- Email: [email protected]
Membrane-based detection of wetting phenomenon in direct contact membrane distillation
Author’s Accepted Manuscript Membrane-based detection of wetting phenomenon in direct contact membrane distillation Farah Ejaz Ahmed, Boor Singh Lalia, Raed Hashaikeh www.elsevier.com/locate/memsci
PII: DOI: Reference:
S0376-7388(17)30569-0 http://dx.doi.org/10.1016/j.memsci.2017.04.035 MEMSCI15199
To appear in: Journal of Membrane Science Received date: 27 February 2017 Revised date: 17 April 2017 Accepted date: 18 April 2017 Cite this article as: Farah Ejaz Ahmed, Boor Singh Lalia and Raed Hashaikeh, Membrane-based detection of wetting phenomenon in direct contact membrane d i s t i l l a t i o n , Journal of Membrane Science, http://dx.doi.org/10.1016/j.memsci.2017.04.035 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Membrane-based detection of wetting phenomenon in direct contact membrane distillation Farah Ejaz Ahmed, Boor Singh Lalia, Raed Hashaikeh*
Chemical Engineering Department, Khalifa University of Science and Technology, Masdar Institute,
Masdar
City,
P.O.
Box
54224,
Abu
*Corresponding author Email: [email protected] Phone: +971-28109152
April 2017
1
Dhabi,
United
Arab
Emirates
Abstract
Advances in membrane distillation are limited by the wetting phenomenon. Wetting is typically determined by examining the permeate water quality. In this work, we apply an electrically conductive membrane to direct contact membrane distillation combined with an electrochemical system. The membrane was fabricated by heat pressing a carbon cloth with electrospun PVDFHFP mat. The membrane acts not only as a barrier to reject salt with 99.6% rejection, but also as an electrode where in the current through the system is used to detect wetting which allows Na+ and Cl- ions to complete the cell. It was found that the membrane has an LEP of 36 psi and mean pore size of 0.2 µm. A continuous voltage of +1V was applied during the direct contact membrane distillation process process, and a sharp increase in current was observed at the point where wetting was induced. Keywords: conductive membrane; direct contact membrane distillation; wetting; PVDF-HFP; detection
2
1. Introduction Membrane distillation (MD) is a thermally driven separation process that has attracted attention for desalination and water treatment. In MD, a temperature difference, and consequently a vapor pressure difference across a porous hydrophobic membrane drive only vapor molecules to be transported from the feed to the permeate [1]. The efficiency of the MD process depends greatly on the ability of the membrane to resist wetting. In fact, progressive wetting of the membrane eventually prevents separation from continuing, and is one of the factors that deter MD processes from competing with other industrial desalination techniques. Increase in permeate conductivity, or decline in salt rejection, is a typical consequence of membrane wetting as well as fouling and scaling [2]. Wetting occurs when the transmembrane pressure exceeds the liquid entry pressure (LEP) of a membrane. LEP depends on both the pore size and the hydrophobicity of the membrane material, as highlighted by the Cantor-Laplace equation. As stated above, pore wetting is undesirable in MD processes. However, even membranes with high liquid entry pressure (LEP) that resist wetting for some time eventually succumb to wetting. Several studies in literature indicate the decline in salt rejection over time due to membrane wetting [3]. Current literature suggests two distinct categories of wetting detection in MD. The first involves examining water quality of the permeate, while the other involves studying variation of the membrane. Among membrane-based techniques, one way to confirm wetting is by examining the membrane surface on the permeate side for salt deposits. This requires the membrane to be removed and real-time detection is not possible. On the other hand, wetting detection through examining quality of the permeate can be slow, especially when multiple membranes are stacked together as in large scale MD systems. It is also challenging to identify which membrane in the 3
system is responsible for wetting. We suggest the use of an electrically conductive membrane that can be incorporated in a combined MD-electrochemical cell to detect wetting electrolytically. An in-situ membrane-based detection method where the membrane acts as an electrode would allow for feedback and control and would also make immediate detection possible, which is beneficial for large scale MD units. Electrospun nanofiber membranes have been the focus of many recent membrane distillation studies, due to versatility of the fabrication technique, high resulting porosity and interconnected pore structure [4]. Several studies have reported on the successful application of electrospun polyvinylidene fluoride-co-hexafluoropropylene (PVDF-HFP) [5] for MD. This is because PVDF-HFP is hydrophobic and can be electrospun using simple solvents at room temperature to form pores that are in the desirable range for MD. In this work, we demonstrate the fabrication of electrospun PVDF-HFP - carbon cloth membranes for MD and the use of these electrically conductive membranes for in-situ electrolytic wetting detection.
4
2. Materials and Methods 2.1. Fabrication Acetone and D-methylacetamide (DMA) were purchased from Sigma Aldrich. Carbon cloth was purchased from Gelon LIB Co. (China). PVDF–HFP was obtained from Kynar Powerflex LBG. Galwick was obtained from Porous Materials Inc. All materials were used as received. Electrospinning of PVDF-HFP was carried out as described previously [6]. PVDF–HFP was dissolved in a binary mixture of acetone and DMA with a weight ratio of 7:3. A Nanon-01A setup (MECC, Japan) was used for electrospinning at room temperature with 60% relative humidity inside the chamber. The needle tip–collector distance was kept at 15 cm and a voltage of 25 kV was applied between the needle and rotating drum. The PVDF–HFP solution was placed in a syringe with a needle of diameter 0.8 mm. A voltage of 25 kV was applied between the needle and rotating drum. The solution feed rate was kept at 1 ml/h. Electrospun PVDF–HFP nanofibers were collected on the rotating drum and dried in a conventional oven at 40 °C for 24 h.
Figure 1: Schematic that shows fabrication of CC-ES PH membrane by first coating carbon cloth with PVDF-HFP and then pressing electrospun fibers onto pre-coated carbon cloth Carbon cloth (CC) was pretreated with a dilute solution of PVDF-HFP to improve its adhesion to the electrospun layer. This was done by coating 5 wt. % DMA on CC and drying overnight on a
5
heating plate at 45 ℃. The electrospun PVDF-HFP mat was then hot-pressed onto the carbon cloth using a household iron, using a previously described procedure [7]. A schematic of the membrane is shown in Figure 1. The thickness of the resulting membrane was ~400 µm. 2.2. Characterization 2.2.1. Morphology Morphology of carbon cloth before coating with dilute PVDF-HFP and with electrospun PVDFHFP was examined using field-enhanced scanning electron microscopy (NovaNano, FEI) under high vacuum. Samples was coated with a thin layer of gold using a precision etching coating system (Gatan Model 682, Germany). 2.2.2. Pore size distribution Pore size distribution (PSD) of the membranes was analyzed with PMI Capillary Flow Porometer (PMI, Ithaca, NY-USA). Samples were first wet with Galwick (surface tension: 15.9 dynes/cm) and then the liquid was displaced from pores using a pressurized gas. The instrument determines pore size distribution and average pore size from the gas pressure needed to remove liquid from pores. 2.2.3. Liquid Entry Pressure (LEP) LEP measurements were also carried out on the capillary flow porometer, using DI water. A small circular sample with a diameter of 2.5 cm was covered with DI water inside the sealed chamber and a bubble point test was run with air at gradually increasing pressure. The LEP was recorded as the pressure corresponding to the point of initial passage of flow through the membrane.
6
2.2.4. Contact angle Contact angle tests were carried out at room temperature using an EasyDrop Standard drop shape analysis (KRUSS, Germany). A 4 μL droplet of deionized water was produced on the membrane surface and the digital image was used to determine contact angle. The contact angle was measured using sessile drop, tangent line and circle fitting methods and the average of these values was recorded. 2.3. Membrane distillation and wetting detection Direct contact MD was combined with an electrochemical cell in custom built system as depicted in Figure 2. A flat membrane sample of size 3.5 cm x 5.5 cm was used with the electrospun fibers facing the feed side, and the carbon cloth on the permeate side
a )
b )
7
Figure 2: a) schematic of the electrochemical cell, where black is the electrically conductive carbon cloth layer and white is the active electrospun PVDF-HFP; b) photograph of the DCMD cell The feed solution was prepared with tap water and table salt with a concentration of 20,000 ppm. The flow rate on both permeate and feed sides were controlled and kept at 0.6 L/min. Feed was kept at 65 ℃ while the permeate temperature was maintained at 25 ℃. For wetting detection, wetting was intentionally induced after 60 min of distillation by injecting 200 mL ethanol into the feed, bringing the alcohol concentration in the feed to about 10 wt. %. At the given feed temperature, ethanol has a much lower surface tension of 18 mN/m than water (65 mN/m). Adding ethanol to the feed mixture therefore reduces the surface tension and induces wetting during MD [8]. The conductivity of the permeate and flux were recorded. In this electrochemical MD setup, the carbon cloth on the permeate side acted as working electrode, whereas a stainless steel mesh on the other side of the cell acted as counter electrode. A voltage of either -1 V (cathodic) or +1 V (anodic) was applied using Autolab 302N potentiostat and the current through the system was recorded by the software Nova 1.8. The DCMD process was continued for 30 minutes after wetting. TDS was measured using an Accumet XL50 (Fisher Scientific) conductivity meter. Each experiment was carried out twice.
8
3. Results and Discussion 3.1 Membrane morphology
Figure 3: SEM images of a) carbon cloth pre-coated with dilute PVDF-HFP, b) top surface of PVDF-HFP - CC membrane with the inset showing a water droplet on the surface, and c) bottom surface of PVDF-HFP-CC membrane Figure 3a shows an SEM image of the carbon cloth after coating with PVDF-HFP to prepare for the electrospun layer. Microscale PVDF-HFP particles can be seen coating the carbon fibers. After pressing the electrospun layer (Figure 3b), the top surface of the membrane shows fibers fused together, which is typical of heat pressed electrospun PVDF-HFP fibers and contributes to their mechanical strength [7]. The bottom surface (Figure 3c) also shows the nano fibers through the carbon cloth. Upon a closer look, a thin layer of the PVDF-HFP coating that passes across the holes of the woven carbon cloth is also observed. It should be noted that the PVDF-HFP
9
particles in Figure 3a fused into a thin layer upon pressing as shown on the right in Figure 3c and is illustrated by the schematic in Figure 1. The pore size distribution of the fabricated membrane along with that of electrospun PVDF-HFP on its own is shown in Figure 4. The layer of PVDF-HFP cast on carbon cloth does not have a significant effect on the pore size. Pores are well within a suitable range for MD (0.1 - 1 µm) [9].
Figure 4: Pore size distribution of the CC-ES PVDF-HFP membrane and ES PVDF-HFP membrane Liquid entry pressure is another important parameter for MD membranes. A high LEP is desirable as higher the LEP, the more resistant a membrane is to wetting [10]. Table 1 summarizes the key properties of the membrane. It is important to note that the membranes are hydrophobic with a contact angle of about 130° (Figure 3b inset). A small pore size and high contact angle contribute to a high LEP value.
10
Table 1: LEP, bubble point and contact angle of the membrane Membrane
LEP (psi)
Bubble point (µm)
Mean pore size (µm)
Contact angle (°)
ES PH
34 ± 1
0.55 ± 0.1
0.17 ± 0.01
129 ± 4
CC-ES PH
36 ± 1
0.44 ± 0.1
0.2 ± 0.015
130 ± 3
Table 1 also shows the contact angle for the membrane, which as expected is similar to that of hot-pressed electrospun PVDF-HFP, in line with previously reported data [6]. 3.2 Direct contact membrane distillation (DCMD) Induced wetting at t = 60 min
Figure 5: DCMD flux for CC-ES PH membrane
11
a)
b)
Figure 6: Permeate TDS and current for a) anodic and b) cathodic voltage applied to the working electrode
Figure 5 shows the flux through the CC-ES PH membrane when an anodic state is applied, although the flux variation with time was not affected by the electrical state. Flux continues to increase as is typical in the beginning of the MD process as the various temperatures (feed, permeate, cell) stabilize and steady-state is achieved. The flux is about 17 LMH before the onset of wetting. Wetting causes pores to be at least partially filled with water, that hinders the passage of vapor across the membrane thus resulting in a decrease of flux. Decline in flux is due to partial wetting that causes salt to penetrate and therefore TDS also increases. Permeate TDS and current passing through the electrochemical system for both anodic and cathodic cases are also shown in Figure 6. It is clear that at the point of wetting (60 minutes), a sharp increase in current is noticed in both cases. This is because wetting allows Na+ and Cl- ions to pass through the membrane in addition to water, completing the circuit and increasing the conductivity of the electrolyte. Before wetting occurs (i.e. t < 60 min), the current is negligible, especially in the case where a positive voltage is applying to the permeate side. On the other hand, in the case of 12
negative voltage, it can be seen that the current gradually increases even before wetting occurs, after which point there is a sharp increase and the current reaches much higher values than in the anodic case. Although we intentionally caused the membrane to wet at a certain time, membrane wetting is usually a gradual phenomenon, which may be hard to detect with a negative bias, where a slight increase in current is noticed prior to membrane wetting. This difference in behavior between anodic and cathodic wetting detection could be attributed to the different electrocatalytic behavior of the carbon cloth and the stainless steel electrode and shows that the bias strongly affects detection. 4. Conclusion In this work, electrically conductive MD membranes were prepared by pressing electrospun PVDF-HFP on carbon cloth that was pre-coated with a thin layer of PVDF-HFP. The hydrophobicity of the membrane, mean pore size of 0.2 µm and LEP of 36 psi make this membrane a promising candidate for DCMD, demonstrating that the thin coating of PVDF-HFP on the carbon cloth has an insignificant effect on membrane properties and performance with respect to DCMD. The added electrically conductive layer allowed the membrane to be used in combined MD-electrochemical cell where the conductive layer acted as an electrode. A new technique to detect MD wetting electrolytically in real time was demonstrated using the conductive membrane. This technique, along with the development of conductive MD membranes, is likely to open new avenues for wetting detection and control in MD systems.
5. References [1] A. Alkhudhiri, N. Darwish, N. Hilal, Membrane distillation: A comprehensive review, Desalination, 287 (2012) 2-18.
13
[2] M. Gryta, Long-term performance of membrane distillation process, Journal of Membrane Science, 265 (2005) 153-159. [3] M. Gryta, M. Barancewicz, Influence of morphology of PVDF capillary membranes on the performance of direct contact membrane distillation, Journal of Membrane Science, 358 (2010) 158-167. [4] L.D. Tijing, J.-S. Choi, S. Lee, S.-H. Kim, H.K. Shon, Recent progress of membrane distillation using electrospun nanofibrous membrane, Journal of Membrane Science, 453 (2014) 435-462. [5] C.-I. Su, J.-H. Shih, M.-S. Huang, C.-M. Wang, W.-C. Shih, Y.-s. Liu, A study of hydrophobic electrospun membrane applied in seawater desalination by membrane distillation, Fibers and Polymers, 13 (2012) 698-702. [6] F. Ejaz Ahmed, B.S. Lalia, N. Hilal, R. Hashaikeh, Underwater superoleophobic cellulose/electrospun PVDF–HFP membranes for efficient oil/water separation, Desalination, 344 (2014) 48-54. [7] B.S. Lalia, E. Guillen-Burrieza, H.A. Arafat, R. Hashaikeh, Fabrication and characterization of polyvinylidenefluoride-co-hexafluoropropylene (PVDF-HFP) electrospun membranes for direct contact membrane distillation, Journal of Membrane Science, 428 (2013) 104-115. [8] A. Razmjou, E. Arifin, G. Dong, J. Mansouri, V. Chen, Superhydrophobic modification of TiO2 nanocomposite PVDF membranes for applications in membrane distillation, Journal of Membrane Science, 415–416 (2012) 850-863. [9] B.S. Lalia, V. Kochkodan, R. Hashaikeh, N. Hilal, A review on membrane fabrication: Structure, properties and performance relationship, Desalination, 326 (2013) 77-95. [10] B.B. Ashoor, S. Mansour, A. Giwa, V. Dufour, S.W. Hasan, Principles and applications of direct contact membrane distillation (DCMD): A comprehensive review, Desalination, 398 (2016) 222-246.
14
Highlights
An electrically conductive MD membrane is prepared by pressing electrospun polymer nanofibers on carbon cloth.
Membrane has a mean pore size of 0.2 µm and LEP of 36 psi, with salt rejection of 99.6%.
Membrane is used in combined DCMD-electrochemical system.
Real time wetting detection technique relies on measuring current through the system.
15
PII: DOI: Reference:
S0376-7388(17)30569-0 http://dx.doi.org/10.1016/j.memsci.2017.04.035 MEMSCI15199
To appear in: Journal of Membrane Science Received date: 27 February 2017 Revised date: 17 April 2017 Accepted date: 18 April 2017 Cite this article as: Farah Ejaz Ahmed, Boor Singh Lalia and Raed Hashaikeh, Membrane-based detection of wetting phenomenon in direct contact membrane d i s t i l l a t i o n , Journal of Membrane Science, http://dx.doi.org/10.1016/j.memsci.2017.04.035 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Membrane-based detection of wetting phenomenon in direct contact membrane distillation Farah Ejaz Ahmed, Boor Singh Lalia, Raed Hashaikeh*
Chemical Engineering Department, Khalifa University of Science and Technology, Masdar Institute,
Masdar
City,
P.O.
Box
54224,
Abu
*Corresponding author Email: [email protected] Phone: +971-28109152
April 2017
1
Dhabi,
United
Arab
Emirates
Abstract
Advances in membrane distillation are limited by the wetting phenomenon. Wetting is typically determined by examining the permeate water quality. In this work, we apply an electrically conductive membrane to direct contact membrane distillation combined with an electrochemical system. The membrane was fabricated by heat pressing a carbon cloth with electrospun PVDFHFP mat. The membrane acts not only as a barrier to reject salt with 99.6% rejection, but also as an electrode where in the current through the system is used to detect wetting which allows Na+ and Cl- ions to complete the cell. It was found that the membrane has an LEP of 36 psi and mean pore size of 0.2 µm. A continuous voltage of +1V was applied during the direct contact membrane distillation process process, and a sharp increase in current was observed at the point where wetting was induced. Keywords: conductive membrane; direct contact membrane distillation; wetting; PVDF-HFP; detection
2
1. Introduction Membrane distillation (MD) is a thermally driven separation process that has attracted attention for desalination and water treatment. In MD, a temperature difference, and consequently a vapor pressure difference across a porous hydrophobic membrane drive only vapor molecules to be transported from the feed to the permeate [1]. The efficiency of the MD process depends greatly on the ability of the membrane to resist wetting. In fact, progressive wetting of the membrane eventually prevents separation from continuing, and is one of the factors that deter MD processes from competing with other industrial desalination techniques. Increase in permeate conductivity, or decline in salt rejection, is a typical consequence of membrane wetting as well as fouling and scaling [2]. Wetting occurs when the transmembrane pressure exceeds the liquid entry pressure (LEP) of a membrane. LEP depends on both the pore size and the hydrophobicity of the membrane material, as highlighted by the Cantor-Laplace equation. As stated above, pore wetting is undesirable in MD processes. However, even membranes with high liquid entry pressure (LEP) that resist wetting for some time eventually succumb to wetting. Several studies in literature indicate the decline in salt rejection over time due to membrane wetting [3]. Current literature suggests two distinct categories of wetting detection in MD. The first involves examining water quality of the permeate, while the other involves studying variation of the membrane. Among membrane-based techniques, one way to confirm wetting is by examining the membrane surface on the permeate side for salt deposits. This requires the membrane to be removed and real-time detection is not possible. On the other hand, wetting detection through examining quality of the permeate can be slow, especially when multiple membranes are stacked together as in large scale MD systems. It is also challenging to identify which membrane in the 3
system is responsible for wetting. We suggest the use of an electrically conductive membrane that can be incorporated in a combined MD-electrochemical cell to detect wetting electrolytically. An in-situ membrane-based detection method where the membrane acts as an electrode would allow for feedback and control and would also make immediate detection possible, which is beneficial for large scale MD units. Electrospun nanofiber membranes have been the focus of many recent membrane distillation studies, due to versatility of the fabrication technique, high resulting porosity and interconnected pore structure [4]. Several studies have reported on the successful application of electrospun polyvinylidene fluoride-co-hexafluoropropylene (PVDF-HFP) [5] for MD. This is because PVDF-HFP is hydrophobic and can be electrospun using simple solvents at room temperature to form pores that are in the desirable range for MD. In this work, we demonstrate the fabrication of electrospun PVDF-HFP - carbon cloth membranes for MD and the use of these electrically conductive membranes for in-situ electrolytic wetting detection.
4
2. Materials and Methods 2.1. Fabrication Acetone and D-methylacetamide (DMA) were purchased from Sigma Aldrich. Carbon cloth was purchased from Gelon LIB Co. (China). PVDF–HFP was obtained from Kynar Powerflex LBG. Galwick was obtained from Porous Materials Inc. All materials were used as received. Electrospinning of PVDF-HFP was carried out as described previously [6]. PVDF–HFP was dissolved in a binary mixture of acetone and DMA with a weight ratio of 7:3. A Nanon-01A setup (MECC, Japan) was used for electrospinning at room temperature with 60% relative humidity inside the chamber. The needle tip–collector distance was kept at 15 cm and a voltage of 25 kV was applied between the needle and rotating drum. The PVDF–HFP solution was placed in a syringe with a needle of diameter 0.8 mm. A voltage of 25 kV was applied between the needle and rotating drum. The solution feed rate was kept at 1 ml/h. Electrospun PVDF–HFP nanofibers were collected on the rotating drum and dried in a conventional oven at 40 °C for 24 h.
Figure 1: Schematic that shows fabrication of CC-ES PH membrane by first coating carbon cloth with PVDF-HFP and then pressing electrospun fibers onto pre-coated carbon cloth Carbon cloth (CC) was pretreated with a dilute solution of PVDF-HFP to improve its adhesion to the electrospun layer. This was done by coating 5 wt. % DMA on CC and drying overnight on a
5
heating plate at 45 ℃. The electrospun PVDF-HFP mat was then hot-pressed onto the carbon cloth using a household iron, using a previously described procedure [7]. A schematic of the membrane is shown in Figure 1. The thickness of the resulting membrane was ~400 µm. 2.2. Characterization 2.2.1. Morphology Morphology of carbon cloth before coating with dilute PVDF-HFP and with electrospun PVDFHFP was examined using field-enhanced scanning electron microscopy (NovaNano, FEI) under high vacuum. Samples was coated with a thin layer of gold using a precision etching coating system (Gatan Model 682, Germany). 2.2.2. Pore size distribution Pore size distribution (PSD) of the membranes was analyzed with PMI Capillary Flow Porometer (PMI, Ithaca, NY-USA). Samples were first wet with Galwick (surface tension: 15.9 dynes/cm) and then the liquid was displaced from pores using a pressurized gas. The instrument determines pore size distribution and average pore size from the gas pressure needed to remove liquid from pores. 2.2.3. Liquid Entry Pressure (LEP) LEP measurements were also carried out on the capillary flow porometer, using DI water. A small circular sample with a diameter of 2.5 cm was covered with DI water inside the sealed chamber and a bubble point test was run with air at gradually increasing pressure. The LEP was recorded as the pressure corresponding to the point of initial passage of flow through the membrane.
6
2.2.4. Contact angle Contact angle tests were carried out at room temperature using an EasyDrop Standard drop shape analysis (KRUSS, Germany). A 4 μL droplet of deionized water was produced on the membrane surface and the digital image was used to determine contact angle. The contact angle was measured using sessile drop, tangent line and circle fitting methods and the average of these values was recorded. 2.3. Membrane distillation and wetting detection Direct contact MD was combined with an electrochemical cell in custom built system as depicted in Figure 2. A flat membrane sample of size 3.5 cm x 5.5 cm was used with the electrospun fibers facing the feed side, and the carbon cloth on the permeate side
a )
b )
7
Figure 2: a) schematic of the electrochemical cell, where black is the electrically conductive carbon cloth layer and white is the active electrospun PVDF-HFP; b) photograph of the DCMD cell The feed solution was prepared with tap water and table salt with a concentration of 20,000 ppm. The flow rate on both permeate and feed sides were controlled and kept at 0.6 L/min. Feed was kept at 65 ℃ while the permeate temperature was maintained at 25 ℃. For wetting detection, wetting was intentionally induced after 60 min of distillation by injecting 200 mL ethanol into the feed, bringing the alcohol concentration in the feed to about 10 wt. %. At the given feed temperature, ethanol has a much lower surface tension of 18 mN/m than water (65 mN/m). Adding ethanol to the feed mixture therefore reduces the surface tension and induces wetting during MD [8]. The conductivity of the permeate and flux were recorded. In this electrochemical MD setup, the carbon cloth on the permeate side acted as working electrode, whereas a stainless steel mesh on the other side of the cell acted as counter electrode. A voltage of either -1 V (cathodic) or +1 V (anodic) was applied using Autolab 302N potentiostat and the current through the system was recorded by the software Nova 1.8. The DCMD process was continued for 30 minutes after wetting. TDS was measured using an Accumet XL50 (Fisher Scientific) conductivity meter. Each experiment was carried out twice.
8
3. Results and Discussion 3.1 Membrane morphology
Figure 3: SEM images of a) carbon cloth pre-coated with dilute PVDF-HFP, b) top surface of PVDF-HFP - CC membrane with the inset showing a water droplet on the surface, and c) bottom surface of PVDF-HFP-CC membrane Figure 3a shows an SEM image of the carbon cloth after coating with PVDF-HFP to prepare for the electrospun layer. Microscale PVDF-HFP particles can be seen coating the carbon fibers. After pressing the electrospun layer (Figure 3b), the top surface of the membrane shows fibers fused together, which is typical of heat pressed electrospun PVDF-HFP fibers and contributes to their mechanical strength [7]. The bottom surface (Figure 3c) also shows the nano fibers through the carbon cloth. Upon a closer look, a thin layer of the PVDF-HFP coating that passes across the holes of the woven carbon cloth is also observed. It should be noted that the PVDF-HFP
9
particles in Figure 3a fused into a thin layer upon pressing as shown on the right in Figure 3c and is illustrated by the schematic in Figure 1. The pore size distribution of the fabricated membrane along with that of electrospun PVDF-HFP on its own is shown in Figure 4. The layer of PVDF-HFP cast on carbon cloth does not have a significant effect on the pore size. Pores are well within a suitable range for MD (0.1 - 1 µm) [9].
Figure 4: Pore size distribution of the CC-ES PVDF-HFP membrane and ES PVDF-HFP membrane Liquid entry pressure is another important parameter for MD membranes. A high LEP is desirable as higher the LEP, the more resistant a membrane is to wetting [10]. Table 1 summarizes the key properties of the membrane. It is important to note that the membranes are hydrophobic with a contact angle of about 130° (Figure 3b inset). A small pore size and high contact angle contribute to a high LEP value.
10
Table 1: LEP, bubble point and contact angle of the membrane Membrane
LEP (psi)
Bubble point (µm)
Mean pore size (µm)
Contact angle (°)
ES PH
34 ± 1
0.55 ± 0.1
0.17 ± 0.01
129 ± 4
CC-ES PH
36 ± 1
0.44 ± 0.1
0.2 ± 0.015
130 ± 3
Table 1 also shows the contact angle for the membrane, which as expected is similar to that of hot-pressed electrospun PVDF-HFP, in line with previously reported data [6]. 3.2 Direct contact membrane distillation (DCMD) Induced wetting at t = 60 min
Figure 5: DCMD flux for CC-ES PH membrane
11
a)
b)
Figure 6: Permeate TDS and current for a) anodic and b) cathodic voltage applied to the working electrode
Figure 5 shows the flux through the CC-ES PH membrane when an anodic state is applied, although the flux variation with time was not affected by the electrical state. Flux continues to increase as is typical in the beginning of the MD process as the various temperatures (feed, permeate, cell) stabilize and steady-state is achieved. The flux is about 17 LMH before the onset of wetting. Wetting causes pores to be at least partially filled with water, that hinders the passage of vapor across the membrane thus resulting in a decrease of flux. Decline in flux is due to partial wetting that causes salt to penetrate and therefore TDS also increases. Permeate TDS and current passing through the electrochemical system for both anodic and cathodic cases are also shown in Figure 6. It is clear that at the point of wetting (60 minutes), a sharp increase in current is noticed in both cases. This is because wetting allows Na+ and Cl- ions to pass through the membrane in addition to water, completing the circuit and increasing the conductivity of the electrolyte. Before wetting occurs (i.e. t < 60 min), the current is negligible, especially in the case where a positive voltage is applying to the permeate side. On the other hand, in the case of 12
negative voltage, it can be seen that the current gradually increases even before wetting occurs, after which point there is a sharp increase and the current reaches much higher values than in the anodic case. Although we intentionally caused the membrane to wet at a certain time, membrane wetting is usually a gradual phenomenon, which may be hard to detect with a negative bias, where a slight increase in current is noticed prior to membrane wetting. This difference in behavior between anodic and cathodic wetting detection could be attributed to the different electrocatalytic behavior of the carbon cloth and the stainless steel electrode and shows that the bias strongly affects detection. 4. Conclusion In this work, electrically conductive MD membranes were prepared by pressing electrospun PVDF-HFP on carbon cloth that was pre-coated with a thin layer of PVDF-HFP. The hydrophobicity of the membrane, mean pore size of 0.2 µm and LEP of 36 psi make this membrane a promising candidate for DCMD, demonstrating that the thin coating of PVDF-HFP on the carbon cloth has an insignificant effect on membrane properties and performance with respect to DCMD. The added electrically conductive layer allowed the membrane to be used in combined MD-electrochemical cell where the conductive layer acted as an electrode. A new technique to detect MD wetting electrolytically in real time was demonstrated using the conductive membrane. This technique, along with the development of conductive MD membranes, is likely to open new avenues for wetting detection and control in MD systems.
5. References [1] A. Alkhudhiri, N. Darwish, N. Hilal, Membrane distillation: A comprehensive review, Desalination, 287 (2012) 2-18.
13
[2] M. Gryta, Long-term performance of membrane distillation process, Journal of Membrane Science, 265 (2005) 153-159. [3] M. Gryta, M. Barancewicz, Influence of morphology of PVDF capillary membranes on the performance of direct contact membrane distillation, Journal of Membrane Science, 358 (2010) 158-167. [4] L.D. Tijing, J.-S. Choi, S. Lee, S.-H. Kim, H.K. Shon, Recent progress of membrane distillation using electrospun nanofibrous membrane, Journal of Membrane Science, 453 (2014) 435-462. [5] C.-I. Su, J.-H. Shih, M.-S. Huang, C.-M. Wang, W.-C. Shih, Y.-s. Liu, A study of hydrophobic electrospun membrane applied in seawater desalination by membrane distillation, Fibers and Polymers, 13 (2012) 698-702. [6] F. Ejaz Ahmed, B.S. Lalia, N. Hilal, R. Hashaikeh, Underwater superoleophobic cellulose/electrospun PVDF–HFP membranes for efficient oil/water separation, Desalination, 344 (2014) 48-54. [7] B.S. Lalia, E. Guillen-Burrieza, H.A. Arafat, R. Hashaikeh, Fabrication and characterization of polyvinylidenefluoride-co-hexafluoropropylene (PVDF-HFP) electrospun membranes for direct contact membrane distillation, Journal of Membrane Science, 428 (2013) 104-115. [8] A. Razmjou, E. Arifin, G. Dong, J. Mansouri, V. Chen, Superhydrophobic modification of TiO2 nanocomposite PVDF membranes for applications in membrane distillation, Journal of Membrane Science, 415–416 (2012) 850-863. [9] B.S. Lalia, V. Kochkodan, R. Hashaikeh, N. Hilal, A review on membrane fabrication: Structure, properties and performance relationship, Desalination, 326 (2013) 77-95. [10] B.B. Ashoor, S. Mansour, A. Giwa, V. Dufour, S.W. Hasan, Principles and applications of direct contact membrane distillation (DCMD): A comprehensive review, Desalination, 398 (2016) 222-246.
14
Highlights
An electrically conductive MD membrane is prepared by pressing electrospun polymer nanofibers on carbon cloth.
Membrane has a mean pore size of 0.2 µm and LEP of 36 psi, with salt rejection of 99.6%.
Membrane is used in combined DCMD-electrochemical system.
Real time wetting detection technique relies on measuring current through the system.
15