Thin-film composite tri-bore hollow fiber (TFC TbHF) membranes for isopropanol dehydration by pervaporation

Thin-film composite tri-bore hollow fiber (TFC TbHF) membranes for isopropanol dehydration by pervaporation

Journal of Membrane Science 471 (2014) 155–167 Contents lists available at ScienceDirect Journal of Membrane Science journal homepage: www.elsevier...

9MB Sizes 0 Downloads 125 Views

Journal of Membrane Science 471 (2014) 155–167

Contents lists available at ScienceDirect

Journal of Membrane Science journal homepage: www.elsevier.com/locate/memsci

Thin-film composite tri-bore hollow fiber (TFC TbHF) membranes for isopropanol dehydration by pervaporation Dan Hua a,b, Yee Kang Ong b, Peng Wang b, Tai-Shung Chung a,b,n a b

NUS Graduate School for Integrative Science and Engineering, National University of Singapore, Singapore 117456, Singapore Department of Chemical & Biomolecular Engineering, National University of Singapore, Singapore 117585, Singapore

art ic l e i nf o

a b s t r a c t

Article history: Received 21 May 2014 Received in revised form 17 July 2014 Accepted 26 July 2014 Available online 14 August 2014

For the first time, this paper reports the fabrication of novel thin-film composite tri-bore hollow fiber (TFC TbHF) membranes for pervaporation dehydration of isopropanol (IPA). The interfacial polymerization conditions were first optimized using the conventional single-bore hollow fiber (SbHF) substrate. Then, the effects of spinning parameters such as bore fluid composition, bore fluid flow rate, air gap and dope flow rate on TFC TbHF membranes' geometry, morphology and pervaporation performance were systematically investigated. It was found that the bore fluid composition is the most critical factor. Moreover, fine TFC TbHF membranes with an outer diameter of approximately 1 mm exhibit better separation performance for IPA dehydration than thick TFC TbHF membranes with an outer diameter of approximately 2 mm. The optimal TFC TbHF membrane shows a flux of 2.65 kg m  2 h  1 with a separation factor (water/IPA) of 261 for the dehydration of an IPA/water (85/15 wt%) mixture at 50 1C. Most importantly, the newly designed TFC TbHF membrane not only shows enhanced separation performance and mechanical strengths as compared with the conventional TFC SbHF spun from the identical spinning conditions, but also exhibits satisfying long-term stability, suggesting its a great potential for pervaporation applications. & 2014 Elsevier B.V. All rights reserved.

Keywords: Multi-bore hollow fiber (MbHF) Tri-bore hollow fiber (TbHF) Thin film composite (TFC) Isopropanol Pervaporation

1. Introduction Alcohols are of great importance in industry because they are widely used as good solvents, cleaning agents, raw materials or chemical intermediates for organic synthesis, and potentially clean liquid fuels [1–7]. Accordingly, the dehydration of alcohol/water mixtures is a critical issue in the production and recycle of alcohols. However, the alcohol (ethanol, isopropanol, or butanol) and water can form an azeotropic mixture, rendering the purification of alcohols by conventional methods such as distillation inefficient and uneconomic. Considering the requirements for energy-saving and environmentally friendly processes in the chemical industry nowadays, the pervaporation process is of considerable interest due to its high separation efficiency, ability to break azeotropes, lower energy consumption, as well as flexible process control and module fabrication [7–10]. The pervaporation membrane is the heart of pervaporation processes. Based on its configuration, it can be categorized as flat sheet, hollow fiber (HF) or tubular membrane. Among them, the polymeric HF membrane has been widely applied in various n Corresponding author at: National University of Singapore, Department of Chemical and Biomolecular Engineering, 10 Kent Ridge Crescent, Singapore 117585, Singapore. Tel.: þ 65 6516 6645; fax: þ65 6779 1936. E-mail address: [email protected] (T.-S. Chung).

http://dx.doi.org/10.1016/j.memsci.2014.07.059 0376-7388/& 2014 Elsevier B.V. All rights reserved.

separation processes due to its superior advantages such as higher surface area, higher packing density, excellent flexibility and ease of fabrication as well as scale-up [8,11–13]. Polymeric pervaporation HF membranes including single-layer asymmetric HFs and dual-layer composite HFs are usually fabricated by means of spinning via a non-solvent induced phase inversion process [13–21]. Since heat treatment, silicone coating and chemical crosslinking are often needed to improve the separation efficiency which incur extra costs and the dual-layer spinning technique is complicated, the thin-film composite hollow fiber (TFC HF) membrane fabricated via interfacial polymerization has drawn attention in recent years for pervaporation due to its excellent separation performance and ease of fabrication [22–28]. The conventional HF membrane has single-bore geometry. Extensive research works have been focused on the design and fabrication of single-bore hollow fiber (SbHF) membranes [11]. Fine polymeric SbHFs are found to suffer from the tendency of fiber breakage during continuous operations and thus lack longterm stability [29,30]. To enhance the mechanical properties, mixed matrix SbHF membranes were fabricated by the incorporation of inorganic fillers such as carbon nanotubes, graphene based nanomaterials, and layered silicate into the polymer matrix [31–33]. Besides, a new generation of multi-bore hollow fiber (MbHF) membranes is emerging recently [34–43]. The new membranes also possess improved mechanical properties as

156

D. Hua et al. / Journal of Membrane Science 471 (2014) 155–167

compared with traditional SbHFs owing to the presence of spokes as the mechanical reinforcement, and thus reduce fiber breakage and increase operation reliability. Comparing with the mixed matrix SbHFs, the polymeric MbHFs are free from issues such as filler agglomeration and poor compatibility between the inorganic filler and the polymer matrix. Two MbHF membranes have been commercialized for ultrafiltration (UF) applications. They are (1) Multibores polyethersulfone (PES) HF membranes consisting of seven bores developed by inge GmbH [35] and (2) tri-bore PES or polyvinylidene fluoride (PVDF) HF membranes launched by Hyflux [36]. Both of them claimed greater durability and mechanical strength while maintaining the same permeate production rate as compared with the respective SbHF membranes [35,36]. Moreover, some showed extra benefits such as a  75% reduction in energy consumption compared with the traditional UF membrane in the pretreatment of reverse osmosis plants for seawater desalination [37,38]. Besides UF, the applications of MbHFs were also extended to other membrane processes. Chung and his co-workers fabricated a series of MbHF membranes with various dimensions, configurations and pore sizes for UF, membrane distillation (MD) and forward osmosis (FO) applications [39–42]. The detailed MbHF spinneret designs, membrane formation mechanisms and spinning parameters

were fully explored. Their MbHFs for MD not only showed high permeation fluxes and energy efficiency but also possess superior stability and robustness. In addition, TFC MbHFs for FO were fabricated via a spinning process followed by interfacial polymerization. These fibers showed comparable separation performance with other FO membranes. Spruck et al. also fabricated TFC seven-bore HF membranes which showed good mechanical strengths and acceptable performance in nanofiltration for desalination and water softening applications [43]. In view of the aforementioned advantages of MbHF membranes and TFC membranes, we aim to design and explore thin-film composite tri-bore hollow fiber (TFC TbHF) membranes for the application of IPA dehydration. To our best knowledge, this is a pioneering work to fabricate polymeric outer-selective TFC TbHF membranes for pervaporation dehydration. Firstly, the effects of interfacial polymerization conditions on pervaporation performance of TFC HF membranes were studied by using conventional SbHF substrates. A most suitable procedure for the subsequent fabrication of TFC TbHF membranes was then identified based on the aforementioned study. Afterwards, the effects of spinning conditions on membrane morphology of the TbHF substrates and their TFC TbHF performance for pervaporation dehydration of IPA were systematically investigated. The outcome of this study may

Fig. 1. The scheme of fabricating TFC HF with different methods: (1) the central red solid line box: the interfacial polymerization process; (2) the green dashed line box: precoating HPEI and then interfacial polymerization; (3) the blue dotted line box: post-PDMS coating after the interfacial polymerization. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 2. The parameters (right) for describing the configuration of the TbHF substrate (left).

D. Hua et al. / Journal of Membrane Science 471 (2014) 155–167

provide useful insights towards the design of next generation HF membranes for pervaporation applications.

2. Experimental methods

157

Table 1 Effects of amine solution composition on IPA dehydration performance of TFC SbHF a. Condition Composition of the Total flux Water (g m  2 h  1) concentration at aqueous amine solution the permeate (wt %)

Separation factor (Water/IPA)

2.1. Materials A commercial poly (ether imide) known as Ultems 1010 was supplied by SABIC Innovative Plastics. N-methyl-2-pyrrolidone (NMP, analytical grade, Merck) and ethanol (analytical grade, Fisher) were acquired as the solvent and non-solvent respectively to prepare the spinning solution, while NMP, 1-butanol (BuOH, analytical grade, Fisher) as well as deionized water were used to prepare the bore fluids. A hyper-branched polyethyleneimine (HPEI, 50 wt% in aqueous solutions, Sigma-Aldrich) with a molecular weight of 60 kg/mol was used to prepare the surface coating solution. Sylgards184 silicone elastomer and its curing agent for polydimethylsiloxane (PDMS) coating were purchased from Dow Corning Singapore Pte. Ltd. The reaction monomers for the interfacial polymerization, namely, m-phenylenediamine (MPD, reagent grade, Tokyo Chemical Industry Co., Ltd.) and trimesoyl chloride (TMC, reagent grade, Sigma-Aldrich) were dissolved in aqueous and organic phases, respectively. In addition, sodium dodecyl sulfate (SDS, reagent grade, Fluka), triethylamine (TEA, reagent grade, Sigma-Aldrich), and cetyltrimethylammonium bromide (CTAB, reagent grade, Sigma-Aldrich) were procured as additives for the interfacial polymerization. Moreover, methanol (reagent grade, Merck) and n-hexane (analytical grade, Fisher) were employed as solvents in this study. Isopropanol (IPA, reagent grade, Fisher) and deionized water were used to prepare the feed mixture for pervaporation studies. All chemicals were used as received.

Substrate Nil 1 MPD (2 wt%) 2 MPD(2 wt%)þ SDS (0.1 wt%)b 3 MPD (2 wt%)þ CTAB(0.1 wt%)b

61727 404 2682 7 37 25017 61

37.3 7 2.6 90.4 7 0.3 84.17 1.0

3.4 7 0.4 51.2 7 1.4 307 2.1

26377 38

83.0 7 1.0

287 2.0

a Substrate spinning conditions: bore fluid composition: NMP/H2O (95/5 wt%), bore fluid flow rate: 5 ml/min, air gap: 2.5 cm, dope flow rate: 5 ml/min. b TEA was used as the base source with a weight amount of 0.5 wt%.

end sealed by epoxy were immersed into the aqueous solution for 3 min at room temperature. They were then blotted with tissue paper and dipped into the organic solution for 1 min for the interfacial polymerization. Finally, the resultant fibers were annealed at 65 1C for 15 min to stabilize the selective TFC layer. In the case of applying HPEI as the pre-treatment (referred to as HPEI-TFC), the substrate was first immersed into an aqueous 1 wt% HPEI solution for 2 min, and then dried at 65 1C for 10 min. Thereafter, the polyamide selective layer was synthesized as described above. The resultant sample is referred to as HPEI-TFC. In the case of PDMS post-treatment (referred to as TFC-PDMS), the TFC-HF was immersed into a 3 wt% PDMS/hexane solution for 1 min and then the coated fiber was cured for at least 24 h in air. The scheme of all methods to fabricate TFC-HF membranes is illustrated in Fig. 1. 2.4. The fabrication of HF modules

2.2. The fabrication of TbHF substrates The TbHF substrates were prepared via a dry-jet wet spinning process using a specifically designed tri-bore spinneret with blossom geometry [41]. The detailed spinning process has been described elsewhere [15]. A homogeneous dope solution of Ultems/ethanol/NMP with a ratio of 23/5/72 wt% was prepared, poured into an ISCO syringe pump and degassed overnight before the spinning process. This dope composition has been found to be useful for producing SbHFs with suitable pore sizes for interfacial polymerization [22]. Different bore fluids were prepared by mixing NMP with water or n-butanol as the non-solvent at certain ratios. The dope solution and bore fluid were concurrently extruded out of the spinneret and the nascent fibers entered into the water coagulation tank with a certain air gap. The TbHF substrates were then collected by a collection drum and rinsed in tap water for at least 3 days to remove the residual solvent with water changed daily. Then, the TbHF substrates were solvent exchanged with three consecutive times of methanol followed by hexane. Finally, they were air dried for further characterizations, postmodifications and pervaporation tests. 2.3. The fabrication of TFC HF membranes The TFC HF membranes were prepared by conducting interfacial polymerization on the HF substrates, where the monomer MPD in the aqueous phase reacts with another monomer TMC in the organic phase to form a polyamide selective layer on the outer surface of the substrates [44–47]. The aqueous phase consists of 2 wt% MPD in deionized water with or without 0.1 wt% SDS (or CTAB) and 0.5 wt% TEA in certain conditions; while the organic phase has 0.15 wt% TMC in hexane. The TbHF substrates with one

The HF modules were prepared by loading the HF membranes into a module holder assembled from two Swagelok stainless steel male run tees and a 3/8 in. perfluoroalkoxy tube, and both ends of the male run tee were sealed with epoxy. Each module contained one HF with an effective length of 14–15 cm. All modules were cured for 48 h at ambient temperature before evaluation. 2.5. Characterizations of HF substrates and TFC HFs The morphologies of TbHF and SbHF membranes were observed by an optical microscope (microscope: Olympus, SZX16; digital camera: Olympus, CMAD3) and a field-emission scanning electron microscope (FESEM, JEOL JSM-6700LV). The dimensions of TbHF membranes could be determined from the microscopic images, including the outer diameter, inner diameter, wall thickness, as well as spoke thickness as shown in Fig. 2. The shaded area represents the polymeric cross-section area. The membrane surface topology was examined using a Nanoscope V atomic force microscope (AFM) from Bruker Dimension ICON. A scanning size of 10 μm  10 μm was chosen and the mean surface roughness (Ra) was used to quantify the surface roughness. To study the surface hydrophilicity of original and HPEI coated HF substrates, the water contact angle was measured by a Sigma 701 Tensiometer from KSV Instruments Limited. Each sample was measured at least thrice. The membrane mechanical properties of the fabricated HFs were characterized using an Instron tensiometer (Model 5542, Instron Corp.) at room temperature, where the HFs were clamped at the both ends with an initial gauge length of 50 mm and a testing rate of 10 mm/min. With the aid of a slow beam positron apparatus, experiments were carried out by both Doppler broadening energy spectroscopy (DBES) and

158

D. Hua et al. / Journal of Membrane Science 471 (2014) 155–167

Fig. 3. FESEM images of SbHF outer surfaces made from different amine solutions, (a) substrate, (b) 1st amine solution, (c) 2nd amine solution, (d) 3rd amine solution.

Fig. 4. FESEM images of cross-section and outer surface of TFC SbHFs: (a) substrate, (b) only TFC, (c) TFC-PDMS, (d) HPEI-TFC. (The spinning conditions of the SbHF substrate: bore fluid composition: NMP/H2O (95/5 wt%), bore fluid flow rate: 5 ml/min, air gap: 2.5 cm, dope flow rate: 5 ml/min.).

positron annihilation lifetime spectroscopy (PALS) to study the depth profile of the selective layer as well as the free volume of TFC TbHF membranes. 2.6. Pervaporation experiments A laboratory scale pervaporation unit was employed to evaluate the membrane performance and the details of the apparatus have been depicted by Liu et al. [14]. A 2 L feed solution of IPA/water

mixture (85/15 wt%) was circulated through the shell side of the module with a flow rate of 30 L/h at a feed temperature of 50 1C, while the lumen side (permeate side) of the module was vacuumed at a pressure below 1 mbar throughout the experiments. The system was stabilized for 2 h before the sample collection. Thereafter, permeate samples were collected by a cold trap immersed in liquid nitrogen. The samples were then weighted and their compositions were analyzed by gas chromatography (Hewlett–Packard GC 7890, HP-INNOWAX column, thermal

D. Hua et al. / Journal of Membrane Science 471 (2014) 155–167

conductivity detector (TCD)). At least three permeate samples were collected and their average was reported. The flux was calculated by the following equation: J¼

Q At

ð1Þ

wherein, Q, A and t are the weight of the permeate sample collected, the effective membrane surface area, and the time interval during the sample collection, respectively. The separation factor is defined by the equation below:

β¼

ywi =ywj xwi =xwj

ð2Þ

wherein, yw and xw are the weight fractions of the component in the permeate and feed, respectively; the subscripts i and j refer to components of water and IPA, respectively. The intrinsic properties of an asymmetric membrane could be reflected by the driving force normalized terms; namely, molebased permeance and selectivity [48]. Therefore, the flux and separation factor were converted to the mole-based permeance and selectivity based on the solution-diffusion model: P Ji P~ i ¼ i ¼ p l M i ðxi γ i psat i  yi p Þ

ð3Þ

wherein, P~ i and Pi are the mole-based permeance and permeability of component i, respectively. l is the membrane thickness, Ji, Mi, γi, and psat are the flux, molecular weight, activity coefficient, i and the saturated vapor pressure of component i, respectively, xi and yi are the mole fractions of component i in the feed and permeate, pp is the total pressure at the permeate side, which could be assumed as zero due to the vacuum condition. Both of γi Table 2 The IPA dehydration performance of TFC SbHF membranes with different treatments a. Coating conditions

Total flux Water concentration at (g m  2 h  1) the permeate (wt%)

Separation factor (Water/IPA)

Nil (substrate) TFC (condition 1) TFC-PDMS HPEI-TFC

61727 404 2682 7 37 2480 7 57 23567 26

37.3 7 2.6 90.4 7 0.3 90.8 7 0.5 97.17 0.2

3.4 7 0.4 51.2 7 1.4 55.9 7 3.6 192.2 7 12.1

a Substrate spinning conditions: bore fluid composition: NMP/H2O (95/5 wt%), bore fluid flow rate: 5 ml/min, air gap: 2.5 cm, dope flow rate: 5 ml/min.

159

and psat were determined by the Wilson equation and the Antoine i equation, respectively, with the aid of the AspenTech Process Modeling software (version 7.2). Then, the mole-based selectivity of the membrane (α) was defined as the ratio of the permeability of components i and j:

α ¼ P~ i =P~ j

ð4Þ

3. Results and discussion 3.1. Effect of TFC selective layer Since the polyamide selective layer plays an important role in determining the overall separation performance, identifying a suitable interfacial polymerization protocol for pervaporation dehydration of IPA is necessary. To search for the proper protocol, SbHFs spun from the same dope were used prior to applying to the novel TbHFs.

3.1.1. Interfacial polymerization conditions Interfacial polymerizations with and without TEA additive (i.e., a reaction accelerator) and surfactant (i.e., a wetting agent) have been reported for pervaporation dehydration of alcohols [25,26]. Accordingly, three different amine solutions were conducted on SbHF substrates to compare their effects on the polyamide selective layer for pervaporation dehydration of IPA. As shown in Table 1, the TFCSbHF interfacially polymerized from the 1st amine solution and TMC has greater dehydration performance in terms of flux and separation factor than those TFC-SbHFs synthesized from TMC with the 2nd and 3rd amine solutions. Fig. 3 compares their surface morphology. The membrane synthesized from the 1st amine solution and TMC has a dense polyamide layer consisting of “nodular structure”. However, the TFC HFs manufactured from TMC and amine solutions containing surfactants have “flake-like” surface morphology. This is because surfactants facilitate the diffusion of MPD to the organic phase, enlarge the contact area and enhance the reaction. The resulting polyamide layer with this morphology was reported to have a larger free volume and loose structure [49–51]. As a result, it has a lower separation factor compared to the TFC HF synthesized from the condition 1. Therefore, the condition 1 was applied to the subsequent sections for the interfacial polymerization of TbHF substrates.

Fig. 5. 3-D AFM images and mean surface roughness of SbHF outer surfaces (10  10 μm2) prepared under different coating methods (a) substrate (b) only TFC (c) TFCPDMS (d) only HPEI; (e) HPEI-TFC. (The spinning conditions of the SbHF substrate: bore fluid composition: NMP/H2O (95/5 wt%), bore fluid flow rate: 5 ml/min, air gap: 2.5 cm, dope flow rate: 5 ml/min.).

160

D. Hua et al. / Journal of Membrane Science 471 (2014) 155–167

Fig. 6. The cross-section, outer surface as well as inner surface morphologies of TbHF substrates spun from the bore fluid compositions of NMP/H2O (80/20 wt%), NMP/H2O (95/5 wt%), NMP/BuOH (80/20 wt%), NMP/BuOH (95/5 wt%). Other spinning conditions: bore fluid flow rate: 5 ml/min , air gap: 0.5 cm, dope flow rate: 5 ml/min.

Fig. 7. Schematic representation of the concept of phase inversion processes for the formation of thick and fine TbHF substrates.

D. Hua et al. / Journal of Membrane Science 471 (2014) 155–167

3.1.2. Effect of HPEI or PDMS coating To further improve the pervaporation performance of TFC HF membranes, two treatments were explored to alleviate defects in the polyamide selective layer. Fig. 4 shows the morphologies of four types of fibers; namely, (1) the hollow fiber substrate, (2) TFC hollow fiber (i.e., interfacial polymerization under condition 1), (3) HPEI-TFC (i.e., HPEI pre-treatment and then interfacial polymerization), and (4) TFC-PDMS (i.e., interfacial polymerization and then PDMS coating), while Table 2 summarizes their pervaporation performance. The HPEITFC hollow fiber has the most impressive pervaporation performance. Compared to the TFC hollow fiber, it has about fourfold separation factor without significantly compromising the flux. This is likely due to the fact that the HPEI coated substrate has a hydrophilic smooth surface with smaller pores [24,25]. As a consequence, not only does the modified substrate facilitate the absorbance of MPD for interfacial polymerization but also promote the formation of a less defective polyamide layer. The surface contact angle of the original HF substrate reduced from 88.472.21 to 40.472.91 after the HPEI coating, which proves the enhancement of surface hydrophilicity. The AFM images of these membrane surfaces further confirm our hypotheses. As shown in Fig. 5, the surface roughness decreases after the HPEI coating. Moreover, the amine groups of HPEI can react with TMC, minimize the interstitial space and thus further increase the separation factor [24,25]. Similarly, the roughness of the TFC hollow fiber decreases after the PDMS coating. However, since there is no chemical reaction between TFC and PDMS, the defects in the rough polyamide layer may not be fully sealed by the PDMS coating. Thus, the TFC-PDMS hollow fiber does not show much improvement in separation factor. As a result, the HPEI-TFC method was chosen to fabricate the TFC TbHF membranes in the following study.

3.2. The effects of spinning parameters on TbHF substrates and pervaporation performance of TFC TbHFs In order to fabricate TbHF substrates with a desirable pore-size distribution and open-cell substructure to maximize the separation Table 3 Effects of bore fluid composition on IPA dehydration performance of TFC TbHF membranes a. Bore fluid composition

NMP/H2O (80/20 wt%) NMP/H2O (95/5 wt%) NMP/BuOH (80/20 wt%) NMP/BuOH (95/5 wt%)

Total flux (g m  2 h  1)

Water concentration at the permeate (wt%)

Separation factor (Water/IPA)

943 7 11

77.8 7 0.5

20.17 0.6

24757 80

97.1 7 0.2

197.0 7 14.6

27307 47

95.1 7 0.3

113.0 7 6.5

3080 7 112

98.0 7 0.3

274.0 7 40.5

a Other spinning conditions: bore fluid flow rate: 5 ml/min, air gap: 0.5 cm, dope flow rate: 5 ml/min.

161

performance of TFC TbHFs, it is necessary to investigate the key spinning factors for the formation of TbHFs [11]. Therefore, the effects of bore fluid composition, bore fluid flow rate, air-gap distance as well as dope flow rate on TbHF morphology and its pervaporation performance after interfacial polymerization were systematically investigated.

3.2.1. Bore fluid composition Fig. 6 depicts the morphologies of TbHF substrates spun from bore fluids with different compositions. The bore fluid composition has significant impact toward the fiber dimension and morphology. The outer diameter, wall thickness, and spoke thickness of TbHFs spun from a strong bore fluid (namely, NMP/H2O (80/20 wt%) which contains 20 wt% H2O (strong non-solvent)) are much larger than those TbHFs spun from three other weaker bore fluids consisting of more NMP (solvent) or BuOH (weak nonsolvent). As a result, the former is a large and thick TbHF, while the latter three are fine TbHFs. All the fine TbHFs show partial deformed spokes. However, they have a denser outer surface and a much more porous inner surface compared with the thick TbHF. The mechanism of forming fine and thick TbHFs can be elucidated from Fig. 7. During the dry-jet wet-spinning process, the phase inversion begins at the lumen side of the nascent fiber once the bore fluid contacts with the polymer dope solution at the exit of the spinneret. Meanwhile, a partial phase inversion occurs at the outer surface of the nascent fiber due to the moisture induced phase inversion in the air gap region. The nascent fiber would have a complete phase inversion at its outer surface once it is fully immersed into the external coagulation bath of water [52–54]. Since NMP/H2O (95/5 wt%), NMP/BuOH (80/20 wt%) and NMP/ BuOH (95/5 wt%)) are weak bore fluids, while H2O is a strong coagulant, these result in a much slower phase inversion at the lumen side than at the shell side of the fiber [55–57]. The resultant TbHF substrate has a porous inner surface and a dense outer surface. The slow phase inversion in the air-gap region also makes the nascent fiber stretchable by gravity. As a consequence, fine TbHFs are produced. It is worth noting that the coagulation strength of BuOH is weaker than water [57]; therefore, the inner surface of TbHFs spun from NMP/BuOH has a more porous structure than those spun from NMP/water at the same composition. On the contrary, a faster phase inversion occurs at the lumen side of the fiber when a stronger bore fluid (i.e. NMP/H2O (80/ 20 wt%)) is applied. It not only quickly solidifies the inner dimension of the nascent fiber but also tightens its inner skin structure. As a result, the resultant dry-jet wet-spun TbHF has a large diameter and a dense inner skin. Moreover, since the solidifying rate of the lumen side is faster than that at the shell side, the net mass transfer direction of the solvent has the tendency to move towards the shell side of the fiber, yielding a TbHF substrate with a denser inner surface and a more porous outer surface. As shown in Table 3, the thick TFC TbHF membrane has a much lower flux compared with the fine ones due to its less porous structure (typically at inner surface) and thicker walls as

Table 4 R parameters, TFC layer thicknesses and positron lifetime results of the fine and thick TFC TbHF membranes a. Sample

R1

L1 (nm)

τ3 (ns)

I3 (%)

R (Å)

FFV (%)

Thick TbHF Fine TbHF

0.4194 70.0004 0.4190 70.0003

199 712 101 74

2.234 70.025 1.595 70.037

12.85 70.12 10.53 70.27

3.067 0.02 2.45 7 0.04

2.78 7 0.07 1.177 0.08

a The TbHF substrates for the thick and fine TFC TbHFs were spun from bore fluids of NMP/H2O (80/20 wt%) and NMP/H2O (95/5 wt%), respectively. R1 is the R parameter obtained from VEPFIT; L1 is the thickness of the top selective layer obtained from VEPFIT; τ3 is the o-Ps lifetime obtained from PALS; I3 is the intensity of τ3 obtained from PALS; R is the mean free volume radius obtained from PALS; FFV is the fractional free volume obtained from PALS.

162

D. Hua et al. / Journal of Membrane Science 471 (2014) 155–167

Fig. 8. (a) The cross-section, outer surface as well as inner surface morphologies of TbHF substrates spun from the bore fluid flow rates of 3, 4 and 5 ml/min when using NMP/H2O 95/5 wt% as the bore fluid. Other spinning conditions: air gap: 0.5 cm, dope flow rate: 5 ml/min. (b) The cross-section, outer surface as well as inner surface morphologies of TbHF substrates spun from the bore fluid flow rates of 3, 4 and 5 ml/min when using NMP/BuOH 95/5 wt% as the bore fluid. Other spinning conditions: air gap: 0.5 cm, dope flow rate: 5 ml/min.

D. Hua et al. / Journal of Membrane Science 471 (2014) 155–167

163

Table 5 Effects of bore fluid rate on IPA dehydration performance of TFC TbHF membranesa. Bore fluid composition

Bore rate (ml/min)

Total flux (g m  2 h  1)

Water concentration at the permeate (wt%)

Separation factor (Water/IPA)

NMP/H2O (95/5 wt%)

3 4 5 3 4 5

20757 34 2348 7 29 24757 80 2099 7 54 25157 74 3080 7 112

88.9 7 0.5 95.17 0.2 97.17 0.2 95.4 7 0.3 97.6 7 0.4 98.0 7 0.3

45.4 72.2 110.7 73.8 197.0 714.6 120.7 78.2 233.6739.1 274.0 740.5

NMP/BuOH (95/5 wt%)

a

Other spinning conditions: air gap: 0.5 cm, dope flow rate: 5 ml/min.

Fig. 9. The cross-section, outer surface as well as inner surface morphologies of TbHF substrates spun at the air gap distances of 0.5, 2.5 and 5.0 cm. Other spinning conditions: bore fluid composition: NMP/H2O (95/5 wt%), bore fluid flow rate: 5 ml/min, dope flow rate: 5 ml/min.

Table 6 Effects of air gap on IPA dehydration performance of TFC TbHF membranesa. Air gap (cm)

Total flux (g m  2 h  1)

Water concentration at the permeate (wt%)

Separation factor (Water/IPA)

0.5 2.5 5

2475 780 2647 749 2877 7110

97.1 70.2 97.8 70.2 96.9 70.3

197.0 714.6 260.7 726.9 180 719.4

a Other spinning conditions: bore fluid composition: NMP/H2O (95/5 wt%), bore fluid flow rate: 5 ml/min, dope flow rate: 5 ml/min.

mentioned above. Moreover, the lower separation factor may be ascribed to the relatively large pores at the fiber outer surface, which is unfavorable for forming a uniform and defect-free TFC layer [24]. The difference in TFC layer between thick and fine TbHFs has been further verified by PAS DBES and PALS. Firstly, the R parameter curves of TFC TbHF membranes as a function of positron incident energy were obtained from PAS DBES. By fitting these curves using the VEPFIT program with a three-layer model, R1 and thickness L1 of the selective layer of TFC membranes could be obtained [22]. As presented in Table 4, the TFC selective layer of the fine TFC TbHF is much thinner than that of the thick TFC TbHF

(104 74 vs. 199 712 nm), thus the former has a higher flux than the latter. Since the o-Ps lifetime τ3 and its intensity I3 obtained from PALS analyses correspond to the free volume size and concentration, respectively, the mean free volume radius (R) and fractional free volume (FFV) can be calculated according to established semiempirical correlation equations from the pick-off annihilation [58– 60]. Table 4 shows the calculated results and indicates that the fine TFC TbHF have a smaller R and FFV than the thick TFC TbHF. The smaller R implies that the fine TFC TbHF has a higher rejection against IPA than the thick TFC TbHF. In addition, the smaller FFV does not affect the fine TFC TbHF because it has a thinner selective layer. As a result, the fine TbHF substrates spun from weak bore fluids were more preferred for fabricating TFC membranes with good overall pervaporation performance.

3.2.2. Bore fluid flow rate The influences of bore fluid flow rate on TbHF morphology were investigated and shown in Fig. 8(a and b). An increase in bore fluid flow rate results in an enlargement in outer diameter and inner diameter, as well as a reduction in wall thickness. In addition, the large surface pores on the inner skin become small

164

D. Hua et al. / Journal of Membrane Science 471 (2014) 155–167

Fig. 10. The cross-section, outer surface as well as inner surface morphologies of TbHF substrates spun from dope flow rate of 5, 6 and 7 ml/min. Other spinning conditions: bore fluid composition: NMP/H2O (95/5 wt%), bore fluid flow rate: 5 ml/min , air gap: 2.5 cm.

Table 7 Effects of dope flow rate on IPA dehydration performance of TFC TbHF membranesa. Dope flow rate Total flux (ml/min) (g m  2 h  1)

Water concentration at the Separation factor permeate (wt%) (Water/IPA)

5 6 7

97.8 7 0.2 93.6 7 0.5 90.8 7 0.3

26477 49 23347 81 2081 7 75

260.7 7 26.9 81.8 7 7.5 55.5 7 2.1

a Other spinning conditions: bore fluid composition: NMP/H2O (95/5 wt%), bore fluid flow rate: 5 ml/min, air gap: 2.5 cm.

and scattered at a higher bore fluid flow rate, which may be due to the fact that these large inner surface pores are formed because of non-solvent intrusion from the outer surface [54,56]. This is proven by Fig. 8 where the enlarged outer cross-section and inner surface morphologies change as a function of bore fluid flow rate. The outward-pointed macrovoids (i.e., the tails of macrovoids face to the outer surface [54]) across the fiber walls become smaller at a higher bore fluid flow rate. As a result, a high bore-fluid flow rate would retard the non-solvent intrusion and reduce the pore sizes. Moreover, a higher bore fluid flow rate also causes more severe deformation of the spokes because of slow phase inversion and stress unbalance [61]. The flux and separation factors of fine TFC TbHFs prepared under various bore fluid flow rates are shown in Table 5. A higher bore fluid flow rate leads to an increase in flux, which is consistent with aforementioned observations because of thinner wall and less transport resistance. A higher bore fluid flow rate also enhances the separation factor by radially expanding the nascent fiber, which may result in an increase in hoop elongation and induce molecular orientation along the circumference [62]. In addition, the effects of bore fluid flow rate on pervaporation performance are valid for both TFC TbHFs when using NMP/H2O (95/5 wt%) and NMP/BuOH (95/5 wt%) as bore fluids. Even though the bore fluid of NMP/BuOH (95/5 wt%) produces TFC TbHFs with better separation performance than the NMP/H2O (95/5 wt%), the

former has weaker or more bended spokes than the latter because H2O is a much stronger coagulant than BuOH. As a result, the TbHF substrate spun from the bore fluid of NMP/H2O (95/5 wt%) at a flow rate of 5 ml/min is preferred. 3.2.3. Air gap Fig. 9 shows the effects of air gap on cross-section, outer and inner skin morphologies of the TbHF substrates. The increase in air gap results in TbHFs with a smaller outer diameter and a thinner wall due to the elongational stretch induced by the gravitational force [14,52]. The spokes of TbHF become severely deformed at the air gap of 5 cm possibly due to unbalanced flow stresses and slow phase inversion. Nevertheless, a dense outer layer and a porous inner layer are observed for all the TbHFs. As shown in Table 6, the flux of the TFC TbHF continuously increases at a higher air gap due to the decrease of wall thickness and consequently reduced transport resistance. The up and down trend of separation factor with air gap distance is caused by two competing factors: the elongational stress induced by the gravity may bring about molecular orientation at the skin (i.e., selective) layer and enhance the separation factor, while a large air gap may overstretch the skin and create defects that favor flux but impair the separation factor [14,52,63]. Therefore, the TbHF substrate spun from an air gap distance of 2.5 cm is selected for the subsequent studies considering its separation performance and morphology in maintaining the mechanical integrity of the TbHFs. 3.2.4. Dope flow rate The dope flow rate also influences the TbHF morphology as demonstrated in Fig. 10. The thicknesses of both wall and spokes increase with an increase in dope flow rate. At the optimum air gap of 2.5 cm, the spokes of all TbHFs are able to maintain their geometry. However, a higher dope flow rate results in a less porous inner surface because a thick wall may retard non-solvent intrusion from the outer surface. As a consequence, the flux of the

D. Hua et al. / Journal of Membrane Science 471 (2014) 155–167

165

Table 8 Comparisons of mechanical properties and pervaporation performance of TFC SbHF and TFC TbHF membranesa. Comparisons Geometries Mechanical properties

PV performance

polymeric cross-section area (mm2) Outer diameter (mm) Elongation at break (%) Maximum tensile Stress(MPa) Young's Modulus (MPa) Total flux (g m  2 h  1) Separation factor

SbHF

thick TbHF

fine TbHF

0.33 1.00 42.8 72.8 11.8 70.5 350.5 7 9.0 2356 726 192.2 7 12.1

1.53 2.18 42.7 73.2 14.9 7 0.3 411.4 710.8 10087 56 22 70.6

0.30 1.02 50.0 7 2.9 15.0 70.7 426.2 7 9.1 26477 49 260.7 7 26.9

a Spinning conditions: dope flow rate: 5 mL min  1; bore fluid flow rate: 5 mL min  1; air gap: 2.5 cm; coagulant bath: water; bore fluid composition NMP/H2O (95/5 wt %) for the SbHF and fine TbHF and NMP/H2O (80/20 wt%) for the thick TbHF; take up speed: free fall.

Table 9 Performance benchmark of the fine TFC TbHF with other polymeric membranes for IPA pervaporative dehydration. Membrane

T/oC Feed (IPA wt%)

Flux/ Separation factor g m  2 h  1 (water/IPA)

mole based water permeance (mol m  2 h  1k Pa  1)

mole-based selectivity (water/IPA)

Ref.

Heat treated polyacrylonitrile (PAN) HF P84 co-polyimide hollow fiber Heat treated P84/PES dual layer HF Chitosan/polyacrylonitrile (PAN) composite HF Torlons/P84 blended HF Matrimids HF Torlon 4000 T-MV/Ultem dual layer HF 6FDA-ODA-NDA/Ultems HF Cellulose/polysulfone dual-layer HF MPD-TMC polyamide TFC/torlon HF HPEI-TMC polyamide TFC/ torlon HF HPEI-HGOTMS polyamide TFC/ Ultems HF EDA-TMC polyamide/modified PTFE flat sheet membrane TAEA- TMC polyamide TFC/mPAN EDA-TMC polyamide TFC/ mPAN TFC flat sheet membrane MPD-TMC polyamide TFC /CPA-5 flat sheet membrane with heat treatment HPEI/MPD-TMC polyamide TFC/Ultems1010 TbHF

25 60 60 25 60 80 60 60 25 50 50 50 70

90 85 85 90 85 84 85 85 95 85 85 85 70

186 883 570 145 1000 1800 765 480 4.4 1374 1980 3519 1720

1116 10585 125 2991 185 132 1944 2332 94981 53 349 278 177

5.5 3.4 2.1 4.3 3.8 2.7 3.0 1.8 0.2 7.8 12.3 21.8 3.5

1022 13227 145 2739 216 155 2275 2750 66731 62 407 324 364

[13] [14] [15] [16] [17] [18] [19] [20] [21] [24] [24] [23] [70]

70 25

70 90

340 213

1150 105

0.7 5.8

2394 96

[71] [72]

16

90

140

40

11.0

59

[73]

50

85

2647

261

15.8

294

This work

overstretch of polymer chains when the dope flow rate exceeds a critical value [65–67]. Based on the above studies on TbHFs, the fine TbHF spun from the bore fluid composition of NMP/H2O (95/5 wt%), air gap of 2.5 cm, the dope flow rate of 5 ml/min is the most preferred membrane for IPA dehydration, owing to its excellent separation performance and desirable geometry.

3.3. A comparison of TFC TbHF and TFC SbHF membranes

Fig. 11. Long-term pervaporation performance of the fine TFC TbHF for the dehydration of aqueous IPA (IPA/water 85/15 wt%) at 50 1C.

TFC TbHF decreases at a higher dope flow rate due to the increased wall thickness and less porous inner wall (shown in Table 7). The separation factor also decreases possibly because of higher substructure resistance [64] and slight defects created by the

Table 8 compares the fine and thick TFC TbHF membranes with the conventional TFC SbHF membranes in three aspects including geometries, mechanical properties as well as the pervaporation performance. In terms of geometry, the outer diameters of the fine TbHF and SbHF are similar, which are smaller than that of the thick TbHF. However, the fine TFC TbHF membrane outperforms the other two in all aspects in terms of mechanical strength and IPA dehydration performance. It has the highest tensile strength, Young's Modulus, total flux and separation factor. In contrast, the thick TFC TbHF shows the worst pervaporation performance. Clearly, designing the TbHF substrate to have a proper morphology is essential to ensure the resultant TFC TbHF membranes with superior performance.

166

D. Hua et al. / Journal of Membrane Science 471 (2014) 155–167

3.4. Benchmarking and long-term pervaporation performance of the fine TFC TbHF Table 9 compares the pervaporation performance of the fine TFC TbHF with other pervaporation membranes for IPA dehydration available in literatures. The fabricated fine TFC TbHF shows a high permeance with a reasonable selectivity among the reported data. Moreover, it is worth mentioning that the performance of a membrane with a low permeance but a very high selectivity is reported to easily fall into the pressure-ratio-limited region and thus increases the capital investment due to the requirement of a larger membrane area [68,69]. As a result, the newly fabricated fine TFC TbHF membrane with a high permeance and a moderate selectivity is desirable for industrial applications in comparison to those membranes with low permeance and very high separation factors. Moreover, the long-term performance stability is also an important parameter to evaluate the membrane. Therefore, the dehydration of aqueous IPA using the fine TFC TbHF membrane was carried out and the permeate compositions were continuously monitored for more than 200 h at 50 1C. From the results shown in Fig. 11, a slight decline in flux but constant product purity are observed, evidencing a relatively stable performance of the fine TFC TbHF throughout the monitored period. The slight decline in flux is possibly due to the stabilization of the TFC layer or the TbHF substrates. It will be studied in the future.

4. Conclusions We have molecularly designed TFC TbHF membranes with excellent pervaporation performance for IPA dehydration and superior mechanical strength to traditional SbHF membranes. The optimal TFC TbHF membrane shows a flux of 2.65 kg m  2 h  1 with a separation factor of 246 for water/IPA separation at 50 1C using 85/15 wt% IPA/water as the feed. The following conclusions can be drawn from this study: (1) The interfacial polymerization conditions play important roles in determining the morphology of the TFC layer and its performance for pervaporation dehydration. The pretreatment of HPEI on substrates not only smoothens the substrate surface but also produces the TCF layer with fewer defects. (2) The effects of spinning conditions on the geometry and morphology of TbHFs have been determined. Bore fluid composition plays the most important role. To prepare highperformance TFC TbHFs for pervaporation dehydration of IPA, the preferred conditions to prepare TbHF substrates are as follows: spinning from a dope made of Ultems/ethanol/NMP (23/5/72 wt%) with a bore fluid composition of NMP/H2O (95/ 5 wt%), a bore fluid flow rate of 5 ml/min, an air gap of 2.5 cm, and a dope flow rate of 5 ml/min. (3) In comparison with the TFC SbHF with a similar outer diameter, the fine TFC TbHF shows superior pervaporation performance and mechanical strength. (4) The newly developed fine TFC TbHF shows both satisfactory separation performance and long-term stability. It may have potential for the industrial IPA dehydration application.

Acknowledgments The authors are grateful to National Research Foundation, Prime Minister's Office, Singapore for funding this research under

its Competitive Research Program for the project entitled, “New Biotechnology for Processing Metropolitan Organic Wastes into Value-Added Products” (CRP Award No. NRF-CRP 5–2009-5 (NUS Grant number R-279-000-311-281)). Special thanks to Dr. Guimin Shi, Dr. Jian Zuo, Dr. Zhengzhong Zhou and Mr. Yupan Tang for their valuable suggestions. References [1] L.M. Vane, A review of pervaporation for product recovery from biomass fermentation processes, J. Chem. Technol. Biotechnol. 80 (2005) 603–629. [2] M.L. Gimenes, L. Liu, X. Feng, Sericin/poly(vinyl alcohol) blend membranes for pervaporation separation of ethanol/water mixtures, J. Membr. Sci. 295 (2007) 71–79. [3] Y. Huang, R.W. Baker, J.G. Wijmans, Perfluoro-coated hydrophilic membranes with improved selectivity, Ind. Eng. Chem. Res. 52 (2012) 1141–1149. [4] P.S. Rachipudi, M.Y. Kariduraganavar, A.A. Kittur, A.M. Sajjan, Synthesis and characterization of sulfonated-poly(vinyl alcohol) membranes for the pervaporation dehydration of isopropanol, J. Membr. Sci. 383 (2011) 224–234. [5] G. Liu, D. Yang, Y. Zhu, J. Ma, M. Nie, Z. Jiang, Titanate nanotubes-embedded chitosan nanocomposite membranes with high isopropanol dehydration performance, Chem. Eng. Sci. 66 (2011) 4221–4228. [6] S. Atsumi, J.C. Liao, Metabolic engineering for advanced biofuels production from Escherichia coli, Curr. Opin. Biotechnol. 19 (2008) 414–419. [7] G. Liu, W. Wei, W. Jin, Pervaporation membranes for biobutanol production, ACS Sustain. Chem. Eng. 2 (2014) 546–560. [8] X. Feng, R.Y.M. Huang, Liquid separation by membrane pervaporation: a Review, Ind. Eng. Chem. Res. 36 (1997) 1048–1066. [9] L.Y. Jiang, Y. Wang, T.S. Chung, X.Y. Qiao, J.Y. Lai, Polyimides membranes for pervaporation and biofuels separation, Prog. Polym. Sci. 34 (2009) 1135–1160. [10] M.Y. Teng, K.R. Lee, S.C. Fan, D.J. Liaw, J. Huang, J.Y. Lai, Development of aromatic polyamide membranes for pervaporation and vapor permeation, J. Membr. Sci. 164 (2000) 241–249. [11] N. Peng, N. Widjojo, P. Sukitpaneenit, M.M. Teoh, G.G. Lipscomb, T.S. Chung, J. Y. Lai, Evolution of polymeric hollow fibers as sustainable technologies: past, present, and future, Prog. Polym. Sci. 37 (2012) 1401–1424. [12] Y. Su, G.G. Lipscomb, H. Balasubramanian, D.R. Lloyd, Observations of recirculation in the bore fluid during hollow fiber spinning, AIChE J. 52 (2006) 2072–2078. [13] H.A. Tsai, Y.S. Ciou, C.C. Hu, K.R. Lee, D.G. Yu, J.Y. Lai, Heat-treatment effect on the morphology and pervaporation performances of asymmetric PAN hollow fiber membranes, J. Membr. Sci. 255 (2005) 33–47. [14] R.X. Liu, X.Y. Qiao, T.S. Chung, The development of high performance P84 copolyimide hollow fibers for pervaporation dehydration of isopropanol, Chem. Eng. Sci. 60 (2005) 6674–6686. [15] R.X. Liu, X.Y. Qiao, T.S. Chung, Dual-layer P84/polyethersulfone hollow fibers for pervaporation dehydration of isopropanol, J. Membr. Sci. 294 (2007) 103–114. [16] H.A. Tsai, W.H. Chen, C.Y. Kuo, K.R. Lee, J.Y. Lai, Study on the pervaporation performance and long-term stability of aqueous iso-propanol solution through chitosan/polyacrylonitrile hollow fiber membrane, J. Membr. Sci. 309 (2008) 146–155. [17] M.M. Teoh, T.S. Chung, K.Y. Wang, M.D. Guiver, Exploring Torlon/P84 copolyamide-imide blended hollow fibers and their chemical cross-linking modifications for pervaporation dehydration of isopropanol, Sep. Purif. Technol. 61 (2008) 404–413. [18] L.Y. Jiang, T.S. Chung, R. Rajagopalan, Dehydration of alcohols by pervaporation through polyimide matrimids asymmetric hollow fibers with various modifications, Chem. Eng. Sci. 63 (2008) 204–216. [19] Y. Wang, S.H. Goh, T.S. Chung, P. Na, Polyamide-imide/polyetherimide duallayer hollow fiber membranes for pervaporation dehydration of C1–C4 alcohols, J. Membr. Sci. 326 (2009) 222–233. [20] N. Widjojo, T.S. Chung, Pervaporation dehydration of C2–C4 alcohols by 6FDAODA-NDA/Ultems dual-layer hollow fiber membranes with enhanced separation performance and swelling resistance, Chem. Eng. J. 155 (2009) 736–743. [21] Z. Mao, X. Jie, Y. Cao, L. Wang, M. Li, Q. Yuan, Preparation of dual-layer cellulose/polysulfone hollow fiber membrane and its performance for isopropanol dehydration and CO2 separation, Sep. Purif. Technol. 77 (2011) 179–184. [22] J. Zuo, T.S. Chung, Design and synthesis of a fluoro-silane amine monomer for novel thin film composite membranes to dehydrate ethanol via pervaporation, J. Mater. Chem. A 1 (2013) 9814. [23] J. Zuo, Y. Wang, T.S. Chung, Novel organic–inorganic thin film composite membranes with separation performance surpassing ceramic membranes for isopropanol dehydration, J. Membr. Sci. 433 (2013) 60–71. [24] J. Zuo, Y. Wang, S.P. Sun, T.S. Chung, Molecular design of thin film composite (TFC) hollow fiber membranes for isopropanol dehydration via pervaporation, J. Membr. Sci. 405–406 (2012) 123–133. [25] G.M. Shi, T.S. Chung, Thin film composite membranes on ceramic for pervaporation dehydration of isopropanol, J. Membr. Sci. 448 (2013) 34–43. [26] P. Sukitpaneenit, T.S. Chung, Fabrication and use of hollow fiber thin film composite membranes for ethanol dehydration, J. Membr. Sci. 450 (2014) 124–137.

D. Hua et al. / Journal of Membrane Science 471 (2014) 155–167

[27] S.H. Huang, C.H. Wu, K.R. Lee, J.Y. Lai, Polysulfonamide thin-film composite membranes applied to dehydrate isopropanol by pervaporation, Appl. Mech. Mater. 377 (2013) 222–226. [28] H.A. Tsai, L.H. Chung, K.R. Lee, J.Y. Lai, The preparation of polyamide/ polyacrylonitrile composite hollow fiber membranes for pervaporation, Appl. Mech. Mater. 377 (2013) 246–249. [29] S. Elmore, G. Glenn Lipscomb, Analytical approximations of the effect of a fiber size distribution on the performance of hollow fiber membrane separation devices, J. Membr. Sci. 98 (1995) 49–56. [30] P. Le-Clech, A. Fane, G. Leslie, A. Childress, MBR focus: the operators' perspective, Filtr. Sep. 42 (2005) 20–23. [31] P.S. Goh, A.F. Ismail, S.M. Sanip, B.C. Ng, M. Aziz, Recent advances of inorganic fillers in mixed matrix membrane for gas separation, Sep. Purif. Technol. 81 (2011) 243–264. [32] X.-Y. Song, Y.-P. Shi, J. Chen, A novel extraction technique based on carbon nanotubes reinforced hollow fiber solid/liquid microextraction for the measurement of piroxicam and diclofenac combined with high performance liquid chromatography, Talanta 100 (2012) 153–161. [33] V.M. Magueijo, L.G. Anderson, A.J. Fletcher, S.J. Shilton, Polysulfone mixed matrix gas separation hollow fibre membranes filled with polymer and carbon xerogels, Chem. Eng. Sci. 92 (2013) 13–20. [34] P. Apetel, J.-M. Espenan, Progress for extruding semi-permeable membrane having separated hollow passgeways, US Patent 5171493; 1992. [35] 〈http://www.inge.ag/index_en.php? section=multibore_membrane&pic=produkte〉. [36] 〈http://www.hyfluxmembranes.com/kristal_membrane.html〉. [37] D. Gille, W. Czolkoss, Ultrafiltration with multi-bore membranes as seawater pre-treatment, Desalination 182 (2005) 301–307. [38] K.A. Bu-Rashid, W. Czolkoss, Pilot tests of multibore UF membrane at Addur SWRO desalination plant, Bahrain, Desalination 203 (2007) 229–242. [39] N. Peng, M.M. Teoh, T.S. Chung, L.L. Koo, Novel rectangular membranes with multiple hollow holes for ultrafiltration, J. Membr. Sci. 372 (2011) 20–28. [40] P. Wang, T.S. Chung, Design and fabrication of lotus-root-like multi-bore hollow fiber membrane for direct contact membrane distillation, J. Membr. Sci. 421–422 (2012) 361–374. [41] P. Wang, L. Luo, T.S. Chung, Tri-bore ultra-filtration hollow fiber membranes with a novel triangle-shape outer geometry, J. Membr. Sci. 452 (2014) 212–218. [42] L. Luo, P. Wang, S. Zhang, G. Han, T.S. Chung, Novel thin-film composite tribore hollow fiber membrane fabrication for forward osmosis, J. Membr. Sci. 461 (2014) 28–38. [43] M. Spruck, G. Hoefer, G. Fili, D. Gleinser, A. Ruech, M. Schmidt-Baldassari, M. Rupprich, Preparation and characterization of composite multichannel capillary membranes on the way to nanofiltration, Desalination 314 (2013) 28–33. [44] J.E. Cadotte, Interfacially synthesized reverse osmosis membrane, US Patent 4277344; 1981. [45] R.B. Hodgdon, Using all aliphatic polyamides from polyamino compounds and compounds having two or more acid halide groups, US Patent 5616249; 1997. [46] P.S. Singh, S.V. Joshi, J.J. Trivedi, C.V. Devmurari, A.P. Rao, P.K. Ghosh, Probing the structural variations of thin film composite RO membranes obtained by coating polyamide over polysulfone membranes of different pore dimensions, J. Membr. Sci. 278 (2006) 19–25. [47] A.K. Ghosh, E.M.V. Hoek, Impacts of support membrane structure and chemistry on polyamide-polysulfone interfacial composite membranes, J. Membr. Sci. 336 (2009) 140–148. [48] R.W. Baker, J.G. Wijmans, Y. Huang, Permeability, permeance and selectivity: a preferred way of reporting pervaporation performance data, J. Membr. Sci. 348 (2010) 346–352. [49] M. Duan, Z. Wang, J. Xu, J. Wang, S. Wang, Influence of hexamethyl phosphoramide on polyamide composite reverse osmosis membrane performance, Sep. Purif. Technol. 75 (2010) 145–155. [50] Y. Mansourpanah, K. Alizadeh, S.S. Madaeni, A. Rahimpour, H. Soltani Afarani, Using different surfactants for changing the properties of poly(piperazineamide) TFC nanofiltration membranes, Desalination 271 (2011) 169–177. [51] Y. Cui, X.Y. Liu, T.S. Chung, Enhanced osmotic energy generation from salinity gradients by modifying thin film composite membranes, Chem. Eng. J. 242 (2014) 195–203.

167

[52] T.S. Chung, X. Hu, Effect of air-gap distance on the morphology and thermal properties of polyethersulfone hollow fibers, J. Appl. Polym. Sci. 66 (1997) 1067–1077. [53] H.A. Tsai, C.Y. Kuo, J.H. Lin, D.M. Wang, A. Deratani, C. Pochat-Bohatier, K. R. Lee, J.Y. Lai, Morphology control of polysulfone hollow fiber membranes via water vapor induced phase separation, J. Membr. Sci. 278 (2006) 390–400. [54] N. Widjojo, T.S. Chung, Thickness and air gap dependence of macrovoid evolution in phase-inversion asymmetric hollow fiber membranes, Ind. Eng. Chem. Res. 45 (2006) 7618–7626. [55] J.G. Wijmans, J.P.B. Baaij, C.A. Smolders, The mechanism of formation of microporous or skinned membranes produced by immersion precipitation, J. Membr. Sci. 14 (1983) 263–274. [56] J.J. Qin, T.S. Chung, Effects of orientation relaxation and bore fluid chemistry on morphology and performance of polyethersulfone hollow fibers for gas separation, J. Membr. Sci. 229 (2004) 1–9. [57] F. Tasselli, E. Drioli, Tuning of hollow fiber membrane properties using different bore fluids, J. Membr. Sci. 301 (2007) 11–18. [58] S.J. Tao, Positronium annihilation in molecular substances, J. Chem. Phys. 56 (1972) 5499–5510. [59] M. Eldrup, D. Lightbody, J.N. Sherwood, The temperature dependence of positron lifetimes in solid pivalic acid, Chem. Phys. 63 (1981) 51–58. [60] H. Chen, W.S. Hung, C.H. Lo, S.H. Huang, M.L. Cheng, G. Liu, K.R. Lee, J.Y. Lai, Y. M. Sun, C.C. Hu, R. Suzuki, T. Ohdaira, N. Oshima, Y.C. Jean, Free-volume depth profile of polymeric membranes studied by positron annihilation spectroscopy: layer structure from interfacial polymerization, Macromolecules 40 (2007) 7542–7557. [61] S. Bonyadi, T.S. Chung, W.B. Krantz, Investigation of corrugation phenomenon in the inner contour of hollow fibers during the non-solvent induced phaseseparation process, J. Membr. Sci. 299 (2007) 200–210. [62] Y. Wang, M. Gruender, T.S. Chung, Pervaporation dehydration of ethylene glycol through polybenzimidazole (PBI)-based membranes. 1. Membrane fabrication, J. Membr. Sci. 363 (2010) 149–159. [63] T.S. Chung, The limitations of using Flory-Huggins equation for the states of solutions during asymmetric hollow-fiber formation, J. Membr. Sci. 126 (1997) 19–34. [64] R.Y.M. Huang, X. Feng, Resistance model approach to asymmetric polyetherimide membranes for pervaporation of isopropanol/water mixtures, J. Membr. Sci. 84 (1993) 15–27. [65] T.S. Chung, J.J. Qin, J. Gu, Effect of shear rate within the spinneret on morphology, separation performance and mechanical properties of ultrafiltration polyethersulfone hollow fiber membranes, Chem. Eng. Sci. 55 (2000) 1077–1091. [66] I.D. Sharpe, A.F. Ismail, S.J. Shilton, A study of extrusion shear and forced convection residence time in the spinning of polysulfone hollow fiber membranes for gas separation, Sep. Purif. Technol. 17 (1999) 101–109. [67] A.F. Ismail, M.I. Mustaffar, R.M. Illias, M.S. Abdullah, Effect of dope extrusion rate on morphology and performance of hollow fibers membrane for ultrafiltration, Sep. Purif. Technol. 49 (2006) 10–19. [68] R.W. Baker, Membrane Technology and Applications, England, John Wiley and Sons, 2004. [69] Y. Huang, T.C. Merkel, R.W. Baker, Pressure ratio and its impact on membrane gas separation processes, J. Membr. Sci. 463 (2014) 33–40. [70] Y.L. Liu, C.H. Yu, J.Y. Lai, Poly(tetrafluoroethylene)/polyamide thin-film composite membranes via interfacial polymerization for pervaporation dehydration on an isopropanol aqueous solution, J. Membr. Sci. 315 (2008) 106–115. [71] C.L. Li, S.H. Huang, D.J. Liaw, K.R. Lee, J.Y. Lai, Interfacial polymerized thin-film composite membranes for pervaporation separation of aqueous isopropanol solution, Sep. Purif. Technol. 62 (2008) 694–701. [72] S.H. Huang, G.J. Jiang, D.J. Liaw, C.L. Li, C.C. Hu, K.R. Lee, J.Y. Lai, Effects of the polymerization and pervaporation operating conditions on the dehydration performance of interfacially polymerized thin-film composite membranes, J. Appl. Polym. Sci. 114 (2009) 1511–1522. [73] J. Albo, J. Wang, T. Tsuru, Application of interfacially polymerized polyamide composite membranes to isopropanol dehydration: effect of membrane pretreatment and temperature, J. Membr. Sci. 453 (2014) 384–393.