Hideyo Tabata

Effect of additives on some properties of silicon oxynitride ceramics

Silicon oxynitride ceramics are formed by reaction sintering of silicon nitride and silica with certain metal oxide additives. The reaction rate during sintering and the subsequent properties of silicon oxynitride are affected by the quantity and kinds of additives. The reaction rate increased for addition of equal molar amounts of ZrO2, ZrO2 (+2.8 mol % Y2O3), AlO1.5, LnO1.5, CeO2, MgO, in that order (where Ln=Nd, Sm, Gd, Dy, Er, Yb and Y). The lanthanide oxide (1.5 mol %)-doped silicon oxynitride ceramics had a high fracture toughness, because crack deflection occurred due to the precipitation of an intergranular crystalline phase with a high thermal expansion coefficient compared with silicon oxynitride. The oxidation rate was higher with an increasing quantity of additive. In samples containing an intergranular crystalline phase, stability of the crystalline phase is an important factor and could impair the oxidation resistance of silicon oxynitride ceramics.

Joining of silicon nitride ceramics by hot pressing

The joining of hot-pressed silicon nitride ceramics, containing Al2O3 and Y2O3 as sintering aids, has been carried out in a nitrogen atmosphere. Uniaxial pressure was applied at high temperature during the joining process. Polyethylene was used as a joining agent. Joining strength was measured by four-point bending tests. The effects of joining conditions such as temperature (from 1400 to 1600°C), joining pressure (from 0.1 to 40 MPa), holding time (from 0.5 to 8 h) and surface roughness (Rmax) of the joining couple (about 0.12, 0.22 and 1.2μm) on the joining strength were examined. The joining strength was increased with increases in joining temperature, joining pressure and holding time. Larger surface roughness caused lower joining strength. The higher joining strength was attributed to a larger true contact area. The area was increased through plastic deformation of the joined couple at elevated temperatures. The highest joining strength attained was 567 MPa at room temperature, which was about half the value of the average flexural strength of the original body. The high temperature strength measured at 1200° C did not differ very much from the room-temperature value.

Preparation of translucent mullite ceramics

Des ceramiques de mullite hautement translucides ont ete preparees par filtrage sans pression ni additif, avec elimination des phases secondaires et agrandissement approprie de la taille des grains de mullite

X-Ray Topographic Observation of Stacking Fault in Stoichiometric MgAl2O4 Spinel Single Crystals

Stacking faults and dislocations in stoichiometric MgAl2O4 single crystals are investigated by means of X-ray topography and chemical etching. The correspondence of an etch pit to a dislocation line and of an etch groove to a stacking fault is confirmed. The fault vector of the stacking fault is determined to be a/4 , and two fault systems in the spinel lattice are established to be of the type (100) a/4[110] and (100) a/4[110]. A model of atomic arrangement around the stacking fault is presented, and it is concluded that the stacking fault is constructed through a relative displacement of cation packing within the perfect anion sub-lattice.

Growth and morphology of chrysoberyl single crystals

Abstract Chrysoberyl (BeAl 2 O 4 ) single crystals were grown from PbO-PbF 2 -B 2 O 3 and PbO-PbF 2 fluxes. The crystal habit was considerably influenced by the presence of boron oxide in the flux. The crystals grown from the binary (PbO-PbF 2 ) flux were thick plates parallel to the (001) plane and bounded by (010) and (111) surfaces, whereas prismatic and equidimensional crystals with (011), (010) and (111) surfaces were grown from the ternary (PbO-PbF 2 -B 2 O 3 ) flux. The appearance of (011) surfaces taking the place of (001) was explained in terms of substitutional adsorption of boron ions at beryllium sites. The selectivity of adsorption in different surfaces is attributed to the difference in surface density of the beryllium sites in various crystallographic planes.

Spinel twin formation in flux grown MgAl2O4 crystals

Abstract An attempt to grow artificial spinel twins in flux grown MgAl 2 O 4 crystals has been made with the addition of BeAl 2 O 4 in the melt. The mechanism of the twin formation is interpreted to be due to the introduction of stacking disorder between the oxygen close-packing layers taking place as a result of the epitaxial growth of crysoberyl layers on the growing spinel surface.

Preparation of spinel fibers by directional solidification of MgAl2O4Mg2SiO4 eutectic

Abstract Eutectic directional solidification of the pseudobinary system MgAl 2 O 4 Mg 2 SiO 4 was carried out by the floating zone method, and yielded a typical colony microstructure at a solidification rate of 120 mm/h. Spinel fibers in a forsterite matrix were formed in the central part of each colony. A bundle of fibers consisting of spinel single phase was obtained by dissolving the forsterite matrix first in a hydrochloric acid, and then in a NaOH aqueous solution. Most of the spinel fibers were less than 2 μm in diameter and longer than 50 μm, and were grown along the 〈100〉 crystallographic direction.

Thermal decomposition of silicon nitride along with powder characteristics.

窒化ケイ素の熱分解過程と粉末粒子中の酸素の存在状態の関係,および, 比表面積など他の粉末物性値との関係を検討した。窒化ケイ素の真空中での熱分解反応は1400℃まではおだやかであった。1450℃以上では一定期間の初期過程ののち, 急激に分解して遊離のケイ素を生成した。初期過程においては粉末の酸素含有量は減少したが, これをすぎると, 分解生成物中の窒化ケイ素に対する酸素のモル比はほぼ一定になった。このモル比からα-窒化ケイ素中への酸素の固溶限界は組成式Si3N3.932-3.94200.101-0.086で表わされた。熱分解の初期過程においては,窒化ケイ素粒子の酸素に富んだ表面層の分解除去によって遊離のケイ素の生成をともなうことなしに酸素含有量を低減することができた。しかしながら, 酸素含有量を固溶限界以下にすることは, 熱処理雰囲気を変えても不可能であった。また, 初期過程では窒化ケイ素のα→β転移は起こらなかった。非晶質粉末では,初期過程においても結晶化してα相になったが, 結晶性は悪かった。比表面積は熱分解反応の進行にともなって減少した。

Influence of size and shape on homogeneity of powder compacts formed by cold isostatic pressing (part 1): Press forming of thick cylinders

The influence of size and diameter/thickness (D/T) ratio on the homogeneity of cylindrical silicon carbide powder compacts formed by cold isostatic pressing has been studied. The degree of inhomogeneity was estimated by the measurement of spatial distribution of bulk density, Vickers microhardness and pore size distribution for the specimens cut out from the compacts. The degree of inhomogeneity in the density and hardness was relatively smaller than that in uniaxially die-pressed compacts. However, the distribution patterns of the density and hardness which were similar to each other showed the presence of high density and high hardness layer near the surface of the compacts. Furthermore, inhomogeneity was found locally in the inner part of the compacts. A considerable decrease in volume fraction of the smaller sized pores was recognized at the corner edge of the cylindrical compacts. The volume of the inhomogeneous layer near the upper surface in the compacts with a larger D/T ratio tends to be reduced with decreasing D/T ratio. For the same D/T ratio, however, the homogeneity did not depent on the size of the compacts. This behavior could be explained by the difference in damping behavior of the applied pressure between the radial and axial directions.

Failure Probabilities of Joints Consisting of Different Kinds of Brittle Materials

The influence of thermal stress in joints, which is induced during cooling by the difference in thermal expansion coefficient between the joined brittle materials, on the failure probabilities of the joints was estimated. A numerical analysis by the finite-element method was carried out on SiC/Si3N4 joints. It was assumed that the strength of each material obeys the Weibull distribution function and that the joining strength is high enough to prevent rupture at the joining interface. Since the tensile stress develops predominantly at the side of SiC which has a larger thermal expansion coefficient than that of Si3N4, the Weibull modulus of SiC was taken into consideration for calculating the failure probability of the joint. The dimension of the joint and the Weibull modulus of SiC member were shown to have great influence on the failure probability. Therefore, these two factors must be taken into consideration in addition to the fracture strength of material to be joined for the evaluation of the joining strength of various brittle materials.