Sea and brackish water desalination through a novel PVDF-PTFE composite hydrophobic membrane by vacuum membrane distillation

Abstract

The present study aims to evaluate the performance of porous hydrophobic Polyvinylidene fluoride − Polytetrafluoroethylene (PVDF-PTFE) composite membranes for desalination by vacuum membrane distillation (VMD) technique. The effect of operating parameters such as feed NaCl concentration (10,000 to 40,000 mg/L), feed temperature (50 °C to 80 °C), and downstream pressure (80 to 120 mmHg) on water permeation rate was studied. The increase in feed temperature enhanced the water permeation rate due to a rise in driving force across the membrane. For a constant downstream pressure of 80 mmHg, feed temperature of 80 °C and feed flow rate of 80 L/h, the membrane exhibited a maximum water flux of 3 kg/m²h with 99.86% salt rejection when aqueous NaCl concentration of 10,000 mg/L was charged as feed. Membrane characterization was performed using various analytical tools to determine physico-chemical properties such as pore size, structural elucidation, thermal stability, crystallinity, and hydrophobicity of the membrane material. Further, a temperature and concentration polarization coefficient-based analysis was performed by solving the mass and heat transport model equations using MATLAB software. The proposed research study promotes the application of VMD for recovering potable water from highly saline sea/brackish water and alleviates brine disposal issues.

Introduction

The growing shortage of potable water resources has created a major threat to populations belonging to arid and semi-arid regions of the world [1]. Conventional flash distillation and reverse osmosis (RO) were identified as common techniques for the desalination process. Despite being abundantly used, multistage flash distillation and RO-based desalination processes were found to be disadvantageous, as they exhibit drawbacks such as fouling/scaling problems, energy-intensive nature, and low water recovery [2]. To meet the escalating demand for freshwater, it is necessary to develop a process that could reduce the overall energy requirement and increase process efficiency. Membrane distillation (MD) has emerged as an alternative desalination strategy that brings down capital costs and energy consumption to a large extent [3]. Almost 100% of non-volatiles are rejected in the MD process with no limitation on feed concentration when compared to the pressure-driven RO process which offers less potential for treating of highly saline solution with lower water recovery [4]. The separation of volatile components from the feed mixture is facilitated through a microporous hydrophobic membrane with a system operated below the boiling point of the feed liquid. Polymer materials exhibiting properties of low surface energy, high thermal stability, chemical stability, and inertness are usually preferred for MD application [5]. Polytetrafluoroethylene (PTFE) and polyvinylidene fluoride (PVDF) are considered to be the prominent commercial membrane materials for vacuum membrane distillation (VMD) due to their higher thermal stability and hydrophobicity [6]. In the recent times, many researchers have been actively working on optimizing the procedure for plasma/absorption-grafting on commercial Polyproplene (PP), PVDF and PTFE membranes to enhance the hydrophobicity required in MD systems [7, 8]. Similar study by Li et al., (2020) stated that hydrophobic properties were improved by incorporating graphene oxide (GO) nanosheets in the PVDF nanofibers for VMD process [9]. However, these studies reveal that the process of manufacturing the composite nanofibers is highly tedious and difficult to scale up. Among the hydrophobic polymers, PTFE has better chemical resistance with low surface energy along with good mass transfer resistance and operability. Considerable work has been reported on seawater desalination using the VMD technique. Zhang et. al (2016) performed VMD for sodium chloride (NaCl) solution with three different commercial hydrophobic membranes and simulated the results with model validation [10]. A theory model was developed for the estimation of permeate flux. The process parameters like NaCl concentration were varied from 10 to 25 g/L and feed temperature from 56 to 72 °C along with the feed flow rate as 10–80 L/h. The downstream pressure was changed from 517 to 622 mm Hg. The model helped to predict the flux besides the overall process efficiency. Dong et. al (2014) prepared a super hydrophobic nanofibrous electro-spun membrane by varying the PTFE concentration in the dope solution from 0–12 wt% [11]. The membrane exhibited the highest contact angle of 152.2°. The steady-state flux of 18.5 kg/m²h with 99.9% salt rejection was obtained for 3.5 wt% NaCl. A mathematical model was developed to estimate the vapor flux under various operating conditions of permeate pressure from 3 to 15 kPa with a feed temperature of 50–70 °C along with its validation. Zhelun et. al (2019) prepared PTFE/PVDF nanocomposite membranes by varying PTFE concentration from 0 to 40% [12]. For the feed temperature of 30 °C and permeate pressure of 97.54 kPa with 35 g/L NaCl solution, the pure water flux obtained was 5.61 kg/m²h with 99.9% salt rejection. The results indicate the applicability of the prepared membranes for desalination. Tong et. al (2016) worked on synthesizing of composite PVDF hollow fibers with an active layer coated with tetrafluoroethylene and 2,2,4-trifluoro-5-trifluoro methoxy-1,3-dioxole [13]. They passed a 35 g/L NaCl solution in the VMD system under a vacuum of 0.09 MPa at a feed temperature of 70 °C. A flux of 10 kg/m²h was reported with greater than 99.9% salt rejection. Fan et. al (2012) used an indigenous porous PVDF membrane for VMD application through vapor induced phase separation method [14]. The study compared the performance of the prepared PVDF membrane with a commercial PTFE membrane by passing a solution of 3.5 wt% sodium chloride. A permeation flux of 22.5 kg/m²h with a rejection of 99.9% was achieved at a feed temperature of 70 °C and downstream pressure of 31.5 kPa. Cao et al. presented a simulation study based on Aspen Plus for 35 g/kg of salt concentration [15]. A mathematical model was developed and verified. The process performance was analyzed to find that the concentration and temperature polarization effects are significant at higher feed temperatures or downstream pressures. Zhang et al. studied the 3D simulation of an aqueous NaCl solution of 10 g/L concentration using computational fluid dynamics (CFD) for VMD application [16]. The reported flux results of the studies are summarized and shown in Table 1. The phase transition in the membrane process was studied by simulating the vapor–liquid distribution throughout the membrane module. The variation in pressure, temperature, and phase transition was noticed near the boundary layer on the membrane side. A few studies were also discussed on the impact of temperature polarization coefficient (TPC) and concentration polarization coefficient (CPC) on the MD design. The MD is deemed to appropriately designed when the TPC in the range of 0.6 to 1.0 [17]. The work by Mohammad et al. (2020) stated that the TPC in DCMD system for saline water desalination was in the range of 0.88 to 0.94 [18]. Moreover, Alsalhy et al. (2018) has studied the effect of feed temperature and flowrate on TPC, it was observed that the TPC is inversely proportional to feed temperatuare while directly correlated to feed flowrate [17]. Ibrahim et al. (2012) demonstrated the predicted and experimental influence of feed and permeate velocities on permeate flux. As the feed and permeate flow velocities increases, there was fall in concentration and temperature polarization that lead to higher permeate flux [19]. Based on the existing literature, it is asserted that PVDF and PTFE emerge as the most optimal polymers for VMD applications, attributed to their exceptional chemical resistance and robust durability. These properties enable PVDF to withstand aggressive chemical environments often encountered in VMD systems, ensuring prolonged operational reliability. On the other hand, PTFE plays a pivotal role by virtue of its non-stick characteristics and exceptional resistance to high temperatures. In the context of VMD, PTFE contributes to heightened membrane performance, effectively preventing fouling and thereby ensuring the unimpeded and efficient transport of vapor throughout the membrane in the distillation process. The synergistic use of PVDF and PTFE in VMD application enhances overall membrane system durability, chemical resilience, and operational efficiency. Recognizing the constraints and disadvantages associated with conventional desalination approaches, namely multistage flash distillation and reverse osmosis, characterized by issues like fouling, elevated energy consumption, and suboptimal water recovery, there exists a compelling imperative to investigate alternative methodologies. In light of the rising demand for freshwater, this study is centered on VMD as an auspicious approach, with the objective of alleviating total energy demands and augmenting process efficiency. Thus, the present study attempts to investigate the performance of PVDF-PTFE composite membranes for seawater desalination using the VMD technique. This study explores VMD as a promising alternative, aiming to reduce overall energy requirements and enhance process efficiency. Specifically, the research emphasizes exploring the performance and efficacy of the PVDF-PTFE composite membranes, specifically in the challenging conditions posed by seawater desalination processes. In this context, the study evaluated the membrane performance at varied operating conditions of feed concentration, downstream pressure, and feed temperature. Additionally, a comprehensive analysis employing temperature and concentration polarization coefficients was performed by solving the mass and heat transport model equations using MATLAB software tool. Through these efforts, the research aims to contribute valuable insights into the potential of PVDF-PTFE composite membrane for enhancing the effectiveness of VMD in sea and brackrish water desalination applications.

Table 1. Reported VMD Flux values of different membranes for desalination

Membrane material	Flux (kg/m2h)	Operating conditions	References
Commercial PTFE	10.89	Tb,f = 72 °C, 20 g/L NaCl, Flow rate = 80 L/h	[10]
PVDF–PTFE nanofibrous membranes	18.50	Tb,f = 60 °C, 3.5 wt.% NaCl, Permeate pressure = 9 kPa, Flow rate = 80 L/h	[11]
PVDF-PTFE nanocomposite membrane	5.61	Tb,f = 30 °C, 3.5 wt.% NaCl, Vacuum pressure = 97.54 kPa,	[12]
Hyflon AD60/PVDF composite hollow fiber membrane	10	Tb,f = 70 °C, 35 g/L NaCl, vacuum = 0.09 MPa	[13]
PVDF membrane	22.4	Tb,f = 73 °C, 3.5 wt.% NaCl, downstream pressure = 31.5 kPa, Flow rate = 54 L/h	[14]
Hydrophobic hollow fiber membrane	12.5	Tb,f = 70 °C, 3.5 wt.% NaCl, downstream pressure = 0.09 MPa, Flow rate = 90 L/h	[15]
PTFE hollow fiber membrane	17.2	Tb,f = 70 °C, Downstream pressure = 3.0 kPa	[16]
PVDF-PTFE flat sheet composite membrane	3	Tb,f = 80 °C, feed flow rate = 80 L/h, NaCl concentration = 10,000 mg/L, Downstream pressure = 80 mmHg	Present work

Experimental

Materials and methods

PVDF polymer was procured from M/s Solvay, Vadodara, Gujarat, while a PTFE membrane filter of 0.22 µm pore size, used as a substrate was procured from Merck Specialities Pvt. Ltd., Mumbai, India. N,N-Dimethyl formamide (DMF). Sodium chloride (NaCl) was obtained from Sri Vishnu Sai Scientific Pvt. Ltd., Hyderabad to prepare synthetic solutions of brackish water (10,000 ppm) and seawater (20,000 to 40,000 ppm) for desalination trials. The demineralized water used to prepare NaCl solution of known concentration was obtained from an inbuilt laboratory cascaded reverse osmosis system. The feed and permeate samples were analyzed for salt concentration using digital conductivity (Model no.: DCM 900; Supplier: Global Electronics, Hyderabad) and TDS meters (Model no. TDS-3; Supplier: HM DIGITAL INDIA PVT. LTD., Rajasthan).

Synthesis of PVDF/PTFE microporous composite membrane

Microporous PVDF on a commercial PTFE substrate was prepared by the solution casting and phase inversion technique. A PVDF polymer of 12 g was added to 88 ml of DMF solvent and the solution was stirred until complete dissolution of the polymer in the solvent was ensured. The obtained bubble-free homogenous polymer solution was spread onto a PTFE porous substrate affixed on a glass plate using an automatic casting machine as illustrated in Fig. 1. The thickness of the active PVDF membrane layer was adjusted by moving the stainless steel metal blade in the casting machine. After casting the polymer solution, the PTFE substrate with the glass plate was submerged in a non-solvent ice-cold water bath to obtain a composite membrane by phase inversion method. The final membranes were stored in 0.5% sodium metabisulphite solution to prevent microbial contamination. The characteristics of the membrane are provided in Table 2.

Fig. 1 [Images not available. See PDF.]

Flow chart on preparation of microporous PVDF-PTFE composite membrane

Table 2. Membrane characteristics

S. No	Parameter	Properties
1	Material	PVDF-PTFE composite
2	Pore size (μm)	1.2
3	Thickness (μm)	210
4	Porosity (%)	69.5%
5	Contact angle	125°
6	LEP (MPa)	30

Membrane characterization

The morphological studies of the membranes were performed using scanning electron microscopy (Hitachi S-3000N, Japan). Before the study, the membrane films were frozen in the liquid nitrogen. Prior to imaging the sputter coating for the sample is done by coating of Pt metal of 0.02 – 0.03 nm thickness onto a polymeric membrane. The crystallinity of the composite membrane was examined by a wide-angle X-ray diffractometer of Siemens Model D 5000 (Siemens, Washington DC, USA). The transformations in the physical structure of the membrane were assessed by varying the diffraction angle ranging from 2° – 80°. The structural analysis of the composite membrane was carried out by KBr pellet method using Nicolet 740 Fourier Transform Infrared Spectrophotometerprocured from Boston, USA. The inter and intra-molecular bonding within the polymers was recorded at a varying wave number from 400 to 4000 cm^–1. Thermal degradation of the membrane was measured under air using a Seiko 220TG/DTA instrument purchased from Tokyo, Japan. The hydrophobic and hydrophilic characteristics of the composite membrane were measured by the Sessile drop method using Goniometer-VCA 2000 supplied by AST products, USA. A water droplet made from a few µL was placed over the dry surface of the membrane using a micro syringe to record the contact angle. The average contact angles were recorded at three to five different locations on the membrane surface. The porosity (ε) for the composite membrane was measured quantitatively. A known amount of membrane film was taken and kept in distilled water for 12 h. Later, the membrane film was removed from the water and adhered droplets on the surface were wiped-off using tissue paper and further dried for 24 h in an oven maintained at 60 °C. The dried membranes were weighed and membrane porosity was calculated using below equation:

1

$$\%Porosity=\frac{W_1-W_2}{{\mathrm{\rho }}_{\text{w}}\times \mathrm{A}\times \mathrm{\delta }}\times 100$$

The liquid entry pressure (LEP) for the membrane is calculated using Youngs Laplace equation as shown below [20]:

2

$$LEP=\frac{-2\beta {\gamma }_{L}COS\theta }{r_{\mathit{max}}}$$

Experimental procedure

Experiments on desalination were performed using a bench-scale vacuum membrane distillation system assembled in-house. Figure 2a, b show the schematic and actual photograph of the experimental set-up for condensation of the volatile components from the feed solution. Feed NaCl solution of varying concentrations ranging from 10 to 40 g/L was prepared by dissolution of a known quantity of NaCl in deionized water. The prepared salt solution was fed into the 5 L jacketed tank where the temperature was maintained by circulating hot water outside the tank using a diaphragm pump. The composite membrane with an effective area of 0.015 m² was placed inside the rectangular plate and frame membrane holdup cell of dimensions 24 cm (length) × 18 cm (breadth). The temperature of the feed solution was monitored using temperature controllers placed at the entrance of the membrane module. The feed solution flow rate was adjusted using a rotameter located in the feed line between the feed pump and the membrane cell. The exit side of the membrane module is connected to the evacuated system to create a vacuum across the system. The difference in partial pressure acts as a driving force for water vapor transport from the feed side through the membrane to the downstream side. The vapors from the system were allowed to condense in an ice-cold bath. Bench scale experiments were performed by varying the parameters like feed concentration, permeate pressure, and feed temperature. The experiments were carried out during the daytime only by varying the operating parameters such as feed concentration, feed temperature, and downstream pressure. The experiments were performed for a total period of 40 h.

Fig. 2 [Images not available. See PDF.]

(a) Process flow diagram of vacuum-driven membrane distillation system, and (b) Actual photograph of membrane distillation test skid

The water permeation flux (J) and salt rejection (%R) were calculated by Eqs. (3 and 4):

3

$$J=\frac{W}{A\times t}$$where W is the amount of product collected (kg), A is the effective membrane area (m²), and t is the operation time (h).

4

$$\%R=\frac{C_F-C_P}{C_F}\times 100$$where, C_F and C_P are the concentrations of sodium and chloride ions in feed and permeate samples, respectively.

Mass and heat transfer phenomena in vacuum-driven membrane distillation

For a binary system (NaCl-water mixture) with only one solute (NaCl), the mass transport of volatile water molecules through a porous barrier mainly depends on pore size, thickness, tortuosity, porosity of membrane, etc. The operating parameters including the concentration of solute in the feed solution, operating temperature and downstream pressure affect the performance of the membrane process. The mass transport mechanism of water vapor molecule is generally governed by Knudsen diffusion (K_n) i.e. the mean distance traveled by the water vapor molecule (λ_v) would be higher than the membrane pore diameter (d_p) [21]:

5

$$K_n={\lambda }_{\nu }/d_p$$

where λ_v can be determined using the following expression [22]:

6

$${\lambda }_{\nu }=\frac{K_B{T}_b}{\sqrt{2\pi }{P}_m{\sigma }_{\nu }^2}$$where the Boltzmann constant (k_B) is 1.38 × 10^–23(J/K), the collision factor (ν) is 2.641 × 10^–10 (m) for water vapor, T_b is the bulk surface temperature (K), P_m is the mean pressure within membrane pores. The Knudsen number is calculated as 1.4. For a non-ideal NaCl-water mixture, the pressure drop (ΔP) across the membrane can be given as follows:

7

$$\mathrm{\Delta }P=P_1-P_2$$where, P₁ and P₂ are the partial vapor pressures of water molecules in the feed solution and the vapor phase on the downstream side of the membrane, respectively. The partial pressure of water vapor at the upstream side of the membrane is expressed as follows:

8

$$P_1=x_w{\gamma }_w{p}_{\nu }$$where x_w is the mole fraction of water on the membrane surface, γ_w is the activity coefficient for a non-ideal NaCl-water binary mixture which further depends on the solute (NaCl) concentration at the membrane surface (n_sm) [23]:

9

$${\gamma }_w=1-0.5n_{\mathit{sm}}-10n_{\mathit{sm}}^2$$

The vapor pressure of the water molecule (P_v in kPa) is given by Antoine equation [24]:

10

$$p_v=\text{exp}\left(16.53,-,\frac{3985.44}{T-38.9974}\right)$$

The transport of water vapor molecules from the bulk side of the feed solution through the membrane to the downstream side increases the salt concentration and reduces the temperature at the membrane surface. This results in the accumulation of macromolecules over the membrane surface on the feed side forming a mass transfer boundary layer besides the existence of a thermal boundary layer leading to the occurrence of concentration and temperature polarization phenomenon in membrane distillation. The polarization effects are further described by mass and heat transfer coefficients that can be obtained from Sherwood (k_m.d_h/D_Kn) and Nusselt (h_f.d_h/K) numbers given by Sider-Tate empirical correlations, respectively [25, 26]:

11

$$Sh=1.86{\left(R,e,.,S,c,.,d_h,/,L\right)}^{1/3}{\left({\mu }_1,/,{\mu }_b\right)}^{0.14}$$

12

$$Nu=1.86{\left(R,e,.,P,r,.,d_h,/,L\right)}^{1/3}{\left({\mu }_1,/,{\mu }_b\right)}^{0.14}$$where d_h is the hydraulic diameter, L is the length of the feed flow channel, µ₁ and µ_b are the viscosities of water at the feed/membrane interface and bulk feed, respectively. The Reynolds number (Re), Prandtl number (Pr), and Schmidt number (Sc) are expressed as follows:

13

$$Re=\rho v{d}_h/\mu$$

14

$$Pr=C_p\mu /k$$

15

$$Sc=\mu /\rho Dk$$

where ρ is the bulk density at feed temperature (kg/m³), v is the velocity of feed (m/s),d_h is the hydraulic diameter (m), µ is the dynamic viscosity of feed (kg/m.s), C_p is the specific heat capacity of feed (J/kgK), k is the thermal conductivity of feed (W/mK) and Dis the diffusivity of feed (m²/s).

The mass transfer expression for water vapor molecules through the membrane is expressed by Eq. (16) [25, 26]:

16

$$J=\rho k_m\text{ln}\left(\frac{n_{s,i}}{n_{s,b}}\right)$$where, k_m is the mass transfer coefficient (m/s), n_s,I and n_s,b are the molar fraction of solute at interface and bulk, respectively.

The energy balance on the feed-membrane side can be expressed as follows [25, 26]:

17

$$J{\mathrm{\Delta }H}_v=h_f\left(T_b,-,T_i\right)$$where, J is the water vapor flux (kg/m²h), h_f is the heat transfer coefficient (W/m²K), T_b and T_i are the temperatures of feed at bulk and interface (K), respectively, ΔH_v is the heat of vaporization of water (J/kg).

The molar fraction of solute (n_s,i) in the feed solution near the membrane surface and the temperature at the feed-membrane interface (T_i) were determined using the above model equations and the corresponding concentration polarization coefficient (CPC) and temperature polarization coefficient (TPC) are evaluated using the following expressions [27, 28]:

18

$$CPC=\frac{n_{s,i}}{n_{s,b}}$$and

19

$$\text{TPC}=\frac{{\text{T}}_{\text{i}}-{\text{T}}_{\text{v}}}{{\text{T}}_{\text{b}}-{\text{T}}_{\text{v}}}\text{TPC}=\frac{{\text{T}}_{\text{i}}-{\text{T}}_{\text{p}}}{{\text{T}}_{\text{b}}-{\text{T}}_{\text{p}}}$$

Results and discussion

Membrane characterization

Scanning electrode microscope analysis

The surface and cross-section morphological structures of pristine PVDF and PVDF-PTFE composite membranes are shown in Fig. 3. As can be seen in Fig. 3a, the surface morphology of pristine PVDF membrane appears to be porous in structure with pores evenly distributed over the membrane surface whereas, the composite membrane consisted of larger microporous structure with pores distributed over the surface of polymer matrix Fig. 3c. It can be clearly seen that, the surface morphological property of composite membrane vary drastically. By coating PVDF polymer over the PTFE substrate, the pore diameter of the composite membrane is enhanced and found to be higher than that of pristine PVDF membrane. The casting of PVDF polymer on a PTFE substrate resulted in formation of round shape large pores over the surface of a composite membrane, while smaller pores with sponge like appearance are observed in case of pristine PVDF membrane. The cross sectional morphologies as shown in Fig. 3b and d of both the membranes showed a sponge-like structure with clear differentiation between the active polymer layer and bottom support layer [29, 30]. The use of PTFE as a substrate material for polymer solution casting is considered to be advantageous for VMD process, since the hydropbhobic nature of the substrate material prevents the pore wetting phenomena besides providing good mechanical strength. In addition, the combination of PVDF and PTFE provides a varying range of pore sizes which finally enhances the membrane properties and performance.

Fig. 3 [Images not available. See PDF.]

Surface and cross-sectional morphology of (a, b) pristine PVDF and (c, d) PVDF-PTFE composite membrane

X-Ray diffraction analysis

Figure 4 shows the x-ray diffractive peak of (a) pristine PVDF and (b) PVDF-PTFE composite membrane. The XRD pattern of the pristine PVDF membrane exhibited an amorphous nature while the PVDF-PTFE composite membrane showed a semi-crystalline structure. The high-intensity broad peak appearing at 2θ value of 18.9° in Fig. 4a corresponds to the Miller indices of (0, 2, 0) crystal peak of the α-crystal form of PVDF [31], while the composite membrane in Fig. 4b shows three significant peaks at 2θ = 16°, 22°, 24°. This indicates the presence of PTFE in the composite membrane [32]. Further, it can be observed that the peak intensity at 2θ value of 24° corresponds to the Miller indices of (1, 1, 0) crystal peak of β phase structure for composite membrane. The study confirms the composite membrane signifies the crystallinity behavior of the membrane.

Fig. 4 [Images not available. See PDF.]

XRD Spectra of (a) pristine PVDF and (b) PVDF-PTFE composite membrane

Fourier-transform infrared spectroscopy analysis

Figure 5 shows the FTIR spectra of (a) pristine PVDF and (b) PVDF-PTFE composite membranes. As shown in Fig. 5a, the pristine PVDF membrane exhibited symmetrical and asymmetrical vibrations of the -CH₃ band at 3026 cm⁻¹ and 2984 cm⁻¹, respectively, while the peak appearing at 1441 cm⁻¹ corresponds to bending or scissoring vibrations of the -CH₂ group and the peak at 1179 cm⁻¹ is due to -CF stretching vibration within the PVDF matrix. As can be seen in Fig. 5b, the PVDF-PTFE composite membrane showed absorption bands appearing at 2846.71 cm⁻¹ and 2911.89 cm⁻¹ represent the symmetric and asymmetric stretching vibrations of the -CH₂ group [33], respectively. In addition, the bands originating at 720.83 cm⁻¹ represent the α-type crystal of the PVDF membrane [34]. Further, the bands at 1740.84 cm⁻¹ denote the stretching vibration of C = C stretching vibration [35]. The bands appearing at 1104.13 and 1035.27 cm⁻¹ are mainly due to asymmetrical and symmetric stretching vibration of the -CF₂ group in PTFE whereas the asymmetric bending of CH₂ is observed at 1464.76 cm⁻¹ [36]. The study showed the physical bonding of PVDF and PTFE polymers.

Fig. 5 [Images not available. See PDF.]

FTIR spectra of (a) Pristine PVDF and (b) PVDF-PTFE composite membrane

Thermogravimetric analysis

Figure 6 shows the weight loss variations of the membrane with temperature for pristine PVDF and PVDF-PTFE composite membranes. As seen in the figure, the membranes have undergone a two-step thermal decomposition process. The PVDF membrane exhibited first weight loss at 450 °C followed by second weight loss at around 500 °C. The PVDF-PTFE composite membrane begins to decompose at a lower temperature than the pristine PVDF membrane. It can be observed that the weight reduction at about 290 to 480 °C is due to the cleavage of C-H bonds of the PVDF matrix whereas at higher temperatures at around 490 to 600 °C the decomposition of the main chain of PTFE polymer matrix was observed [37, 38]. The study reveals that the thermal stability of the pristine PVDF membrane was higher than the composite membrane over the range of temperatures studied. Moreover, the pristine PVDF membrane exhibited a gradual weight loss trend after 500 °C, while the composite membrane showed sudden weight loss after 480 to 490 °C operating temperature. The hydrophobic PTFE support layer provides sufficient mechanical strength and stability to the membrane besides helping to avoid pore wetting, while the active PVDF layer allows transportation of water vapors at higher temperature profiles in VMD process.

Fig. 6 [Images not available. See PDF.]

TGA Thermograms of pristine PVDF and PVDF-PTFE composite membranes

Contact angle measurement

The hydrophobicity of the PVDF-PTFE composite membrane was measured by Goniometer with a contact angle image shown in Fig. 7. The measurements were performed 5 times and the mean value of the contact angle was calculated to be 125°. According to the Wenzel model, the surface roughness transforms the membrane material property to be hydrophilic or hydrophobic [39]. If the contact angle of the surface is lower than 90°, the roughing renders it to a hydrophilic nature whereas on the other hand surface with a higher contact angle, roughening transforms a hydrophobic surface into a superhydrophobic surface. In the present study, the PVDF / PTFE membrane with a hydrophobic nature and higher water contact angle (> 90°) renders the surface hydrophobic.

Fig. 7 [Images not available. See PDF.]

Water contact angle of PVDF-PTFE composite membrane

Effect of feed composition

Figure 8 shows the effect of feed concentration on permeate flux and percentage of salt rejection. Each experiment was performed for a period of 3 h and the sample was tested for salt presence at the end of the experiment. The study was done using the PVDF-PTFE composite membrane by varying salt concentration from 10,000 ppm to 40,000 ppm at a constant downstream pressure of 80 mmHg, feed temperature of 70 °C, and feed flow rate of 80 L/h. The increase in feed concentration resulted in a reduction in solvent flux from 2.5 kg/m²h to 1.95 kg/m²h. At lower feed concentrations, their is heightened prevalence of hydration bonds facilitating a more pronounced interaction to promote easier evaporation, subsequently manifesting in elevated permeation flux. The increase in solute concentration decreases the vapor pressure of the solution due to the interference of solute molecules with water molecules to break free for moving through the membrane into the vapor phase. There was a noticeable reduction in flux value at a higher salt concentration of 40,000 ppm. This phenomenon can further be explained by concentration and temperature polarization near the membrane surface [40]. At a constant feed temperature, the increase in salt concentration will lower the water activity coefficient of the salt solution thereby resulting in a reduction in pressure gradient across the membrane surface which further leads to less permeation of water vapor through pores of the membrane. The decline in permeate flux at elevated salt concentrations can occasionally precipitate crystallization and scaling on the membrane surface. However, it has been discerned that the impact of scaling in the context of water desalination through VMD is comparatively subdued and does not had a significant effect on percentage of salt rejection. Hence, the study showed that opting for membrane distillation for water desalination could be considered the most advantageous process when compared to the reverse osmosis process for drinking water purification. In addition, the lower hydrostatic pressure requirement makes membrane distillation a competitive process for water purification [41].

Fig. 8 [Images not available. See PDF.]

Effect of feed concentration on flux and percentage of salt rejection

Effect of permeate pressure

Figure 9 shows the influence of downstream pressure on water flux and NaCl rejection at a constant feed NaCl solution concentration of 10,000 ppm, operating feed temperature of 80 °C, and feed flow rate of 80 L/h using PVDF-PTFE composite membrane. The experiment on the effect of downstream pressure was performed for a period of 3 h and the sample was analyzed for percentage of salt rejection at the end of each experiment. It was observed that, with increasing permeate pressure from 80 to 120 mmHg, the permeate flux decreases from 3 to 2.66 kg/m².h. Notably, the results reveal a distinct correlation between permeate pressure levels and permeate flux. The lower transmembrane pressure difference across the membrane result in reduced permeate flux. It is essential to recognize that this nuanced relationship underscores the sensitivity of the system to variations in downstream pressure, shedding light on the intricate dynamics governing the VMD process. Furthermore, the study showed a reduction in the salt rejection percentage as the permeate pressure increased. The salt rejection was found to reduce from 99.85% to 99.52%.

Fig. 9 [Images not available. See PDF.]

Effect of permeate pressure on flux and percentage of salt rejection

Effect of feed temperature

Figure 10 shows the effect of feed temperature on water flux and percentage of salt rejection across the membrane at a constant feed concentration of 10 g/L of NaCl solution and downstream pressure of 80 mm Hg. The inlet temperature of the brine solution was varied from 50 ℃ to 80 ℃ for an inlet feed flow rate of 80 L/h. As can be seen in Fig. 10, the inlet temperature of the brine solution has a strong influence on the water flux. The increase in feed temperature from 50 ℃ to 80 ℃ increased the water vapor flux from 1.47 to 3 kg/m²h with percentage of salt rejection varying from 99.27 to 99.86%. This is mainly due to enhancement in driving force i.e. the vapor pressure gradient across the membrane which can be explained by the Antoine equation [24, 25], elucidating the intricate thermodynamic interactions governing the membrane distillation process. Higher temperatures contribute to an increase in the kinetic energy of water molecules in the feed solution. Consequently, the increased kinetic energy of the water molecules accelerates the rate of evaporation of water vapor through the pore of the membrane, leading to higher water flux [42].

Fig. 10 [Images not available. See PDF.]

Effect of feed temperature on permeate flux and percentage of salt rejection

Effect of operating parameters on temperature and concentration polarization

The heat and mass transfer in membrane distillation emerge due to uneven mixing of hot fluid in the feed near the boundary layer on the surface of the membrane. The applied vacuum on the downstream side of the membrane facilitates the vaporization of water molecules from the surface of the membrane further resulting in a temperature gradient i.e. the temperature near the membrane surface is lower when compared to feed bulk. This phenomenon is termed as temperature polarization [43, 44]. Figure 11a shows the effect of feed temperature on the temperature polarization coefficient for a VMD system operated with commercially available PVDF-PTFE composite membrane at a constant flow rate of 80 l/h and and constant downstream pressure of 80 mmHg. It can be observed that the rise in feed temperature from 50 °C to 80 °C increased TPC from 0.83 to 0.96. The increase in TPC can be attributed to the rise to driving force across the membrane which is resulting from rise in feed temperature. In addition, the increased feed temperature reduces the feed viscosity consequently enhancing the mass transfer within the system [27, 45]. The observed TPC value of near to 1 indicates less impact of temperature polarization on the process which ultimately leads to well designed MD cell. A study reported by Politano et al. used silver nanofiller incorporated microporous PVDF membrane to overcome the impact of temperature polarization effect in the MD process for seawater desalination [46]. MD also experiences concentration polarization phenomena i.e. the solute concentration near the membrane surface is higher than the bulk feed solution. It is believed that the impact of concentration polarization is minimal in the MD process when compared to temperature polarization. However, the effect of concentration polarization results in the accumulation of salt crystal particles over the membrane surface [47]. Figure 11b shows the variation in CPC with feed solute concentration. Evaluation of CPC was done for varying feed concentration from 10,000 to 40,000 ppm and constant feed temperature and feed flow rate of 323 K and 80 l/h, respectively. It can be observed that with an increase in NaCl concentration, the CPC was found to decrease from 1.034 to 1.031. This is due to an increase in solute concentration near the membrane surface leading to an increase in boundary layer thickness. The present observations in the study align with the results furnished by Gayatri et al., where they identified a minimal decrease in CPC with increase in feed NaCl concentration [48].

Fig. 11 [Images not available. See PDF.]

Effect of feed temperature on (a) temperature polarization coefficient and (b) feed concentration on concentration polarization coefficient

Conclusions

This study has attempted the use of an indigenously synthesized microporous polyvinylidene fluoride (PVDF) membrane deposited on a commercial polytetrafluoroethylene (PTFE) substrate for desalination using the vacuum-driven membrane distillation (VMD) technique. The membrane was characterized for physicochemical properties using SEM, FTIR, TGA, and XRD tools besides hydrophobicity by measuring the contact angle for the membrane. It was observed that the membrane porosity plays a dominant role during mass or heat transport of water vapor molecules from the bulk of the feed side through the membrane to the permeate side. The performance of membranes was investigated at varying operating conditions of salt concentration, downstream pressure, feed temperature, and flow rate. The process has achieved a high water flux of 3 kg/m²h with a salt rejection of 99.86% for the PVDF-PTFE composite membrane. The percentage of salt rejection was found to remain stable (up to 99%) even when operated at high feed temperatures which indicates the potential of VMD as a suitable process for desalination application. The mass and heat transport mechanism of water vapor through the membrane was examined considering the vapor transference route to be Knudsen molecular diffusion. The prediction of the molar fraction of solute and the temperature at the feed-membrane interface was done by solving the mass and heat transport model equations using MATLAB software tool. The study showed that the system performance is not majorly affected by the concentration or temperature polarization coefficients. If powered by solar energy, the presently developed VMD membrane process could prove to be an economical alternative to multi-stage flash distillation or multiple-effect evaporation for the desalination of sea and brackish water. Besides water desalination, the synthesized membrane has great potential for applications including oil/water separation, dehumidification, air purification and effluent treatment by membrane bioreactor and biomedical field.

Acknowledgements

The authors acknowledge Director, CSIR-IICT for supporting research work with manuscript communication no. IICT/Pubs./2022/309.

Author contributions

Dr. M. Madhumala has carried out experimental trials, done formal analysis and drafted the manuscript. Ms. S. Srishti has helped in conducting the experiments and anlysed the water samples for conductivity and total dissolved content. Ms. S. Fatima has contributed by solving the mass and heat transfer model equations using MATLAB tool to determine the concentration and temperature polarization coefficients for the system. Dr. S. Sridhar has helped in editing the draft.

Funding

The authors are thankful to the Department of Science and Technology, New Delhi for providing funds to perform this research work under research grant code: GAP-789.

Data availablity

Data will be made available by the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Abbreviations

J

Water permeation flux (kg/m2s)

%R

Percentage salt rejection (–)

W

Amount of product collected (kg)

A

Effective membrane area (m²)

t

Time of operation (h)

C_F

Concentrations of sodium chloride ions in feed sample (ppm)

C_P

Concentrations of sodium chloride ions in permeate sample (ppm)

K_n

Knudsen number (–)

λv

Mean distance traveled by the water vapor molecule (m)

dp

Membrane pore diameter (µm)

k_B

Boltzmann constant (J/K)

ν

Collision factor (M)

LEP

Lquid entry pressure (MPa)

β

Geometric coefficient (–)

cosθ

Contact angle (°)

r_max

Pore radii (m)

γ_L

Surface tension of liquid (mN/m)

T_b

Bulk surface temperature (K)

P_m

Mean pressure within the membrane pores (Pa)

ΔP

Pressure drop across the membrane (Pa)

P₁

Partial vapor pressure of water molecule in the feed solution (Pa)

P₂

Partial vapor pressure of water vapor molecule on the downstream side of the membrane (Pa)

x_w

Mole fraction of water (–)

γ_w

Activity coefficient for a non-ideal NaCl-water binary mixture (–)

n_sm

Solute concentration at the membrane surface (mol/m³)

P_v

Vapor pressure of the water molecule (Pa)

d_h

Hydraulic diameter (m)

L

Length of the feed flow channel (m)

µ₁

Viscosity of water at the feed/membrane interface (Pa.s)

µ_b

Viscosity of water at bulk feed (Pa.s)

Re

Reynold number (–)

Pr

Prandtl number (–)

Sc

Schmidt number (–)

ρ

Bulk density (kg/m³)

v

Velocity of feed (m/s)

d_h

Hydraulic diameter (m)

µ

Dynamic viscosity of feed (kg/m.s)

k

Thermal conductivity of feed (W/mK)

D

Diffusivity (m²/s)

k_m

Mass transfer coefficient (kg/(m².sPa))

n_s

_I: Molar fraction of solute at interface (–)

n_s

_b: Molar fraction of solute at bulk (–)

J

Water vapor flux (kg/m²h)

h_f

Heat transfer coefficient (W/m²K)

T_b

Bulk feed temperature (K)

T_i

Temperature of feed at interface (K)

ΔH_v

Heat of vaporization of water (J/kg)

C_p

Specific heat capacity (J/kgK)

T_w

Boiling point of water (K)

T_f

Feed temperature (K)

T_p

Permeate side temperature (K)

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Pandey, S; Mohapatra, G; Arora, R. Groundwater quality, human health risks and major driving factors in arid and semi-arid regions of Rajasthan, India. J Clean Prod; 2023; 15, 427 [DOI: https://dx.doi.org/10.1016/j.jclepro.2023.139149]

2. Pangarkar, BL; Sane, MG; Guddad, M. Reverse osmosis and membrane distillation for desalination of groundwater: a review. J Mater Sci; 2011; 2011, pp. 1-9.

3. Bamasag, A; Almatrafi, E; Alqahtani, T; Phelan, P; Ullah, M; Mustakeem, M; Obaid, M; Ghaffour, N. Recent advances and future prospects in direct solar desalination systems using membrane distillation technology. J Clean Prod; 2023; 20, 385 [DOI: https://dx.doi.org/10.1016/j.jclepro.2022.135737]

4. Yang, G; Zhang, J; Peng, M; Du, E; Wang, Y; Shan, G; Ling, L; Ding, H; Gray, S; Xie, Z. A mini review on antiwetting studies in membrane distillation for textile wastewater treatment. Process; 2021; 9, 243.[COI: 1:CAS:528:DC%2BB3MXhvF2ns7bP] [DOI: https://dx.doi.org/10.3390/pr9020243]

5. Jiao, B; Cassano, A; Drioli, E. Recent advances on membrane processes for the concentration of fruit juices: a review. J Food Eng; 2004; 63, pp. 303-324. [DOI: https://dx.doi.org/10.1016/j.jfoodeng.2003.08.003]

6. Teoh, MM; Chung, TS; Yeo, YS. Dual-layer PVDF/PTFE composite hollow fibers with a thin macrovoid-free selective layer for water production via membrane distillation. Chem Eng J; 2011; 171, 2 pp. 684-691.[COI: 1:CAS:528:DC%2BC3MXotVekt7Y%3D] [DOI: https://dx.doi.org/10.1016/j.cej.2011.05.020]

7. Giraldo-Mejía, H; Quintero, YM; Mery, F; Rodriguez, F; Curcio, E; Estay, H; Garcia, A. Plasma-grafting surface modifications to enhance membrane hydrophobicity for brine membrane distillation. Desalination; 2023; 1, 567 [DOI: https://dx.doi.org/10.1016/j.desal.2023.116942]

8. Dufour, T. From basics to frontiers: a comprehensive review of plasma-modified and plasma-synthesized polymer films. Polymers; 2023; 15, 17 3607.[COI: 1:CAS:528:DC%2BB3sXhvFaqtL%2FF] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37688233][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10490058][DOI: https://dx.doi.org/10.3390/polym15173607]

9. Li, H; Shi, W; Zeng, X; Huang, S; Zhang, H; Qin, X. Improved desalination properties of hydrophobic GO-incorporated PVDF electrospun nanofibrous composites for vacuum membrane distillation. Sep Purif Technol; 2020; 2, 230 [DOI: https://dx.doi.org/10.1016/j.seppur.2019.115889]

10. Zhang, H; Liu, M; Sun, D; Li, B; Li, P. Evaluation of commercial PTFE membranes for desalination of brine water through vacuum membrane distillation. Chem Eng Process Process Intensif.; 2016; 110, pp. 52-63.[COI: 1:CAS:528:DC%2BC28Xhs1Srur%2FI] [DOI: https://dx.doi.org/10.1016/j.cep.2016.09.023]

11. Dong, ZQ; Ma, XH; Xu, ZL; You, WT; Li, FB. Superhydrophobic PVDF–PTFE electrospun nanofibrous membranes for desalination by vacuum membrane distillation. Desalination; 2014; 347, pp. 175-183.[COI: 1:CAS:528:DC%2BC2cXhtVGqu7fF] [DOI: https://dx.doi.org/10.1016/j.desal.2014.05.015]

12. Zhelun, L; Rana, D; Matsuura, T; Lan, CQ. The performance of polyvinylidene fluoride-polytetrafluoroethylene nanocomposite distillation membranes: an experimental and numerical study. Sep Purif Technol; 2019; 226, pp. 192-208. [DOI: https://dx.doi.org/10.1016/j.seppur.2019.05.102]

13. Tong, D; Wang, X; Ali, M; Lan, CQ; Wang, Y; Drioli, E; Wang, Z; Cui, Z. Preparation of Hyflon AD60/PVDF composite hollow fiber membranes for vacuum membrane distillation. Sep Purif Technol; 2016; 157, pp. 1-8.[COI: 1:CAS:528:DC%2BC2MXhvFWhtrnO] [DOI: https://dx.doi.org/10.1016/j.seppur.2015.11.026]

14. Fan, H; Peng, Y. Application of PVDF membranes in desalination and comparison of the VMD and DCMD processes. Chem Eng Sci; 2012; 79, pp. 94-102.[COI: 1:CAS:528:DC%2BC38XpvFKktL4%3D] [DOI: https://dx.doi.org/10.1016/j.ces.2012.05.052]

15. Cao, W; Mujtaba, IM. Simulation of vacuum membrane distillation process for desalination with Aspen Plus. Ind Eng Chem Res; 2015; 54, 2 pp. 672-680.[COI: 1:CAS:528:DC%2BC2cXitFCrsL3N] [DOI: https://dx.doi.org/10.1021/ie502874c]

16. Zhang, Y; Peng, Y; Ji, S; Wang, S. Numerical simulation of 3D hollow-fiber vacuum membrane distillation by computational fluid dynamics. Chem Eng Sci; 2016; 152, pp. 172-185.[COI: 1:CAS:528:DC%2BC28XhtVWlurfJ] [DOI: https://dx.doi.org/10.1016/j.ces.2016.05.040]

17. Alsalhy, QF; Ibrahim, SS; Hashim, FA. Experimental and theoretical investigation of air gap membrane distillation process for water desalination. Chem Eng Res Des; 2018; 130, pp. 95-108.[COI: 1:CAS:528:DC%2BC2sXhvF2jtrbJ] [DOI: https://dx.doi.org/10.1016/j.cherd.2017.12.013]

18. Mohammad Ameen, NA; Ibrahim, SS; Alsalhy, QF; Figoli, A. Highly saline water desalination using direct contact membrane distillation (DCMD): experimental and simulation. Water; 2020; 12, 6 1575. [DOI: https://dx.doi.org/10.3390/w12061575]

19. Ibrahim, SS; Alsalhy, QF. Modeling and simulation for direct contact membrane distillation in hollow fiber modules. AIChE J; 2013; 59, pp. 589-603.[COI: 1:CAS:528:DC%2BC38XotVWjtbY%3D] [DOI: https://dx.doi.org/10.1002/aic.13845]

20. Franken, A; Nolten, J; Mulder, M; Bargeman, D; Smolders, C. Wetting criteria for the applicability of membrane distillation. J Membr Sci; 1987; 33, pp. 315-328.[COI: 1:CAS:528:DyaL1cXotVWj] [DOI: https://dx.doi.org/10.1016/S0376-7388(00)80288-4]

21. Khayet, M; Godino, MP; Mengual, JI. Modelling transport mechanism through a porous partition. J Non-Equilibrium Thermodyn; 2001; 26, pp. 1-14.[COI: 1:CAS:528:DC%2BD3MXktFyku7k%3D] [DOI: https://dx.doi.org/10.1515/JNETDY.2001.001]

22. Lawson, KW; Lloyd, DR. Membrane distillation. J Membr Sci; 1997; 124, pp. 1-25.[COI: 1:CAS:528:DyaK2sXpvVGktw%3D%3D] [DOI: https://dx.doi.org/10.1016/S0376-7388(96)00236-0]

23. Gryta, M. Concentration of NaCl solution by membrane distillation integrated with crystallization. Sep Sci Technol; 2002; 37, 15 pp. 3535-3558.[COI: 1:CAS:528:DC%2BD38XoslKnsLY%3D] [DOI: https://dx.doi.org/10.1081/SS-120014442]

24. pp.649–652, "Introduction to Material and Energy Balances" by G.V. Reklaitis, John Wiley & Sons, 1983

25. Alkhudhiri, A; Darwish, N; Hilal, N. Membrane distillation: A comprehensive review. Desalination; 2012; 87, pp. 2-18. [DOI: https://dx.doi.org/10.1016/j.desal.2011.08.027]

26. Mericq, JP; Laborie, S; Cabassud, C. Vacuum membrane distillation of sweater reverse osmosis brines. Water Res; 2010; 44, 18 pp. 5260-5273.[COI: 1:CAS:528:DC%2BC3cXht1KhtbfJ] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/20659753][DOI: https://dx.doi.org/10.1016/j.watres.2010.06.052]

27. Abu-Zeid, MAE; Zhang, Y; Dong, H; Zhang, L; Chen, H-L; Hou, L. A comprehensive review of vacuum membrane distillation technique. Desalination; 2015; 356, pp. 1-14.[COI: 1:CAS:528:DC%2BC2cXhvVamsbnJ] [DOI: https://dx.doi.org/10.1016/j.desal.2014.10.033]

28. Samuel, O. Olatunji and lucy mar camacho, heat and mass transport in modeling membrane distillation configurations: a review. Front Energy Res; 2018; 6, pp. 1-18.

29. Ting, H; Xiaogang, L; Qian, W; Yue, Z; Xiaozu, W; Zhaohui, W; Naser, T. Zhaoliang, Poly(vinylidene fluoride) (PVDF) membrane fabrication with an ionic liquid via non-solvent thermally induced phase separation (N-TIPs) C. Appl Water Sci; 2022; 12, 42 pp. 1-11.

30. Tianshu, L; Xiaoji, Z; Yizhuo, S; Renbi, B. Anticorrosion performance of PVDF membranes modified by blending PTFE nanoemulsion and prepared through usual non-solvent-induced phase inversion method. Membranes; 2021; 11, 420. [DOI: https://dx.doi.org/10.3390/membranes11060420]

31. Luo, X; Yang, J; Zhang, R; Li, H; Guo, Y; Mu, R; Zhao, Y. Microstructures and properties of PVDF membranes: effect of the solvent. Ferroelectr; 2022; 595, pp. 95-108.[COI: 1:CAS:528:DC%2BB38Xit1GqurnI] [DOI: https://dx.doi.org/10.1080/00150193.2022.2079461]

32. RaviKumar, C; Song, JI. Harvesting of flow current through implanted hydrophobic PTFE surface within silicone-pipe as liquid nanogenerator. Sci Rep; 2022; 12, 3700. [DOI: https://dx.doi.org/10.1038/s41598-022-07614-5]

33. Baig, N; Arshad, Z; Ali, SA. Synthesis of a biomimetic zwitter ion icpenta polymer to fabricate high-performance PVDF membranes for efficient separation of oil-in-water nano-emulsions. Sci Rep; 2022; 12, 5028 pp. 1-15.

34. Shen, X; Xie, T; Wang, J; Wang, F. Improved fouling resistance of poly(vinylidene fluoride) membrane modified with poly(acryloyl morpholine)-based amphiphilic copolymer. Colloid Polym Sci; 2017; 295, 7 1211.[COI: 1:CAS:528:DC%2BC2sXovFCktLs%3D] [DOI: https://dx.doi.org/10.1007/s00396-017-4117-6]

35. Jaleh, B; Gavary, N; Fakhri, P; Muensit, N; Taheri, SM. Characteristics of PVDF membranes irradiated by electron beam. Membranes (Basel); 2015; 5, 1 pp. 1-10. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25569360][DOI: https://dx.doi.org/10.3390/membranes5010001]

36. Piwowarczyk, J; Jedrzejewski, R; Moszynski, D; Kwiatkowski, K; Niemczyk, A; Baranowska, J. XPS and FTIR studies of polytetrafluoroethylene thin films obtained by physical methods. Polym; 2019; 11, 1629 pp. 1-13.

37. Wang, R; Xu, G; He, Y. Structure and properties of polytetrafluoroethylene (PTFE) fibers. E-Polym.; 2017; 17, 3 pp. 215-220.[COI: 1:CAS:528:DC%2BC2sXlsVWntr0%3D] [DOI: https://dx.doi.org/10.1515/epoly-2016-0059]

38. Marshall, JE; Zhenova, A; Roberts, S; Petchey, T; Zhu, P; Dancer, CEJ; McElroy, CR; Kendrick, E; Goodship, V. On the Solubility and Stability of Polyvinylidene Fluoride. Polym.; 2021; 13, 1354 pp. 1-31.

39. Wang, B; Zhang, Y; Shi, L; Li, J; Guo, Z. Advances in the theory of superhydrophobic surfaces. J Mater Chem; 2012; 22, pp. 20112-20127.[COI: 1:CAS:528:DC%2BC38XhtlagsbnO] [DOI: https://dx.doi.org/10.1039/c2jm32780e]

40. Suleman, M; Asif, M; Jamal, SA. Temperature and concentration polarization in membrane distillation: a technical review. Desalin Water Treat; 2021; 229, pp. 52-68.[COI: 1:CAS:528:DC%2BB3MXitFGgt7%2FO] [DOI: https://dx.doi.org/10.5004/dwt.2021.27398]

41. Drioli, E; Ali, A; Macedonio, F. Membrane distillation: recent developments and perspectives. Desalination; 2015; 356, pp. 56-84.[COI: 1:CAS:528:DC%2BC2cXhsl2gtbfI] [DOI: https://dx.doi.org/10.1016/j.desal.2014.10.028]

42. Khayet, M.; Matsuura, T. Membrane distillation principles and applications. The Netherlands, 2011. Elsevier.

43. Alklaibi, AM; Lior, N. Membrane-distillation desalination: Status and potential. Desalination; 2005; 171, 2 pp. 111-131.[COI: 1:CAS:528:DC%2BD2MXhtlGm] [DOI: https://dx.doi.org/10.1016/j.desal.2004.03.024]

44. El-Bourawi, MS; Ding, Z; Ma, R; Khayet, M. A framework for better understanding membrane distillation separation process. J Membr Sci; 2006; 285, 1–2 pp. 4-29.[COI: 1:CAS:528:DC%2BD28XhtFSntLfF] [DOI: https://dx.doi.org/10.1016/j.memsci.2006.08.002]

45. Triki, Z; Fergani, Z; Lekmine, S; Tahraoui, H; Amrane, A; Zamouche, M; Kebir, M; Aymen, A; Khezami, L; Zhang, J. Numerical modelling and performance evaluation of vacuum membrane distillation for energy-efficient seawater desalination: towards energy-efficient solutions. Water; 2023; 15, 3612 pp. 1-25.

46. Politano, A; Argurio, P; Profio, GD; Sanna, V; Cupolillo, A; Chakraborty, S; Arafat, HA; Curcio, E. Photothermal membrane distillation for seawater desalination. Adv Mater; 2017; 29, 1603504. [DOI: https://dx.doi.org/10.1002/adma.201603504]

47. Lokare, OR; Vidic, RD. Impact of operating conditions on measured and predicted concentration polarization in membrane distillation. Environ Sci Technol; 2019; 53, 20 pp. 11869-11876.[COI: 1:CAS:528:DC%2BC1MXhvVWqurfP] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31545033][DOI: https://dx.doi.org/10.1021/acs.est.9b04182]

48. Gayathri, N; Yongjun, C; Sanghyun, J; TaeMun, H; Saravanamuthu, V. Experiments and modeling of a vacuum membrane distillation for high saline water. J Ind Eng Chem; 2014; 4, 20 pp. 2174-2183.