Environ. Sci. Technol. 2010, 44, 8230–8235
Tailoring the Structure of Thin Film Nanocomposite Membranes to Achieve Seawater RO Membrane Performance MARY LAURA LIND,† DANIEL EUMINE SUK,‡ THE-VINH NGUYEN,§ AND ERIC M. V. HOEK* Department of Civil & Environmental Engineering and California NanoSystems Institute, University of California, Los Angeles (UCLA), Los Angeles, California, United States
Received May 9, 2010. Revised manuscript received September 19, 2010. Accepted September 20, 2010.
Herein we report on the formation and characterization of pure polyamide thin film composite (TFC) and zeolite-polyamide thin film nanocomposite (TFN) reverse osmosis (RO) membranes. Four different physical-chemical post-treatment combinations were applied after the interfacial polymerization reaction to change the molecular structure of polyamide and zeolite-polyamide thin films. Both TFC and TFN hand-cast membranes were more permeable, hydrophilic, and rough than a commercial seawater RO membrane. Salt rejection by TFN membranes was consistently below that of hand-cast TFC membranes; however, two TFN membranes exhibited 32 g/L NaCl rejections above 99.4%, which was better than the commercial membrane under the test conditions employed. The nearly defect-free TFN films that produced such high rejections were achieved only with wet curing, regardless of other post-treatments. Polyamide films formed in the presence of zeolite nanoparticles were less cross-linked than similarly cast pure polyamide films. At the very low nanoparticle loadings evaluated, differences between pure polyamide and zeolite-polyamide membrane water and salt permeability correlated weakly with extent of cross-linking of the polyamide film, which suggests that defects and molecularsieving largely govern transport through zeolite-polyamide thin film nanocomposite membranes.
Introduction Fresh water is essential to human survival and is integral in the global economy for its uses in agricultural irrigation, industrial processes, oil and gas exploration, and electricity production (1). Continuous population growth and industrial development stress the limited supply of freshwater. This water stress cannot be eliminated by conservation efforts alone. Hence, production of fresh water from alternative * Corresponding author tel: (310) 206-3735; fax: (310) 206-2222; e-mail:
[email protected]. † Current address: Arizona State University School of Engineering of Transport, Matter and Energy, P.O. Box 876106 Tempe, AZ 852876106, United States. ‡ Current address: Samsung Cheil Industries Inc., E-Project Team, 332-2 Gocheon-dong, Uiwang-si, Gyeonggi-do, Korea 437-711. § Current address: Vietnam National University-Hochiminh City, Hochiminh City University of Technology, Faculty of Environment, 268 Ly Thuong Kiet Street, District 10, Hochiminh City, Vietnam. 8230
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 21, 2010
sources such as reclaimed wastewater, brackish groundwater, and ocean water must be considered. Commercially available reverse osmosis (RO) membranes can produce high quality water from such alternative water sources, but improvements on existing RO membranes are needed to further reduce operating costs, energy demand, and chemical consumption. Improved RO membranes might exhibit higher water permeability, solute selectivity, or fouling resistance. Mixed matrix membranessin which a filler material is embedded within a polymeric matrixsare already used in a variety of industrial and environmental processes including fuel cells, pervaporation, and gas separations (2-6). This concept has added a new degree of freedom in the development of membranes with novel separation performance, i.e., selection of the unique properties of the filler material, which may include enhanced permeability, selectivity, stability, surface area, or catalytic activity. More recently, mixed matrix membranes are being explored to tailor the performance and add new functionality to membranes for water purification applications. Thin film nanocomposite membranes for reverse osmosis applications have been developed that incorporate pure metal, metal oxide, and zeolite molecularsieve nanoparticles. These recent efforts are briefly reviewed here. Kwak et al. deposited titanium dioxide nanoparticles onto hand-cast polyamide composite membranes and demonstrated differences in hydrophilicity and flux (7). Lee et al. incorporated silver nanoparticles into a thin polyamide layer during interfacial polymerization (8). While these membranes exhibited some antibacterial effects, they produced nanofiltration-like selectivity (96-97% rejection of 2000 ppm magnesium sulfate) without significant changes in water permeability. Lee et al. similarly incorporated 30-nm titanium dioxide nanoparticles into polyamide thin films during the interfacial polymerization reaction (9). In brackish water chemistry, the water flux of the titanium dioxide-polyamide membranes increased with nanoparticle loading up to 5 wt % with minimal changes in salt rejection; however, at particle loadings above 5 wt % salt rejection by the membranes decreased dramatically suggesting the flux enhancement largely resulted from defects in the polyamide created by the solid nanoparticles (9). Singh et al. incorporated 16-nm silica nanoparticles into polyamide thin film composite membranes (10). These membranes were tested using 500 ppm dioxane solutions and the structure of the polymer was investigated with small angle neutron scattering. In later work from the same laboratory, Jadav et al. incorporated 16- and 3-nm silica nanoparticles into the polyamide layer of thin film composite membranes (11). The resultant silica-polyamide membranes exhibited brackish water desalination performance, but salt selectivity decreased with nanoparticle loading indicating defect formation in the polyamide thin film as the principle mechanism of flux enhancement (11). Jeong et al. demonstrated that incorporating zeolite molecular-sieve nanoparticles into polyamide thin films (during interfacial polymerization) could double the water flux without reducing observed rejection of 2 g/L sodium chloride, magnesium sulfate, and polyethylene glycol solutions (12). A key feature of this work is that zeolite nanoparticle size was designed to be the same size as polyamide film thickness, thereby, creating a “percolation threshold” through the thin film with a single particle. It was hypothesized that zeolite molecular-sieves improved membrane permeability while maintaining good selectivity by acting as preferential flow paths for water transport. However, a series of zeolite-polyamide coating films prepared with impermeable 10.1021/es101569p
2010 American Chemical Society
Published on Web 10/13/2010
TABLE 1. Post-Treatment Conditions for Membrane Synthesisa fabrication conditions
cure temperature [C]
cure time [s]
cure medium
rinse A.1 [s]
rinse A.2 [s]
rinse A.3 [s]
rinse B.1 [s]
1-A 1-B 2-A 2-B
90 90 90 90
120 120 360 360
water water oven oven
120 120 -
30 30 -
120 120 -
600 600
a
Rinse A.1 ) 0.2 g/L NaOCl. Rinse A.2 ) 1 g/L Na2S2O5. Rinse A.3 ) 90 °C deionized water. Rinse B.1 ) 2 g/L NaHCO3.
zeolite nanoparticles (internal pores filled with the polymer template) produced fluxes intermediate between the pure polyamide membrane and the TFN film prepared with pore opened zeolites at the same zeolite loading (12). This result offered indirect evidence that another mechanism, besides molecular-sieving, could be responsible for the enhanced membrane performance. Subsequently, two additional mechanisms for the observed flux enhancement were proposed (13, 14): (1) defect formation due to zeolite nanoparticle aggregation in the organic monomer solution and (2) heat release from the zeolites by hydration during the interfacial polymerization reaction, which changed the cross-linked structure of the polyamide films. To date, nanocomposite RO membranes have exhibited separation performance suitable for brackish water RO or nanofiltration applications. If defect formation is the primary mechanism responsible for flux enhancement, then it should be very difficult to produce nanocomposite RO membranes with seawater RO membrane salt selectivity. However, if either molecular sieving or altered polyamide film structure is the primary mechanism responsible for flux enhancement, then seawater RO selectivity should be attainable. In this paper, methods for fabricating TFN membranes with water permeability and salt selectivity rivaling commercial seawater RO membranes were explored. Specifically, four different post-treatment regimes were applied to pure polyamide and zeolite-polyamide thin film composite membranes to change their molecular structure and minimize defect formation. Very low zeolite loadings were used to minimize the effects of molecular sieving on the separation performance of the membrane. Hand-cast TFC and TFN membrane separation performance, interfacial properties, and polyamide thin film structure were characterized and compared to a commercially fabricated seawater RO membrane.
Experimental Section Nomenclature. The following nomenclature is used in this manuscript. Polysulfone coated polyester membranes (PS, NanoH2O) were used as the support on which the polyamide was deposited. During the synthesis of the thin polyamide film meta-phenylene diamine (MPD, 98% Sigma-Aldrich) was dissolved in deionized water. The organic solvent Isoparrafin-G (Isopar-G, Gallade Chemical, Santa Ana, CA) was used to dissolve trimesoyl chloride (TMC, Sigma-Aldrich). Nanoparticles of Linde type A (LTA, NanoH2O) zeolites, approximately 250 nm in diameter, were used. The chemical formula for fully hydrated LTA is Na12[(AlO2)12(SiO2)12] · 27H2O. Chemicals used for post-treatment were sodium hypochlorite solution (NaOCl, available chlorine 10-13%, Sigma-Aldrich), sodium metabisulfite (Na2S2O5, Fisher Chemical), and sodium bicarbonate (NaHCO3, Sigma-Aldrich). Membrane solute rejection was tested using sodium chloride (NaCl). Thin film composite (TFC) membranes are pure polyamide thin films without nanoparticles, whereas thin film nanocomposite (TFN) membranes are polymerized with nanoparticles present in the Isopar-G-TMC monomer solution. Membranes are designated in the following manner: TFC-1-A means a thin film composite membrane fabricated
with curing conditions 1 and rinsing conditions A (found in Table 1). A commercially available membrane, SWC3+ (Hydranautics) was used for comparison with the hand-cast membranes. Membrane Preparation. All steps for preparation of the interfacially polymerized polyamide thin film, other than solution preparation, were performed in a vertical sash fume hood. The methods presented below are a modification of a previously published summary of commercially relevant polyamide composite membrane post-treatments (15). TFC membranes were prepared as follows. First, the polysulfone support membrane was taped to an 8 × 5 × 0.25 in. borosilicate glass plate with Fisherbrand laboratory tape. Two separate monomer solutions were prepared: MPD in water and TMC in organic. The MPD in deionized water solution, 3.4% (wt/wt %), was prepared in a glass jar and the jar was wrapped in aluminum foil to prevent light-oxidation of the MPD. A 0.15% (wt/wt %) solution of TMC in Isopar-G was prepared in a 500-mL pyrex solution bottle. Both solutions were stirred with a magnetic stirrer at room temperature for a minimum of 3 h prior to use. After stirring, the MPD solution was poured into a pyrex dish on an approximately 20 degree incline. The plate with the PS support taped to it was then placed into the MPD solution for 2 min, the support side was placed into the solution such that the backside of the plate not immersed in the solution. After 2 min the plate was lifted from the solution and the excess solution was allowed to drain from the surface. The plate was then placed on a rubber mat, with the glass down and the MPD-soaked support facing up. An air-knife (model 110012SS-316, Exair, Cincinnati OH) with filtered, compressed air was used to remove excess solution from the support membrane. The air-knife was operated with a pressure of 20 psi and held approximately 0.5 in. from the membrane surface; a complete pass of the 8-in. length of the sample took a minimum of 15 s. Excess MPD solution was wiped from the taped edges and back of the glass plate. Then the plate and membrane were immersed vertically in the TMC solution for 1 min in a custom-fabricated 250-mL container. After 1 min in the TMC solution, the plate and membrane were removed from the TMC container and held vertically for 2 min. After vertical holding, the membranes were subjected to either wet or dry curing followed by a series of post-treatment rinses (Table 1). Postcure exposure was to either of two different combinations of rinses (200 ppm NaOCl in deionized water, followed by 1 g/L Na2S2O5 in deionized water, or 2 g/L NaHCO3 in deionized water). Hot water cures and rinses were performed in a 90 °C deionized water bath in a pyrex dish on a hot plate. Dry oven cures were performed at 90 °C, with the membrane placed vertically on a rack in the center of the oven (737F Isotemp Oven, Fisher Scientific). For all rinses, the membrane was removed from the glass support and the solutions were kept in 1-L glass beakers. Chlorination of polyamide membranes is known to structurally degrade the membranes (16). Here the NaOCl is believed to first scavenge unreacted MPD that has not yet escaped the structure; then the Na2S2O5 is believed to neutralize any of VOL. 44, NO. 21, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8231
the unreacted NaOCl and prevent it from further attacking the polyamide. The NaHCO3 rinse is a basic solution (pH ∼11). Zeolite-polyamide nanocomposite thin films were cast identically to TFC membranes, except 0.2% (wt/wt%) of colloidal zeolite nanoparticles were dispersed in the IsoparG-TMC solution. The TMC solution including nanoparticles was ultrasonicated for 40 min at 20 °C immediately prior to use in the interfacial polymerization reaction. Membrane Separation Performance. The synthesized thin films were evaluated for permeability of pure water and NaCl using a custom-fabricated 6-cell reverse osmosis testing system described by Jin et al. (17). The membranes were compacted for a minimum of 12 h at ∼59 bar (850 psi), until the pure water flow reached a steady state. After compaction, pure water flow rate was measured at ∼55 bar (800 psi). Next, the feedwater solution was changed to a 32 g/L NaCl solution. Permeate flow rate and conductivity of feed and permeate samples were measured after the system performance was stable for at least 30 min. Water flux was determined from permeate water flow rate as Jw )
Qp Am
(1)
where Qp is the permeate water flow rate and Am is the effective membrane area (0.00194 m2). Feed and permeate conductivities were used to calculate the observed salt rejection from Kp Xs ) 1 Kf
(2)
where Kf and Kp were the feed and permeate conductivity. Calculation of Membrane Transport Coefficients. Transport through RO membranes is generally considered to occur by a solution-diffusion type mechanism where water, Jw, and salt, Js, flux were calculated by Jw ) A(∆p - ∆π)
(3)
Js ) B∆c
(4)
Here, ∆p is the (applied) trans-membrane hydraulic pressure, ∆π is the trans-membrane osmotic pressure, and ∆c () cm - cp) is the (real) trans-membrane concentration gradient, where cm is the feed-side membrane surface salt concentration and cp is the permeate salt concentration. Also, A and B are the water and solute permeability coefficients. Trans-membrane osmotic pressure for NaCl was determined from ∆π ) 2RT(cm - cp)
(5)
where R is the universal gas constant and T is the temperature. The osmotic coefficient has been assumed to equal 1, this is reasonable because at the solute concentrations used in this study the error between the Gibbs (nonlinear) and van’t Hoff (linear) equations is only ∼4.34% (18). Using this as a correction factor in our osmotic pressure calculations does not change our conclusions on structure-performance relationships. The membrane surface salt concentration was estimated using
()
cm Jw ) 1 - Xs + Xsexp cf ks
(6)
where cf is the feed concentration, Xs () 1 - cp/cf) is the observed salt rejection, and ks is the salt mass transfer coefficient. The channel average mass transfer coefficient in 8232
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 21, 2010
the laboratory scale crossflow membrane filtration system was estimated by 1
ks ) 1.85(ReSc) /3
D dh
(7)
Here, Re is the Reynolds number, Sc is the Schmidt number, D is the solute diffusivity, and dh () 2Hc, where Hc is the crossflow channel height) is the hydraulic diameter of the crossflow channel (19). Next, the real membrane salt rejection could be calculated directly from
(
Rs ) 1 -
)
cp 1 - Xs )1cm 1 - Xs + Xsexp(Jw /ks)
(8)
Combining the measured flux and rejection with the calculated mass transfer coefficient, the pure water permeability coefficient during salt water experiments was A)
Jw ∆p - 2RTRscm
(9)
which was obtained by combining eq 3 with eqs 5-8. The salt permeability coefficient was calculated from B ) Jw
(1 - Rs) Rs
(10)
which was derived by substituting Js ) Jwcp and the parenthetical expression from eq 5 into eq 4, and dividing both sides by cm. The water-salt selectivity of a membrane can be characterized by the ratio of water to solute permeability () A/B). Membrane Surface Characterization. Root-mean squared (RMS) surface roughness and surface area difference (SAD) were quantified by atomic force microscopy, AFM, (Nanoscope IIIa; Digital Instruments, Santa Barbara, CA, USA). For AFM analysis membrane coupons were dried flat in a desiccator overnight prior to analysis. Attenuated total reflection Fourier transform infrared spectroscopy (ATRFTIR) was performed (FT/IR 670 plus, Jasco, Easton, MD, USA) with a variable angle ATR attachment coupled to a germanium crystal operated at 45 degrees. Prior to ATRFTIR measurement the samples were dried flat in a desiccator for a minimum of 24 h. Membrane surface chemical composition was analyzed with X-ray photoelectron spectroscopy (XPS). XPS data were collected using a Surface Science Instruments M-Probe system that has been described previously (20). Ejected electrons were collected at an angle of 35° from the surface normal, and the sample chamber was maintained at 99.4% rejection of 32 g/L NaCl) opens up new possibilities for tailoring the performance of seawater RO membrane materials and processes.
Acknowledgments Financial support for this research was provided in part by the U.S, Environmental Protection Agency (Award 87888.01) and the UCLA California NanoSystems Institute. We are grateful to the Molecular Materials Research Center of the Beckman Institute at the California Institute of Technology for providing access to the XPS instrument. We are very grateful to Prof. Bruce Dunn in the UCLA Department of Material Science & Engineering for his assistance in providing access to the ATR-FTIR instrument.
Note Added after ASAP Publication This paper was published ASAP on October 13, 2010. Equation 2 was changed. The revised paper was reposted on October 28, 2010.
Literature Cited (1) Perlman, H. U.S. Department of the Interior/U.S. Geological Survey Water Q&A: Water use. http://ga.water.usgs.gov/edu/ qausage.html#HDR1 (accessed April 21, 2010). (2) Tin, P. S.; Chung, T. S.; Jiang, L. Y.; Kulprathipanja, S. Carbonzeolite composite membranes for gas separation. Carbon 2005, 43, 2025–2027. (3) Zimmerman, C. M.; Singh, A.; Koros, W. J. Tailoring mixed matrix composite membranes for gas separations. J. Membr. Sci. 1997, 137 (1-2), 145–154. (4) Okumus, E.; Gurkan, T.; Yilmaz, L. Development of a mixedmatrix membrane for pervaporation. Sep. Sci. Technol. 1994, 29, 2451–2473. (5) Vu, D. Q.; Koros, W. J.; Miller, S. J. Mixed matrix membranes using carbon molecular sieves - I. Preparation and experimental results. J. Membr. Sci. 2003, 211 (2), 311–334. (6) Chang, H. Y.; Lin, C. W. Proton conducting membranes based on PEG/SiO2 nanocomposites for direct methanol fuel cells. J. Membr. Sci. 2003, 218 (1-2), 295–306. (7) Kwak, S. Y.; Kim, S. H.; Kim, S. S. Hybrid organic/inorganic reverse osmosis (RO) membrane for bactericidal anti-fouling. 1. Preparation and characterization of TiO2 nanoparticle selfassembled aromatic polyamide thin-film-composite (TFC) membrane. Environ. Sci. Technol. 2001, 35, 2388–2394. (8) Lee, S. Y.; Kim, H. J.; Patel, R.; Im, S. J.; Kim, J. H.; Min, B. R. Silver nanoparticles immobilized on thin film composite polyamide membrane: Characterization, nanofiltration, antifouling properties. Polym. Adv. Technol. 2007, 18, 562–568. (9) Lee, H. S.; Im, S. J.; Kim, J. H.; Kim, H. J.; Kim, J. P.; Min, B. R. Polyamide thin-film nanofiltration membranes containing TiO2 nanoparticles. Desalination 2008, 219 (1-3), 48–56.
(10) Singh, P. S.; Aswal, V. K. Characterization of physical structure of silica nanoparticles encapsulated in polymeric structure of polyamide films. J. Colloid Interface Sci. 2008, 326 (1), 176– 185. (11) Jadav, G. L.; Singh, P. S. Synthesis of novel silica-polyamide nanocomposite membrane with enhanced properties. J. Membr. Sci. 2009, 328 (1-2), 257–267. (12) Jeong, B. H.; Hoek, E. M. V.; Yan, Y. S.; Subramani, A.; Huang, X. F.; Hurwitz, G.; Ghosh, A. K.; Jawor, A. Interfacial polymerization of thin film nanocomposites: A new concept for reverse osmosis membranes. J. Membr. Sci. 2007, 294 (1-2), 1–7. (13) Lind, M. L.; Ghosh, A. K.; Jawor, A.; Huang, X. F.; Hou, W.; Yang, Y.; Hoek, E. M. V. Influence of zeolite crystal size on zeolitepolyamide thin film nanocomposite membranes. Langmuir 2009, 25, 10139–10145. (14) Lind, M. L.; Jeong, B. H.; Subramani, A.; Huang, X. F.; Hoek, E. M. V. Effect of mobile cation on zeolite-polyamide thin film nanocomposite membranes. J. Mater. Res. 2009, 24, 1624–1631. (15) Hiroki, T.; Sugita, K.; Oto, K. Composite semipermeable membrane, and production process thereof. U.S. Patent 7,279,097, 2007. (16) Kwon, Y. N.; Tang, C. Y.; Leckie, J. O. Change of chemical composition and hydrogen bonding behavior due to chlorination of crosslinked polyamide membranes. J. Appl. Polym. Sci. 2008, 108, 2061–2066. (17) Jin, X.; Huang, X. F.; Hoek, E. M. V. Role of Specific Ion Interactions in Seawater RO Membrane Fouling by Alginic Acid. Environ. Sci. Technol. 2009, 43, 3580–3587. (18) Crittenden, J.; Trussell, R. R.; Hand, D. W.; Howe, K. J.; Tchobanoglous, G. Water Treatment Principles and Design, 2nd ed.; John Wiley & Sons, Inc.: Hoboken, NJ, 2005; pp 1447-1448. (19) Hoek, E. M. V.; Kim, A. S.; Elimelech, M. Influence of crossflow membrane filter geometry and shear rate on colloidal fouling in reverse osmosis and nanofiltration separations. Environ. Eng. Sci. 2002, 19, 357–372. (20) Bansal, A.; Li, X. L.; Yi, S. I.; Weinberg, W. H.; Lewis, N. S. Spectroscopic studies of the modification of crystalline Si(111) surfaces with covalently-attached alkyl chains using a chlorination/alkylation method. J. Phys. Chem. B 2001, 105, 10266–10277. (21) Tang, C. Y. Y.; Kwon, Y. N.; Leckie, J. O. Probing the nano- and micro-scales of reverse osmosis membranes - A comprehensive characterization of physiochemical properties of uncoated and coated membranes by XPS, TEM, ATR-FTIR, and streaming potential measurements. J. Membr. Sci. 2007, 287 (1), 146–156. (22) Wenzel, R. N. Surface roughness and contact angle. J. Phys. Colloid Chem. 1949, 53, 1466–1467. (23) Urban, M. W. Vibrational Spectroscopy of Molecules and Macromolecules on Surfaces; John Wiley & Sons, Inc.: New York, 1993. (24) Kyotani, T.; Shimotsuma, N.; Kakui, S. FTIR-ATR study of the surface of a tubular zeolite NaA membrane ultrasonically reacted with water and acetic acid. Anal. Sci. 2006, 22, 325–327. (25) Alfaro, S.; Rodriguez, C.; Valenzuela, M. A.; Bosch, P. Aging time effect on the synthesis of small crystal LTA zeolites in the absence of organic template. Mater. Lett. 2007, 61, 4655–4658. (26) Breck, D. W. Zeolite Molecular Sieves; John Wiley & Sons: New York, 1974; pp 415. (27) Singh, P. S.; Joshi, S. V.; Trivedi, J. J.; Devmurari, C. V.; Rao, A. P.; Ghosh, P. K. Probing the structural variations of thin film composite RO membranes obtained by coating polyamide over polysulfone membranes of different pore dimensions. J. Membr. Sci. 2006, 278 (1-2), 19–25. (28) Clarizia, G.; Algieri, C.; Regina, A.; Drioli, E. Zeolite-based composite PEEK-WC membranes: Gas transport and surface properties. Microporous Mesoporous Mater. 2008, 115 (1-2), 67–74.
ES101569P
VOL. 44, NO. 21, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8235