Printer Friendly

Development of hierarchical surface roughness on porous poly (vinylidene fluoride) membrane for membrane distillation process.

1 | INTRODUCTION

Membrane distillation (MD) is an advanced separation method, which allows the transport of vapor molecules through a porous hydrophobic membrane by the vapor pressure gradient. [1] In the simplest form, direct contact membrane distillation (DCMD), a transmembrane vapor pressure gradient can be introduced through a temperature difference between the feed and a colder aqueous solution and vapor would be condensed into the liquid, consequently. [2] The potential advantages of MD technology are the theoretically 100% rejection of nonvolatile species and moderate operating conditions, which enables MD to be coupled with low-grade and renewable energy sources to provide the energy. [3,4] So far, MD technology has been employed for the separation of nonvolatile and volatile species from aqueous solutions in water desalination [2] brine concentration, [5] water recovery through wastewater treatment, [6,7] heavy metal removal, [8] food processing applications, [9] and removal of trace organic compounds such as benzene. [10]

Although the porous membrane is merely used as a physical support, its characteristics for MD separation are very crucial. [11] However, one of the obstacles to MD development and commercialization, despite of the potential advantages is the lack of suitable membranes. [12] The intrinsically hydrophobic membranes, including polytetrafluoroethylene (PTFE) and polyvinylidene fluoride (PVDF), were originally fabricated for MF applications, which are inadequate for MD applications. [13] One of the significant challenges in MD operation is the membrane pore wetting, which happens when the feed solution intrudes into the pores if the hydrostatic pressure exerted on the membrane exceeds the pore entry pressure, known as the liquid entry pressure (LEP). [14] Besides, fouling and scaling on the membrane can induce pore wetting. [15]

Wetting would lead to the reduction of permeate flux as well as the deterioration of the separation efficiency by salt passage into the product through the created liquid bridge and diffusive transport. [16] Regarding the Laplace (Cantor) equation, [17] the risk of pore wetting can be decreased using a more hydrophobic membrane with a smaller maximum pore size. However, increasing the membrane hydrophobicity is more effective to improve the membrane pore wetting resistance. [18] The higher LEP value allows to apply higher feed pressure and shear rates, which suppress the polarization effect and lead to higher permeation flux. [19] Alternatively, it provides the opportunity to increase the membrane pore size without the occurrence of pore wetting, and thus, a higher flux can be achieved. [20] It could also restrict the propensity of membrane fouling and biofouling. [21,22] Another advantage of higher hydrophobicity is the minimization of heat loss through conduction due to the presence of air pockets between the liquid and the membrane surface. [19]

So far, different strategies have been developed to improve the membrane surface hydrophobicity, based on increasing the surface roughness and decreasing the membrane surface energy. [23,24] The formation of hierarchical structure with multi-level surface roughness on the solid surface, [19] enhances the surface roughness and water contact angle dramatically. The hierarchical structure reduces the surface wettability due to the presence of trapped air inside the surface grooves, which limits the contact area between the solid surface and the liquid. [25] Augmented surface hydrophobicity can be achieved using various techniques, including sol-gel, [26] etching, [17] crystallization of polymers, [27] spraying, [28] electrohydrodynamics, [29] electrodeposition, [30] layer by layer assembly, [31] and phase separation processes. [32,34]

Phase separation is found to be a cost-effective and straightforward approach to increase the surafee roughness and thus, hydrophobicity, without the need for low surface energy modifiers. [32,33] In this method, highly hydrophobic polymeric coating is created on substrates through a physical treatment which involved no chemical reaction. [35] Erbil et al [32] formed a superhydrophobic coating through phase separation of polypropylene, which shows a gel-like porous structure with increased surface roughness and a water contact angle of 160[degrees]. Zhao et al [36] fabricated Superhyrophobic coating with hierarchical roughness in one step by vapor induced phase separation of bisphenol A polycarbonate (PC) solution under humid condition. Zhang et al [33] proposed a convenient method for phase separation of the PC/dichloromethane solution with the suitable selection of solvent and nonsolvent. Their prepared coating exhibited micro and nanometer roughness and showed superhydrophobicity with a water contact angle of up to 160o.

In the present paper, for the first time, the phase separation of PC solution was applied to the PVDF membrane surface to increase the surface hydrophobicity. The membranes were examined thoroughly using various characterization techniques, and the performance of pristine and coated membranes was compared in DCMD experiments.

2 | EXPERIMENTAL

2.1 | Materials

Polyvinylidene fluoride (PVDF) powder ([M.sub.w] = 534 000 by GPC), dimethylacetamide (DMAc) with 99.9% purity, bisphenol A polycarbonate (PC) granules (melt index 10-12 g/10 minutes), chloroform (HPLC grade), and 1,4-dimethylbenzene (p-xylene) were provided by Sigma-Aldrich.

2.2 | Preparation of the PVDF membrane

Firstly, to obtain a homogenous solution, the PVDF powder was dried in a 100[degrees]C oven for 24 hours and then dissolved in DMAc at 50[degrees]C using a magnetic stirrer. Left stagnant for 24 hours at room temperature, degassing of the prepared solution was done, and then it was cast slowly onto a clean glass surface via a film applicator, which provided an even surface with 200 [micro]m thickness. After 10 seconds solvent evaporation, the cast film was subsequently immersed into 21[degrees]C reverse osmosis (RO) process water bath. The flat sheet membrane, which was immediately coagulated and peeled off the glass surface, was left in the coagulation bath for 24 hours for complete removal of the solvent. After washing the solidified membrane, it was air-dried at room temperature overnight. PVDF of 12 wt% concentration was found as the optimum concentration for the fabrication of a microfiltration membrane. It was found that beyond the 12 wt% polymer concentration, the membrane structure becomes dense and the permeation flux was reduced.

2.3 | Modification of the PVDF membrane

The PC granules were dissolved in chloroform at the different concentrations of 25, 50, and 100 mg/mL and were stirred continuously at room temperature, to obtain a homogenous and clear solution. The PVDF membranes were firstly dipped into the PC solution for 5 seconds to coat the membrane surface, and then were immersed into the p-xylene bath, as the nonsolvent, and stayed there for 1 minute to ensure a complete exchange of solvent and nonsolvent. The prepared membranes were left at room temperature for about 20 minutes for evaporation of the solvent residue, till the white coating appeared on the membrane surface.

2.4 | Membrane characterization

To investigate the effect of surface modification on the resultant membrane morphology, scanning electron microscopy (SEM) was done using the FEI Quanta 200 scanning electron microscope (Thermo Fisher Scientific). The membrane sample was broken in liquid nitrogen to assess its cross-section. The sample was then goldcoated in the Baltec SCD 050 sputter coater. The higher magnification images of the coated membrane surface were taken using a variable-pressure scanning electron microscope (Zeiss Sigma VP, Carl Zeiss, Germany). The membrane sample was chromium-coated in the Emitech K575X high-resolution sputter coater (EM technologies Ltd., Kent, England), before imaging.

The contact angle of water with the surface of membrane samples was determined via the sessile drop method using the KSV Cam 200 contact angle goniometer (KSV Instruments, Finland). The analysis was performed five times for each membrane sample, and the average of obtained angles was reported.

The pristine and coated membranes were investigated in terms of their average pore size via the wet/dry flow method using the capillary flow Porolux porometer (IB-FT GmbH, Germany) and the Porolux 100 software. The membrane sample with an area of 2.68 [cm.sup.2] was placed in the test cell after being wetted by a liquid with low surface tension (16 dyne/cm). Compressed air was applied through the membrane, and the measurements were performed.

The overall membrane porosity was acquired through the gravimetric method. The membrane sample was firstly weighed ([W.sub.d]) and then immersed in ethanol for 12 hours. The free ethanol on the wet membrane surface was removed using a tissue paper, and the membrane was weighed again ([W.sub.w]). To calculate the membrane porosity ([epsilon]), Equation (1) was applied [37]:

[epsilon] = ([w.sub.w] - [w.sub.d])/[[rho].sub.i]/([w.sub.w] - [w.sub.d])/[[rho].sub.i] + [w.sub.d]/[[rho].sub.p] * 100% (1)

where [[rho].sub.i] and [[rho].sub.p] are the densities of ethanol (0.7893 g [cm.sup.-3]) and polymer (1.78 g [cm.sup.-3]), respectivelyJ3xl At least three membrane samples were examined under the same condition, and the average of the obtained values was reported as the membrane porosity.

The pristine and coated membranes was analyzed in terms of their surface chemical composition via the attenuated total reflectance Fourier-transform infrared (ATRFTIR) spectroscopy using the Nicolet 5700 spectrometer (Thermo Fisher Scientific). 64 scans were conducted in the wave number range of 680 to 3180 [cm.sup.-1] with 4 [cm.sup.-1] resolution.

The topography and surface roughness of the membrane samples were analyzed using an atomic force microscope (Nanosurf Easyscan, Switzerland), on a 2.5 [micro]m * 2.5 [micro]m scan area in tapping mode. In addition, to observe the surface roughness profile of the pristine and coated membranes, the 3D images were generated using Phenom ProSuite surface roughness software and Phenom scanning electron microscope.

As shown in Figure 1A, the [LEP.sub.w] (water liquid entry pressure) values of the pristine and coated membranes were obtained using a dead-end filtration module in a home-made setup. The dry membrane sample was placed in the stainless steel cell between the feed side (upper chamber) and the permeate side (lower chamber). Pressure was applied to the feed side at a constant increase rate of 0.1 bar per 10 minutes from a compressed air cylinder. The pressure was increased continuously until the first water droplet got through the membrane and started dropping from the permeate side. This pressure was recorded as LEP, which was performed three times for each membrane and the average of the obtained values was reported.

2.5 | DCMD experiments

The MD performance of the pristine and coated membranes was investigated by a DCMD apparatus equipped with a module with an effective membrane area of 11.3 [cm.sup.2]. The utilized set-up is schematically shown in Figure 1B.

The membrane was placed in the membrane cell with a mesh spacer positioned on both sides of the membrane to act as a support and turbulence promoter. The synthetic feed solution used in the experiments was 3.5 wt% sodium chloride solution, resembling the seawater salinity, and the permeate was RO water. The feed and permeate solutions were circulated counter-currently at an equal flow rate of 0.4 L/min.

During the experiment, the feed and permeate temperatures were kept constant at 65[degrees]C using a water bath heater and 25[degrees]C using a chiller, respectively. In the permeate tank, the conductivity was observed applying a conductivity meter to detect the incidence of pore wetting or any possible leakage. The overflow from the permeate tank is the membrane permeation, which was collected continuously in a beaker located on a digital balance to calculate the permeate flux, J (kg/[m.sup.2]h), using the equation below. [20]:

where [DELTA]W (kg) is the mass of the collected permeate, A ([m.sup.2]) is the membrane filtration area, and [DELTA]t (h) is the time of permeate collection. The salt rejection (R) was calculated according to the following relation [20]:

R = (1 - [C.sub.p]/[C.sub.f]) * 100% (3)

where [C.sub.f] and [C.sub.p] are the salt concentration (g/L) in the feed and permeate, respectively.

3 | RESULTS AND DISCUSSION

3.1 | ATR-FTIR

The ATR-FTIR spectra of the pristine and coated membranes are shown in Figure 2. The characteristic absorption peaks were identified in both spectra. For the pristine PVDF membrane, the peaks at 879, 1178, and 1403 [cm.sup.-1] correspond to the C-C, C[F.sub.2], and C[H.sub.2] vibrations, respectively. [39] The same peaks are also seen in the coated membrane spectrum, however, with less absorbance intensity. Figure 2 also indicates that new peaks arose on the coated membrane surface at 1506 and 1767 [cm.sup.-1], which are attributed to CC and CO stretching, respectively, due to the presence of PC. [40] Furthermore, the peaks at 1193 and 1014 [cm.sup.-1] are assigned to O-CC asymmetric stretching and C-O bond, respectively. [40]

3.2 | Membrane morphology and hydrophobicity

Figure 3 presents the morphology of the pristine PVDF membrane surface and cross-section. With respect to Figure 3A, the membrane revealed asymmetric structure with macrovoids beneath the thin top layer, which is well known for membranes prepared through nonsolvent induced phase inversion process. [41] Figure 3B,C shows the surface of fabricated membrane, in which micro pores are obvious especially in Figure 3C. Table 1 presents the average pore size of the membranes, determined by capillary flow porometery. According to the results, the pristine membrane has an average pore size of 0.196 [micro]m, which is between 0.01 and 1 [micro]m, [2] and thus, it is appropriate to be used in DCMD.

Figure 4 indicates the SEM images of the membranes coated with different concentrations of the PC solutions. According to this figure, by increasing the PC solution concentration, the membrane surface was gradually covered with polycarbonate deposits which shifted the relatively smooth surface of the pristine membrane toward rough membrane surface. By treating the dip-coated membrane with nonsolvent, the nonsolvent cracks into the PC skin layers to conduct the solvent-nonsolvent exchange leading to precipitation of the polymer.133' The precipitation of polymer is initiated and leads to the formation of polymer nuclei on the PVDF membrane surface, and then the polymer-rich phase aggregated around the polymer nuclei and grow in all direction which results in the formation of various morphological structures. [36]

As revealed in Figure 4A1, only slight PC deposits formed on the membrane surface coated with 25 mg/mL PC solution, and changes in the surface topography cannot be observed. This is due to the insufficient polymer content in the PC solution, which initiated the formation of polymer nuclei, however, without the existence of further polymer to grow. By increasing the concentration of PC solution to 50 mg/mL, the polymer precipitated out of the solution and subsequently acted as the nuclei, around which the polymer-rich phase aggregated in order to decrease the surface tension. [36] It can be seen that PC aggregates were not completely distributed on the membrane surface, which was related to the insufficient polymer content and separate polymer nucleation. Further increase in the concentration of the PC solution to 100 mg/mL brought about the formation of big aggregates around the precipitated polymer nuclei with relatively uniform and continuous structures of PC deposits, which cover the membrane surface.

Figure 4(#2) presented the cross-sectional morphologies of the coated membranes with different PC concentrations. Regarding Figure 4A2, the membrane crosssection consists of a dense top layer and a sublayer containing finger-like voids. By increasing the PC concentration, the membrane cross-sectional structure has not been altered considerably, and PC deposits were distributed mainly on the membrane surface.

Figure 5 indicates the magnified SEM images of the PC deposits obtained by membrane coating using the 100 mg/mL PC solution. Regarding Figure 5A,B, the coated membrane surface was covered with connected PC aggregates consisting of micro papillae formed on the surface. According to higher magnification SEM images provided in Figure 5C,D, the presence of nanoscale surface roughness on the micro papillae created a hierarchical roughness on the surface of the membrane.

3.3 | Membrane surface topography

AFM analysis was conducted to investigate the surface topography of the pristine and coated membranes, and the obtained three-dimensional images are presented in Figure 6. Due to the high surface roughness of the membrane coated with 100 mg/mL PC solution, the AFM facility was not capable of providing the data of sharp height variations of the surface. The mean roughness ([R.sub.a]) and root mean square (RMS) roughness ([R.sub.q], the SD of surface heights [42]) are presented in Figure 6. Figure 7 also shows the three-dimensional images of the pristine and coated membranes surface roughness topography obtained using the surface roughness software and scanning electron microscope. This figure shows that the coated membranes presented rougher surfaces compared to the pristine membrane due to the presence of PC deposits and aggregates on the surfaces, as mentioned before. According to these results, by increasing the PC concentration, the membrane surface roughness increases, which demonstrates that polymer content is a crucial factor in the formation and growth of the aggregates, and thus, final topography of the membrane surface. [33]

3.4 | The water contact angle of membranes

Figure 8A illustrates the determined contact angles of water droplet on the pristine and coated membrane surfaces. The results indicate that the water contact angle augmented from 82[degrees] for the pristine membrane to 130[degrees] for the membrane coated with 100 mg/mL PC solution. High water contact angle of the membranes coated with solutions containing 50 and 100 mg/mL PC concentration can be attributed to the formation of hierarchical structures with micro and nanoscale roughness due to the PC deposits. By increasing the surface roughness, the water repellency characteristic of the surface improved due to the presence of trapped air inside the grooves, which limits the area of contact between the water droplet and the solid surface. [25] According to the SEM images, the higher surface contact angle of the membranes coated with higher PC concentrations is due to the better distribution of the PC deposits and the more complex topography of the aggregates formed on the membranes surface. Thus, the presence of PC deposits results in formation of a rough surface, which leads to higher membrane surface contact angle and hydrophobicity.

3.5 | Membrane LEP

The pore wetting resistance of the pristine and coated membranes is assessed by their LEP value. According to the Laplace (cantor) equation, [17] a membrane with high hydrophobicity and a small maximum pore size possesses a high LEP value, and thus, higher pore wetting resistance in MD operation. In the present study, LEP measurements were performed, and the corresponding LEP values are shown in Figure 8B.

Regarding Figure 8B, the LEP of the membranes coated with 25, 50, and 100 mg/mL PC solution increased to 245, 302, and 305 kPa, respectively, compared to the pristine membrane with the LEP of 235 kPa. This effect can be attributed to higher surface hydrophobicity and also lower average pore size of the coated membranes (presented in Table 1), which contributed to higher LEP values. [19]

The membrane coated with 50 mg/mL PC solution was utilized in the DCMD experiment because its LEP value was higher than that of the membrane coated with 25 mg/mL PC solution, while had a slight LEP difference with the membrane coated with 100 mg/mL PC solution. Besides, the presence of a thick layer of PC deposits on the surface of the membrane coated with 100 mg/mL PC solution (Figure 4C2) would make an extra mass transfer resistance, which leads to lower permeability.

3.6 | DCMD performance of the membranes

The pristine membrane and the membrane coated with 50 mg/mL PC solution were utilized in the DCMD experiment using an aqueous solution of sodium chloride with 3.5 wt% concentration as feed, which resembles the seawater salinity. As shown in Figure 9, the coated membrane showed a lower permeation flux compared to the pristine one. The lower flux can be attributed to the reduction of about 14% and 8% in the average pore size and porosity of the coated membrane, respectively, as reported in Table 1. Besides, the higher thickness of the coated membrane can impose a resistance for mass transfer through the membrane, too.1201 However, regarding Figure 9, the pristine membrane exhibited a sharp increase in the permeate conductivity (from 15.6 to 1348 microS/cm) over time, while the slope of changes in the permeate conductivity was less for the coated membrane (from 15.6 to 546 microS/cm). This result for permeate conductivity led to the salt rejection of 97.5% and 99% for the pristine and coated membranes, respectively. The higher permeate conductivity for the pristine membrane can be attributed to the total or partial pore wetting, which assists the salt transportation through the membrane pores. [19] The coated membrane was more resistant to pore wetting and consequently exhibited lower permeate conductivity due to the higher LEP, which restricted the penetration of the water into the pores. It was reported in literature that the mass of water passing through the wetted pores is in orders of magnitude larger than the mass of water vapor flux through the nonwetted pores. [43] Therefore, the higher permeate flux obtained for the pristine membrane may also be related to the penetration of the liquid feed carrying salt into the membrane pores.

Table 2 illustrates a comparison of the present study results with other investigations on MD membranes. The present coated membrane achieved a salt rejection of 99% and an average flux of 5.4 kg kg [m.sup.-2] [h.sup.-1], which is within the range of membranes reported in the literature. The membrane morphological characteristics, operating parameters, and MD module design are effective parameters on the permeate flux.

Due to the use of more chemicals and usually more complex fabrication methods, which often lead to higher membrane production cost and consequently membrane limited application, the reported fluxes in some of the literature are higher than that in the present study. While in the present study, the hydrophobic membrane was fabricated by utilizing only two commercially available polymers through a simple and straightforward method, which is more capable to be commercialized.

Figure 10A,B and Figure 1[degrees]C,D show the SEM images of the surface of the pristine and coated membranes after 20 hours DCMD operation, respectively. According to this figure, higher surface hydrophobicity, obtained by the presence of PC deposits on the membrane surface, decreased the fouling and scaling of the membrane surface with salt crystals. Regarding the pristine membrane surface in Figure 3B,C, it can be seen in Figure 10A,B that salt crystals are deposited on the pristine membrane surface after DCMD operation. According to Figure 10C, D, increasing the membrane surface hydrophobicity leads to less fouling and scaling propensity, which is known as an effective parameter in membrane wetting. [23] Thus, for a long period of DCMD operation, a more stable operation can be expected for the coated membrane.

Figure 10E,F shows the presence of PC deposits on the coated membrane surface before and after 20 hours DCDM experiment. This figure demonstrates the mechanical and thermal stability of the coated membrane after a long term DCMD operation without any significant changes in the surface topography and hydrophobicity of the membrane. The hydrophobic modification of the membrane surface presented in this work was achieved via a simple and rapid phase separation through manipulating the surface roughness. Therefore, it is a feasible and robust modification method which can be used for any type of membrane to increase the surface hydrophobicity. This result illustrates the capability of using the highly hydrophobic membrane prepared by this method for a long-term MD operation.

4 | CONCLUSION

In this work, the flat sheet PVDF membrane was prepared through phase inversion process. The hydrophobicity of the membrane surface was successfully increased by manipulating the surface roughness, through the phase separation of the PC solution at different concentrations. A series of characterization tests as well as the DCMD performance of the pristine and coated membranes were carried out. The results show that hierarchical roughness was developed on the membrane surface due to the formation of PC deposits. The water contact angle and LEP values of the membranes increased from 82oand 235 kPa for the pristine membrane to 130o and 305 kPa for the coated membrane with the highest PC concentration, respectively. According to the DCMD experiment the coated membrane showed lower permeate conductivity and less fouling propensity, and thus, proper for long-term MD operation in comparison with the pristine membrane. It is believed that this simple and practicable method can be applied to increase membrane hydrophobicity for industrial applications and would be a potential candidate for MD processes.

DOI: 10.1002/pen.25412

ACKNOWLEDGMENT

The authors acknowledge the Riddet Institute, a national Centre of Research Excellence, for providing experimental facilities, and MMIC (Manawatu Microscopy and Imaging Centre) for the imaging support at Massey University.

ORCID

Seyed Mahmoud Mousavi [ID] https://orcid.org/0000-0003 4648-295X

Received: 6 February 2020 | Revised: 14 April 2020 | Accepted: 22 April 2020

REFERENCES

[1] K. Tahvildari, A. Zarabpour, M. Ghadiri, A. Hemmati, Polym. Eng. Sci. 2014, 54(11), 2553.

[2] M. S. El-Bourawi, Z. Ding, R. Ma, M. Khayet, J. Membr. Sci. 2006, 285(1-2), 4.

[3] K. W. Lawson, D. R. Lloyd, J. Membr. Sci. 1997, 124, 1.

[4] W. G. Shim, K. He, S. Gray, I. S. Moon, Sep. Purif. Technol. 2015, 143, 94.

[5] R. Bouchrit, A. Boubakri, A. Hafiane, S. A.-T. Bouguecha, Desalination. 2015, 376, 117.

[6] P. Jacob, P. Phungsai, K. Fukushi, C. Visvanathan, J. Membr. Sci. 2015, 475, 330.

[7] N. M. Mokhtar, W. J. Lau, A. F. Ismail, J. Water Process Eng. 2014, 2, 71.

[8] P. P. Zolotarev, V. V. Ugrozov, I. B. Volkina, V. M. Nikulin, J. Hazard. Mater. 1994, 37(1), 77.

[9] B. Jiao, A. Cassano, E. Drioli, J. Food Eng. 2004, 63(3), 303.

[10] F. A. Banat, J. Simandl, Chem. Eng. Sci. 1996, 51(8), 1257.

[11] D. Sun, Z. S. Zheng, M. Q. Liu, B. B. Li, F. Huang, D. Y. Li, Polym. Eng. Sci. 2017, 57(12), 1311.

[12] E. Drioli, A. Ali, F. Macedonio, Desalination. 2015, 356, 56.

[13] L. Eykens, K. De Sitter, C. Dotremont, L. Pinoy, B. Van der Bruggen, Separ. Purif Technol. 2017, 182, 36.

[14] A. Alkhudhiri, N. Darwish, N. Hilal, Desalination. 2012, 287, 2.

[15] M. Gryta, J. Membr. Sci. 2008, 325(1), 383.

[16] E. Guillen-Burrieza, M. O. Mavukkandy, M. R. Bilad, H. A. Arafat, J. Membr. Sci. 2016, 515, 163.

[17] W.T. Xu, Z.P. Zhao, M. Liu, K.C. Chen, J. Membr. Sci. 2015, 491, 110.

[18] Y. Lv, X. Yu, S.T. Tu, J. Yan, E. Dahlquist, J. Membr. Sci. 2010, 362(1-2), 444.

[19] Razmjou A, Arifin E, Dong G, Mansouri J, Chen V. J. Membr. Sci. 2012, 415-416, 850.

[20] J. Zhang, Z. Song, B. Li, Q. Wang, S. Wang, Desalination. 2013, 324, 1.

[21] H. Zhang, R. Lamb, J. Lewis, Sci. Technol. Adv. Mater. 2005, 6 (3-4), 236.

[22] B. J. Privett, J. Youn, S. A. Hong, J. Lee, J. Han, J. H. Shin, M. H. Schoenfisch, Langmuir. 2011, 27(15), 9597.

[23] M. Rezaei, D. M. Warsinger, M. C. Duke, T. Matsuura, W. M. Samhaber, WaterRes. 2018, 139, 329.

[24] D. Zhao, J. Zuo, K.J. Lu, T.S. Chung, Desalination. 2017, 413, 119.

[25] X. Zhang, N. Zhao, S. Liang, X. Lu, X. Li, Q. Xie, X. Zhang, J. Xu, Adv. Mater. 2008, 20(15), 2938.

[26] F. Liu, L. Wan, W. Zhou, H. Li, Z. Qiu, S. Zhang, J. Shen, Surf. Coat. Technol. 2017, 321, 90.

[27] X. Lu, C. Zhang, Y. Han, MacromoL Rapid Commun. 2004, 25 (18), 1606.

[28] S. Pan, N. Wang, D. Xiong, Y. Deng, Y. Shi, Appl. Surf. Sci. 2016, 389, 547.

[29] L. Jiang, Y. Zhao, J. Zhai, Angew. Chem. Int. Ed 2004, 43(33), 4338.

[30] D. Zhang, L. Li, Y. Wu, B. Zhu, H. Song, Appl. Surf. Sci. 2019, 473, 493.

[31] K. S. Liao, A. Wan, J. D. Batteas, D. E. Bergbreiter, Langmuir. 2008, 24(8), 4245.

[32] H. Y. Erbil, A. L. Demirel, Y. Avci, O. Mert, Science. 2003, 299 (5611), 1377.

[33] Y. Zhang, H. Wang, B. Yan, et al., J. Mater. Chem 2008, 18(37), 4442.

[34] X. Li, G. Chen, Y. Ma, L. Feng, H. Zhao, L. Jiang, F. Wang, Polymer. 2006, 47(2), 506.

[35] L. Tian, Z. Yuanyuan, M. Yingying, H. Ran, Appl. Clay Sci. 2015, 118, 337.

[36] N. Zhao, J. Xu, Q. Xie, L. Weng, X. Guo, X. Zhang, L. Shi, MacromoL Rapid Commun. 2005, 26(13), 1075.

[37] X. Li, Y. Wang, X. Lu, C. Xiao, J. Membr. Sci. 2008, 320(1-2), 477.

[38] W. Zhang, Y. Li, J. Liu, B. Li, S. Wang, J. Membr. Sci. 2017, 535, 258.

[39] Q. Chen, Z. Yu, Y. Pan, G. Zeng, H. Shi, X. Yang, F. Li, S. Yang, Y. He, J. Mater. Sci.--Mater. Electron 2017, 28(4), 3865.

[40] A. Idris, Z. Man, A. Maulud, M. Khan, Membranes. 2017, 7(2), 21.

[41] P. Van de Witte, P. J. Dijkstra, J. Van den Berg, J. Feijen, J. Membr. Sci. 1996, 117(1-2), 1.

[42] M. M. A. Shirazi, D. Bastani, A. Kargari, M. Tabatabaei, Desalin. Water Treat. 2013, 51(31-33), 6003.

[43] R. Thomas, E. Guillen-Burrieza, H. A. Arafat, J. Membr. Set. 2014, 452, 470.

[44] J. A. Prince, G. Singh, D. Rana, T. Matsuura, V. Anbharasi, T. S. Shanmugasundaram, J. Membr. Sci. 2012, 397, 80.

[45] Z. Li, Y. Peng, Y. Dong, H. Fan, P. Chen, L. Qiu, Q. Jiang, Appl. Surf. Sci. 2014, 317, 338.

[46] S. Meng, J. Mansouri, Y. Ye, V. Chen, J. Membr. Sci. 2014, 450, 48.

[47] J. A. Prince, D. Rana, G. Singh, T. Matsuura, T. J. Kai, T. S. Shanmugasundaram, Chem. Eng. J. 2014, 242, 387.

Sima Rabiei (1) | Seyed Mahmoud Mousavi (1) | Anthony H.J. Paterson (2)

(1) Chemical Engineering Department, Faculty of Engineering, Ferdowsi University of Mashhad, Mashhad, Iran

(2) School of Food and Advanced Technology, Massey University, Palmerston North, New Zealand

Correspondence

Seyed Mahmoud Mousavi, Chemical Engineering Department, Faculty of Engineering, Ferdowsi University of Mashhad, Mashhad, Iran. Email: mmousavi@um.[a.sub.c].ir

Caption: FIGURE 1 Schematic illustration of the experimental setups used for, A, the LEPw measurement and B, DCMD

Caption: FIGURE 2 ATR-FTIR spectra of the pristine and coated membranes [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIGURE 3 SEM images of the pristine membrane: A, Cross-section; B, surface (1300*); and C, surface (x6000)

Caption: FIGURE 4 SEM images of the surface and cross section of the membranes coated with, A, 25 mg/mL; B, 50 mg/mL; and C, 100 mg/ ml PC solution

Caption: FIGURE 5 High-resolution SEM surface images of the membrane coated with 100 mg/mL PC concentration: A, PC deposits at x6000; B, size of the PC aggregates; C, PC deposits at *3 00 000; and D, PC deposits at x5 00 000 [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIGURE 6 AFM images of, A, pristine membrane and membranes coated with, B, 25 mg/mL, and C, 50 mg/mL PC solution [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIGURE 7 Surface roughness profiles of, A, the pristine membrane and the membranes coated with, B, 25 mg/mL; C, 50 mg/mL; and D, 100 mg/mL PC solution [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIGURE 9 Permeate flux and conductivity vs time for the pristine membrane and the membrane coated with 50 mg/mL PC solution at the conditions: feed: 3.5 wt% NaCl solution, feed temperature: 65[degrees], permeate temperature: 25[degrees]C, feed and permeate flow rates: 0.4 L/min [Color figure can be viewed at wileyonlinelibrary.com]

Caption: FIGURE 10 SEM images of the membrane surface; the presence of salt crystals on the pristine membrane, A,B; and the coated membrane with 50 mg/mL PC solution, C,D, after 20 hours DCMD test, and the presence of PC deposits on the coated membrane with 50 mg/mL PC solution before, E, and after, f, 20 hours DCMD test [Color figure can be viewed at wileyonlinelibrary.com]
TABLE 1 Effect of PC solution concentration on the
membrane morphology

                                   Mean pore
Membrane          Porosity (%)     size ([micro]m)

Uncoated PVDF     79               0.196
25 mg/mL PC       76               0.193
50 mg/mL PC       73               0.168
100 mg/mL PC      71               0.104

TABLE 2 Comparison of the DCMD performance of the prepared membrane
in the present study with other membranes

                                       Membrane
                                       properties

                                       [epsilon]      [d.sub.p]
Membrane                               (%)            ([micro]m)

Flat sheet PVDF coated with PC         73             0.168
(this work)

Flat sheet PVDF (M0-10) [43]           54.27          0.14

Flat sheet PVDF-clay nanofiber [44]    81             0.64

Flat sheet PVDF/PDMS-Si[O.sub.2]       71.1           0.175
composite [45]

Flat sheet PVDF/PDMS-Si[O.sub.2]       68.52          0.36
(spray coating) [20]

Flat sheet FTCS-Ti[O.sub.2]/           --             --
PEG/PVDF [46]

Flat sheet PVDF/PVP/SMM                --             0.06
(M1+N3) [47]

                                      Membrane       Operating
                                      properties     parameters

                                       CA
Membrane                               ([degrees])   Feed

Flat sheet PVDF coated with PC         113           3.5 wt% NaCl
(this work)

Flat sheet PVDF (M0-10) [43]           128           10 g/L NaCl

Flat sheet PVDF-clay nanofiber [44]    154.2         3.5 wt% NaCl

Flat sheet PVDF/PDMS-Si[O.sub.2]       154           35 g/L NaCl
composite [45]

Flat sheet PVDF/PDMS-Si[O.sub.2]       156           3.5 wt.% NaCl
(spray coating) [20]

Flat sheet FTCS-Ti[O.sub.2]/           159           10 wt% NaCl
PEG/PVDF [46]

Flat sheet PVDF/PVP/SMM                148.4         3.5 wt% NaCl
(M1+N3) [47]

                                       Operating parameters

                                       [T.sub.f]   Flow
Membrane                               CC)         rate

Flat sheet PVDF coated with PC         65          0.4 L/min
(this work)

Flat sheet PVDF (M0-10) [43]           60          0.25 L/s

Flat sheet PVDF-clay nanofiber [44]    62          1.8 L/min

Flat sheet PVDF/PDMS-Si[O.sub.2]       73          54 L/h
composite [45]

Flat sheet PVDF/PDMS-Si[O.sub.2]       70          --
(spray coating) [20]

Flat sheet FTCS-Ti[O.sub.2]/           60          0.2 m/s
PEG/PVDF [46]

Flat sheet PVDF/PVP/SMM                65          1.2 L/m
(M1+N3) [47]

                                       Flux (kg     Salt
                                       [m.sup.2]    rejection
Membrane                               [h.sup.1])   (%)

Flat sheet PVDF coated with PC         5.4          99
(this work)

Flat sheet PVDF (M0-10) [43]           3.5          --

Flat sheet PVDF-clay nanofiber [44]    2.4          >99.9

Flat sheet PVDF/PDMS-Si[O.sub.2]       12.4         >99.9
composite [45]

Flat sheet PVDF/PDMS-Si[O.sub.2]       8            >99.9
(spray coating) [20]

Flat sheet FTCS-Ti[O.sub.2]/           10           --
PEG/PVDF [46]

Flat sheet PVDF/PVP/SMM                12           >99.9
(M1+N3) [47]

FIGURE 8 A, Water contact angle and B, LEP values of the
pristine and coated membranes [Color figure can be viewed at
wileyonlinelibrary.com]

(A)

            Water contacts
             angle (deg.)

Original         82
25 mg/ml         83
50 mg/ml        113
100 mg/ml       130

(B)

             LEP (kPA)

Original        235
25 mg/ml        245
50 mg/ml        302
100 mg/ml       305

Note: Table made from bar graph.
COPYRIGHT 2020 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2020 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Rabiei, Sima; Mousavi, Seyed Mahmoud; Paterson, Anthony H.J.
Publication:Polymer Engineering and Science
Date:Jul 1, 2020
Words:5861
Previous Article:Double yielding behavior of in situ microfibrillar polyolefin elastomer/poly(lactic acid) composites: Effect of microfibrillar morphology.
Next Article:Synthesis of polyamide-hydroxyapatite nanocomposites.
Topics:

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |