Toward Knowledge-Based Grain-Boundary Engineering of Transparent Alumina Combining Advanced TEM and Atomistic Modeling
Transition element dopants (e.g., Y, La) are commonly used as sintering aids in polycrystalline alumina ceramics, which segregate to the grain boundaries and control the grain-boundary mobility. However, due to the extremely thin (<2nm) layer of segregated dopants, the experimental characterization of the segregated alumina grain boundaries is a complex task. Computational studies have focused only on tilt grain boundaries, which are only a small fraction in a sintered alumina sample. In this study, a quantitative characterization of the segregation of Y and La at general high angle grain boundaries in transparent alumina is carried out using a unique combination of advanced TEM and near coincidence grain-boundary atomistic simulations. The result show that high angle grain boundaries may lead to enhanced grain growth in comparison to symmetric tilt twin grain boundaries due to the reduced configuration entropy for dopant segregation and higher order grain-boundary complexions. On the other hand, multidoping with different dopants was shown to be more beneficial than single doping due to its contribution in increasing the configurational entropy for segregation. The advanced TEM analysis showed Y and La distributions and concentrations on a series of general grain boundaries in very good agreement with the atomistic simulations. This validation of atomistic modeling technique used in this study means, as it a generic method, can be used as a predictive tool to design ceramic microstructure and properties.