Spray-coating of nanoporous carbon membranes for air separation

Spray-coating of nanoporous carbon membranes for air separation

Journal of Membrane Science 161 (1999) 1–5 Rapid communication Spray-coating of nanoporous carbon membranes for air separation Madhav Acharya, Henry...

102KB Sizes 9 Downloads 36 Views

Journal of Membrane Science 161 (1999) 1–5

Rapid communication

Spray-coating of nanoporous carbon membranes for air separation Madhav Acharya, Henry C. Foley ∗ Center for Catalytic Science and Technology, Department of Chemical Engineering, University of Delaware, Newark, DE 19716 USA Received 5 March 1999; received in revised form 3 May 1999; accepted 6 May 1999

Abstract Nanoporous carbons (NPC) can be fabricated in the form of supported thin films with minimal defects and remarkably high size selectivity of oxygen over nitrogen. Spray coating of porous stainless steel disks with a solution of poly(furfuryl) alcohol (PFA) in acetone was used to synthesize nanoporous carbon membranes in a reproducible manner, and is the first reported case of this technique being used for supported carbon membrane synthesis. The disks were dried and then pyrolyzed in an inert atmosphere to a final temperature of 600◦ C. The resulting membranes were tested with binary oxygen/nitrogen mixtures. In experiments with a binary feed of oxygen and nitrogen, the membranes were found to have good oxygen over nitrogen selectivities (up to 4) and O2 fluxes on the order of 10−9 mol/m2 s Pa. Attempts were made to modify membrane performance further by post-pyrolysis heating, additional coatings or by hydrocarbon pyrolysis. None of these post-treatments produced membranes superior to those derived directly from the initial spray coating and pyrolysis. ©1999 Elsevier Science B.V. All rights reserved. Keywords: Gas separations; Inorganic membranes; Membrane preparation and structure; Microporous and porous membranes; Nanoporous carbon

Ceramic membrane technology is being developed rapidly for several gas separations. Since membrane processes can be more energy efficient and, therefore, more economical than other methods of gas separation, considerable research is being undertaken to determine new materials and methods for membrane synthesis. One prominent example is that of the surface selective flow (SSF) membrane developed by Air Products and Chemicals. This technology has been commercialized and is able to produce enriched hydrogen

∗ Corresponding author. Present address: Mobil Technology Company, Paulsboro, NJ 08066, USA. Tel.: +1-302-831-6856; fax: +1302-831-2085; e-mail: [email protected]

streams at high pressure from mixed hydrocarbon and hydrogen feeds [1–3]. Among the problems that must be overcome before broader application of ceramic membrane technology can be realized is the fabrication of these materials in a manner that is both reproducible and scalable for manufacturing. Fabrication of ceramic membranes in high yield is a generic problem [4]. The SSF membrane consists of a nanoporous carbon (NPC) on an underlying support, which is reported to be produced by involving the pyrolysis of a polymer latex coating [1,2]. More typical are the results of another group of researchers, who recently reported their results on high flux silica membranes. In contrast to the SSF membrane materials, these silica materials required synthe-

0376-7388/99/$ – see front matter ©1999 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 6 - 7 3 8 8 ( 9 9 ) 0 0 1 7 3 - 8


M. Acharya, H.C. Foley / Journal of Membrane Science 161 (1999) 1–5

sis in a clean room environment in order to eliminate membrane defects [5]. The difference in the apparent degree of difficulty between the methods of preparation used in these two studies defines the spectrum that can be encountered in the production of ceramic membranes. This makes clear the need for more research into the means of fabrication of ceramic membrane materials [6]. Readily-implemented production methods are more likely to lead to rapid commercialization. That nanoporous carbon molecular sieves with pore sizes less than 0.5 nm can separate nitrogen from oxygen has been known for some time and is the basis for nitrogen pressure swing adsorption (NPSA) [7,8]. This separation, done over packed beds of nanoporous carbon, is considered to be based more upon the kinetics of diffusion than on the thermodynamics of adsorption [9,10]. Entropic effects, at the narrowest points of the complex pore networks of NPC, may provide the necessary discrimination between oxygen and nitrogen [11,12]. To use nanoporous carbons in the form of a highly size selective membrane requires new methods of materials fabrication to prepare defect-free membranes with high reproducibility. Previous work on polymerderived NPC molecular sieves, has yielded the synthesis parameters required to obtain specific separations [13–16]. Furthermore, research done with polymerderived nanoporous carbon membranes makes it clear that this is a promising new technology [17–21]. Our first efforts at preparing nanoporous carbon membranes yielded encouraging results in size-selective separation or molecular sieving [22]. Reproducibility proved to be problematic, however, and hampered more rapid research and development of these materials. In this communication, we describe a spray coating system for the production of thin layers of nanoporous carbon on the surface of a porous stainless steel support that overcomes these problems. The entire synthesis process is done in a standard fume hood, with no added precautions required for particulates removal. Once prepared, these membranes are robust, stable and require no special handling or storage. Good performance for oxygen–nitrogen separation, reproducibility and simplicity are the hallmarks of this preparation method. Porous stainless steel supports of 0.2 ␮m pore size were obtained from Mott Metallurgical Corporation in

the form of disks (diameter 1.87500 , thickness 0.03900 ). The supports were cleaned with chloroform in a sonication bath for 15 min and then dried in air. The carbon precursor was a solution of 50 : 50 poly(furfuryl) alcohol in acetone. This was sprayed onto the support in the form of a fine mist using an external mix airbrush (Badger Model 250) with nitrogen gas. The stainless steel support was mounted on an electric motor shaft and rotated horizontally. The rotation speed (500 rpm) and nitrogen gas flowrate (100 m3 /min) were adjusted to provide uniform coating of the solution, based on simple visual inspection, on the surface of the disk. The total spray time for each coating was less than 15 s, and the weight of pyrolyzed carbon added in each coating was approximately 20–30 mg. Three membranes were coated by this process and pyrolyzed at 600◦ C. These membranes are labeled as SPCM-2, SPCM-3 and SPCM-4. Permeation experiments were carried out by placing these membranes in a flanged module with Viton® gaskets, as has been described elsewhere [22]. A 1 : 1 mixture of O2 and N2 flowed over the coated side of the membrane, while helium flowed as a sweep gas on the opposite side in order to entrain the permeate for analysis. A molecular Sieve 5A column (80/100 mesh, Supelco) was used for separation and analysis of oxygen and nitrogen. The maximum pressure differential used in the experiments was 7.3 bar and the module showed no sign of leakage at this pressure over prolonged periods of experimentation. The permeances of oxygen and nitrogen varied with the number of coatings of NPC applied to the support. For oxygen the permeances ranged from 3 × 10−10 to 1.2 × 10−9 mol/m2 s Pa while for nitrogen they were from just under 1 × 10−10 to slightly above 4 × 10−10 mol/m2 s Pa. No pressure dependence was detected in the permeances indicating the absence of Pouiselle flow and cracks (Fig. 1). Membrane performance did not deteriorate with testing (results were obtained over a period of 3 months), nor did it suffer when the membranes were left exposed to air for a period of 1 week and were then re-tested. The selectivity for permeation of oxygen over that of nitrogen from the binary mixtures was between 2.7 and 3.7. Secondary heat-treatments of membrane SPCM-2 at 600◦ C for 1 and 2 h, without the additional polymer precursor, did not change the selectivity very much, but did lead to small increases in permeance

M. Acharya, H.C. Foley / Journal of Membrane Science 161 (1999) 1–5


Fig. 1. Permeances of oxygen and nitrogen through SPCM-2, 3 and 4 as a function of topside pressure.

Fig. 2. Permeances of oxygen and nitrogen through nanoporous carbon membrane as a function of pressure; ‘first-test’ after synthesis, ‘second-test’ and ‘third-test’ after heating the membrane for 1 and 2 h, respectively.

(Fig. 2). This small but measurable increase in permeance is most likely due to an increase in the porosity of the membrane, as has been discussed earlier by Mariwala and Foley [14]. SPCM-3 was coated once more with PFA solution and pyrolyzed

at 600◦ C for an additional hour. After a fourth coating, the permeances of both gases dropped but the selectivity remained fairly constant. Going beyond this number of coatings can lead to losses in selectivity.


M. Acharya, H.C. Foley / Journal of Membrane Science 161 (1999) 1–5

Fig. 3. Permeances of oxygen and nitrogen on SPCM-2 before (solid symbols) and after carbon deposition (open symbols) (20% propene in He, 600◦ C, 2 h).

Fig. 4. O2 /N2 selectivity versus O2 permeance through all three nanoporous carbon membranes.

Hydrocarbon deposition on carbon molecular sieves has been utilized successfully to reduce pore sizes and to increase selectivity for oxygen over nitrogen [23]. We attempted to do the same for membrane

with SPCM-2. A stream of 20% propylene in helium was flowed over the membrane at a temperature of 600%C for 2 h. Rather than improving separative performance, both the permeances and the selectivity de-

M. Acharya, H.C. Foley / Journal of Membrane Science 161 (1999) 1–5

creased (Fig. 3). The high concentration of propylene in the gas stream most likely resulted in blockage of pores responsible for the O2 /N2 separation. Fig. 4 shows the O2 /N2 selectivitty versus O2 permeance for all three membranes. The objective in future work would be to increase selectivity while keeping permeance the same. The spray coating method for synthesis of nanoporous carbon membranes from poly(furfuryl) alcohol on stainless steel works well, even in the relatively simplistic mode of implementation used in this study. That, however, is the point—simple and relatively forgiving methods of fabrication are likely to be those that will be used for the commercial production of ceramic membranes. The advantage of the liquid phase precursor comes to the fore when considering new means to deliver the polymer to the support surface in a controlled and reproducible fashion. Spray coating is well characterized so process design and control can be done straightforwardly, in contrast to the synthesis of hollow-fiber membranes. Further work is underway to develop this and related methods of membrane synthesis.

Acknowledgements Funding for this research was provided by the DuPont Company, the State of Delaware Research Partnership and the Department of Energy, Office of Basic Energy Sciences. References [1] M.B. Rao, S. Sircar, Nanoporous carbon membranes for separation of gas mixtures by surface selective flow, J. Memb. Sci. 85 (1993) 253–264. [2] M.B. Rao, S. Sircar, Performance, Performance and pore characterization of nanoporous carbon membranes for gas separation, J. Memb. Sci. 110 (1996) 109–118. [3] D.J. Parrillo, C. Thaeron, S. Sircar, Separation of bulk hydrogen sulfide-hydrogen mixtures by selective surface flow membrane, AIChE J. 43 (1997) 2239–2245. [4] H. Verweij, Progress in inorganic membranes, CHEMTECH 26 (1996) 37–41. [5] R.M. de Vos, H. Verweij, High-selectivity, high-flux silica membranes for gas separation, Science 279 (1998) 1710– 1711. [6] R.T. Yang, Gas Separation by Adsorption Processes, Butterworth, Boston, 1987.


[7] D.M. Ruthven, Principles of Adsorption and Adsorption Processes, Wiley, New York, 1984. [8] M. Suzuki, Adsorption Engineering, Elsevier, Amsterdam, 1990. [9] H. Juntgen, K. Knoblauch, K. Harder, Fuel 60 (1981) 817. [10] A.I. La Cava, V.A. Koss, D. Wickens, Gas Sep. Purif. 3, 1989, 180. [11] A Singh, W.J. Koros, Significance of entropic selectivity for advanced gas separation membranes, Ind. Eng. Chem. Res. 35 (1996) 1231–1234. [12] M.S. Kane, J.F. Goellner, H.C. Foley, S.J.L. Billinge, R. DiFrancesco, Symmetry breaking in nanostructure development of carbogenic molecular sieves: effects of morphological pattern formation on oxygen, Symmetry breaking in nanostructure development of carbogenic molecular sieves: effects of morphological pattern formation on oxygen and nitrogen transport, Chem. Mat. 8 (1996) 2159– 2171. [13] R.K. Mariwala, Microporous Carbogenic Materials: Synthesis, Characterization and Engineering for Adsorption and Catalysis, Ph.D. Thesis, University of Delaware, 1993. [14] R.K. Mariwala, H.C. Foley, Evolution of ultramicroporous adsorptive structure in poly(furfuryl alcohol) derived carbogenic molecular sieves, Ind. Eng. Chem. Res. 33, 1994, 607–615. [15] D.S. Lafyatis, J. Tung, H.C. Foley, Poly(furfuryl alcohol)derived carbon molecular sieves: dependence of adsorptive properties on carbonization temperature, time, time and poly(ethylene glycol) additives, Ind. Eng. Chem. Res. 30 (1991) 865–873. [16] H.C. Foley, Carbogenic molecular sieves: synthesis, properties, properties and applications, Micr. Mat. 4 (1995) 407–433. [17] C.W. Jones, W.J. Koros, 2. Regeneration following organic exposure, Carbon 32 (1994) 1427–1432. [18] C.W. Jones, W.J. Koros, 1. Preparation, characterization based on polyimide precursors, Carbon 32 (1994) 1419–1425. [19] C.W. Jones, W.J. Koros, Carbon composite membranes – a solution to adverse humidity effects, Ind. Eng. Chem. Res. 34 (1995) 164–167. [20] C.W. Jones, W.J. Koros, Characterization of ultramicroporous carbon membranes with humidified feeds, Ind. Eng. Chem. Res. 34 (1995) 158–163. [21] V.C. Geiszler, W.J. Koros, Effects of polyimide pyrolysis conditions on cabon molecular sieve membrane properties, Ind. Eng. Chem. Res. 35 (1996) 2999–3003. [22] M. Acharya, B.A. Raich, H.C. Foley, M.P. Harold, J.J. Lerou, Metal supported cabogenic molecular sieve membranes – synthesis, Metal supported cabogenic molecular sieve membranes – synthesis and applications, Ind. Eng. Chem. Res. 36 (1997) 2924–2930. [23] A.L. Cabrera, J.E. Zehner, C.G. Coe, T.R. Gaffney, T.S. Farris, J.N. Armor, Two-step hydrocarbon deposition with a single hydrocarbon, Carbon 31 (1993) 969–976.