Polymeric Membranes for Gas and Vapour Separation

One of the main concerns of modern society is to maintain a high standard of living and to create a sustainable future. In this context we are facing a number of problems related to the need for more energy efficient and environmentally friendly use of our limited resources.

Membrane technology is seen as a potential alternative for currently used processes, because of its small footprint, its energy efficiency and its modular design [1,2]. Important separations where membrane technology may be used are

• H₂/CO₂ separation to produce hydrogen for fuel cells
• CO₂/N₂ separation applied to flue gas or lime oven exhaust gases for CO₂ sequestration
• CO₂/CH₄ separation for natural gas treatment or for biogas upgrading
• O₂/N₂ separation to produce oxygen enriched air or pure nitrogen. Organic vapour recovery from air at gasoline stations or oil platforms by membranes reduce emissions to the atmosphere and enhances process efficiency and economy.

Types of Separation Membranes

The selective separation of gases and/or vapours requires thin films are able to separate on the basis of molecular properties. Such membranes may be inorganic and porous, such as zeolites or mesoporous silica, separating on the basis of molecular dimensions or condensability of the permeating species. They can be metallic, mostly for hydrogen separation on the basis of the possibility to undergo chemical cleavage and recombination reaction. The most common and abundant group of commercially available membranes is formed by polymeric membranes, separating on the basis of the so-called solution-diffusion mechanism.

Permeability vs Selectivity

A limitation of polymeric membranes for gas and vapour transport is the so-called permeability-selectivity trade-off, first reported by Robeson et al. in 1991 [3] and then further updated [4]. This trade-off basically determines that if one is looking for new materials with a higher permeability, the price to pay is a lower selectivity, and vice versa. While originally introduced on the basis of empirical considerations only, a physical background of this trade-off has also been given [5].

In spite of this intrinsic limitation, one of the main targets for research in polymeric membranes for gas separation has therefore become the search for materials which exceed this upper bound. Molecular design and modelling techniques are nowadays used, on the one hand, to support and understand experimental findings and, on the other hand, to predict the membrane performance [6].

Membrane Formation

The simplest way of polymeric membrane preparation is the solvent evaporation method, which starts from a homogeneous polymer solution and generally gives relatively thick membranes. These are good for fundamental studies of transport phenomena, but they are not very suitable for practical use because of their relatively low permeance, inversely proportional to the film thickness. Common glassy or rubbery polymers give homogeneous dense films, but in the case of block-copolymers, solvent evaporation under the proper conditions may lead to three-dimensional morphologies with unique transport properties [7].

Integrally skinned dense membranes with a thin selective layer can be prepared by phase inversion. In this method, first introduced by Loeb and Sourirajan [8], a polymer solution is brought in contact with an appropriate nonsolvent, leading the precipitation of the polymer. Up to date this method is still more an art than a science, but under the appropriate conditions the membrane formed will have a top layer with carefully controlled transport properties and a porous support layer providing its mechanical resistance. In the dry-wet phase inversion method, the polymer is coagulated after short exposure to the air and ultra-thin membranes may be formed with an effective thickness down to ca. 50 nm. [9]. In the so-called dry phase inversion method the coagulation step is not present and phase inversion takes place because the nonsolvent is already present in the casting solution and is less volatile than the solvent. This method gives somewhat less control over the thickness of the selective skin compared to the dry-wet method [10].

Composite membranes are typically prepared by a post treatment of a porous support, for instance by dip-coating with a dilute polymer solution and subsequent solvent evaporation [11], by phase interfacial polymerization [12]. In the case of hollow fibres the composite membrane can be prepared by direct spinning with a triple orifice spinneret [13].

Fig. 1. Ultra-thin skin of an asymmetric membrane [9]

Membrane Materials

The number of possible membrane materials [14] is nearly infinite and varies from highly selective less permeable glassy polyimides and polyetherimides, to polysulfones and poly­ethersulfones, to highly permeable rubbery polydimethylsiloxane. Rubbery polymers are in general more permeable than glassy polymers and are more selective for condensable species.

Still relatively few polymers are industrially used. The first generation cellulose esters are still remarkably popular, in spite of their very modest separation properties, because of their robustness and relatively stable performance under different conditions. More recently there is an increasing interest in high free volume glassy polymers like PTMSP [15], polynorbornenes and related polymers [16,17] and polymers with intrinsic microporosity, [18,19], having unusually high permeabilities and similar solution selectivity as the rubbery polymers.

A different but closely related topic is the pyrolysis of polymeric precursor membranes for production of highly permeable and selective carbon membranes [20]. Glassy perfluoropolymers like Teflon AF [21] or Hyflon AD [6] combine a high free volume with a high resistance to most organic vapours [22], opening perspectives for more demanding applications.

Attempts to exceed the performance limits of pure organic polymers, defined by the Robeson upper bound, have led to the extensive study of Mixed Matrix Membranes [23,24], in which usually inorganic porous fillers are dispersed in the polymeric matrix. Also dense fillers may influence the transport properties positively [25]. In these systems the major challenge is to achieve good compatibility between the different materials.

Transport Phenomena / Mechanisms

Transport of gases and vapours in dense polymeric membranes is normally governed by the solution-diffusion mechanism [26]. The ideal selectivity, a_i,j, is the ratio between the permeability of the species i and j, and consists of a diffusion term and a solubility term:

Selectivity in rubbery polymers is generally solubility controlled, whereas in glassy polymers it is usually diffusion controlled. An exception is represented by some high free volume polymers like PTMSP, PIMs etc., which may exhibit reverse selectivity typical for rubbers. In the majority of the cases the mixed gas selectivity is lower than the ideal selectivity, although in these high free volume polymers selective condensation of the more permeable more condensable species may effectively block the less permeable species and thus exhibit higher selectivity than the ideal selectivity. When comparing results, care must be taken that these may slightly depend on the specific measurement technique used [27]. Time lag measurements are particularly powerful since they provide both permeability and diffusivity data. When performed with care, they may reveal unusual phenomena, such as the clustering of alcohols in perfluoropolymer membranes [28].

A special case is facilitated transport, in which special additives or functional groups in the membrane actively and selectively enhance the transport of one of the species in a mixture. Examples are facilitated transport of olefins by silver salts [29], or CO₂ by amine-bearing groups [30]. Currently ionic liquids are receiving particular interest [31] as additives to enhance gas transport. Different techniques are studied to obtain stable membranes [32,33,34].

Fig. 2. Correlation of the penetrant’s critical volume and the transport properties of an ionic liquid gel membrane with 80% of IL.

The free volume plays a key role in the transport through dense polymer membranes and its knowledge is important for the understanding of the membrane performance. Different probing techniques, based on experimental or computational methods, can give an accurate picture of the FV distribution in membrane materials [35]. Glassy polymers are in a non equilibrium state and physical aging leads to a reduction of the free volume [36], and as a consequence of the permeability, accompanied by an increase in selectivity [37]. This effect is faster in thin films [38] and asymmetric membranes prepared by phase inversion can therefore have a significantly higher selectivity than the corresponding thick films [9].

The need for better membranes and membrane materials is clearly demonstrated by funding of various specific projects and networks under the EUs’ framework programmes [39].

References

1. P. Bernardo, G. Golemme, E. Drioli, Ind. Eng. Chem. Res. 48 (2009) 4638.
2. R.W. Baker, Ind. Eng. Chem. Res. 41 (2002) 1393.
3. L.M. Robeson, J. Membr. Sci. 62 (1991) 165.
4. L.M. Robeson, J. Membr. Sci. 320 (2008) 390.
5. B.D. Freeman, Macromolecules 32 (1999) 375.
6. M. Macchione, J.C. Jansen, G. De Luca, E. Tocci, M. Longeri, E. Drioli, Polymer 48 (2007) 2619.
7. K.-V Peinemann, M. Konrad, V. Abetz, Desalination, 199 (2006) 124.
8. S. Loeb, S. Sourirajan, Adv. Chem., 28 (1963) 117.
9. J.C. Jansen, M.G. Buonomenna, A. Figoli, E. Drioli, J. Membrane Sci., 272 (2006) 188.
10. J.C. Jansen, M. Macchione and E. Drioli, J. Membr. Sci., 255 (2005), 167.
11. J.C. Jansen, F. Tasselli, E. Tocci, E. Drioli, Desalination, 192 (2006) 207.
12. S.Sridhar, B. Smith, S. Mayor, B. Prathab, T.M. Aminabhavi, J. Mater. Sci., 42 (2007), 9392.
13. S.-G. Li, G.H. Koops, M.H.V. Mulder, T. van den Boomgaard, C.A. Smolders, J. Membr. Sci., 94 (1994) 329.
14. Yampolskii, Yu., Pinnau, I., Freeman, B.D. (Eds.), Materials Science of Membranes for Gas and Vapor Separation; John Wiley & Sons: Chichester, England, 2006.
15. K. Nagai, T. Masuda, T. Nakagawa, B.D. Freeman, I. Pinnau, Prog. Polym. Sci. 26 (2001) 721.
16. M. Gringolts M. Bermeshev, Yu. Yampolskii, L. Starannikova, V. Shantarovich, E. Finkelshtein, Macromolecules 43 (2010) 7165.
17. Gringol'ts, ML, Bermeshev, MV, Syromolotov, AV, Starannikova, LE, Filatova, MF, Makovetskii, KL, Finkel'shtein, ES, Petroleum Chem., 50 (2010) 352.
18. P. M Budd, B.S. Ghanem, S. Makhseed, N.B. McKeown, K.J. Msayib, C.E. Tattershall, Chem. Commun. (2004) 230.
19. P.M. Budd, E.S. Elabas, B.S. Ghanem, S. Makhseed, N.B. McKeown, K.J. Msayib, C.E. Tattershall, D. Wang, Adv. Mater. 2004, 16, 456.
20. C.W. Jones, W.J. Koros, Carbon 32 (1994) 1419.
21. H. Zhang, S. G. Weber, Teflon AF materials, Top. Curr. Chem. 308 (2012) 307.
22. I. Pinnau, Z. He, A.R. Da Costa, K.D. Amo, R. Daniels, US Pat. 6,361,582 B1 (2002).
23. M.A. Aroon, A. F. Ismail, T. Matsuura, M. M. Montazer-Rahmati, Sep. Purif. Technol., 75 (2010) 229.
24. T.-S. Chung, L. Y. Jiang, Y. Li, S. Kulprathipanja, Prog. Polym. Sci., 32 (2007) 483.
25. Merkel, T.C., Freeman, B.D., Spontak, R.J., He, Z., Pinnau, I., Meakin, P., Hill, A.J., Science, 296 (2002) 519.
26. J.G. Wijmans, R.W. Baker, J. Membr. Sci., 107 (1995) 1.
27. K. Friess, J.C. Jansen, O. Vopicka, A. Randová, V. Hynek, M. Sípek, L. Bartovská, P. Izák, M. Dingemans, J. Dewulf, H. Van Langenhove, E. Drioli, J. Membr. Sci., 338 (2009) 161.
28. J.C. Jansen, K. Friess, E. Drioli, J. Membr. Sci., 367 (2011) 141.
29. S.U. Hong, J.H. Jin, J. Won, Y.S. Kang, Adv. Mater. 12 (2000) 968.
30. H. Matsuyama, A. Terada, T. Nakagawara, Y. Kitamura, M. Teramoto, J. Membr. Sci., 163, (1999) 221.
31. P. Scovazzo, J. Membr. Sci. 343 (2009), 199.
32. J.C. Jansen, K. Friess, G. Clarizia, J. Schauer, P. Izák, Macromolecules. 44 (2011) 39.
33. J.E. Bara, E.S. Hatakeyama, D.L. Gin, R.D. Noble, Polym. Adv. Technol.,19 (2008) 1415.
34. O. Vopička, V. Hynek, K. Friess, P. Izák, Eur. Polym. J., 46 (2010) 123.
35. J.C. Jansen, M. Macchione, E. Tocci, L. De Lorenzo, Yu.P. Yampolskii, O. Sanfirova, V.P. Shantarovich, M. Heuchel, D. Hofmann, E. Drioli, Macromolecules 42 (2009) 7589.
36. Y. Huang, X. Wang, D.R. Paul, J. Membr. Sci., 277 (2006) 219.
37. Y. Huang, D.R. Paul, Polymer 45 (2004) 8377.
38. P Y. Huang and D. R. Paul, Macromolecules, 39 (2006) 1554.
39. FP7 Project NMP3-SL-2009-228631: DoubleNanoMem - Nanocomposite and Nano­structured Polymeric Membranes for Gas and Vapour Separations; FP7 Project NMP3-SL-2009-228652: SelfMem - Self-Assembled Polymer Membranes; FP7 Project NMP3-SL-2009-228701: NASA-OTM - NAnostructured Surface Activated ultra-thin Oxygen Transport Membranes; FP6 Network of Excellence NMP3-CT-2004-500623: NanoMempro - Expanding membrane macroscale applications by exploring nanoscale material properties.