1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9

Gas and vapour separation in polymeric membranes


Polymer membranes for gas separation keep getting massive interest in chemical engineering research [1]. They can be a nightmare for the process engineer, who prefers to see them as a black box, but they are among the most interesting research topics in membrane technology from the Materials Science point of view. Due to their macromolecular structure, their structural and conformational variety (e.g. molar mass distribution, branching...) and an inherent non-equilibrium state, the properties of polymer membranes depend on their preparation history and they may continue to change over time. Furthermore, the intimate contact between the permeating species and the polymer matrix causes that the performance varies as a function of the operating conditions. Such behaviour requires detailed knowledge of polymer structure-property relationships and thorough studies to understand the underlying phenomena and to anticipate the possible behaviour of novel systems.

Detailed and systematic studies significantly enhanced the understanding and allowed the control of transport phenomena in polymer membranes of various types. Pure gas and vapour permeation measurements in a time-lag setup yield values for the permeability, diffusion and solubility constants of gases in dense polymeric membranes while a variable volume setup with Mass Spectrometric gas analyzer provides mixed gas permeability data.


Polymers of intrinsic microporosity (PIMs)

In collaboration with the University of Cardiff, it was recently demonstrated that increasing stiffness of the polymer chain in PIMs (Fig. 1) effectively shifts the behaviour from the solution diffusion mechanism more and more towards that of a molecular sieving mechanism [2,3].

 Figure1. PIMs with increasing rigidity of the polymer backbone

The microporous PIMs easily retain residual casting solvent, which is removed by soaking with methanol and subsequent drying. After this treatment the membranes were shown to undergo aging, resulting in a gradual decrease of permeability. Mixed matrix membranes (MMMs) were designed to reduce this aging. MMMs based on fully organic nanofillers of so called cage molecules were designed in collaboration with the Universities of Manchester and of Liverpool to favour polymer/filler compatibility. Porosity of such structures causes a significant increase of permeability while they are rigid enough to limit the permeability decline in time [4].

Ionic liquid membranes


Addition of ionic liquids (ILs) was used to tailor the permselective properties of membranes prepared from intrinsically low permeable polymers [5]. The ILs plasticize the polymer and in semi-crystalline polymers they also reduce crystallinity, resulting in a drastic increase in permeability.

Since solubility of the gases is nearly constant, the permeability increase as a function of the IL content derives mainly from the higher diffusion coefficient. Correlation with the elastic or modulus (Fig. 2) was used for the first time to show the transition from diffusion controlled to solubility controlled permeation  [6].

At high IL content solubility controlled permeation takes place and such membranes are highly permeable for vapours and readily condensable gases.

Figure 2. Correlation of the permeability and young’s modulus as a function of the IL content, showing the transition from diffusion controlled to solubility controlled permeability

 Perfluoropolymer membranes

Figure 3. Structure of the glassy perfluoropolymer Hyflon AD

Studies on amorphous glassy perfluoropolymers revealed a totally different behaviour than the above two membrane types. These chemically and thermally resistant polymers also belong to the medium to high free volume polymers in which the FV distribution can be estimated for instance by photochromic probes [7]. These polymers are totally insoluble in most common organic solvents and they do not swell significantly in alcohols. Transport of the latter is dominated by the stronger alcohol-alcohol interactions than the polymer-alcohol interactions. Time lag measurements revealed very unusual behaviour of alcohols, witnessing a strong clustering of alcohols in the void space [8]. Appropriate data elaboration procedures and fitting of the transient with multiple time lag curves give a quantitative approximation of the average alcohol cluster size.



1. P. Bernardo et. al. Ind. Eng. Chem. Res., 48 (2009) 4638-4663. DOI: 10.1021/ie8019032
2. M. Carta et al., Science, 339 (2013) 303-307.
DOI: 2010.1126/science.1228032
3. C.G. Bezzu et al., Adv. Mater. 24 (2012) 5930-5933. DOI: 10.1002/adma.201202393
4. A.F. Bushell et al., Angew. Chem., Int. Edition, 125 (2013) 1291-1294. DOI: 10.1002/anie.201206339
5. J.C. Jansen et al., Macromolecules. 44 (2011) 39-45. DOI: 10.1021/ma102438k6.
6. K. Friess et al., J. Membr. Sci.415-416 (2012) 801-809. DOI: 2010.1016/j.memsci.2012.05.072
7. J.C. Jansen et al., Macromolecules, 42 (2009) 7589-7604. DOI: 10.1021/ma901244d
8. J.C. Jansen et al., J. Membr. Sci. 367 (2011) 141-151. DOI: 10.1016/j.memsci.2010.10.063


Istituto per la Tecnologia delle Membrane CNR-ITM
Via P. Bucci cubo 17/C c/o Università della Calabria
87036 Rende (CS)   Phone: +39 0984 492027

Codice Fiscale 80054330586  -  Partita IVA 02118311006 
Il Cnr è soggetto allo split payment









Indirizzo Posta Elettronica Certificata (PEC) istituzionale: protocollo.itm@pec.cnr.it

2019 CNR-ITM - WebMaster Romeo DE LUCA