Bendable Zeolite Membranes: Synthesis and Improved Gas

Jun 1, 2015 - †Department of Chemistry and Biochemistry and ‡William G. Lowrie Department of Chemical and Bimolecular Engineering, The Ohio State ...
1 downloads 12 Views 8MB Size
Article pubs.acs.org/Langmuir

Bendable Zeolite Membranes: Synthesis and Improved Gas Separation Performance Bo Wang,† W. S. Winston Ho,‡ Jose D. Figueroa,§ and Prabir K. Dutta*,† †

Department of Chemistry and Biochemistry and ‡William G. Lowrie Department of Chemical and Bimolecular Engineering, The Ohio State University, Columbus, Ohio 43210, United States § National Energy Technology Laboratory, US Department of Energy, 626 Cochran Mill Road, Pittsburgh, Pennsylvania 15236, United States S Supporting Information *

ABSTRACT: Separation and sequestration of CO2 emitted from fossil energy fueled electric generating units and industrial facilities will help in reducing anthropogenic CO2, thereby mitigating its adverse climate change effects. Membrane-based gas separation has the potential to meet the technical challenges of CO2 separation if high selectivity and permeance with low costs for large-scale manufacture are realized. Inorganic zeolite membranes in principle can have selectivity and permeance considerably higher than polymers. This paper presents a strategy for zeolite growth within the pores of a polymer support, with crystallization time of an hour. With a thin coating of 200−300 nm polydimethylsiloxane (PDMS) on the zeolite−polymer composite, transport data for CO2/N2 separation indicate separation factors of 35−45, with CO2 permeance between 1600 and 2200 GPU (1 GPU = 3.35 × 10−10 mol/(m2 s Pa)) using dry synthetic mixtures of CO2 and N2 at 25 °C. The synthesis process results in membranes that are highly reproducible toward transport measurements and exhibit long-term stability (3 days). Most importantly, these membranes because of the zeolite growth within the polymer support, as contrasted to conventional zeolite growth on top of a support, are mechanically flexible.

1. INTRODUCTION Worldwide carbon dioxide emissions in 2012 from consumption of fossil fuels was estimated to be 32.3 gigatons.1 Concentration of atmospheric CO2 has increased from 280 ppm in the preindustrial era to over 400 ppm now and correlated with global climate change.2 Carbon capture and sequestration technology along with renewables, improved energy efficiency, nuclear, and other approaches can mitigate the effects of climate change. There are many competing technologies under consideration for capture of CO2, including amine scrubbing, other solvent absorption, adsorption, oxycombustion, chemical looping, and membranes.3 Even though some of these technologies are being tested at a small pilot scale (0.5−1 MW) for postcombustion applications, and efforts are currently ongoing to do so on a commercial scale, suites of more cost-effective technologies are needed. Membranes are an attractive alternative, since there is no thermal regeneration process involved for gas-to-gas separation. Polymeric membranes have shown promise because of their separation properties and large-scale manufacturability, leading to reduced costs.4 However, polymeric membranes because of their solution-diffusivity mechanisms of gas separation exhibit an upper bound (Robeson plot) in intrinsic performance, with selectivity decreasing with increased permeance.5 Zeolite membranes do not have this limitation. Zeolites are crystalline © 2015 American Chemical Society

microporous aluminosilicates with molecular porosity; for example, the faujasite (FAU) family of zeolites have a threedimensional architecture with 13 Å supercages connected through 7.4 Å windows.6 Calculations have suggested that using FAU, CO2/N2 separation with selectivity exceeding 500 and with CO2 permeability of 10 000 barrer (1 barrer = 3.35 × 10−16 mol m/(m2 s Pa), permeance units are expressed in GPU, with barrer and GPU being the same for a 1 μm membrane) can be achieved.7 Extensive research is being carried out on zeolite membranes,8−11 with development of novel ways to deposit zeolites on supports.12 Growth of zeolite crystals within the pores of alumina supports, a method referred to as pore plugging, and gas separation properties with such materials have also been reported.13−16 Polymer supports are also being used for zeolite membranes.17 The potential benefit of zeolite membranes has not been completely realized because of the difficulty of synthesizing high performance membranes reliably and cost-effectively.8,9 Moreover, the long synthesis times, batch processing and the brittle nature of the membranes result in 1000-fold higher cost of zeolite membranes over polymeric membranes.9 Thus, not surprising, there is only one example of Received: April 9, 2015 Revised: May 30, 2015 Published: June 1, 2015 6894

DOI: 10.1021/acs.langmuir.5b01306 Langmuir 2015, 31, 6894−6901

Article

Langmuir

Zeolite Y Membrane Synthesis. Secondary growth of zeolite membrane used gel composition of 8.3 Na2O:1 Al2O3:6.4 SiO2:483.9 H2O.18,20 First, 2.208 g of Al(OH)3 and 7.29 g NaOH were dissolved in 85.24 g of H2O. Then 13.85 g of Ludox SM-30 was slowly added into the mixture. The gel was sealed in a polypropylene bottle to age for 4 h at room temperature before use. After aging, the opaque gel was moved to a round-bottom flask, brought to reflux at ∼100 °C, and 40 mL of H2O was removed from the gel via distillation in 1 h. The zeolite seeded PES support was locked in a flat Teflon holder and immersed into the flask after water removal was complete and brought again to reflux. Then, 40 mL of H2O was dripped back into the concentrated gel and support over 1 h of reflux, and the synthesis process was completed. After synthesis, the grown membrane was cooled to room temperature and soaked in pure water to remove excess ions. HDA and PDMS Coating. Spin-coating was used to deposit a thin layer of polydimethylsiloxane (PDMS) or hexadecylamine (HDA) on grown zeolite membrane. To prepare the solution for spin-coating, HDA was dissolved in DI water to make 1 mM dispersion followed by pH adjustment to 4 with HCl. After HDA coating, the membranes were dried at room temperature before PDMS coating. Fresh PDMS monomer solution was prepared each time. PDMS monomer stock solution was diluted with heptane, followed by addition of cross-linker and catalyst with the ratio of 100:1:0.5 (PDMS:cross-linker:catalyst). Zeolite membrane samples were taped on a flat spin-coating support before coating. For spin-coating, the solution was dropped on the whole membrane area for 3 s before starting the spinning process. The spinning process involves a period of 5 s at 2000 rpm followed by 4000 rpm for 1 min. After PDMS coating, membrane sample was polymerized at room temperature overnight before use.

a commercial use of zeolite membrane for pervaporation application in dehydration of ethanol.9 Given this background, this paper presents a zeolite− polymer composite membrane concept with the following attributes. Zeolites are synthesized within a porous polymer support, with the polymer acting as a scaffold for growth of interconnected zeolite particles. Synthesis of membranes for CO2/N2 separation with reproducible performance, long-term stability, and, most important, flexibility has been realized. In an earlier study, we noted that zeolite membranes grown on top of flexible polymer supports were susceptible to cracking and loss of transport properties if not maintained in a flat geometry during all stages of membrane fabrication and use.18 The flexibility of the membrane being reported here without loss in performance and its rapid growth, within an hour, makes it feasible to employ conventional polymer membrane manufacturing techniques for large-scale fabrication of zeolite−polymer composite membranes.

2. EXPERIMENTAL SECTION Chemical. Ludox SM-30 colloidal silica (SiO2, 30%), Ludox HS-30 colloidal silica (SiO2, 30%), aluminum isopropoxide (Al(O−CH(CH3)2)3, 98%), tetramethylammonium bromide ((CH3CH2CH2)4N(2Br), 98%), and hexadecylamine (HDA, CH3(CH2)15NH2) were purchased from Aldrich (Milwaukee, WI). Aluminum hydroxide (Al(OH)3, 76.5%) was purchased from Alfa Aesar. Sodium hydroxide pellet (NaOH, 99.0%) was purchased from Fisher Scientific. Tetramethylammonium hydroxide ((TMAOH), 25% aqueous) was purchased from SACHEM Inc. Polydimethylsiloxane (PDMS) was provided by Wacker Silicones, Inc. (Dehesive 944). Helium (4.5 grade), carbon dioxide (4.0 grade), and nitrogen (4.5 grade) were purchased from Praxair. All chemicals were used as received. Poly(ether sulfone) (PES) 300 kDa membrane was purchased from Millipore Biomax. H2O used in this study was purified by a Millipore ultrapure water system. Nanocrystalline Zeolite Y Seeds. The synthesis solution composition of nanozeolite Y was 0.048 Na2O:2.40 (TMA)2O(2OH):1.2 (TMA)2O(2Br):4.35 SiO2:1.0 Al2O3:249 H2O, where TMA+ is tetramethylammonium cations.19 The silicon source, 26.2 g of Ludox HS-30, and 10.46 g of TMAOH were mixed, sealed with parafilm, and stirred at room temperature for 30 min before use. The alumina source, 12.5 g of aluminum isopropoxide, was dissolved in mixture of 76.5 g of H2O and 52.3 g of TMAOH solution by heating in a water bath at 70 °C. After cooling down to room temperature, 13.1 g of TMABr was added to alumina source solution followed by mixing with the silicon source. The clear sol was aged at room temperature with stirring for 3 days, followed by heating at 100 °C in oil bath with stirring for 4 days. In the synthesis process, nanozeolite remained suspended in solution. Nanozeolite Y particles were captured by ultracentrifugation using Thermo Scientific Sorvall MX-150 from product mixture and washed until pH of supernatant was 7. Then nanozeolite particles were sodium ion exchanged by stirring the zeolite suspension in 0.1 M NaCl solution overnight. The ion exchanged product was washed with DI water and stored as 1 wt % aqueous stock solution. Zeolite Y Seed Deposition. Nanozeolite deposition on PES support were prepared with vacuum assisted dip-coating.18 PES supports were soaked in distilled water overnight followed by in isopropanol for 1 h and in water for 1 h before use. Before dip-coating, nanozeolite stock solution was ultrasonicated for 1 h and then diluted with distilled water to the required concentration. In ultrasonication process, water in sonicator needs to be changed every 20 min to keep temperature low. 20 mL of prepared nanozeolite suspension was moved to a crystal dish before use. In dip-coating process, PES support surface was soaked in nanozeolite seed dispersion for 3 s. After coating, the support was dried at room temperature overnight and stored in plastic sample bags.

3. RESULTS Zeolite Membrane Synthesis. Figure 1 details the procedure, which involved deposition of nanozeolite seeds

Figure 1. Synthesis procedure of bendable zeolite membrane: (a) reflux apparatus for zeolite membrane synthesis; (b) zeolite membrane fabrication procedure.

onto the poly(ether sulfone) (PES) support (dimension 4 cm × 2 cm). The seeded PES was then placed in a gel 8.5 Na2O:1 Al2O3:10.9 SiO2:487 H2O from which water (40% v/v) had been removed via distillation in a reflux apparatus (shown in Figure 1a). Membrane growth occurred in the dehydrated gel while water was added back with the system under reflux, and completed in 60 min, as outlined in Figure 1b.18,20 Nanozeolite Seed Loading Optimization. Figure 2 shows the details of the PES support. As seen from the cross-section SEM of Figure 2a, the support is made up of a 100 μm PES layer on a thicker nonwoven fabric support. X-ray diffraction shows that the PES layer is amorphous (data not 6895

DOI: 10.1021/acs.langmuir.5b01306 Langmuir 2015, 31, 6894−6901

Article

Langmuir

Figure 3. SEM micrographs of polymer support coated with (a) 5, (b) 10, and (c) 50 μg/mL nanozeolite seed dispersions and their corresponding zeolite membrane after secondary growth following procedure outlined in Figure 1b (a → d; b → e; c → f).

nanozeolite crystals are found within the pores of the PES support. These seeded supports were put through the procedure outlined in Figure 1b. The morphology of the membranes formed are shown in Figure 3d−f. With nanozeolite dispersion of 50 μg/mL (Figure 3f), a continuous zeolite membrane layer on top of the support was noted. Only isolated zeolite crystals were observed on zeolite membrane grown from 10 μg/mL zeolite dispersion coated PES support (Figure 3e). With zeolite dispersion concentration of 5 μg/mL, no zeolite crystals were observed on the support after secondary growth (Figure 3d). We have primarily focused on the membranes made with the 10 μg/mL nanozeolite dispersion. In an earlier study,18 we have reported on a membrane grown with a 900 μg/mL nanozeolite dispersion, which resulted in a 200−300 nm continuous zeolite layer on top of the PES support, comparable to the morphology shown with the 50 μg/mL loaded sample (Figure 3f). Characteristics of Zeolite Membrane Prepared with 10 μg/mL Nanozeolite Seed. The scanning electron micrograph of the sample grown with 10 μg/mL nanozeolite coating (Figure 3e) shows that after zeolite growth there were isolated crystals on the PES support, with the porous structure of PES evident. Figure 4 presents a more detailed view of this sample. Figure 4a is a STEM view of a cross-sectional FIB cut of the membrane and indicates growth of a material within the membrane. Figures 4b is the S and Si elemental maps, indicating that a silicon containing species has grown within the pores of the PES (S map is indicative of PES). Figure 4c is Xray diffraction pattern of the membrane, and peaks marked with an asterisk are indicative of growth of FAU zeolite (the rest of the peaks arise from the nonwoven fabric backing, as seen in Figure 2b). After removing the nonwoven fabric backing of the support with a sharp blade, the PES was dissolved by Nmethylpyrrolidone, leaving behind a white film. Figure S2 shows photographs of the membrane before and after dissolution (sample without the nonwoven fabric), and a film with the same geometrical features as the original PES sample is observed, suggesting that a continuous layer was present within the PES. Figure 4d is a top-view SEM image of the continuous inorganic layer, which on closer inspection is a collection of crystals (Figure 4e). The side view of this layer is shown in Figure 4f along with higher magnification image in Figure 4g and a TEM image in Figure 4h. A network of submicron zeolite crystals has grown through the entire PES network, with a three-dimensional interconnectivity between the zeolite crystallites.

Figure 2. (a) Side-view SEM of PES support. (b) XRD of PES support (including PES layer and nonwoven fabric). (c) Top-view SEM of PES support. (d) TEM of nanozeolite seed particles.

shown), but the backing gives rise to a crystalline structure, as seen in Figure 2b. Figure 2c is a top-view SEM of the support, with pore size of ∼70 nm and a porosity of 15% (provided by the manufacturer). Figure 2d is a high-resolution TEM of the nanozeolite seeds that were used for seeding. With a range of crystallite size around 30−40 nm, these seeds are expected to penetrate inside the porous PES support. Nanozeolite dispersions with concentrations of 5, 10, and 50 μg/mL were used to coat the PES and SEM of these surfaces are shown in Figure 3a−c, respectively. No zeolite crystals could be observed on top of PES support coated with 5 μg/mL nanozeolite dispersion. As zeolite dispersion concentration increases, more nanozeolite crystals are evident on the PES support. Figure S1 shows a higher resolution SEM image of the PES support for the 10 μg/mL seeded sample, in which the 6896

DOI: 10.1021/acs.langmuir.5b01306 Langmuir 2015, 31, 6894−6901

Article

Langmuir

Figure 4. Structural characterization of membrane made with 10 μg/mL nanozeolite seed loading and the procedure outlined in Figure 1b: (a) side view STEM of a cross-sectional FIB cut of membrane; (b) S and Si elemental maps on a cross-sectional FIB cut of membrane; (c) XRD of membrane; (d) low-magnification and (e) high-magnification top-view SEM of zeolite layer with PES dissolved; (f) side-view SEM of zeolite layer with PES dissolved; (g) high-magnification side-view SEM of zeolite layer close to surface with PES dissolved; (h) HRTEM showing the connectivity of crystals that are grown within the polymer matrix.

the sample (ZYM/PES) shows poor separation properties with CO2/N2 selectivity of 1 and CO2 permeance of 2636 GPU, indicating the presence of defects between the zeolite crystallites. A thin silicone cover layer deposited with 2 wt % PDMS does seal these defects, with CO2 permeance of 1728 GPU and CO2/N2 selectivity of 38 (2%PDMS/ZYM/PES). Thicker PDMS layer (using 3.5 wt % PDMS) lowers gas permeance of membrane, and membranes coated with a 1 wt % PDMS do not cover all the defects. Thus, all further studies were done with membranes prepared with the 2 wt % PDMS coating. The thickness of the 2 wt % PDMS layer is of the order of ∼200−300 nm, as shown in Figure 6a. Separation characteristics of the membranes were evaluated for ten independently prepared membranes. Figure 6b is a plot of the transport data (black squares). The average permeance of CO2 was 1838 ± 159 GPU with a CO2/N2 selectivity of 38 ± 3. Treating the zeolite−PES membrane with n-hexadecylamine (HDA) followed by PDMS treatment improved the CO2/N2 separation factor to 52 ± 3 (blue diamonds in Figure 6b). In order to test the mechanical stability, the membranes were bent around a radius of 1.5 in., as shown in Figure 6c, without sacrificing the transport properties (red circles in Figure 6b, average CO2 permeance of 1714 ± 107 GPU and CO2/N2 selectivity of 37 ± 1). However, upon bending around a curvature of 1 in., the CO2/N2 selectivity dropped to 1. Figure S3 shows the optical micrographs of the pristine sample and the

High-resolution diffraction of the internal region within the polymer supports, shown in Figure 5, indicates crystalline content in the zeolitic layer, as evidenced by the diffraction spots, whereas the underlying PES layer was amorphous.

Figure 5. Transmission electron microscopy and diffraction patterns of side view of the membrane: Cross-section TEM of bendable zeolite membrane with diffraction patterns collected at different positions of membrane (top picture shows the diffractions spots as evidence of crystallinity; bottom picture is from the PES support).

CO2/N2 Separation Performance of Zeolite Membrane. The above-synthesized membranes were used for CO2/N2 separation, and Table 1 compares the transport properties at various stages of the synthesis. All transport data were taken after membrane performance reached a steady state, and this time varied with the different samples. As synthesized,

Table 1. Gas Transport Measurement Results of Zeolite Membrane Samplesa sample name (hours)

CO2 permeance (GPU)

N2 permeance (GPU)

total gas permeance (GPU)

CO2/N2 selectivity

PES (2 h) 2%PDMS/PES (3 h) ZYM/PES (2 h) HDA/ZYM/PES (2 h) 1%PDMS/ZYM/PES (3 h) 2%PDMS/ZYM/PES (10 h) 3.5% PDMS/ZYM/PES (10 h) 2%PDMS/HDA/ZYM/PES (8 h)

3675b 939 2636 2048 1316 1728 602 1590

3118b 67 2198 1864 924 46 21 31

6794b 1006 4833 3912 2239 1774 623 1620

1b 14 1 1 1 38 29 52

All data with as-prepared samples with no bending. Gas transport experiments are performed at 25 °C with feed gas composition of dry 20% CO2 + 80% N2. “Hours” indicates the time at which the data reported in the table were recorded. bData reported in ref 18. a

6897

DOI: 10.1021/acs.langmuir.5b01306 Langmuir 2015, 31, 6894−6901

Article

Langmuir

curvature, large pieces are observed. Upon bending around 1 in. curvature, there is extensive fragmentation of the in-grown zeolite layer. Figure 6c shows the long-term transport behavior of one of the membranes. As synthesized, and after PDMS coating, the permeance is low (300 GPU) and increases to reach 1700 GPU after 10 h, followed by a downward trend after 40 h with permeance reaching 1400 GPU at 68 h. The CO2/N2 selectivity remains stable after 10 h at 35, with a slight increase to 37 after 40 h. Transport results under different feed gas compositions and temperatures are shown in Figure 8. As shown in Figure 8a, CO2/N2 selectivity increased with CO2 mole fraction, as did CO2 permeance. The dependence of the transport properties on temperature is shown in Figure 8b, with both CO2 and N2 permeance increasing with temperature, while CO2 /N 2 selectivity decreased.

4. DISCUSSION In this paper, our strategy has been to grow the zeolite layer within a porous polymer support. This approach confers on the zeolite polymer composite several novel properties, including rapid synthesis (∼1 h), reproducible gas separation properties, and bendability. From the rapid synthesis perspective, several issues are relevant. The choice of the porosity of the PES support is critical so that nanozeolite seeds can be lodged within the pores. Optimal results were obtained within ∼30−40 nm zeolite seeds within the 70 nm pores of PES (Figure 3b and Figure S1) . These optimally seeded supports grow within the polymer support within 1 h by using a highly nucleated gel as reactant.20 After physically removing the nonwoven fabric backing, and further dissolution of the PES, a monolithic selfstanding 2 μm thick continuous zeolite membrane (Figure 4d,e and Figure S2) was obtained and consists of interconnected zeolite crystals (Figure 4f−h) that have grown with the porous PES layer acting as the template. This zeolite layer is defective with no CO 2/N 2 selectivity (ZYM/PES in Table 1). Intercrystalline defects has been widely reported to exist between micron-sized crystals in zeolite membranes, resulting in decreased membrane selectivity.21 Longer crystallization time,22 multiple growth,23 modification of charge of zeolite surface by additives, as well as modification of support,24,25 and other strategies for defect patching26−28 have been reported to successfully heal defects. In the polymer membrane literature, one of the most effective methods for sealing defects in the skin of the membrane is by coating with a thin layer of highly permeable silicone rubber. The transport properties of such coated

Figure 6. Gas separation performance of membrane: (a) side-view SEM of PDMS coated membrane; (b) gas transport measurement results of (■) membrane, as-synthesized (●) after bending of membrane around 1.5 in. radius, and (◆) membrane coated with HDA; (c) picture of membrane bending process; (d) long-term (●) CO2 permeance and (■) CO2/N2 selectivity of one of the membranes.

samples bent around 1.5 and 1 in. There appears to be striations observed with the sample bent around 1 in. Figure 7 compares the SEM of the pristine sample and samples bent around 1.5 and 1 in. curvature, all after PES dissolution. With the pristine sample, and the sample bent around 1.5 in.

Figure 7. SEM micrographs of the dissolved part of the membranes: (a) as-synthesized, and after bending around different radii of curvature; (b) bending around 1.5 in. curvature and (c) 1 in. curvature. 6898

DOI: 10.1021/acs.langmuir.5b01306 Langmuir 2015, 31, 6894−6901

Article

Langmuir

The permeance and CO2/N2 selectively change with feed gas as shown in Figure 8, suggesting that gas separation is occurring via the zeolite framework. The mechanism of CO2 selectivity over N2 stems from the higher quadrupole moment of CO2,33 which facilitates interaction with the zeolite framework, and its surface diffusion transport, whereas the N2 transport through the zeolite is blocked due to pore restriction.34 Increase of gas permeance with higher temperatures as shown in Figure 8 is because of the higher surface diffusivity, leading to a loss in overall selectivity since N2 permeance outpaces that of CO2. As for the long-term transport measurements (Figure 6d), our hypothesis is that the initial increase in permeance is due to the removal of trapped water at the PDMS−zeolite interface. Zeolites were used “as prepared” and are saturated with water. Removal of this water is essential for the gases to access the interior of the zeolite. It is unclear why the permeance drops, but this phenomenon was noted with zeolite membranes on alumina supports.34 The transport properties of these membranes are remarkably reproducible, even after bending the membrane around a 1.5 in. radius of curvature (Figure 6b). Such an observation is unprecedented in the zeolite membrane area.8,9 The performance is better than current polymer membranes with CO2 permeance of 1000 GPU and CO2/N2 selectivity of 50 at 30 °C.4 If zeolite membranes are grown on the top of the polymer, then with bending around 1.5 in. curvature the CO2/N2 selectivity drops from 72 to 2.18 With the current membranes, the limit of bending is exceeded with 1 in. curvature, since doing so drops the CO2/N2 selectivity from 38 to 1, and comes from the extensive cracking of the inorganic layer (Figure S3 and Figure 7). After dissolving the PES, the inorganic layer is fragile. Cracks observed in the SEM of Figure 7a,b of the pristine sample and, after bending around 1.5 in. radius, are due to the fragility of the membrane and develop during sample preparation. These cracks are distinct from sample bent around the 1 in. curvature, where the layer is considerably more fragmented with rough edges. Our hypothesis for the stability of the zeolite polymer composite to bending is that the internal polymer layer running through the entire zeolite matrix accommodates the tensile forces on the zeolite layer that are developed upon bending. A parallel is drawn with the considerable improvement in bendability of concrete with internal reinforcing by steel or polymer fibers35,36 and flexibility of silicon on plastic substrates.37 However, there is a limit to the bending, as observed with the samples bent around the 1 in. radius curvature. Both the flexibility and the rapid (1 h) growth of the zeolite− PES composite membrane provide the opportunity to adapt the rapid roll-to-roll method used for polymer membrane manufacture to the present composite membrane. One visualizes a process where a seeded PES support is rolled through gel, with a residence time of an hour, resulting in the grown zeolite−PES composite that is washed and coated with a PDMS layer. The flexibility of the membranes coupled with the rapid synthesis time and roll-to-roll fabrication will significantly lower the costs of manufacture as compared to conventional zeolite membranes that are presently fabricated in a batch process. There have been extensive studies of mixed matrix membranes (MMM), in which zeolite crystallites are dispersed in polymers with different structures.38 MMM dual-layer hollow fiber arrangement with zeolite beta exhibited a novel particle distribution, but the skin was still defective, with poor transport

Figure 8. Effect of feed gas composition and temperature on membranes. Gas separation performance of membrane with (a) different feed gas compositions and (b) different temperatures.

membranes are characteristic of the polymer.29,30 Sealing of nanodefects in silica and zeolite membranes with PDMS has been reported.31 In this study, a ∼200−300 nm PDMS coating was essential for satisfactory transport measurements (Table 1). We propose that PDMS fills the intercrystalline defects between the interconnected zeolite crystals. It is unlikely that PDMS layer is responsible for the improved CO2/N2 selectivity to 38, since with PDMS/PES alone, the CO2/N2 selectivity is of the order of 14 and comparable to previous literature reports.32 The possibility that PDMS penetrates into the zeolite pores is unlikely, since the size of PDMS is estimated to be 0.8 nm,31 higher than the 0.74 zeolite pore opening. So, the PDMS is filling up the defects between the crystals. The transport results with zeolite coated with HDA (hexadecylamine) support this hypothesis. HDA treatment will lead to amino group interacting with the zeolite surface (electrostatic), and the hexadecyl chain should be pointing away from the zeolite, thereby making the interface hydrophobic. Such an interface will facilitate PDMS wetting of the zeolite surface, and we propose that this leads to void elimination and improved gas transport, resulting in higher CO2/N2 selectivity (38 to 52, sample 2%PDMS/HDA/ZYM/PES in Table 1), as shown in Figure 6b. 6899

DOI: 10.1021/acs.langmuir.5b01306 Langmuir 2015, 31, 6894−6901

Article

Langmuir properties.39 Ideal MMM exhibit transport by gas molecules passing through adjacent zeolite crystallites that set up a percolation path. Under these circumstances, high loading of zeolite crystals in the polymer should promote better zeolite packing. However, even though increasing zeolite loading in the polymer is simple in practice, reduced transport results are obtained since considerable voids exist between zeolite crystallites that are inaccessible to the polymer.38−40 The bendable zeolite membranes developed in this study are different from MMM because the zeolite crystallites in the bendable membrane are grown into a three-dimensional network, aided by the scaffolding provided by the PES. Interconnection of zeolite crystals significantly improved gas transport through zeolite frameworks. The closest parallel between MMM and the present study is a membrane sample made from PDMS coated on a seeded PES support. We reported that with 3.5 wt % PDMS coating on a 200 nm thick nanozeolite seed coating (prepared from a 900 μg/mL seed solution) the CO2/N2 selectivity was 15.5,18 significantly lower than the CO2/N2 selectivity of 38 being reported in this paper. Our current hypothesis is that the PDMS is not playing a role in the separation and merely blocking defects.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The U.S. Department of Energy under Award DE-FE0007632 supported this work. However, any opinions, findings, conclusions, or recommendations expressed herein are those of the author(s) and do not necessarily reflect the views of the DOE. The authors thank Hendrik Colijn for his help collecting TEM images. We also thank chemistry department machine shop, electric shop, and glass shop at The Ohio State University for their help.



(1) International Energy Statistics, EIA; http://www.eia.gov/cfapps/ ipdbproject/iedindex3.cfm?tid=90&pid=44&aid=8 (accessed Mar 16, 2015). (2) US Department of Commerce, N. ESRL Global Monitoring Division, Global Greenhouse Gas Reference Network; http://www. esrl.noaa.gov/gmd/ccgg/trends/index.html (accessed Mar 16, 2015). (3) Figueroa, J. D.; Fout, T.; Plasynski, S.; McIlvried, H.; Srivastava, R. D. Advances in CO2 Capture technologyThe U.S. Department of Energy’s Carbon Sequestration Program. Int. J. Greenhouse Gas Control 2008, 2 (1), 9−20. (4) Merkel, T. C.; Lin, H.; Wei, X.; Baker, R. Power Plant PostCombustion Carbon Dioxide Capture: An Opportunity for Membranes. J. Membr. Sci. 2010, 359 (1−2), 126−139. (5) Robeson, L. M. The Upper Bound Revisited. J. Membr. Sci. 2008, 320 (1−2), 390−400. (6) Auerbach, S. M.; Carrado, K. A.; Dutta, P. K. Handbook of Zeolite Science and Technology. In Handbook of Zeolite Science and Technology; Marcel Dekker, Inc.: New York, 2003. (7) Krishna, R.; van Baten, J. M. In Silico Screening of Zeolite Membranes for CO2 Capture. J. Membr. Sci. 2010, 360 (1−2), 323− 333. (8) Caro, J.; Noack, M.; Kölsch, P.; Schäfer, R. Zeolite Membranes − State of Their Development and Perspective. Microporous Mesoporous Mater. 2000, 38 (1), 3−24. (9) Gascon, J.; Kapteijn, F.; Zornoza, B.; Sebastián, V.; Casado, C.; Coronas, J. Practical Approach to Zeolitic Membranes and Coatings: State of the Art, Opportunities, Barriers, and Future Perspectives. Chem. Mater. 2012, 24 (15), 2829−2844. (10) Li, S.; Tuan, V. A.; Falconer, J. L.; Noble, R. D. Effects of Zeolite Membrane Structure on the Separation of 1,3-Propanediol from Glycerol and Glucose by Pervaporation. Chem. Mater. 2001, 13 (5), 1865−1873. (11) López, F.; Bernal, M. P.; Mallada, R.; Coronas, J.; Santamaría, J. Preparation of Silicalite Membranes on Stainless Steel Grid Supports. Ind. Eng. Chem. Res. 2005, 44 (20), 7627−7632. (12) Varoon, K.; Zhang, X.; Elyassi, B.; Brewer, D. D.; Gettel, M.; Kumar, S.; Lee, J. A.; Maheshwari, S.; Mittal, A.; Sung, C.-Y.; et al. Dispersible Exfoliated Zeolite Nanosheets and Their Application as a Selective Membrane. Science 2011, 334 (6052), 72−75. (13) Miachon, S.; Landrivon, E.; Aouine, M.; Sun, Y.; Kumakiri, I.; Li, Y.; Prokopová, O. P.; Guilhaume, N.; Giroir-Fendler, A.; Mozzanega, H.; et al. Nanocomposite MFI-Alumina Membranes via Pore-Plugging Synthesis: Preparation and Morphological Characterisation. J. Membr. Sci. 2006, 281 (1−2), 228−238. (14) Guillou, F.; Rouleau, L.; Pirngruber, G.; Valtchev, V. Synthesis of FAU-Type Zeolite Membrane: An Original in Situ Process Focusing on the Rheological Control of Gel-like Precursor Species. Microporous Mesoporous Mater. 2009, 119 (1−3), 1−8. (15) Deng, Z.; Nicolas, C.-H.; Daramola, M. O.; Sublet, J.; Schiestel, T.; Burger, A. J.; Guo, Y.; Giroir-Fendler, A.; Pera-Titus, M. Nanocomposite MFI-Alumina Hollow Fibre Membranes Prepared via Pore-Plugging Synthesis: Influence of the Porous Structure of

5. CONCLUSION A 2 μm thick film of interconnected zeolite crystals is grown within the pores of a poly(ether sulfone) support by appropriate choice of a dilute seeding solution, followed by a rapid zeolite growth process of an hour, and shows excellent CO2/N2 separation properties. Examination of ten separate membranes showed average permeance of CO2 was 1838 ± 159 GPU with a CO2/N2 selectivity of 38 ± 3. Most importantly, these membranes could be bent around a radius of 1.5 cm curvature without any loss in the transport properties. We have reported in a previous study that zeolite membranes made on top of PES support were susceptible to cracking upon bending with loss in transport properties. The rapid zeolite growth as well as the flexibility of the membranes reported in the current study opens up new ways of continuous manufacture of membranes that exploit the unique porosity of zeolites for gas separation. These membranes have the potential to be fabricated into spiral-wound modules with high surface area and significantly lower cost. In addition, this innovative flexible inorganic membrane platform is compatible with other microporous structures.



ASSOCIATED CONTENT

S Supporting Information *

High-resolution SEM images of the seed crystals within the PES pores, optical micrographs of the dissolution process, and of the cracking upon bending. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.5b01306.



REFERENCES

AUTHOR INFORMATION

Corresponding Author

*(P.K.D.) E-mail [email protected]. Author Contributions

This manuscript is based on the PhD thesis of Bo Wang, who did the bulk of the work. The paper was written through contributions of all authors. All authors have given approval to the final version of the manuscript. 6900

DOI: 10.1021/acs.langmuir.5b01306 Langmuir 2015, 31, 6894−6901

Article

Langmuir Hollow Fibres on the Gas/Vapour Separation Performance. J. Membr. Sci. 2010, 364 (1−2), 1−8. (16) Tawalbeh, M.; Tezel, F. H.; Kruczek, B.; Letaief, S.; Detellier, C. Synthesis and Characterization of Silicalite-1 Membrane Prepared on a Novel Support by the Pore Plugging Method. J. Porous Mater. 2013, 20 (6), 1407−1421. (17) Zhan, Z.; Shao, J.; Peng, Y.; Wang, Z.; Yan, Y. High Performance Zeolite NaA Membranes Synthesized on the Inner Surface of zeolite/PES−PI Blend Composite Hollow Fibers. J. Membr. Sci. 2014, 471, 299−307. (18) Wang, B.; Sun, C.; Li, Y.; Zhao, L.; Ho, W. S. W.; Dutta, P. K. Rapid Synthesis of Faujasite/polyethersulfone Composite Membrane and Application for CO2/N2 Separation. Microporous Mesoporous Mater. 2015, 208, 72−82. (19) Holmberg, B. A.; Wang, H.; Norbeck, J. M.; Yan, Y. Controlling Size and Yield of Zeolite Y Nanocrystals Using Tetramethylammonium Bromide. Microporous Mesoporous Mater. 2003, 59 (1), 13−28. (20) Severance, M.; Wang, B.; Ramasubramanian, K.; Zhao, L.; Ho, W. S. W.; Dutta, P. K. Rapid Crystallization of Faujasitic Zeolites: Mechanism and Application to Zeolite Membrane Growth on Polymer Supports. Langmuir 2014, 30 (23), 6929−6937. (21) Caro, J.; Albrecht, D.; Noack, M. Why Is It so Extremely Difficult to Prepare Shape-Selective Al-Rich Zeolite Membranes like LTA and FAU for Gas Separation? Sep. Purif. Technol. 2009, 66 (1), 143−147. (22) Gu, X.; Dong, J.; Nenoff, T. M. Synthesis of Defect-Free FAUType Zeolite Membranes and Separation for Dry and Moist CO2/N2 Mixtures. Ind. Eng. Chem. Res. 2005, 44 (4), 937−944. (23) Zhu, G.; Li, Y.; Zhou, H.; Liu, J.; Yang, W. FAU-Type Zeolite Membranes Synthesized by Microwave Assisted in Situ Crystallization. Mater. Lett. 2008, 62 (28), 4357−4359. (24) Noack, M.; Kölsch, P.; Dittmar, A.; Stöhr, M.; Georgi, G.; Schneider, M.; Dingerdissen, U.; Feldhoff, A.; Caro, J. Proof of the ISS-Concept for LTA and FAU Membranes and Their Characterization by Extended Gas Permeation Studies. Microporous Mesoporous Mater. 2007, 102 (1−3), 1−20. (25) Huang, A.; Wang, N.; Caro, J. Seeding-Free Synthesis of Dense Zeolite FAU Membranes on 3-Aminopropyltriethoxysilane-Functionalized Alumina Supports. J. Membr. Sci. 2012, 389, 272−279. (26) Nomura, M.; Yamaguchi, T.; Nakao, S. Silicalite Membranes Modified by Counterdiffusion CVD Technique. Ind. Eng. Chem. Res. 1997, 36 (10), 4217−4223. (27) Chen, Z.; Yin, D.; Li, Y.; Yang, J.; Lu, J.; Zhang, Y.; Wang, J. Functional Defect-Patching of a Zeolite Membrane for the Dehydration of Acetic Acid by Pervaporation. J. Membr. Sci. 2011, 369 (1−2), 506−513. (28) Lee, H.; Dutta, P. K. Charge Transport through a Novel Zeolite Y Membrane by a Self-Exchange Process. J. Phys. Chem. B 2002, 106 (46), 11898−11904. (29) Henis, J. M. S.; Tripodi, M. K. A Novel Approach to Gas Separations Using Composite Hollow Fiber Membranes. Sep. Sci. Technol. 1980, 15 (4), 1059−1068. (30) Feng, X.; Shao, P.; Huang, R. Y. M.; Jiang, G.; Xu, R.-X. A Study of Silicone Rubber/polysulfone Composite Membranes: Correlating H2/N2 and O2/N2 Permselectivities. Sep. Purif. Technol. 2002, 27 (3), 211−223. (31) Chiu, W. V.; Park, I.-S.; Shqau, K.; White, J. C.; Schillo, M. C.; Ho, W. S. W.; Dutta, P. K.; Verweij, H. Post-Synthesis Defect Abatement of Inorganic Membranes for Gas Separation. J. Membr. Sci. 2011, 377 (1−2), 182−190. (32) Wang, D.; Li, K.; Teo, W. K. Gas Permselection Properties in Silicone-Coated Asymmetric Polyethersulfone Membranes. J. Appl. Polym. Sci. 1997, 66 (5), 837−846. (33) Graham, C.; Pierrus, J.; Raab, R. E. Measurement of the Electric Quadrupole Moments of CO2, CO and N2. Mol. Phys. 1989, 67 (4), 939−955. (34) Kusakabe, K.; Kuroda, T.; Murata, A.; Morooka, S. Formation of a Y-Type Zeolite Membrane on a Porous A-Alumina Tube for Gas Separation. Ind. Eng. Chem. Res. 1997, 36 (3), 649−655.

(35) Olonisakin, A. A.; Alexander, S. D. Mechanism of Shear Transfer in a Reinforced Concrete Beam. Can. J. Civ. Eng. 1999, 26 (6), 810− 817. (36) Tjiptobroto, P.; Hansen, W. Mechanism for Tensile Strain Hardening in High Performance Cement-Based Fiber Reinforced Composites. Cem. Concr. Compos. 1991, 13 (4), 265−273. (37) Xu, X.; Subbaraman, H.; Chakravarty, S.; Hosseini, A.; Covey, J.; Yu, Y.; Kwong, D.; Zhang, Y.; Lai, W.-C.; Zou, Y.; et al. Flexible SingleCrystal Silicon Nanomembrane Photonic Crystal Cavity. ACS Nano 2014, 8 (12), 12265−12271. (38) Dong, G.; Li, H.; Chen, V. Challenges and Opportunities for Mixed-Matrix Membranes for Gas Separation. J. Mater. Chem. A 2013, 1 (15), 4610−4630. (39) Jiang, L. Y.; Chung, T. S.; Cao, C.; Huang, Z.; Kulprathipanja, S. Fundamental Understanding of Nano-Sized Zeolite Distribution in the Formation of the Mixed Matrix Single- and Dual-Layer Asymmetric Hollow Fiber Membranes. J. Membr. Sci. 2005, 252, 80−100. (40) Hussain, M.; König, A. Mixed-Matrix Membrane for Gas Separation: Polydimethylsiloxane Filled with Zeolite. Chem. Eng. Technol. 2012, 35 (3), 561−569.

6901

DOI: 10.1021/acs.langmuir.5b01306 Langmuir 2015, 31, 6894−6901