Dense ␤-Si 3 N 4 with various Y 2 O 3 /SiO 2 additive ratios were fabricated by hot pressing and subsequent annealing. The thermal conductivity of the sintered bodies increased as the Y 2 O 3 /SiO 2 ratio increased. The oxygen contents in the ␤-Si 3 N 4 crystal lattice of these samples were determined using hot-gas extraction and electron spin resonance techniques. A good correlation between the lattice oxygen content and the thermal resistivity was observed. The relationship between the microstructure, grain-boundary phase, lattice oxygen content, and thermal conductivity of ␤-Si 3 N 4 that was sintered at various Y 2 O 3 /SiO 2 additive ratios has been clarified.

and subsequent acid treatments that removed the secondary phases. The lattice oxygen contents of these crystals were determined by the hot-gas extraction method to be 0.258 ± 0.006 and 0.158 ± 0.003 wt% for the additive compositions of Y 2 O 3 :SiO 2 = 1:2 and 2:1, respectively. The oxygen dissolved in the ␤-Si 3 N 4 crystal lattice as much as in the ␣-Si 3 N 4 crystral lattice prepared by the chemical vapor deposition process and in the AlN crystal lattice that exhibited high thermal conductivity.

Rare earth (RE) halides (fluorides, chlorides, bromides, and iodides) have been studied by X-ray photoelectron spectroscopy. The binding energies (BE’s) of the RE 3d and 4d peaks for the chlorides, bromides, and iodides are related to the atomic charge calculated on the basis of Pauling’s scale of electronegativity (Pauling charge). Except for Y and Lu, the observed RE 3d and 4d BE’s for the fluorides are lower than those expected. For Y and Lu, the observed RE 3d and 4d BE’s for all halides are related to the Pauling charge. For the fluorides and chlorides, the BE’s of the ligand peak show variations with increasing atomic number of RE’s and a specific “tetrad effect” is observed in plots of BE versus RE atomic number and versus 1/R (where R is the mean distance from the ligand atom to the neighboring RE atom), except for the BE’s of the F 1s peaks from the light RE fluorides. For the iodides and bromides, no or little characteristic variations are seen in the BE’s of I and Br.

An application of diffuse reflectance infrared Fourier transform (DRIFT) spectrometry to determine powdery sample with an unknown particle size has been proposed. The usefulness of this technique has been demonstrated for the analysis of the α-component in Si3N4 powders. The particle size of the α-component in Si3N4 powder could be determined by measuring the peak intensity ratio between two peaks (500 and 690 cm−1) on the DRIFT spectrum, and the concentration could be determined by comparing the peak intensity to that of standard α-Si3N4 with same particle size. The particle sizes and concentrations obtained by the DRIFT method were in good agreement with those obtained by the light-scattering method and x-ray diffractometry, respectively.

The behavior of vacancy-type defects and displaced Si atoms in Si(100) caused by self-ion implantation has been investigated by variable-energy positron annihilation spectroscopy and Rutherford backscattering spectroscopy/channeling. It is found that the recovery process of the defects strongly depends on the morphology of the implanted region. The divacancies produced by an implantation of 2×1014Si+⋅cm−2, which is less than the critical value required for amorphization, aggregate into large vacancy clusters by annealing at 300 °C. These vacancy clusters diffuse towards the surface at temperatures above 600 °C and anneal out at around 800 °C. The specimen implanted with 2×1015Si+⋅cm−2, in which a complete amorphization takes place in the damaged region, shows a different annealing characteristic. In the first stage (∼600 °C), the amorphous zone is transformed into crystalline material by solid phase epitaxial growth, although large vacancy clusters still remain. These agglomerate clusters continue to grow in a second annealing stage which takes place at around 700 °C. Annealing at 900 °C is required to eliminate these vacancy-type defects.