An In Situ Approach to Create Porous Ceramic Membrane: Polymerization of Acrylamide in a Confined Environment Xinwei Chen z and Liang Hong w,y z Department of Chemical and Biomolecular Engineering, National University of Singapore, Singapore 117576, Singapore y Institute of Materials Research and Engineering, Singapore 117602, Singapore A new processing method for the fabrication of porous ceramic membrane has been developed via a three-step procedure: by uniformly distributing a solid vinyl monomer (e.g., acrylamide) into a green object of ceramic through wet chemistry mixing and compression molding; polymerizing the monomer in a highly compact surrounding, leading to the formation of embedded chain assemblies of polyacrylamide; and removing the polymer via carbonization and calcination. This in situ pore-forming strategy grants less tortuous and well-interconnected pore chan- nels in contrast to the approach of using polymeric porogen such as starch or cellulose. The weight% of initiator and duration of polymerization were scrutinized using thermal analysis, electron microscopy, and Hg porosimetry to understand their influences on the porous structure of sintered ceramic membrane, e.g., a thin Yttria-stabilized zirconia disc, and ultimately its gas perme- ability. The advantage of this pore-forming method lies in the fact that the monomer can be homogenously distributed in the green object in a confined space and the polymer chains formed during the in situ solid state polymerization can develop space occupancy through chain penetration and association, thus leaving behind interconnecting pore channels and more open pores after they were removed eventually. In this study, it has been shown that the resulting porous ceramics manifests a marked improvement (20–80%) in gas permeability over those fabricated by using starch as the pore-former. Furthermore, the porous ceramics fabricated by the new method exhibited higher rapture resistivity on the similar porosity basis. I. Introduction P OROUS ceramic objects have gained remarkable popularity in various engineering and industrial applications over the years due to their thermal and chemical stability in severe envi- ronments such as high temperatures, redox atmospheres, and corrosive liquids. Moreover, ceramic materials offer relevant mechanical reliability and durability necessary for these opera- tions. Major applications of porous ceramics include supports for catalysts, membranes, and filters for waste water purification or bacteria removal, separators in solid oxide fuel cells (SOFCs), and biological media. 1–3 Traditionally, a porous ceramic object is created by sintering a powder assembly in a certain shape (known as a green ceramic object), which is prepared by powder molding with the aid of a suitable polymer binder, at a nondensification temperature. 4 A sintered product could exhibit a high porosity, but it very often lacks sufficient mechanical strength necessary for real industrial applications. Recently, some new processing techniques such as pulse electric current sintering, 5–7 freeze-dry processing, 8–10 and hydroxide decomposition methods 11 were developed to meet various requirements. However, the use of pore-formers such as carbon black or polymer filler, e.g., starch, methylcellulose, and carbon wax, to achieve high porosities remain a popular choice in ceramic processing due to their low cost and easy han- dling. 3,12–16 These combustible components are easily removed by oxidation, leaving void pores to the sintered ceramics. There is normally a percolation volume fraction of the pore- former used against fluid permeability of the sintered ceramic object finally obtained. Below this percolation value, fine pores formed exist separately, and pore channels are severely discon- nected between one another; a low permeability is, therefore, presented. However, with a volume fraction of the pore-former above the percolation value, the desired permeability can be re- alized but at the cost of mechanical properties to a certain ex- tent. A trade-off between these properties is usually carried out for industrial applications. The root cause of this limitation lies in the fact that the pore-former used does not form a continuous phase at a relatively low loading level in the green object due to thermodynamic incompatibility between ceramic particles and the pore-former (or porogen). In general, using an insolu- ble porogen like carbon black, agglomeration of particles in the green object prevents its pore-forming effect. Similarly for the use of polymer porogen, albeit a relatively lower percolation value could be achieved due to an inclusion by the wet chemical means, the self-coiling tendency of the polymer in the green object could still result in a significant reduction in the pore- forming effect. In this study, we demonstrated the concept of forming a poly- meric porogen in situ in a ceramic green body—namely the pro- duction of polymer from its monomer in a compact environment. Figure 1 illustrates the idea of this pore-forming strategy. With this methodology, the drawbacks associated with the conven- tional pore-forming method will be largely lessened, primarily because a smaller-sized monomer can achieve a higher degree of uniformity in the green object than its large-molecular-weight counterpart. Furthermore, shorter molecular chains are gener- ated in this constrained polymerization environment, and thereby minimizing the effects of random coiling and entanglement, which are the thermodynamic tendencies of long polymer chains. Therefore, the generation of thin pore channels and lowering of the percolation threshold can be achieved simultaneously. Polymerization of vinyl monomers in a ceramic environment has been exploited in the gel-casting process, 17,18 a ceramic- forming process in fabricating complex-shaped bodies. How- ever, this type of polymerization differs from our present system in two ways: the polymerization is carried out in a slurry mixture and a high loading of monomer is required. The monomer and initiator chosen in this study are acrylamide and ammonium W.-C. Wei—contributing editor The authors would like to thank the National Research Foundation for their financial support (R-279-000-261-281). w Author to whom correspondence should be addressed. e-mail: chehongl@nus.edu.sg Manuscript No. 26083. Received April 5, 2009; approved August 4, 2009. J ournal J. Am. Ceram. Soc., 93 [1] 96–103 (2010) DOI: 10.1111/j.1551-2916.2009.03373.x r 2009 The American Ceramic Society 96