Of all the metal oxide ceramics, the most thermodynamically stable (as can be seen on an Ellingham diagram) include lithia, lime, alumina, and zirconia; however, lithia and lime are hygroscopic, and this leaves alumina and zirconia as the most stable of the traditional ceramic oxides. Refined primarily from zircon (zirconium silicate), zirconia is remarkably stable and resistant to corrosion from liquid metals. It also has very good mechanical properties which make it ideal for use in cutting tools, milling media, and extrusion dies at low temperatures while having a good degree of ductility at high temperatures.
Thermodynamics of Zirconia
Cubic zirconia, known for its similarity to diamond as a gemstone, has a fluorite structure where every other fluorite cell has a zirconium ion to preserve charge balance; upon cooling to below 2369°C, the structure stretches into a tetragonal form. However, further cooling to around 1170°C results in a tetragonal to monoclinic transformation which, due to the very different geometries of the crystals, results in a 4.5% volumetric expansion and 12% shear in the crystal.
This phase transformation is what makes sintering pure zirconia very difficult; sintering temperatures are generally in the tetragonal regime, and the cooling causes the monoclinic transformation which generally shears the sample to pieces.
Stabilization of the cubic phase came into being in the 1930s by the addition of one of several possible aids including MgO, Y2O3, CaO, and CeO2. By including enough of these additives into the structure so as to avoid the monoclinic phase field, the slow kinetics of the phase transformations can be used to avoid monoclinic transformation by remaining in two-phase regions. Furthermore, in the yttria-zirconia system, a cubic phase is stable at room temperatures beyond 8% Y2O3.
What exactly causes this stabilization? In the case of yttria-stabilized zirconia (YSZ), the yttrium cations are larger than the zirconium cations (1.019 Å versus 0.84 Å according to Shannon 1976) and prevent the crystal from collapsing to a monoclinic structure. The yttria provides a kinetic barrier that inhibits the phase transformation, as the cubic to tetragonal transformation requires that yttria diffuse out of the regions where the tetragonal phase is to form (since the cubic phase is more yttria-rich than the tetragonal). Since this diffusion of yttria must precede the formation of the tetragonal phase, cooling fast enough to prevent the diffusion from occurring results in tetragonal ZrO2.
Moderate cooling rates of YSZ result in limited nucleation of tetragonal zirconia in a cubic zirconia matrix. At room temperature, this tetragonal phase would readily transform into the monoclinic process via a diffusionless martensitic transformation, but the volumetric expansion associated with this transformation would require these tetragonal regions to perform work on the cubic matrix. The energy demands of this transformation are insufficient, and thus the tetragonal phase can persist in the cubic matrix at room temperature.
However, when a crack forms and propogates through such a structure, the crack tip opens enough space for these tetragonal regions to expand and invert to the monoclinic phase. Because this transformation is martensitic, it occurs at the speed of sound and instantly closes the crack as the tetragonal phase expands into the monoclinic--this effect significantly toughens the zirconia by closing any cracks that form, and this is the principle by which transformation-toughened zirconia gains its improved mechanical properties.
Zirconia is an immensely important material in ceramic science due to its various properties. It is a principal constituent of AZS, a very important refractory material, is of great interest as a solid-oxide fuel cell material due to its high amount of vacancies introduced by yttria doping, and is incorporated as a thermal coating on almost every gas turbine engine blade.