Search

Association energies for ferric ion magnesium vacancy dimers and trimers derived from shell-model calculations are combined with oxidation-reduction equilibria data in concentrated crystalline solutions to derive equilibrium defect concentrations for relatively dilute solutions. The several simultaneous non-linear equations required are solved with a numerical iteration method. Calculated results are compared with experimental determinations of solute concentration and oxygen pressure. It is found that with a slight reduction of the shell-model energies, the defect model is in good accord with experimental data at temperatures greater than 1300° C. At lower temperatures, agreement is poor and it is surmised that solute-vacancy clusters may be important.

We have used the HADES program to calculate the energies of various defect aggregates found in MgO containing Al3+ and Fe3+ solutes and compensating cation vacancies. Calculated energies of substitution are compared with heats of solution derived from phasediagram data; from the accuracy of these results, we deduce the validity of the models used for the lattice simulations. We find that our models provide a satisfactory description for Al3+ but are a less precise representation of crystals containing Fe3+; the models used, however, bracket a reasonable range of solute behaviour and important trends are unaffected by reasonable changes in the interionic potentials. The simplest vacancy-solute dimer can have either a <1 0 0> or <1 1 0> orientation; the two constituent defects are closest when the dimer has a <1 1 0> axis, but the <1 0 0> dimer is more stable because of the large displacement and polarization of the oxygen ion between the trivalent ion and vacancy. Trimers with either orientation are about twice as stable as the corresponding dimers. Complex aggregates of solutes and vacancies, which adopt configurations that form nuclei of the mixed-oxide spinel structure, are even more stable and the stability increases with cluster size. Thus we conclude that such clustering is an important phenomenon at low homologous temperatures. Calculated interstitial formation energies in MgO are large (>10eV) and our results for the activation energies for solute motion are of the order of 2 eV.

Isochronal annealing studies were carried out on slow-cooled MgO single crystals. The annealing behaviour of the hydroxyl defects is consistent with a defect model in which brucite precipitates dissolve in MgO forming OH-V-OH defects with a heat of solution equal to 0.45 eV. These OH-V-OH defects further dissociate into VOH-defects and hydroxyl ions with a dissociation energy of 0.82 eV.

High-temperature electron paramagnetic resonance (EPR) spectroscopy with a CO2 laser heat source was used to study the 600 to 1200° C point defects and precipitates in MgO doped with up to 4300ppm Fe. Mass action relations incorporating Debye-Hückel activity coefficient corrections were developed to describe FeMg.-vmg? association and magnesioferrite precipitation.

Scanning transmission electron microscopy (STEM) was used to examine sintered and hot-pressed aluminium-doped silicon carbide. Grain-boundary segregation of Al was observed in both materials. Semi-quantitative analysis was carried out on the Al-saturated hot-pressed material. An estimated heat of segregation between 116 and 174 kJ mol-1 was calculated from the STEM data and equilibrium segregation theory. Strain energy as a driving force was considered negligible. The data are consistent with a segregation model in which the driving force for segregation is the difference in the nature of bonding between the solute and solvent atoms. Fast grain-boundary diffusion of Al was suggested from a loss of Al during heat treatments.

The quenching rate required to “freeze-in” high temperature states depends on the resolution attained in the observational technique used and on the kinetics of the property changes that may occur in the sample. For a successful quench, heat must be removed before observable changes occur. Heat loss during travel to the quench bath must be minimized as this lowers the certainty of the quenched-in temperature state, while the heat removal rate after immersion in the quench bath should be maximized. The required quench rate will determine the necessary sample size and quench medium. Quenching has been experimentally evaluated for a propelled “insulating envelope” sample transfer quench device. Results are used to determine optimum conditions for effective quenching. A procedure is established to guide the choice of sample size and quench medium to attain an adequate quenching rate.

Factors affecting the thermal conductivity of ceramics in the temperature range from 2200–4000 F are considered from the point of view of high-temperature insulation. The effective conductivity of solids and pores over a wide range of temperatures is analyzed. Particular attention is devoted to the effects of structure and phase distribution. This analysis is related to the development of optimum properties in thermal insulation for the temperature range considered.