Carbon-based Membranes for Separation Processes

This chapter gives a historical overview for the development of inorganic and carbon molecular sieve membranes.

The transport mechanism exhibited by most of carbon membranes is the molecular sieving mechanism. The carbon membranes contain constrictions in the carbon matrix, which approach the molecular dimensions of the absorbing species. In this manner, they are able to separate the gas molecules of similar sizes effectively. Another transport mechanism of carbon membrane is selective adsorption-surface diffusion mechanism. Adsorption-selective carbon membranes separate non-adsorbable or weakly adsorbable gases (O₂, N₂, CH₄) from adsorbable gases, such as NH₃, SO₂, H₂S and chlorofluorocarbons (CFCs).

The transport through CMSMs is often studied by using the solution (sorption)-diffusion model, because of the simplicity of the model. In this chapter the above model is applied for single gas permeation, and then more complicated model is applied to predict the separation of binary gas mixtures.

Carbon membranes can be divided into two categories: unsupported and supported carbon membranes. Unsupported membranes have three different configurations: flat (film), hollow fibre and capillary, while supported membranes consist of two configurations: flat and tube. Membranes have been prepared from different polymeric materials under different heating protocols and their performance investigated by various researchers. This chapter summarizes the preparation method of CMSMs in each category and the performance of the CMSMs so prepared in different gas separation applications.

One of the important factors that govern the performance CMSMs are precursor selection. Typical polymeric precursors are polyacrylonitrile, polyimide, phenolic resin, and polyfurfuryl alcohol. Detailed descriptions are made to prepare CMSMs from these precursors. Other factors are pre-treatment of precursor, pyrolysis process and post-treatment. Among those, pretreatent includes oxidation pretreatment, chemical treatment, stretching and others. Post-treatment includes post oxidation, chemical vapour deposition, post pyrolysis, fouling reduction and coating. In this chapter, detailed discussions are made to optimize each factor for obtaining CMSMs of the highest quality.

In this chapter a detailed description is made for the CMSM preparation, characterization and testing procedure. Two examples are given; one for the hollow fiber CMSM made from polyacrylonitrile (PAN) and the other for the flat sheet CMSM made from various types of polyimides (PIs). Each example has been adopted from the PhD thesis written in the laboratories that are currently active in CMSM research. It is the authors’ intention to make readers ready to set up their equipment for the CMSM research by following the procedure described in this chapter.

In this chapter, typical examples of CMSM characterizations are shown. The characterization methods are classified into permeability measurement and physical characterization. Permeability measurement is further classified into gas and liquid permeabilities. Most of CMSMs have been characterized by gas permeation and separation, and many examples have already been shown in the preceding chapters. Therefore, only examples for characterization by liquid permeation experiments are included in this chapter. Physical characterization includes TGA, WAXD, SEM, TEM, AFM, FTIR and pore size measurement by adsorption. There are many other characterization methods, but the above methods have been most commonly used in the CMSM literature.

Membrane module construction is seldom reported in the literature and especially this is the case for CMSMs. There are only two commercially available modules currently, one from Carbon Membranes Ltd (Israel) with hollow fiber configuration and the other from Blue Membranes GmbH (Germany) with honey comb configuration. In this chapter, a detailed description of the latter module is given. Information on the commercial modules based on hollow fiber CMSMs is scarcely available. Therefore, the apparatus used by Haraya and co-workers to prepare asymmetric capillary CMSMs is described.

Recently Holt et al. developed carbon nanotube (CNT) membranes whose pore sizes were smaller than 2 nm by using a microelectro-mechanical system (MEMS)-compatible fabrication process and showed gas and liquid permeation rates orders of magnitude higher than the calculated values based on either Knudsen or Poiseiulle flow mechanism, due most likely to extreme smoothness of the inner wall of carbon nanotubes. This proved the theoretical prediction made earlier by the molecular dynamics simulation. Several attempts have been made since the work of Holt et al. to fabricate the carbon nanotube membrane in a larger scale aiming at industrial applications. There have been other developments in carbon related membranes particularly in the field of nano-technology. One of such developments is carbonization of nanofibers for the membrane adsorption process. Carbon particles and especially carbon nanoparticles have also been incorporated in the mixed matrix membranes to enhance gas separation performance of the membranes.

Most of applications of CMSMs are in the area of gas separation and RO/NF/UF/MF. In particular, interest is growing in the area of light alkenes/alkanes separation. In this chapter, however, emphasis is on CMSMs applications in other processes than gas separation and RO/NF/UF/MF. Those processes include vapor separation, pervaporation, fuel cell applications, water treatment, membrane reactor etc. The preparation, characterization and performance of carbon-based materials in the above applications are described in detail with some examples.

There are only articles concerning the economic evaluation of carbon related membranes. Two articles were chosen in this chapter to demonstrate detailed cost analysis involved in two applications of CMSMs; (1) for the recovery of hydrogen from the natural gas network (Granger, H., Int J Hydrogen Energy 33(9):2379–2388, 2008) and (2) CO₂ removal from landfill gas (Granger, H., J Membr Sci 306:307–317).

This chapter is based on an article written in 2003 by Ismail. He emphasized the importance of inorganic membranes and, in particular, carbon membranes in the future membrane and membrane process development. Although the statistics used in the article have become somewhat old, several important statements made by Ismail seem still valid for the trend in the current membrane technology and also for the future development of inorganic and carbon membranes. Hence, they are reproduced in this chapter with some modification.