- Email: [email protected]
Facile synthesis of highly ordered through-micro-porous polyethylene microfiltration membrane via micro-casting
Materials Letters 198 (2017) 124–127
Contents lists available at ScienceDirect
Materials Letters journal homepage: www.elsevier.com/locate/mlblue
Facile synthesis of highly ordered through-micro-porous polyethylene microfiltration membrane via micro-casting Fan Fan, Lanlan Wang, Hongzhong Liu ⇑ State Key Laboratory for Manufacturing System Engineering, Xi’an Jiaotong University, Xi’an 710049, China
a r t i c l e
i n f o
Article history: Received 3 August 2016 Received in revised form 18 March 2017 Accepted 1 April 2017 Available online 4 April 2017 Keywords: Microfiltration Microstructure Micro-casting Micropillar mold cavity Thin film Functional
a b s t r a c t A polyethylene (PE) microfiltration membrane with highly ordered and identical straight through-pore microstructure was facilely synthesized via micro-casting. Initially, through embedding the micropillars of silicon micropillar template into a polydimethylsiloxane (PDMS) layer under a press force, we built a micropillar mold cavity (MPMC) and infiltrated PE melt into the cavity by submerging the MPMC into the melt to perforate the PE thin film. Then after a release of the PDMS layer, the solidified PE membrane was peeled off from the template. We synthesized membranes with different thicknesses from tailored MPMCs constructed under various press forces, and achieved a perfect peeling off without damage belonging to the thickness of 11 ± 0.3 lm due to its small peeling force. Water filtration experiment demonstrated membrane’s functional performance on separation. Ó 2017 Elsevier B.V. All rights reserved.
1. Introduction For water purification, compared with inorganic microfiltration (MF) membranes [1], polymeric membranes [2,3] are more popular due to their cost-effectiveness and tunable properties [4,5]. However, the conventional polymeric MF membranes synthesized from phase separation [6], electrospinning [7,8] and stretching [9] have inherent defects like tortuous, overlapped and irregular pores, rough surface and great thickness, compromising their separation accuracy and throughput. The track-etched MF membranes [10] possess ideal straight through-pores but the inborn irregular distribution of pores and low porosity impede their applications. Herein, we described a facile method to synthesize polymeric MF membranes with highly ordered and uniform straight through-pores (pore size: 2 lm, pitch: 4 lm) via micro-casting. Low-cost thermoplastic polyethylene (PE) was employed as the membrane material due to its excellent mechanical properties and chemical stability. Firstly, a micropillar mold cavity (MPMC) was constructed by embedding the micropillar tips of a silicon micropillar (SMP) template into a polydimethylsiloxane (PDMS) layer under a press force. Afterwards, by submerging the MPMC in PE melt, PE melt was infiltrated into the cavity. Finally, the solidified membrane was peeled off from the template after a release of the PDMS layer. We synthesized membranes with different thicknesses from tailored MPMCs built by various press forces ⇑ Corresponding author. E-mail address: [email protected] (H. Liu). http://dx.doi.org/10.1016/j.matlet.2017.04.004 0167-577X/Ó 2017 Elsevier B.V. All rights reserved.
and studied their peeling effect difference. Water filtration experiment was performed to evaluate the obtained membrane. 2. Experimental 2.1. Synthesis of membrane Initially, we prepared a SMP template (micropillar diameter: 2 lm, height: 12 lm, pitch: 4 lm and square pattern area: 2.5 cm 2.5 cm) by conventional photolithography and micromachining (Fig. S1). In a hot-press machine (WENHUA CHIPTEK, CHINA), for forming a MPMC (Fig. 1a), a press force precisely applied by the piston pushed a fused silica plate (3.5 cm 3.5 cm) onto a SMP template, to embed the micropillar tips into the soft PDMS layer (thickness: 30 lm, cured at 70 °C for 30 s) on the fused silica plate. Then we further cured the PDMS layer (110 °C for 3 min) to promote the MPMC’s firmness. The PE melt (The added solid PE was heated at 145 °C by the heating plate and melted.) gradually submerged the MPMC (Fig. 1b). For removing the air in the melt and the MPMC and adequately infiltrating the melt into the MPMC, we vacuumized the chamber and keep the vacuum state for 20 min. Then following the entire MPMC was exposed by discharging the melt (Fig. 1c), we ceased heating to cool it naturally. During cooling, the pressure applied on the sample was maintained. When cooled to room temperature, the applied pressure was removed. We took out the sample and released the PDMS layer (Fig. 1d). Then the membrane
F. Fan et al. / Materials Letters 198 (2017) 124–127
125
Fig. 1. Schematic illustration of the membrane synthesis.
thickness was measured by a film thickness tester (F20, Filmetrics Inc., USA). Finally, the PE membrane was grabbed by a clamp attached to a force gauge and peeled off (Fig. 1e). To ensure a constant peeling angle, we moved the force gauge and the SMP template along vertical and horizontal direction respectively by two linear motors at a uniform velocity (0.5 mm/s). We applied press forces (5.1 kgf, 6.4 kgf and 7.8 kgf) to tailor the MPMC’s spacing and repeated the process to synthesize the membranes with thickness of 11 ± 0.3 lm, 10 ± 0.3 lm and 9 ± 0.3 lm respectively. The membrane morphologies were analyzed by a scanning electron microscope (HATACHI SU8010). The membrane porosity was estimated by the typical density method:
Porosity ¼ ½1 ðqa =q0 Þ 100% where qa is the membrane’s apparent density, and q0 is the PE density. 2.2. Evaluation of filtration performance In various industrial fields, a large number of tiny matters ranging from 3 lm to 10 lm in solutions are required to be separated from water. Therefore, for comprehensively evaluating the
membrane capability on filtering the matters bigger than 3 lm, we employed the polystyrene (PS) particle solution (0.15 mg/L, average particle size: 3 lm) as surrogate to perform an integrity test. The membranes were tested under various pressures in a homemade filtration device and at each pressure cycle-test was carried out: each cycle consisted of pure water (200 mL) flux test, PS particle solution (200 mL) filtration (flux and retention ratio test) and back flushing by pure water (300 mL) to clean the membrane. Ten samples were tested under the same condition for averaging the experimental results. 3. Results and discussion 3.1. Synthesis of membranes 3.1.1. Formation of perforation The SMP template is shown in Fig. 2a. Fig. 2b exhibits that the micropillars perforate the PE film. The embedment of the micropillar tips inside the PDMS layer leads to a direct formation of perforation as the PE melt fills in the MPMC. The temperature of PE melt does not influence the PDMS layer, fused silica plate and SMP template, since they can endure much higher temperature. After
Fig. 2. SEM images of (a) silicon micropillar (SMP) template, a typical result of (b) perforate the PE film, (c) the membrane during peeling, (d) and (e) the membrane after peeling off.
126
F. Fan et al. / Materials Letters 198 (2017) 124–127
solidification of the PE and withdrawal of press force, the PDMS layer can be facilely released without damaging micropillar owning to its low surface energy, excellent flexibility [11] and the not deep embedment of micropillars in the PDMS layer. 3.1.2. Peeling off the membrane Fig. 2c and Fig. 2d,e reveal the membrane during peeling and after peeling off, respectively. A peeling off without damaging the membrane and the micropillars is crucial. In spite of a low surface
energy [12] of solid PE that does not adhere to the silicon template, the penetration of micropillars wholly through the membrane thickness determines the most vital role of the thickness in the peeling effect. The thickness contrast (9 ± 0.3 lm, 10 ± 0.3 lm and 11 ± 0.3 lm) of the membranes is shown in Fig. 3a–c. Fig. 3d manifests the correlation between the peeling force and the membrane thickness. As the peeling is performed over the micropillar area the peeling forces are maintained at almost constant plateau values
Fig. 3. SEM images of (a)–(c): membrane thickness contrast, (e) micropillar fracture, (f) pore deformation, (g)–(j) obtained optimized membrane from a perfect peeling off. (d) Correlation between the peeling force and the membrane thickness.
Fig. 4. (a) Variation of Flux and retention ratio with applied pressure. (b) Cycle-test result under various applied pressures. SEM images of (c): a typical membrane surface after filtration, (d) after back flushing.
F. Fan et al. / Materials Letters 198 (2017) 124–127
whereas out of this area they sharply decrease to 0 N. The constant plateau values are caused by invariable number of micropillars in each line of the SMP template and a continuous peeling. An enhancement of only 1 lm can result in a great rise of the peeling forces and the increase rate will be enlarged with the thickness increase (Fig. 3d). Correspondingly, it was observed that deformation of partial pores under the thickness of 10 ± 0.3 lm (Fig. 3f) and even fracture of partial micropillars under the thickness of 9 ± 0.3 lm (Fig. 3e) frequently occurred. Nevertheless, under the thickness of 11 ± 0.3 lm, its small peeling force contributes to a flawless peeling off without damage. The obtained membrane, with porosity estimated at 20.3%, possesses highly ordered and identical through-pores (Fig. 3g–j).
127
The tailored MPMCs at various press forces can produce the membranes with different thicknesses. The optimized membrane with thickness of 11 ± 0.3 lm can be perfectly peeled off ascribed to its smaller peeling force, while the peeling of thinner ones cannot avoid damage. Water filtration evaluation test has demonstrated the obtained membrane’s functional performance of selective permeation and anti-fouling. Acknowledgment The authors acknowledge the financial support of National Natural Science Foundation of China (No. 51305337). Appendix A. Supplementary data
3.2. Evaluation of filtration performance From Fig. 4a, following the Hagen-Poiseuille equation [13], with applied pressure increases the fluxes increase almost linearly. Additionally, high retention ratio (>99%) is acquired. To each applied pressure, the membrane structural advantage endows it with high fluxes and the recovery of pure water after back flushing in previous cycle (Fig. 4b) reveals a reversible fouling tendency. A full capture of PS particles (Fig. 4c) can explain the high retention ratio. An effective back flushing (Fig. 4d) can make the membrane’s reversible fouling tendency better understood. 4. Conclusion We propose an approach of micro-casting to facilely synthesize the PE MF membrane with highly ordered and identical straight through-pore (pore size: 2 lm, pitch: 4 lm) microstructure. Through embedding the micropillars (diameter: 2 lm, pitch: 4 lm, height: 12 lm) of SMP template into a PDMS layer under a press force, we can build a MPMC allowing the infiltration of PE melt by submerging the MPMC into the melt to form perforation.
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.matlet.2017.04. 004. References [1] M.C. Almandoz, C.L. Pagliero, N.A. Ochoa, J. Marchese, Ceram. Int. 41 (2015) 5621–5633. [2] S.H. Woo, J.S. Lee, H.H. Lee, J. Park, B.R. Min, ACS Appl. Mater. Interfaces 7 (2015) 16466–16477. [3] Z. He, D.J. Miller, S. Kasemset, L. Wang, D.R. Paul, B.D. Freeman, J. Membr. Sci. 514 (2016) 659–670. [4] P.F. Ren, Y. Fang, L.S. Wan, X.Y. Ye, Z.K. Xu, J. Membr. Sci. 492 (2015) 249–256. [5] G.D. Kang, Y.M. Gao, J. Membr. Sci. 463 (2014) 145–165. [6] H. Minbu, A. Ochiai, T. Kawase, M. Taniguchi, D.R. Lioyd, T. Tanaka, J. Membr. Sci. 479 (2015) 85–94. [7] H.C. Kim, B.G. Choi, J. Noh, K.G. Song, S.H. Lee, S.K. Maeng, Desalination 346 (2014) 107–114. [8] F.E. Ahmed, B.S. Lalia, R. Hashaikeh, Desalination 356 (2015) 15–30. [9] A. Saffar, P.J. Carreau, A. Ajji, M.R. Kamal, Ind. Eng. Chem. Res. 53 (2014) 14014–14021. [10] B.D. Tompkins, J.M. Dennison, E.R. Fisher, J. Membr. Sci. 428 (2013) 576–588. [11] M. Ma, R.M. Hill, Curr. Opin. Colloids Interface 11 (2006) 193–202. [12] D.K. Owens, R.C. Wendt, J. Appl. Polym. Sci. 13 (1969) 1741–1747. [13] B.S. Lalia, V. Kochkodan, R. Hashaikeh, N. Hilal, Desalination 326 (2013) 77–95.
Contents lists available at ScienceDirect
Materials Letters journal homepage: www.elsevier.com/locate/mlblue
Facile synthesis of highly ordered through-micro-porous polyethylene microfiltration membrane via micro-casting Fan Fan, Lanlan Wang, Hongzhong Liu ⇑ State Key Laboratory for Manufacturing System Engineering, Xi’an Jiaotong University, Xi’an 710049, China
a r t i c l e
i n f o
Article history: Received 3 August 2016 Received in revised form 18 March 2017 Accepted 1 April 2017 Available online 4 April 2017 Keywords: Microfiltration Microstructure Micro-casting Micropillar mold cavity Thin film Functional
a b s t r a c t A polyethylene (PE) microfiltration membrane with highly ordered and identical straight through-pore microstructure was facilely synthesized via micro-casting. Initially, through embedding the micropillars of silicon micropillar template into a polydimethylsiloxane (PDMS) layer under a press force, we built a micropillar mold cavity (MPMC) and infiltrated PE melt into the cavity by submerging the MPMC into the melt to perforate the PE thin film. Then after a release of the PDMS layer, the solidified PE membrane was peeled off from the template. We synthesized membranes with different thicknesses from tailored MPMCs constructed under various press forces, and achieved a perfect peeling off without damage belonging to the thickness of 11 ± 0.3 lm due to its small peeling force. Water filtration experiment demonstrated membrane’s functional performance on separation. Ó 2017 Elsevier B.V. All rights reserved.
1. Introduction For water purification, compared with inorganic microfiltration (MF) membranes [1], polymeric membranes [2,3] are more popular due to their cost-effectiveness and tunable properties [4,5]. However, the conventional polymeric MF membranes synthesized from phase separation [6], electrospinning [7,8] and stretching [9] have inherent defects like tortuous, overlapped and irregular pores, rough surface and great thickness, compromising their separation accuracy and throughput. The track-etched MF membranes [10] possess ideal straight through-pores but the inborn irregular distribution of pores and low porosity impede their applications. Herein, we described a facile method to synthesize polymeric MF membranes with highly ordered and uniform straight through-pores (pore size: 2 lm, pitch: 4 lm) via micro-casting. Low-cost thermoplastic polyethylene (PE) was employed as the membrane material due to its excellent mechanical properties and chemical stability. Firstly, a micropillar mold cavity (MPMC) was constructed by embedding the micropillar tips of a silicon micropillar (SMP) template into a polydimethylsiloxane (PDMS) layer under a press force. Afterwards, by submerging the MPMC in PE melt, PE melt was infiltrated into the cavity. Finally, the solidified membrane was peeled off from the template after a release of the PDMS layer. We synthesized membranes with different thicknesses from tailored MPMCs built by various press forces ⇑ Corresponding author. E-mail address: [email protected] (H. Liu). http://dx.doi.org/10.1016/j.matlet.2017.04.004 0167-577X/Ó 2017 Elsevier B.V. All rights reserved.
and studied their peeling effect difference. Water filtration experiment was performed to evaluate the obtained membrane. 2. Experimental 2.1. Synthesis of membrane Initially, we prepared a SMP template (micropillar diameter: 2 lm, height: 12 lm, pitch: 4 lm and square pattern area: 2.5 cm 2.5 cm) by conventional photolithography and micromachining (Fig. S1). In a hot-press machine (WENHUA CHIPTEK, CHINA), for forming a MPMC (Fig. 1a), a press force precisely applied by the piston pushed a fused silica plate (3.5 cm 3.5 cm) onto a SMP template, to embed the micropillar tips into the soft PDMS layer (thickness: 30 lm, cured at 70 °C for 30 s) on the fused silica plate. Then we further cured the PDMS layer (110 °C for 3 min) to promote the MPMC’s firmness. The PE melt (The added solid PE was heated at 145 °C by the heating plate and melted.) gradually submerged the MPMC (Fig. 1b). For removing the air in the melt and the MPMC and adequately infiltrating the melt into the MPMC, we vacuumized the chamber and keep the vacuum state for 20 min. Then following the entire MPMC was exposed by discharging the melt (Fig. 1c), we ceased heating to cool it naturally. During cooling, the pressure applied on the sample was maintained. When cooled to room temperature, the applied pressure was removed. We took out the sample and released the PDMS layer (Fig. 1d). Then the membrane
F. Fan et al. / Materials Letters 198 (2017) 124–127
125
Fig. 1. Schematic illustration of the membrane synthesis.
thickness was measured by a film thickness tester (F20, Filmetrics Inc., USA). Finally, the PE membrane was grabbed by a clamp attached to a force gauge and peeled off (Fig. 1e). To ensure a constant peeling angle, we moved the force gauge and the SMP template along vertical and horizontal direction respectively by two linear motors at a uniform velocity (0.5 mm/s). We applied press forces (5.1 kgf, 6.4 kgf and 7.8 kgf) to tailor the MPMC’s spacing and repeated the process to synthesize the membranes with thickness of 11 ± 0.3 lm, 10 ± 0.3 lm and 9 ± 0.3 lm respectively. The membrane morphologies were analyzed by a scanning electron microscope (HATACHI SU8010). The membrane porosity was estimated by the typical density method:
Porosity ¼ ½1 ðqa =q0 Þ 100% where qa is the membrane’s apparent density, and q0 is the PE density. 2.2. Evaluation of filtration performance In various industrial fields, a large number of tiny matters ranging from 3 lm to 10 lm in solutions are required to be separated from water. Therefore, for comprehensively evaluating the
membrane capability on filtering the matters bigger than 3 lm, we employed the polystyrene (PS) particle solution (0.15 mg/L, average particle size: 3 lm) as surrogate to perform an integrity test. The membranes were tested under various pressures in a homemade filtration device and at each pressure cycle-test was carried out: each cycle consisted of pure water (200 mL) flux test, PS particle solution (200 mL) filtration (flux and retention ratio test) and back flushing by pure water (300 mL) to clean the membrane. Ten samples were tested under the same condition for averaging the experimental results. 3. Results and discussion 3.1. Synthesis of membranes 3.1.1. Formation of perforation The SMP template is shown in Fig. 2a. Fig. 2b exhibits that the micropillars perforate the PE film. The embedment of the micropillar tips inside the PDMS layer leads to a direct formation of perforation as the PE melt fills in the MPMC. The temperature of PE melt does not influence the PDMS layer, fused silica plate and SMP template, since they can endure much higher temperature. After
Fig. 2. SEM images of (a) silicon micropillar (SMP) template, a typical result of (b) perforate the PE film, (c) the membrane during peeling, (d) and (e) the membrane after peeling off.
126
F. Fan et al. / Materials Letters 198 (2017) 124–127
solidification of the PE and withdrawal of press force, the PDMS layer can be facilely released without damaging micropillar owning to its low surface energy, excellent flexibility [11] and the not deep embedment of micropillars in the PDMS layer. 3.1.2. Peeling off the membrane Fig. 2c and Fig. 2d,e reveal the membrane during peeling and after peeling off, respectively. A peeling off without damaging the membrane and the micropillars is crucial. In spite of a low surface
energy [12] of solid PE that does not adhere to the silicon template, the penetration of micropillars wholly through the membrane thickness determines the most vital role of the thickness in the peeling effect. The thickness contrast (9 ± 0.3 lm, 10 ± 0.3 lm and 11 ± 0.3 lm) of the membranes is shown in Fig. 3a–c. Fig. 3d manifests the correlation between the peeling force and the membrane thickness. As the peeling is performed over the micropillar area the peeling forces are maintained at almost constant plateau values
Fig. 3. SEM images of (a)–(c): membrane thickness contrast, (e) micropillar fracture, (f) pore deformation, (g)–(j) obtained optimized membrane from a perfect peeling off. (d) Correlation between the peeling force and the membrane thickness.
Fig. 4. (a) Variation of Flux and retention ratio with applied pressure. (b) Cycle-test result under various applied pressures. SEM images of (c): a typical membrane surface after filtration, (d) after back flushing.
F. Fan et al. / Materials Letters 198 (2017) 124–127
whereas out of this area they sharply decrease to 0 N. The constant plateau values are caused by invariable number of micropillars in each line of the SMP template and a continuous peeling. An enhancement of only 1 lm can result in a great rise of the peeling forces and the increase rate will be enlarged with the thickness increase (Fig. 3d). Correspondingly, it was observed that deformation of partial pores under the thickness of 10 ± 0.3 lm (Fig. 3f) and even fracture of partial micropillars under the thickness of 9 ± 0.3 lm (Fig. 3e) frequently occurred. Nevertheless, under the thickness of 11 ± 0.3 lm, its small peeling force contributes to a flawless peeling off without damage. The obtained membrane, with porosity estimated at 20.3%, possesses highly ordered and identical through-pores (Fig. 3g–j).
127
The tailored MPMCs at various press forces can produce the membranes with different thicknesses. The optimized membrane with thickness of 11 ± 0.3 lm can be perfectly peeled off ascribed to its smaller peeling force, while the peeling of thinner ones cannot avoid damage. Water filtration evaluation test has demonstrated the obtained membrane’s functional performance of selective permeation and anti-fouling. Acknowledgment The authors acknowledge the financial support of National Natural Science Foundation of China (No. 51305337). Appendix A. Supplementary data
3.2. Evaluation of filtration performance From Fig. 4a, following the Hagen-Poiseuille equation [13], with applied pressure increases the fluxes increase almost linearly. Additionally, high retention ratio (>99%) is acquired. To each applied pressure, the membrane structural advantage endows it with high fluxes and the recovery of pure water after back flushing in previous cycle (Fig. 4b) reveals a reversible fouling tendency. A full capture of PS particles (Fig. 4c) can explain the high retention ratio. An effective back flushing (Fig. 4d) can make the membrane’s reversible fouling tendency better understood. 4. Conclusion We propose an approach of micro-casting to facilely synthesize the PE MF membrane with highly ordered and identical straight through-pore (pore size: 2 lm, pitch: 4 lm) microstructure. Through embedding the micropillars (diameter: 2 lm, pitch: 4 lm, height: 12 lm) of SMP template into a PDMS layer under a press force, we can build a MPMC allowing the infiltration of PE melt by submerging the MPMC into the melt to form perforation.
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.matlet.2017.04. 004. References [1] M.C. Almandoz, C.L. Pagliero, N.A. Ochoa, J. Marchese, Ceram. Int. 41 (2015) 5621–5633. [2] S.H. Woo, J.S. Lee, H.H. Lee, J. Park, B.R. Min, ACS Appl. Mater. Interfaces 7 (2015) 16466–16477. [3] Z. He, D.J. Miller, S. Kasemset, L. Wang, D.R. Paul, B.D. Freeman, J. Membr. Sci. 514 (2016) 659–670. [4] P.F. Ren, Y. Fang, L.S. Wan, X.Y. Ye, Z.K. Xu, J. Membr. Sci. 492 (2015) 249–256. [5] G.D. Kang, Y.M. Gao, J. Membr. Sci. 463 (2014) 145–165. [6] H. Minbu, A. Ochiai, T. Kawase, M. Taniguchi, D.R. Lioyd, T. Tanaka, J. Membr. Sci. 479 (2015) 85–94. [7] H.C. Kim, B.G. Choi, J. Noh, K.G. Song, S.H. Lee, S.K. Maeng, Desalination 346 (2014) 107–114. [8] F.E. Ahmed, B.S. Lalia, R. Hashaikeh, Desalination 356 (2015) 15–30. [9] A. Saffar, P.J. Carreau, A. Ajji, M.R. Kamal, Ind. Eng. Chem. Res. 53 (2014) 14014–14021. [10] B.D. Tompkins, J.M. Dennison, E.R. Fisher, J. Membr. Sci. 428 (2013) 576–588. [11] M. Ma, R.M. Hill, Curr. Opin. Colloids Interface 11 (2006) 193–202. [12] D.K. Owens, R.C. Wendt, J. Appl. Polym. Sci. 13 (1969) 1741–1747. [13] B.S. Lalia, V. Kochkodan, R. Hashaikeh, N. Hilal, Desalination 326 (2013) 77–95.