UC Davis

Nanocrystalline oxides have been studied for their excellent mechanical and optical properties as well as their great chemical stabilities. These properties make them ideal materials for applications such as armored windows, laser gain media, and catalyst supports. Despite their desirable properties, we have a limited understanding of their mechanical behavior at small grain sizes. This makes it difficult to predict and tailor their properties for commercial applications. Furthermore, this class of materials is inherently metastable due to the sizeable energy contributions from their large interfacial areas. This high energy makes them susceptible to coarsening and grain growth which would diminish the properties that make them desirable. This work will focus on addressing these two issues by studying nanocrystalline zinc aluminate (ZnAl2O4) as a model material. This material was chosen due to its relatively high thermal conductivity which makes it a more attractive material, particularly for laser gain media and catalyst supports, than other oxides. The first goal of this work will address the interfacial stabilities of zinc aluminate nanoparticles and fully dense samples. Solid-solid (surface) and solid-vapor (grain boundary) interfacial energies were tailored using dopant segregation as predicted by molecular dynamics simulations: atomistic simulations on a nanoparticle and two grain boundary structures were used to assess the segregation behavior of four dopants [Sc3+ (74.5 pm), In3+ (80.0 pm), Y3+ (90.0 pm), and Nd3+ (98.3 pm)]. The candidate dopants were chosen to induce segregation by maximizing the elastic strain in the lattice (i.e., large ionic radii). All four dopants were estimated to have favorable segregation energies to surfaces and grain boundaries with Y3+ consistently having the lowest energies at interfaces. Accordingly, undoped and doped (0.5 mol% Y2O3) were prepared to experimentally compare their interfacial energies and coarsening behaviors. Surface energies were estimated by water adsorption microcalorimetry which revealed that surface stabilities were effectively improved in doped nanopowders, as predicted in simulations. This behavior was correlated with coarsening inhibition and lower estimated diffusion coefficients, indicating Y3+ improves zinc aluminate stability against coarsening via kinetics and thermodynamics. Similarly, differential scanning calorimetry was used to measure grain boundary energies as a function of grain size for fully dense samples from both compositions. Grain boundary energies were lower for doped samples at each grain size studied which not only predicts limited grain growth for dense samples, but also elevated hardness and toughness. The mechanical performance (i.e., hardness and toughness) of Y-doped and undoped samples were compared to identify the effects of improved grain boundary stabilty. Hardness and toughness values were statistically similar for both compositions, indicating the dopant had negligible effects on zinc aluminate mechanical properties using a concentration of 0.5 mol%. The only exception was at larger grain sizes (above 25 nm) where Y-doped samples had significantly higher hardness. This was presumed to be a result of higher concentrations of Y3+ in these samples due to lower grain boundary area, hence limiting grain boundary mediated deformation or improving dislocation pinning from dopants segregated to dislocation cores. A more pronounced effect on mechanical properties was found when comparing stoichiometric zinc aluminate (Al:Zn = 2.01:1) to Al-rich zinc aluminate (Al:Zn = 2.87:1). Stoichiometric zinc aluminate exhibited elevated hardness with decreasing grain size until grain sizes of ~20 nm, while Al-rich samples underwent further hardening to grain sizes near 12 nm. Distinct cracking patterns were observed in both samples, suggesting the excess Al postponed the softening by stabilizing high-energy grain boundaries. The results from this work show that interfacial stability can be enhanced significantly by doping zinc aluminate with Y3+. This can be used to effectively tune zinc aluminate coarsening behavior at the nanoscale. The present work also showed that excess Al can be used to tune the grain size hardening behavior of nanocrystalline zinc aluminate; however, it remains unclear whether a similar effect can be induced by doping with Y3+ due to the low concentrations used here. These conclusions will prove beneficial as we employ nanocrystalline ceramics for the next generation of armored windows and catalysts.