Abstract
This study is concerned with the production of yttria-stabilised zirconia dip-coated layers on porous ceramic substrates. The important parameters involved in the deposition of a particulate layer of controlled thickness on a substrate and the maintenance of layer integrity during drying have been identified. The porous ceramic substrate to be coated was a cell support for a Rolls Royce Integrated Planar-Solid Oxide Fuel Cell (IP-SOFC). In order to specify the requirements of the coating suspension and subsequent layer, it was necessary to characterize the substrate in terms of surface roughness, porosity, surface pore size distribution and defect structure. Thus a dipped green layer thickness of ~ 100 mum and suspension median particle size of ~ 3 mum were proposed for an ideal aqueous dip suspension. The use of a bimodal particle size distribution with a fine; coarse particle size ratio of ~1:8 and a mass fraction ratio of 1:3 was also identified as potentially being beneficial to layer particle packing density. Initial dipping experiments showed that multiple dipping and drying steps increased the level of cracking in layers. Drying orientation was important, with inversion of freshly dipped samples leading to uncontrollable layer thicknesses. This was through unabsorbed suspension draining back down over the freshly dipped layers, forming drip marks and resulting in an increased level of cracking in these regions. Three series of aqueous suspensions of differing solids volume contents (20-40 vol%) were produced from two powders of differing particle size distribution (nominal D50 particle sizes of 0.5 and 3 mum) to produce two single powder suspension series and one mixed powder suspension series. The suspension particle size distributions and viscosities were characterised prior to dipping. The majority of suspensions showed little or no agglomeration. All measurable suspensions were pseudoplastic. A reduction in solids content reduced suspension viscosity but not the level of pseudoplasticity. The level of pseudoplasticity was inversely related to the particle size. The mixed suspension, although showing intermediate behaviour, showed greater similarity in viscosity and level of pseudoplasticity to the 3 mum rather than the 0.5 mum powder suspensions. Dipping using the three suspension series and different withdrawal velocities was carried out. Layer thicknesses and microstructures of the dip coated substrates were analysed. There was no discernible relationship between sample withdrawal velocity and layer thickness. The important suspension parameters identified for controlling layer thickness were viscosity and solids content, with 30 vol% suspensions producing ~85-145 mum layer thicknesses. The 30 vol% mixed suspension gave crack-free layer thicknesses of ~ 85 mum with high levels of particle packing density, substrate surface planarization and few defects. This was the most suitable suspension of those studied and provides a basis for future research.