybrid PVDF/
solvent for seawater desalination by gap membrane distillation, J Memb Sci 679 graphene membranes for enhancing performance of AGMD process in water
(Aug. 2023), https://doi.org/10.1016/j.memsci.2023.121676. desalination,” Journal of Industrial and Engineering Chemistry, vol. 99, pp. 407–421,
[72] M.R.S. Kebria, et al., Hyper-branched dendritic structure modified PVDF Jul. 2021, doi: 10.1016/j.jiec.2021.04.053.
electrospun membranes for air gap membrane distillation, Desalination 479 (Apr. [76] J. Zahirifar, A. Hadi, J. Karimi-Sabet, A. Dastbaz, Influence of hexagonal boron
2020), https://doi.org/10.1016/j.desal.2019.114307. nitride nanosheets as the additives on the characteristics and performance of PVDF
[73] S. R. Sawant, S. Kalla, and Z. V. P. Murthy, “Enhanced properties of the PVDF for air gap membrane distillation, Desalination 460 (2019) 81–91, https://doi.org/
membrane with carboxylated MWCNT and sodium alginate for membrane 10.1016/J.DESAL.2019.03.004.
distillation,” J Environ Chem Eng, vol. 11, no. 2, Apr. 2023, doi: 10.1016/j. [77] M. Essalhi, M. Khayet, Application of a porous composite hydrophobic/hydrophilic
jece.2022.109259. membrane in desalination by air gap and liquid gap membrane distillation: A
[74] A. Jafari, M.R.S. Kebria, A. Rahimpour, G. Bakeri, Graphene quantum dots comparative study, Sep Purif Technol 133 (2014) 176–186, https://doi.org/
modified polyvinylidenefluride (PVDF) nanofibrous membranes with enhanced 10.1016/j.seppur.2014.07.006.
performance for air Gap membrane distillation, Chem. Eng. Process. - Process Intesif. [78] S. Leaper, et al., Flux-enhanced PVDF mixed matrix membranes incorporating
126 (Apr. 2018) 222–231, https://doi.org/10.1016/j.cep.2018.03.010. APTS-functionalized graphene oxide for membrane distillation, J Memb Sci 554
(2018) 309–323, https://doi.org/10.1016/j.memsci.2018.03.013.

12