|Search for content and authors|
Evolution of grain boundaries and interfaces during sintering and grain growth in doped oxide ceramics: from nanosized powders and atomistic simulations to transparent ceramics and tailored interfaces
|Paul Bowen 1, Uli Aschauer 1, Steve C. Parker , Sandra Galmarini , Dino Spagnoli|
1. Ecole Polytechnique Federale de Lausanne (EPFL), Lausanne 1015, Switzerland
Ceramic properties are inherently linked to their chemical composition and microstructure. The transformation of a ceramic powder from a loose collection of particles into a sophisticated ceramic piece, e.g. dental implants, hip implants, multilayered capacitors, cutting tools or transparent ceramics for laser applications depends heavily on our capacity to control and modify the interfaces during processing and sintering. The use of dopants, which can segregate to both surfaces and grain boundaries during powder synthesis and sintering, are often used to influence these interfacial regions. The significant improvement in analytical techniques namely high resolution transmission electron microscopy (HRTEM, EELS) has given us much insight into the general location of dopants in ceramic microstructures but we still lack the atomistic knowledge which should enable us to design grain boundaries and interfaces for specific properties and applications. A key feature in being able to tailor microstructures from powder processing routes is our capacity to synthesise well controlled powders. The powder characteristics of key importance are size, size distribution, morphology and purity. A brief description of an innovative tubular reactor, the segmented flow tubular reactor (SFTR), that produces powders with better characteristics for ceramics will be presented and the applicability demonstrated with examples of scale-up, powder quality and reproducibility for BaTiO3.
One route towards the goal of tailored microstructures is to use the ever increasing computer power to model at an atomistic level these key interfacial regions. Controlling grain growth and grain boundary composition is a key factor for the production of transparent polycrystalline ceramics such as yttrium aluminium garnet (YAG) and alumina. For applications in high performance lasers YAG is doped with neodymium (Nd), single crystals are limited to around 1%, whereas polycrystalline YAG can accommodate more. To increase laser power but still be able to produce transparent ceramics the location and effect of increased Nd levels is a key issue for progress. The effect of dopants in alumina such as Y for creep resistance, Mg for grain growth control is well established but the detailed mechanisms behind the effects are poorly understood. The level of dopant whereby there is segregation to grain boundaries and eventually the formation of a secondary phase is linked to the type of grain boundary present and their relative surface area, which in turn is linked to the grain size - one of the parameters we are trying to control. Using energy minimisation techniques the location, dopant concentration, grain boundary structure, surface and interfacial energies have been investigated for a number of crystallographic surfaces in YAG and alumina giving us further insight in our quest towards tailored interfaces. Validation of the modelling approach will be demonstrated by comparison with available experimental data and in particular grain boundary segregation/precipitation maps.
Presentation: Invited oral at E-MRS Fall Meeting 2006, Symposium C, by Paul Bowen
See On-line Journal of E-MRS Fall Meeting 2006
Submitted: 2006-05-11 15:19 Revised: 2006-06-07 16:15