| A new concept of interface complexions has recently been proposed. It recognizes grain boundaries as distinct "phases", which have complex and important three-dimensional structure at the atomic level that depends on thermodynamic variables such as composition, temperature, pressure, and crystallographic misorientation. In the present research program, we have performed a fundamental sintering study of Yb/Er-doped and undoped Y2O3 fine particles. Over a narrow temperature range an unusual abrupt increase in grain growth rate and a significant discontinuity in grain boundary mobility have been identified for Yb/Er-doped Y2O3 samples. Based on chemical analysis such a transition is postulated to be associated with impurities (Ca, Si, etc.) segregated to the grain boundary. Consequently, the effect of controlled calcia doping and silica doping on the grain growth behavior of very pure and dense yttria samples in both oxidizing and reducing atmospheres has been investigated in detail for the first time. Calcia doping was found to promote the grain boundary mobility in yttria, which is argued to be responsible for the grain growth transition observed in Yb/Er-doped Y2O3. An unexpected occurrence of abnormal grain growth was observed for the first time in Y2O3 doped with a trace amount of calcia. Conversely, silica has been identified as a grain growth inhibitor in yttria.The structure and chemistry of the grain boundaries in various yttria systems has been correlated with the grain growth kinetic data through a detailed microscopy analysis by HRTEM, HAADF-STEM and EELS, with the aim of identifying and understanding the role of grain boundary complexions in the sintering of yttria ceramics. Analogous to alumina, very pure undoped yttria can be categorized as a type II grain boundary complexion. The grain growth transition in Yb/Er-doped Y2O3 was interpreted as a transition from complexion type III to type IV. In 100 ppm Ca-doped Y2O 3, grain boundary complexion type V was associated with the abnormally growing grains while either type III or type IV was associated with the normal grains. The slowest type of grain boundary complexion, type I, was predicted for Si-doped Y2O3. The idea of grain boundary complexions also provides an alternate explanation for the mechanisms of the two step sintering technique. This new grain boundary complexion concept has important practical implications such as to engineer the properties of grain boundaries and hence to promote the formation of a dense, fine-grained and optically transparent yttria ceramic material by the selection of proper dopants and sintering temperatures which are associated with a less disordered grain boundary complexion. |
