The glass-forming regions have been determined in the GeO₂-Na₂O system by quenching the melts in the compositional range of GeO₂=100-25mol% from above the liquidus to below T_g at various cooling rates, Q₁=1.0×10^-1-Q₇=10⁵K/s. Three glass-forming regions were found but the boundary compositions varied with cooling rate. The reciprocal of the critical cooling rate, -logQ, versus GeO₂ concentration curve shows maxima at concentrations around GeO₂=90, 72 and 35mol%, and minima at GeO₂=85 and 50mol%. The liquidus viscosity, which is highest at GeO₂=100mol%, is maximum at GeO₂=92 and 65mol%, and minimum at GeO₂=78mol%. The slope parameter, E_η/T_L, is also largest at GeO₂=100mol% and becomes maximum at GeO₂=92 and 65mol%, and minimum at 78mol%. The ratio of the constant T₀ in the Fulcher type equation to the liquidus temperature T_L, T₀/T_L, remains small at 0-0.1 in the composition of GeO₂=100-98mol%, but constant in the range 0.5±0.1 in the composition of GeO₂<98mol%. A comparison was made between the compositional dependence of the viscosity parameters (liquidus viscosity, slope parameter and T₀/T_L) and that of the critical cooling rate. From these data, the magnitude of the structural parameter ΔS_f (the fusion entropy per flow unit) was estimated, and the size of the flow unit was elucidated. Extraordinarily high crystallization tendency was observed in the composition of GeO₂=100-99mol%. It was suspected that this is due to an inhomogeneous nucleation resulting from the nonstoichiometry of GeO₂ including oxygen deficiency.

View full abstract

Thermal shock resistance of ZrO₂-(4-17)wt% Y₂O₃ has been investigated by subjecting disk-shaped specimens to a heat cycle of 1100°C/room temperature at a maximum cooling rate of about 500°C/min. Partially stabilized ZrO₂-4wt% Y₂O₃ consisting mainly of monoclinic ZrO₂ phase exhibits poor thermal shock resistance because of its large volume change associated with the martensitic transformation of monoclinic phase→←tetragonal phase. Furthermore, fully stabilized ZrO₂-(12-17)wt% Y₂O₃ with 100wt% cubic phase, which undergoes no martensitic transformation, also shows inferior thermal shock resistance. This is assumed to result from its low fracture toughness. Relatively excellent thermal shock resistance is obtained in partially stabilized ZrO²-7wt% Y₂O₃. This is probably due to its combined effect of small volume change through the martensitic transformation and higher fracture toughness. The effect of Al₂O₃ or SiO₂ addition to ZrO₂-7wt% Y₂O₃ has been also investigated to enhance its thermal shock resistance. ZrO₂-7wt% Y₂O₃-(0.5-4)wt% Al₂O₃ indicated the improved thermal shock resistance. This improvement could be attributed to their higher fracture toughness, together with higher hardness in comparison with those of ZrO₂-7wt% Y₂O₃, and also to the morphological change of monoclinic ZrO₂ phase. In ZrO₂-7wt% Y₂O₃, the monoclinic ZrO₂ grains tended to coalesce into large clusters in the cubic ZrO₂ matrix. As a result, large cracks tend to initiate from the monoclinic phase during the martensitic transformation. On the other hand, the addition of Al₂O₃ results in the isolated spherical grains of monoclinic ZrO₂ phase, leading to improved thermal shock resistance. The addition of (0.5-1)wt% SiO₂ to ZrO₂-7wt% Y₂O₃ causes no morphological change in the monoclinic ZrO₂ grains and consequently results in no improvement in thermal shock resistance.

View full abstract

The firing shrinkage behavior of the alumina green sheet consisting of 96% purity (Al₂O₃-CaO-SiO₂-MgO) which was produced by the doctor blade method was discussed quantitatively. In this study, three commercial types of low soda aluminas were used and two were selected there from and were blended together in certain ratios. Maximum green density peak appeared when the particle radius ratio of two aluminas was more than 2.8 and no peak appeared when less than 2.8. The green density of tape is predictable from TASA (Total Apparent Surface Area) which is calculated from the particle size distribution. The firing volume shinkage of green sheet is given by S_v=(D_g/D_f)*(1-L)*(1-B), where D_f is firing density, D_g is green density, L is ignition loss and B is binder content. This equation shows that the green density governs only firing volume shrinkage, not linear shrinkages. The linear shrinkage is determined by the dimensional distribution parameter from volume shrinkage. The dimensional distribution parameter seems to be determined essentially by the anisotropy of alumina particles in the orientation effects by the tape processing equipment.

View full abstract

δ-BaCO₃ is a new modification of BaCO₃ in which a small amount of sulfate or chromate is incorporated to form a solid solution. The γ-α transformation of BaCO₃ at about 810°C is essential for the formation of δ-phase suggesting the validity of the Hedvall effect. Thus obtained δ-BaCO₃ easily reverts to the low temperature stable phase (γ) on reheating up to 700°C or by contacting with H₂O. This paper deals with the determination of the chemical composition of δ-BaCO₃ obtained by heating with BaSO₄ as an additive. A new method to dissolve the incorporated BaSO₄ was developed, in which a proper amount of strong acid ion exchange resin (abr. H-R) is used. By treating with an acidic suspension of H-R, δ-BaCO₃ decomposed to liberate tiny particles of BaSO₄ as precipitant which undergoes immediately ion exchange between H-R and Ba ion releasing corresponding amounts of SO₄ ion in a solultion. Thus dissolved SO₄ can be analyzed quantitatively by ion chromatography. The determination of the amount of incorporated BaSO₄ in δ-BaCO₃ is based on the following facts;
(1) Ion exchange between H-R and Ba ion proceeds instantaneously (Fig. 1).
(2) Recombination of the exchanged Ba-R₂ with SO₄ ion proceeds very slowly at lower concentrations of SO₄ ion (Fig. 2).
(3) The dissolution trend of BaSO₄ particles by the procedure given here is very sensitive to the activity of BaSO₄; the method is found to be effective for the tiny particles of BaSO₄ without attacking unreacted coarse grains of BaSO₄ (Fig. 4).
By comparing the dissolved amounts of SO₄ ion with those of the theoretical values for the reactant mixtures (Fig. 8), it was concluded that the chemical formula of the δ-BaCO₃ can be represented as Ba (CO₃)_0.90 (SO₄)_0.10.

View full abstract

The milling effects of starting hydrated aluminum sulfate, Al₂(SO₄)₃⋅14-18H₂O, on η→α phase transformation and sinterability of α-Al₂O₃ were studied. Milling of hydrated sulfate lowered the temperatures of dehydration, desulfation and η→α phase transformation by about 30°, 20° and 100°C, respectively. Dehydration of hydrated sulfate produced broken-egg shell-like anhydrous sulfate particles through melting in its water of crystallization of the hydrate. Milled hydrated sulfate became anhydrated particles composed of finer sulfate particles on heating. The anhydrous sulfate desulfated into aggregate grains of η-Al₂O₃ with an irregular pore size distribution. This η-Al₂O₃, finally, resulted in skeletal grains of α-Al₂O₃ in which many cracks were produced. Furthermore, the optimum calcination temperature to prepare α-Al₂O₃ powder for sintering was lowered and the sinterability was improved by milling treatment. The slope of Avrami-plots for η→α phase transformation indicated the two dimensional growth of α-Al₂O₃. The apparent activation energy for the transformation was 110kcal/mol, which remained unchanged on milling. It was confirmed that the enhancement of η→α phase transformation was attributed to accelerated nucleation in η-Al₂O₃ grains, and that the sinterability of α-Al₂O₃ was improved by the increase in the density of green compacts, which was caused by occurrence of many cracks in the skeletal grains of α-Al₂O₃.

View full abstract

A highly sensitive method was developed to determine the ppm level sulfur content in glass. Sulfur in glass was liberated by heating at 1200°C for one hour in purge nitrogen, then volatile sulfur compounds were trapped at liquid argon temperature. Trapped sulfur compounds were led into a gaschromatograph by flush heating and detected by a flame photomeric detector. Sulfur contents determined by this method showed very good and precise linear relationship with the SK_α peak intensities by an X-ray fluorescence spectrometer. This method showed good analytical precision, covering a wide dynamic range from ppm to percent level. Analytical results by this method could be related to the discoloration of Pyrex glass and the redox-state of a glass melt.

View full abstract

The three dimensional model has been developed for effective and precise design of a glass tank furnace. The numerically effective discretization of the partial differential equations and validity of the model are described. Applications to designs of electric melting tank are also shown. This mathematical model consists of the three submodels. The first submodel is the molten glass flow and temperature model. The second is the electric power model. The third is the tracer concentration model. In the first model, an effective numerical analysis technique has been developed, by which converged solutions are obtained with computation time less than that by the other technique already developed so far. Moreover, temperature dependence of physical properties has been also considered. In the second model, various electric conditions such as electrode shapes, arrangements, methods of power supply and connections, have been considered by virtue of superposition principle of governing equations. In the third model, the program system which simulates time dependence of tracer concentration at every position in a tank, has been established. By the three dimensional model, tank design has become more effective in computation time and more precise in considerations of geometrical, material and operational conditions of a tank.

View full abstract

In order to extend the applicability of the laser flash method for measuring thermal diffusivity of materials at high temperature, the correlation factor for heat leak due to radiation has been estimated over the wide temperature range with respect to the relaxation coefficient of the temperature decay after its maximum and the half-rise time to reach the maximum temperature. The thermal diffusivity of boron nitride was determined between 1073 and 1773K using these correlation factors for heat leak due to radiation and compared with the recommended value of TPRC. The absorber of laser beam and the emitter of heat flux for measuring the temperature response curve by an infrared detector are required in thermal diffusivity measurement of good transparent materials such as quartz and sapphire. For this purpose, the platinum thin layers in both side of the columnar specimen were formed by applying and sintering the platinum paste. Such a technique including the correlation factor for heat capacity of thin platinum layers upon the temperature response curve has also been presented and its validity was demonstrated by measuring thermal diffusivity of synthetic sapphire between 1173 and 1773K and comparing the results with the recommended values of TPRC. The correlation factors for heat leak due to radiation or for heat capacity of thin layers estimated in the present work as well as the improved technique with the platinum paste for good transparent materials were found to work well in thermal diffusivity measurement at high temperatures.

View full abstract

Si₃N₄ with Al₂O₃, Y₂O₃, La₂O₃, CeO₂ or Sm₂O₃ was sintered at 1800° to 2000°C under nitrogen of 2 to 4MPa. The density of Si₃N₄ with Y₂O₃, La₂O₃ or CeO₂ increased with increasing amount of additives and increasing sintering temperatures. The density and flexural strength of Si₃N₄ sintered at 2000°C and 4MPa with 10wt% CeO₂ addition were 3.23g/cm³ and about 520MPa, respectively.

View full abstract