Degree Name

Doctor of Philosophy


FTIR-reflectance spectra (20 - 6,000 c m - 1 ) of 3 mol% yittria stabilised zirconia (TZ3Y) and α-alumina have been measured at various temperatures between room temperature and ࣘ 900 °C. The spectra have been fitted using both the Four Parameter Semi Quantum (FPSQ) and Three Parameter Classical (TPC) oscillator formulations of the dielectric function, with the initial assistance of Kramers- Kronig analysis. The dispersion analysis yielded the oscillator parameters for the transverse and longitudinal phonon mode resonances and their lifetimes. Estimates of microwave loss were readily obtained from the dielectric function by direct substitution of the relevant microwave frequency (cm-1). This work was explored as a means of gaining an insight into the mechanisms for the dielectric loss, besides quantifying it. Most of the disparity between dielectric measurements made directly at microwave frequencies (ࣈ 2.45 GHz) by microwave techniques, and those extrapolated from infrared analysis, can be attributed to extrinsic factors such as porosity, which do not contribute to the losses at very high (infrared) frequencies.

Without access to direct microwave measurement techniques, such as B W O interferometry, the order of accuracy of theoretically derived estimates of κ1(T) have been tested principally by comparison between the calculated reflected power values (from κ1(T)) and the experimentally measured values from a dual-directional coupler in the microwave transmission line during heating of the engineering ceramics. Over all microwave forward powers, a difference of about one order of magnitude was observed between experimental measurement and theoretical estimation of reflected power (Tables 5.9,10). Particular care was taken to incorporate effects from skin-loss of microwave power in the metal wall of the cavity and radiation loss from the hot ceramic. The lower theoretical κ1(T) estimates at 28 GHz, as also at 2.45 GHz , are a consequence of the fact that the extrapolated infrared analysis yields only an intrinsic estimate of the dielectric constant.

A chief concern of this thesis then, has been to conduct a thorough and careful study in examining the extent of the applicability of the semi-quantum expression of the dielectric function, to estimating microwave losses in poly-crystalline engineering ceramics that have been raised in temperature above room temperature, through fitting of FTIR reflectance spectra. The results are seen to be consistent with the limited work of Petzelt et al[167, 170], among others, in that the FPSQ dielectric function has been confirmed to hold an inherent sensitivity to the overall shape of a reflectance spectrum, such that it yields negative estimates of loss when extrapolated to microwave frequencies. This only works to confirm the fact that even with the most complete formulation of phonon mechanics to date, resonance mechanisms are not sufficient for estimating microwave absorption in (poly)crystalline solids below ࣈ 30 GHz , particularly as temperature increases.

The allied concern of accurate thermometry in the microwave processing of ceramics, was resolved by the implementation of the hitherto largely theoretical device of multiwavelength pyrometry. The greatest success of this thermometer over conventional ratio pyrometry was achieved by using a spectral emissivity term constructed upon Fresnel-Maxwell field theory rather than empirical relations chosen for computational convenience.

A statistical accuracy of 塲 % from non-linear least squares fitting of the experimental radiance curve, is comparable with that of a standard B-type thermocouple (ࣈࣘ2%)that was specially mounted to suffer minimal microwave and thermal perturbations in its response.

The performance of a conventional two-colour ratio pyrometer was noted to have an uncertainty of up to ± 500 °C if the correct setting for the emissivity-slope correction is not known. Such is the case for engineering ceramics for which there exists little emissivity data. For the engineering ceramics under study here, multi-wavelength pyrometry estimated them to have an essentially wavelength-invariant (grey) emissivity. This was reasonably confirmed by the combined thermocouple-two colour ratio pyrometer measurements of section (4.4.4).