Effect of polymer dope solution temperature on the fabrication of flat-sheet polyvinylidene fluoride (PVDF) membranes: Water filtration and membrane distillation (MD) applications

(1)

Effect of polymer dope solution temperature on the

fabrication of flat-sheet polyvinylidene fluoride

(PVDF) membranes: Water filtration and membrane

distillation (MD) applications

Mohammed Alsultan

Master thesis: 15 ECTS Supervisor: Naser Tavajohi Hassan Kiadeh

Examiner: Passed:

(2)

(3)

I

Abstract

In this project, polyvinylidene fluoride (PVDF) membranes were prepared at different dope solution temperatures (i.e. 25C, 45C, and 65C) by using vapour-induced phase separation. The possibility of dissolving PVDF in organic solvents was investigated using the Hansen solubility parameters. The theoretical calculation and experimental data indicated that dimethylacetamide (DMAc) has the highest solvency power. By increasing the dope solution temperature, the membrane morphology was changed from a dense to a porous structure. The overall porosity and water permeability were increased slightly. Furthermore, X-ray diffraction (XRD) measurements and Fourier transform infrared (FTIR) spectroscopy demonstrated significant changes in the polymorphisms of the prepared membranes. A lower dope solution temperature promoted β-phase polymorphism in the PVDF membrane structure.

(4)

(5)

III

List of abbreviations



d

Dispersion forces



p

Dipole-dipole forces



h

Hydrogen-bonding interactions

ρ

k

The density of kerosene

ρ

p

The polymer density

A

The membrane filtration area (m

2

)

C

p

The concentration of permeate (g/L)

C

f

The concentration of feed (g/L)

DIW

Deionised water

DMAc

N,N-dimethylacetamide

EIPS

Evaporation-induced phase separation

FTIR

Fourier transform infrared

HSPs

Hansen solubility parameters

Jw

Water permeation flux (kg/m

2

_h)

NIPS

Non-solvent-induced phase separation

PVDF

Polyvinylidene fluoride

P

Pressure of 0.1 MPa or 1 bar

RED

Relative energy difference

Ra

Solubility parameter distance

Ro

Hansen solubility parameter sphere

SEM

Scanning electron microscopy

TIPS

Thermally induced phase separation

t

The time taken to collect a certain amount of water (h)

T

p

Permeate temperature (ºC)

T

f

Feed temperature (ºC)

VIPS

Vapor-induced phase separation

V

The volume of water permeated through the membrane (liters)

wt

w

The weight of the wet membrane

wt

d

The weight of the dry membrane

XRD

X-ray diffraction

Author contribution

In this project, the author managed the experiments and analysed the data. The project was supported by Dr Naser Tavajohi. Additionally, he guided and supervised the author throughout the project.

(6)

(7)

V

Abstract ... I

List of abbreviations ... III

Author contribution ... III

1. Introduction ... 1

1.2 Aim of the diploma work ... 1

2. Popular scientific summary including social and ethical aspects ... 1

2.1 Popular scientific summary... 1

2.2 Social and ethical aspects ... 2

3. Experimental ... 2

3.1 Chemicals ... 2

3.2 Solubility parameters ... 2

3.3 Preparation of flat-sheet membranes ... 3

3.4 Characterisation of prepared PVDF flat-sheet membranes ... 3

3.4.1 Membrane morphology study ... 3

3.4.2 Porosity ... 3

3.4.3 Thickness (δ) and liquid entry pressure (LEP) of water ... 4

3.4.4 Fourier transform infrared (FTIR) spectroscopy ... 4

3.4.5 X-ray diffraction (XRD) measurements ... 4

3.4.6 Water filtration permeability... 4

3.4.7 Membrane distillation (MD) performance ... 4

4. Results and discussion ... 5

4.1 Solubility parameter ... 5

4.2 PVDF membrane characterisation ... 7

4.2.1 Morphologies of the prepared flat-sheet membranes ... 7

4.2.2 Porosity (ε), thickness (δ), and liquid entry pressure (LEP) of water ... 8

4.2.3 XRD and FTIR patterns ... 8

4.2.4 Water filtration permeability... 9

4.2.5 Membrane distillation (MD) performance ... 10

5. Conclusions and outlook ... 11

Acknowledgement ... 12

References ... 12

(8)

(9)

1

1. Introduction

Polyvinylidene fluoride (PVDF) is a semi-crystalline polymer containing crystalline and amorphous phases. The crystalline phase provides thermal stability, and the amorphous phase has ﬂexibility. PVDF is widely used in preparing various membranes for microfiltration, ultrafiltration, and nanofiltration due to its excellent mechanical, thermal, and chemical stability. PVDF membrane properties are interesting in the new field of membranes-, such as in membrane reactors, membrane distillation, membrane condensers, and membrane crystallisers [1].

PVDF has five forms of polymorphism: α, β, γ, δ, and ɛ forms [2, 3]. Among the different polymorphisms, the β phase has interesting properties for membrane applications due to its piezoelectric properties. In the β phase, the fluorine and hydrogen atoms are arranged in a zig-zag shape, which is the most stable form from a thermodynamic perspective. For other polymeric membranes, the morphology of the PVDF membrane is a crucial factor in determining its separation efficiency and stability. The PVDF membrane morphology can be tailored during preparation to symmetric, asymmetric, dense, or porous structures. The final morphology is determined based on the target application. PVDF membrane morphologies include symmetrical cellular, symmetrical spherulitic, bicontinuous, and dense structures [3]. Generally, membranes can be fabricated via several methods, such as sintering, track-etching, stretching, and phase inversion. Among the different methods for fabricating membranes, the phase inversion method is the most popular technique for polymeric membrane fabrication. Different phase inversion methods exist, such as non-solvent-induced phase separation (NIPS), thermally-induced phase separation (TIPS), evaporation-induced phase separation (EIPS), and vapour-induced phase separation (VIPS). The final morphology and properties of the membrane are the main factors in determining an appropriate phase inversion method. In all the phase inversion methods, a homogeneous polymer solution is prepared in the first step. The polymer and solvent are then separated by an external factor, such as changing the temperature, introducing a non-solvent, or evaporation [4].

Efforts were made to understand the effects of different manufacturing parameters and their effects on the final morphology and efficiency of the membranes developed by the inversion phase technique.

1.2 Aim of the diploma work

This project aimed to investigate the effect of the polymer solution conditions (i.e. dope solution temperature) on the polymeric membrane morphology, filtration, and membrane distillation performance. In the first step, the solubility parameters of popular solvents in the membrane fabrication industry were calculated. In the second step, three PVDF membranes were fabricated at different dope solution temperatures. The fabricated membranes were characterised based on morphology, crystallinity, and surface chemistry by scanning electron microscopy (SEM), X-ray diffraction (XRD) measurements, and Fourier transform infrared (FTIR) spectroscopy. In the last step, the potential of the fabricated membrane for water purification was investigated.

2. Popular scientific summary including social and ethical

aspects

2.1 Popular scientific summary

Among the different separation units, membrane separation has a better performance based on efficiency, flexibility, control, and scaling. Membrane separation has a promising performance in different fields, such as medical applications [5, 6], water and wastewater treatment [1, 7, 8], desalination [1, 9-11], energy storage [12-14], energy production[15], and gas separation [16-18].

(10)

2 Membranes can be fabricated from organic or inorganic materials in flat-sheet, hollow-fibre, and hollow-capsule shapes with symmetric or asymmetric structures [19]. Membranes can be prepared by different methods, such as sol-gel, interfacial reaction, stretching, extrusion, track-etching, and sintering [4, 19]. The most popular method for fabricating polymeric membranes is phase inversion. The different phase inversion methods include vapour-induced phase inversion, which is interesting for investigating phase inversion phenomena due to the slow rate of phase separation. One of the main reasons for interest in vapour-induced phase inversion is the ability to better control the phase inversion phenomena. VIPS can be used to fabricate symmetric, asymmetric, dense, and porous membranes [20].

The effects of the different fabrication parameters on the performance of membranes produced by VIPS are summarised in a comprehensive review paper by Norafiqah et al. [20]. A key parameter that has been neglected for many years is the influence of the dope solution temperature of the polymer. The dope solution temperature is the temperature at which the polymer dissolves to form the homogenous solution. A few studies investigate the effect of the dope solution temperature [21-24].

2.2 Social and ethical aspects

The project is free from ethical concerns. The project was conducted in the chemistry laboratory at Umeå University. The project contributes to the understanding of the fabrication of PVDF membranes from dope solutions at different temperatures. The project was implemented according to the safety regulations of the chemistry department at Umeå University.

3. Experimental

3.1 Chemicals

PVDF (Solef® 1015, powder) was kindly provided by Solvay Specialty Polymers (Bollate, Italy). N,N-dimethylacetamide was purchased from Sigma Aldrich. Acetone was purchased from Merck. Deionised (DI) water (18.2 MΩ cm, 25°C) was produced with a Milli-Q Plus water purification system (Umeå University). None of the chemicals used in this study were purified further.

3.2 Solubility parameters

The Hansen solubility parameters are considered an effective method for determining the appropriate solvent for dissolving a specific polymer [25]. The relative energy difference (RED) is considered one of the best approaches to estimating the miscibility of the solutions. It can accurately predict whether the polymer can dissolve the solvent according to the following criteria [26]:

• If the RED < 1, the solvent is suitable for that polymer. • If RED = 1, the mixture is under boundary conditions. • If RED > 1, the solvent is poor for that polymer.

Three different forces, dispersion forces, dipole-dipole forces, and hydrogen-bonding, played important roles in determining the solubility parameters (RED).

Using Eq.1 and Eq.2, the RED was determined for all solvents. Thus, the appropriate solvent was chosen to dissolve the PVDF.

Ra=√4(𝑑1− 𝑑2)2+ (𝑝1− 𝑝2) 2

+(ℎ1− ℎ2)2 Eq.1

(11)

3 Here, Ra is the solubility parameter distance; d is the dispersion forces, p is the dipole-dipole

forces, h is hydrogen-bonding interactions, and Ro is the radius of interaction of the Hansen

solubility parameter sphere.

Appendix 1 (Table 1) shows the calculated RED values of PVDF and 58 different solvents.

3.3 Preparation of flat-sheet membranes

In this project, VIPS was used to prepare flat-sheet membranes. Three different dope solutions were prepared by dissolving 15 wt% PVDF powder polymer in 85 wt% DMAc solvent under stirring and at constant temperature (i.e. 25°C, 45°C, and 65°C, separately) over 24 h. The homogeneous dope solutions were then allowed to cool at room temperature for 12 h. Using an automatic film applicator, the homogeneous dope solutions were cast using a glass plate and applicator of 150 μm thickness. The cast film membranes were directly transferred into a home-made vapour incubator and exposed to water vapour at a constant 70% RH for 20 min. Finally, the nascent membranes were immersed in a coagulation bath (water at 25°C) to complete the phase inversion. The membranes were washed thrice with DI water and then dried at room temperature for 2 days to remove the residual solvent.

Figure 1. Schematic of vapour-induced phase separation (VIPS) membrane fabrication; (a) polymer dope solution; (b) automatic film applicator; (c) home-made vapour incubator; (d) coagulation bath, and (e) final microscopic structure of the membranes.

3.4 Characterisation of prepared PVDF flat-sheet membranes

3.4.1 Membrane morphology study

The top, bottom, and cross-sectional morphologies of PVDF flat-sheet membranes were

examined by field emission scanning electron microscope (FESEM, Carl Zeiss). In a

SEM study, samples of each membrane were fractured in liquid nitrogen and coated

with 5 nm-thick platinum with a Quorum Q150T-ES sputter coater.

3.4.2 Porosity

The porosity (ε) of the membrane is defined as the volume of the pores divided by the total volume of the membrane. This can be determined by measuring the weight of the wet membrane using kerosene, which penetrates into the membrane pores. The membrane was kept in kerosene for 24 hours and then weighed. The wet membrane was dried in a vacuum oven at 80°C for 24 h before measuring the dried weight. The PVDF membrane porosity was calculated using Eq. 3. 𝜺 (%) = 𝒘𝒕𝒘−𝒘𝒕𝒅 𝝆𝒌 𝒘𝒕𝒘−𝒘𝒕𝒅 𝝆𝒌 + 𝒘𝒕𝒅 𝝆𝒑 × 𝟏𝟎𝟎 Eq. 3

(12)

4 Where wtw is the weight of the wet membrane, wtd is the weight of the dry membrane, ρk is the

density of kerosene, and ρp is the polymer density.

Three different samples of each membrane were considered to determine the average porosity.

3.4.3 Thickness (δ) and liquid entry pressure (LEP) of water

The membrane thickness was determined with a digital micrometer with a precision of ± 0.1 μm, at 10 positions for each membrane, and the average value was determined with its standard deviation.

The LEP is an important characteristic of PVDF-based membranes. The LEP is defined as the minimum transmembrane pressure required for water to penetrate the large pore size. LEP measurements were performed using a stirred cell 47mm (Millipore). The effective membrane area was (0.00173 m2_{). The container was filled with 25 mL distilled water, and pressure was}

applied gradually to the water at 24°C. The minimum applied pressure before water penetrates the maximum pore size is the LEP value. These experiments were conducted thrice using three different membrane samples produced from different parts of the prepared flat-sheet membranes. The average values and their standard deviations were reported.

3.4.4 Fourier transform infrared (FTIR) spectroscopy

The elemental and functional group analysis of the prepared PVDF flat membranes was performed using attenuated total reflectance-fourier transform infrared spectroscopy (ATR- FTIR) (Model: Bruker Vertex 80v FT-IR spectrometer) coupled with a detector using a scan range from 400-3500 cm-1_{. FTIR spectra were acquired to compare the three different PVDF}

flat-sheet membranes and monitor any remaining residual solvent in the prepared membranes.

3.4.5 X-ray diffraction (XRD) measurements

The crystal structure and phase state of prepared flat-sheet membranes and the physical structures of membranes were characterised by powder XRD (Bruker D8 diffractometer). The scanning angle (2θ) ranged from 5 to 40 using Cu Kα radiation (λ = 1.5418 Å, 30 kV, 40 mA).

3.4.6 Water filtration permeability

The filtration test was conducted using a dead-end stirred cell (Millipore; 47mm; effective area 0.00173 m2_{) to understand the effect of the preparation parameters (i.e. temperatures of dope}

solutions) on the flow through membranes. The filtration cells were operated at a constant regulated pressure of 1 bar (100 kPa) using nitrogen (N2). The DI water was pumped through

the membrane at 25°C. The average water flux was obtained for all the membranes prepared with different dope solution temperatures (i.e. 25°C, 45°C, and 65°C) [27, 28]. The water flux (J) was determined using Eq. 4, and the permeability (P) was determined using Eq. 5. The average value and standard deviation were reported for each membrane.

𝐽 = V

A.t Eq. 4

𝑃𝑊𝑃 = V

A.t.P Eq. 5

Here, V (L) is the volume of water permeated through the membrane, A (m2_{) is the membrane}

filtration area, t (h) is the sampling time, and P (bar) is the applied pressure.

3.4.7 Membrane distillation (MD) performance

MD is a thermally driven process, which transport vapour molecules through porous hydrophobic membranes. The hydrophobicity of the membrane prevents liquid solutions from entering its pores. Various MD modes can be used, such as direct contact membrane distillation (DCMD). Generally, DCMD is used for producing distilled or potable water from aqueous feed solutions containing salt (NaCl), for example, desalination.

In this study, a DCMD experiment was conducted to evaluate the membrane permeability (Jw) and salt rejection factor (SF). As shown in Figure 2, the instrumental setup of DCMD comprises two thermostatic cycles (feed and permeate) connected to a stainless steel membrane module.

(13)

5 The feed compartment was connected to a heating resist, which was maintained at a high temperature, and the permeate compartment was connected to a cooling system, which was maintained at a low temperature. The membrane was placed between the two compartments. The hot feed solution was transported to the top layer and the cold permeate solution to the bottom layer of the membrane. The effective membrane area of the DCMD system was 0.006 m2_.

These experiments were conducted for pure water and an aqueous salt (NaCl) solution of concentration (Cf) 30 g/L at different feed temperatures (Tf) varying from 20ºC to 80ºC and a

permeate temperature (Tp) of 20ºC. The circulation feed and permeate rates were maintained

constant at 0.35 L/h.

The permeate flux (L/m2_{h) was calculated using Eq. 3, and the salt rejection factor was}

calculated using Eq. 6:

SF = (1 −C_Cp

f) . 100 Eq. 6

where Cf and Cp are the concentrations of the feed and permeate solutions, respectively.

Figure 2. Schematic diagram of DCMD setup: (1) Permeate tank; (2) Cooling system (3) peristaltic pumps; (4) thermocouples; (5) flux meter; (6) flat sheet module; (7) feed tank; (8) heating system.

4. Results and discussion

4.1 Solubility parameter

In this work and as a first step, 58 common solvents were selected, considering their solubility parameters. Solvents with RED < 1 were chosen, as shown in Figure 3a. In the second step, solvents with RED < 0.5 were selected, as shown in Figure 3b.

All solvents with RED < 0.5 are candidates for dissolving PVDF. In this study, DMAC was selected as the ideal solvent with the lowest RED value.

As shown in Figure 3, two other solvents can dissolve the polymer and be investigated for membrane fabrication. Due to the time frame, this part of the research postponed to the future projects. 1 2 3 3 4 4 4 4

(14)

6 (a)

(15)

7 (b)

Figure 3. Comparison of the calculated RED values for the solvents with the selected polymer. (a) for RED < 1 and (b) for RED < 0.5.

4.2 PVDF membrane characterisation

4.2.1 Morphologies of the prepared flat-sheet membranes

a.

b.

c.

Figure 4. SEM images of PVDF membranes prepared at different dissolution temperatures: a) 25C, b) 45C, and c) 65C. 10 𝜇m 10 𝜇m 10 𝜇m 5 𝜇m 5 𝜇m 5 𝜇m 2 𝜇m 2 𝜇m 2 𝜇m

(16)

8 The cross-sectional structure analysis was performed using SEM. Figure 4 shows the evolution of the cross-sectional structure with the dope solution temperature. The cross-section of the flat membrane prepared at the lowest temperature (25°C, Figure 4a) had a sponge-like structure. As shown in Figure 4b, as the temperature of the dope solution increased from 25ºC to 45ºC, the sponge-like structure of the cross-section started to disappear, and a finger-like structure started to appear in the internal layer.

This finger-like structure dominates in Figure 4c, corresponding to the flat-sheet membrane prepared at 65ºC. This indecate that the increase in the coagulation rate of the PVDF flat-sheet membrane was due to the increase the dope solution temperature (i.e. 25ºC, 45ºC, and 65ºC). Conversely, the sponge-like structure was more favoured for its slow coagulation rate [29-31].

4.2.2 Porosity (ε), thickness (δ), and liquid entry pressure (LEP) of water

The properties of the prepared flat-sheet membranes are listed in Table 1. A change in membrane properties was related to a structural change from a sponge-like structure to a finger-like structure as the temperature of the dope solution increased. Increasing the temperature of the dope solution increased the velocity of coagulation, facilitating the formation of a finger-like structure along the cross-section of the prepared membranes. Consequently, this formation led to an increase in the proportion of vacuum in the membrane matrix. Conversely, a negative trend was observed for the LEP values. This trend was expected based on the porosity values and the SEM images. The thickness of the membranes increased at a percentage of 6.5%, attributed to the rate of coagulation being modified by varying the temperature of the dope solution (i.e. 25ºC, 45ºC, and 65ºC).

By considering the membrane morphology, the void volume fractions, thicknesses, and LEP values, it was expected that the permeate flux for filtration and membrane distillation would increase with the temperature of the dope solution.

Table 1. Properties of PVDF flat-sheet membranes with different dope solution temperatures.

Dope solution temperature

(ºC)

Porosity

(%)

Thickness

(µm)

LEP

(bar)

25 25.8 ± 0.4 124.4 ± 1.2 1.78 ± 0.12

45 49.4 ± 0.5 128.2 ± 0.9 1.46 ± 0.32

65 55.4 ± 0.5 132.5 ± 2.3 1.15 ± 0.16

4.2.3 XRD and FTIR patterns

As stated earlier, PVDF has five different polymorphisms. Among the different polymorphisms, the β phase has piezoelectric properties. The XRD results shown in Figure 5a indicated a significant change in the polymorphisms of the PVDF membrane caused by changing the dope solution temperature. Increasing the dope solution temperature increased the amount of the α phase. This increase could be attributed to the reduced entanglement of the polymer chain at a high dope solution temperature. The FTIR results shown in Figure 5b indicated a significant change in the polymorphisms of the PVDF membrane caused by changing the dope solution temperature. The peaks at 795 and 975 cm-1 _{indicated the formation}

(17)

9

(a)

(b)

Figure 5. a) XRD and b) FTIR results for PVDF membranes prepared at different dope solution temperatures (i.e. 25C, 45C, and 65C).

4.2.4 Water filtration permeability

Figure 6 indicate the permeate flux of all membranes prepared at different dope solution temperatures (i.e. 25C, 45C, and 65C) during the filtration test. By increasing the dope solution temperature, the overall porosity and water permeability were increased slightly. This increase was consistent with the SEM results. It was observed that by increasing the temperature of the dope solution, the water flow was improved, consistent with the increase in the porosity values (Table 1).

1400 1200 1000 800 600 400 25C 45C  (795 cm-1₎  (840 cm-1₎ Tran smitt an ce (%) Wavenumber (cm-1₎  (1270 cm-1₎  (975 cm-1₎ 65C

(18)

10 Figure 6. Water permeation flux of membranes during filtration at 1 bar.

4.2.5 Membrane distillation (MD) performance

Figure 7a shows the relationship between the feed temperature and DCMD permeate flow. The permeate flux through the PVDF membranes increased significantly with the feed temperature. An increase in Tf from 20ºC to 80ºC resulted in an exponential increase in the permeate flux in

the two cases: distilled water and the 30 g/L salt aqueous solution of NaCl as feed solution (Figure 6 (a and 6)). This increase was attributed to the increase in porosity despite the increase in the thickness of the membranes (Table 1). Moreover, when the temperature difference between the feed and the permeate solution increased, the driving force responsible for mass transport increased, suporting the increase in the permeate flux.

Figure 7b shows the effect of Tf on the salt rejection factor (SF). Generally, the obtained SF

values exceeded 99.41%. It was observed that the SF slightly decreased when the temperature of the dope solution was increased, which was attributed to the decrease in the LEP values caused by increasing the temperature of the dope solution, as shown in Table 1.

0

5

10

15

20

25

30 25°C

45°C

65°C

J

p

,L

/m

2

.h

Dope solution (C)

(19)

11 Figure 7. Effect of permeate temperature on flux for (a) DI water, (b) 30 g/L aqueous solution of NaCl, and (c) salt (NaCl) rejection efficiencies of flat-sheet membrane prepared with different dope solution temperatures.

. Conclusions and outlook

This project has shown that using the Hansen solubility parameters was an effective method for identifying and inspecting suitable solvents for a specific polymer (PVDF). The theoretical calculation indicated that DMAc was the most appropriate solvent for the fabrication of the membrane.

The temperature of the polymer solution was shown to significantly affect the final morphology of the fabricated membranes. Furthermore, based on the XRD and FTIR data, the temperature of the dope solution significantly affected the polymorphisms.

The prepared membranes proved their utility in water filtration and membrane distillation, and they registered modest fluxes compared to commercial membranes. The variation in the filtration results and MD as a function of the dope solution temperature was consistent with the structural variation and characterisation parameters.

For future work, it is suggested that the polymer-solvent relationship is studied in more depth to manufacture membranes with improved performances that are competitive with commercial ones. However, the molecular dynamics simulations are preferred to investigating the effect of the dope solution temperature on the resulting fabricated membranes.

99,0

99,1

99,2

99,3

99,4

99,5

99,6

99,7

99,8

99,9

100,0

40

50

60

70

80 SF

(%

)

T

_f

(ºC)

25 ºC

45 ºC

65 ºC

(c)

(20)

12

Acknowledgement

I want to introduce and express my gratitude to my respected supervisor, Dr Naser Tavajohi Hassan Kiadeh, Assistant Professor, chemistry department, for his guidance, encouragement, useful comments, and for allowing me to work on my thesis. I also thank my honourable friends and research group, Dr Mohammed Essalhi, Dr Mohamad Yahia, and Norafiqah for their suggestions and comments, which helped me to develop basic concepts for this research. I thank all my friends and students who helped me with this thesis and gave me so many good memories. Furthermore, I thank my parents (especially my mother), wife, brothers, sisters, and friends for their encouragement and support in difficult situations.

References

[1] G.-d. Kang, Y.-m. Cao, Application and modification of poly (vinylidene fluoride)(PVDF) membranes–a review, Journal of Membrane Science, 463 (2014) 145-165.

[2] M.-m. Tao, F. Liu, B.-r. Ma, L.-x. Xue, Effect of solvent power on PVDF membrane polymorphism during phase inversion, Desalination, 316 (2013) 137-145.

[3] Z. Cui, N.T. Hassankiadeh, Y. Zhuang, E. Drioli, Y.M. Lee, Crystalline polymorphism in poly (vinylidenefluoride) membranes, Progress in Polymer Science, 51 (2015) 94-126.

[4] B.S. Lalia, V. Kochkodan, R. Hashaikeh, N. Hilal, A review on membrane fabrication: Structure, properties and performance relationship, Desalination, 326 (2013) 77-95.

[5] R.W. Baker, Membrane technology and applications, John Wiley & Sons, 2012.

[6] B. Krause, M. Storr, T. Ertl, R. Buck, H. Hildwein, R. Deppisch, H. Göhl, Polymeric membranes for medical applications, Chemie Ingenieur Technik, 75 (2003) 1725-1732. [7] T. Ahmad, C. Guria, A. Mandal, A review of oily wastewater treatment using ultrafiltration membrane: A parametric study to enhance the membrane performance, Journal of Water Process Engineering, 36 (2020) 101289.

[8] E. Yuliwati, A. Ismail, T. Matsuura, M. Kassim, M. Abdullah, Effect of modified PVDF hollow fiber submerged ultrafiltration membrane for refinery wastewater treatment, Desalination, 283 (2011) 214-220.

[9] L.K. Wang, J.P. Chen, Y.-T. Hung, N.K. Shammas, Membrane and desalination technologies, Springer, 2008.

[10] L. García-Fernández, M. García-Payo, M. Khayet, Effects of mixed solvents on the structural morphology and membrane distillation performance of PVDF-HFP hollow fiber membranes, Journal of membrane science, 468 (2014) 324-338.

[11] D. Bingzhi, Membrane Technology for Water Purification [J], WATER & WASTEWATER ENGINEERING, 1 (1999).

[12] S.M. Aqeel, Z. Huang, J. Walton, C. Baker, Z. Liu, Z. Wang, Polyvinylidene fluoride (PVDF)/polyacrylonitrile (PAN)/carbon nanotube nanocomposites for energy storage and conversion, Advanced composites and hybrid materials, 1 (2018) 185-192.

[13] C. Costa, J.G. Ribelles, S. Lanceros-Méndez, G. Appetecchi, B. Scrosati, Poly (vinylidene fluoride)-based, co-polymer separator electrolyte membranes for lithium-ion battery systems, Journal of power sources, 245 (2014) 779-786.

[14] Y. Lai, L. Wan, B. Wang, PVDF/Graphene Composite Nanoporous Membranes for Vanadium Flow Batteries, Membranes, 9 (2019) 89.

[15] C.Z. Liang, T.-S. Chung, J.-Y. Lai, A review of polymeric composite membranes for gas separation and energy production, Progress in Polymer Science, 97 (2019) 101141.

[16] W.J. Koros, G. Fleming, Membrane-based gas separation, Journal of membrane science, 83 (1993) 1-80.

[17] P. Pandey, R. Chauhan, Membranes for gas separation, Progress in Polymer Science, 26 (2001) 853-893.

(21)

13 [18] L.M. Robeson, Polymer membranes for gas separation, Current Opinion in Solid State and Materials Science, 4 (1999) 549-552.

[19] X. Tan, D. Rodrigue, A review on porous polymeric membrane preparation. Part I: production techniques with polysulfone and poly (vinylidene fluoride), Polymers, 11 (2019) 1160.

[20] N. Ismail, A. Venault, J.-P. Mikkola, D. Bouyer, E. Drioli, N.T.H. Kiadeh, Investigating the potential of membranes formed by the vapor induced phase separation process, Journal of Membrane Science, 597 (2020) 117601.

[21] N. Nawi, M. Bilad, N. Nordin, Effect of dope solution temperature on the membrane structure and membrane distillation performance, in: IOP Conference Series: Earth and Environmental Science, Institute of Physics Publishing, 2018.

[22] F.A. AlMarzooqi, M. Bilad, H.A. Arafat, Development of PVDF membranes for membrane distillation via vapour induced crystallisation, European Polymer Journal, 77 (2016) 164-173.

[23] F. Abdulla AlMarzooqi, M. Roil Bilad, H. Ali Arafat, Improving liquid entry pressure of polyvinylidene fluoride (PVDF) membranes by exploiting the role of fabrication parameters in vapor-induced phase separation VIPS and non-solvent-induced phase separation (NIPS) processes, Applied Sciences, 7 (2017) 181.

[24] M. Buonomenna, P. Macchi, M. Davoli, E. Drioli, Poly (vinylidene fluoride) membranes by phase inversion: the role the casting and coagulation conditions play in their morphology, crystalline structure and properties, European Polymer Journal, 43 (2007) 1557-1572.

[25] C.M. Hansen, Hansen solubility parameters: a user's handbook, CRC press, 2007.

[26] C.M. Hansen, The three dimensional solubility parameter, Danish Technical: Copenhagen, 14 (1967).

[27] Z. Zhang, G. Kang, H. Yu, Y. Jin, Y. Cao, Fabrication of a highly permeable composite nanofiltration membrane via interfacial polymerization by adding a novel acyl chloride monomer with an anhydride group, Journal of Membrane Science, 570-571 (2019) 403-409. [28] M. Khayet, T. Matsuura, Preparation and Characterization of Polyvinylidene Fluoride Membranes for Membrane Distillation, Industrial & Engineering Chemistry Research - IND ENG CHEM RES, 40 (2001).

[29] K. Khulbe, C. Feng, F. Hamad, T. Matsuura, M. Khayet, Structural and performance study of micro porous polyetherimide hollow fiber membranes prepared at different air-gap, Journal of membrane science, 245 (2004) 191-198.

[30] H. Strathmann, K. Kock, P. Amar, R. Baker, The formation mechanism of asymmetric membranes, Desalination, 16 (1975) 179-203.

[31] H. Strathmann, K. Kock, The formation mechanism of phase inversion membranes, Desalination, 21 (1977) 241-255.

Appendix 1

Table 1. Relative energy differences of PVDF and different types of solvent

Structure formula

Solvents d p h RED

C2H4O Acetaldehyde 14,7 8 11,3 0,7

CH₃COOH Acetic acid 14,5 8 13,5 0,81

C2H3N Acetonitrile 15,3 18 6,1 0,82

(22)

14 C8H8O Acetophenone 19,6 8,6 3,7 0,93

CH3COCl Acetyl chloride 15,8 10,6 3,9 0,7

C3H3N Acrylonitrile 16,4 17,4 6,8 0,6 C6H6 Benzene 18,4 0,2 0 1,68 C₄H₁₀O 1-Butanol 16 5,7 15,8 0,95 C6H12O2 Butyl acetate 15,8 3,7 6,3 1,05 C4H6O2 γ-Butyrolactone 19 16,6 7,4 0,64 CCl4 Carbon tetrachloride 17,8 0 0,6 1,65 C6H5Cl Chlorobenzene 19 4,3 2 1,27 CHCl₃ Chloroform 17,8 3,1 5,7 1,1 C6H12O Cyclohexanol 17,4 4,1 13,5 0,9 C6H12O2 Diacetone alcohol 15,8 8,2 10,8 0,54 C6H4Cl2 o-Dichlorobenzene 19,2 6,3 3,3 1,1 C4H10O3 Diethylene glycol 16,6 12 20,7 1,1 (C2H5)2O Diethyl ether 14,5 2,9 5,1 1,3

C4H9NO Dimethyl acetamide DMA 16,8 11,5 10,2 0,13

C3H7NO Dimethyl formamide DMF 17,4 13,7 11,3 0,17 C2H6OS Dimethyl sulphoxide DMSO 18,4 16,4 10,2 0,48 C4H8O2 1,4-Dioxane 19 1,8 7,4 1,2 C2H5OH Ethanol 15,8 8,8 19,4 1,1 C2H7NO Ethanolamine 17 15,5 21,2 1,2 C4H8O2 Ethyl acetate 15,8 5,3 7,2 0,9 C2H4Cl2 Ethylene dichloride 19 7,4 4,1 0,9 C2H6O2 Ethylene glycol 17 11 26 1,7

C6H14O2 Ethylene glycol monobutyl

ether

16 2,1 12,3 1,1 C4H10O2 Ethylene glycol monoethyl

ether 16,2 9,2 14,3 0,59 C3H8O2 Ethylene glycol monomethyl ether 16,2 9,2 16,4 0,76 C5H10O3 Ethyl lactate 16 7,6 12,5 0,62 CH3NO Formamide 17,2 26,2 19 1,8 CH2O2 Formic acid 14,3 11,9 16,6 0,9 C6H14 Hexane 14,9 0 0 1,75 C6H18N3OP Hexamethyl phosphoramide HMPA 18,4 8,6 11,3 0,5 C9H14O Isophorone 16,6 8,2 7,4 0,55 C6H10O Mesityl oxide 16,4 6,1 6,1 0,8 CH3OH Methanol 15,1 12,3 22,3 1,3 CH2Cl2 Methylene dichloride 18,2 6,3 6,1 0,8

C4H8O Methyl ethyl ketone 16 9 5,1 0,7

C6H12O Methyl isobutyl ketone 15,3 6,1 4,1 1

C5H9NO Methyl-2-pyrrolidone NMP 18 12,3 7,2 0,35

C6H5NO2 Nitrobenzene 20 8,6 4,1 0,9

C2H5NO2 Nitroethane 16 15,5 4,5 0,7

CH3NO2 Nitromethane 15,8 18,8 5,1 0,9

(23)

15 C₃H₈O 1-Propanol 16 6,8 17,4 0,99 C4H6O3 Propylene carbonate 20 18 4,1 1 C3H8O2 Propylene glycol 16,8 9,4 23,3 1,4 C4H8O Tetrahydrofuran 16,8 5,7 8 0,75 C5H12N2O Tetramethylurea TMU 16,8 8,2 11,1 0,47

C6H15O4P Triethyl phosphate TEP 16,8 11,5 9,2 0,17

(CH3)3PO4 Trimethyl phosphate TMP 16,8 16 10,2 0,4

C7H8 Toluene 18 1,4 2 1,5

C2HCl3 Trichloroethylene 18 3,1 5,3 1,1

C9H10 2-Vinyl toluene 18,6 1 3,8 1,4

C3H2O3 Vinylenecarbonate 17,3 18,1 9,6 0,59

Solvents used in VIPS

1-N-methylpyrrolidone (NMP)

Molecular formula: C

5

H

9

NO

Properties Value

Molecular weight (g/mol) 99.13

Specific gravity 1.028

Viscosity (cP) 1.666

Miscibility Miscible in water

Boiling point (°C) 202.0

Solubility parameter 23.6

Dipole moment (Debye) 4.09

Dielectric constant 32.20 at 25C

Polarity (water 100) 36

Critical pressure (NM/m2) 4.78

Critical temperature (K) 724

Molar volume 96.4

(24)

16

2-N,N-dimethylformamide (DMF)

Molecular formula: C

3

H

7

NO

3-N,N-dimethylacetamide(DMAc)

Molecular formula: C

4

H

9

NO

Properties Value

Dielectric constant 36.71

Polarity (water 100) 40.4

Critical pressure (mN/m2) 4.48

Molar volume (m3/mol) 77.4

Surface tension (dynes/cm) 36.4

Properties Value

Boiling point (°C) 165

Polarity (water 100) 40.1

Critical pressure (mN/m2) 4.08

(25)

17

4-dimethylsulphideoxide (DMSO)

Molecular formula: C

2

H

6

OS

5-Triethyl phosphate (TEP)

Molecular formula: (C

2

H

5

)

3

PO

4

Properties Value

Boiling point (°C) 462.2 K

Polarity 1.00

Critical pressure (mN/m2)

-Critical temperature (K) 729 K

Properties Value

Miscibility 2.74 M

Polarity 0.69

Critical pressure (mN/m2)

-Critical temperature (K) 750 K

(26)

18

6-2-Pyrrolidone

Molecular formula: C

4

H

7

NO

Solubility parameter calculation for N,N-dimethylacetamide(DMAc)

N,N-Dimethylacetamide

Group _Quantity _Fd,i Fp,i2 Eh,i Vgi

-N< 1 10 391 1194 0 -CO- 1 142 376 478 0 -CH3- 3 205 0 0 0 SUM 767 294257 1672 92.98  8.25 5.834 4.241 (MPa) 16.87 11.93 8.667 Method (MPa) 16.8 11.5 10.2 HSB

Dispersion

Polar

Hydrogen-bonding interactions

Properties Value

Viscosity (cP) 13.3

Polarity (water 100)

-Critical pressure (mN/m2)

-Critical temperature (K) 802.0



_d

= F

_d,i

/ V

_g,i



_P

_{=√FP, i}

2 / V

_g,i



_h

= √

Eh

,i

(27)

19

Table 2. Components of the solubility parameters according to the group

contribution method of Hoftyzer-Van Krevelen.

Structural group Fd (J1/2 cm3/2 /mol) Fp (J1/2 cm3/2 /mol) Eh (J/mol)

-CH3 420 0 0 -CH2- 270 0 0 >CH- 80 0 0 >C< -70 0 0 =CH2 400 0 0 =CH- 200 0 0 =C< 70 0 0 1620 0 0 1430 110 0 -(o,m,p) 1270 110 0 -F 220 0 0 -Cl 450 550 400 -Br 550 0 0 -CN 430 1100 2500 -OH 210 500 20000 -O- 100 400 3000 -COH 470 800 4500 -CO- 290 770 2000 -COOH 530 420 1000 -COO- 390 490 7000 -NH2 280 0 8400 -NH- 160 210 3100 -N< 20 800 5000 -NO2 500 1070 1500 -S- 440 0 0 =PO4 740 1890 13000 Ring 190 0 0

Effect of polymer dope solution temperature on the fabrication of flat-sheet polyvinylidene fluoride (PVDF) membranes: Water filtration and membrane distillation (MD) applications