tures in the range of 540 °C to 1090 °C (1000 °F to 2000 °F). The presence of steam is likely to accelerate the degradation process. Requirements for high separation efficiency impose further materials constraints in terms of pore size distribution and mean pore size in the membrane. A high degree of control during membrane fabrication is necessary to achieve the desired microstructural features. Ceramic membranes consist of a porous support a few millimeters thick, a porous intermediate layer 10 to 100 microns thick with pore diameters in the range of 0.05 to 0.5 microns,2 and the separation layer with a thickness of 1 to 5 microns (Burggraaf et al., 1989). Generally, the separation layer must have pore diameters less than 10 nm for effective separation of gaseous components by diffusion (Krishnan et al., 1993); in some cases a mean pore size of 2.5 nm may be necessary. 3
Current commercially available membranes do not meet all performance requirements for cleanup of coal-gas and flue gas streams, although several manufacturers produce inorganic membranes for micro- and ultrafiltration applications, and some of these have pore diameters less than 10 nm and are capable of separating gaseous components. However, extensive membrane technology has been developed over the past 40 years for nuclear gaseous diffusion applications, and alumina and zirconia membranes have been used for the separation of uranium hexafluoride (UF6) isotopes for the nuclear industry since 1950 (Krishnan et al., 1993). Current DOE programs to develop ceramic membranes for coal-based applications are attempting to leverage this existing knowledge base. Investigators at the Oak Ridge Gaseous Diffusion Plant have produced alumina (ceramic) membranes with pore radii as small as 70 nm. Membrane separation tests have demonstrated a capability to separate hydrogen from gas mixtures (DOE, 1992).
Membrane material research opportunities specific to coal-based systems involve primarily the development of inorganic membranes for separation of coal-derived products and impurities at elevated temperature and in corrosive environments. Improvements can be anticipated in the selectivity and separation efficiency based on enhanced understanding of the relationship between pore structure and the physical chemistry of molecular separations (NRC, 1993). Opportunities also exist for the development of membranes with improved resistance to the environments characteristic of coal-based systems, such that operating lifetimes can be extended. Given the likely increase in concerns over greenhouse gas emissions, the investigation and demonstration of cost-effective separation of methane from very dilute coalbed methane streams using membrane techniques also merit some attention.