The writers have studied the mechanism of the corrosion of molten glass against pyrophyllitic crucible mainly by means of polarization microscope.
The crucible used for this experiment were prepared by firing at 1400° the mixes chiefly consisting of Roseki* and its chamotte with some clayey materials as binder. Two kinds of glass cullet, one being soda-lime glass (Na₂O⋅CaO⋅3SiO₂) and the other soda-lead variety (Na₂O⋅PbO⋅3SiO₂), were also previously prepared by melting the batches of the corresponding composition. Thence, the glass cullets were melted in the crucibles at 1200° and 1400°, each for 0 hrs, ** 2 hrs, 3 hrs and 4 hrs respectively. After cooled down to the room temperature, the boundary parts between the crucible and the glass were microscopically observed, with the results as follows.
In the specimens kept at 1200°, it was observed from the biginning that a “reaction layer” had been formed along the clayey parts and the kaolinitic Rosekis of the crucible where they came into contact with the glass. At the earliest stage, the reaction layer was about 5μ in thickness, chiefly consisting of carnegieite crystals with small amount of nepheline. With the elapsing of time, the reaction layer along the clayey parts diminished, while the reaction layer along the kaolinitic Roseki developed by degrees. On the other hand, in the specimens kept at 1400°, the reaction layer, chiefly consisting of nepheline crystals with small amount of carnegieite, was found merely along the diaspore pseudomorphs from the earliest stage. With the elapsing of time, the reaction layer had diminished by degrees, until it entirely disappeared when kept at 1400° for 3 hrs. Along with the formation of the reaction layer, the marginal part of the crucible had been gradually invaded by molten glass, until it developed into a “corroded layer” of special appearance. Thus it is obvious that materials were exchanged between the crucible and the molten glass during the existence of the reaction layer. After the elimination of the reaction layer, the corrosion proceeded merely through the corroded layer, whose front surface moved gradually towords the interior of the crucible, while the rear surface dissolved down into the molten glass by degrees. The soda-lead glass is apparently more active against the crucible than soda-lime glass, though the sequence of their corrosion process seems to be almost similar.
* Roseki is a kind of rock of hydrothermal origin, chiefly composed of pyrophyllite with small amount of kaolin and diaspore. It is widely distributed in southeastern Japan.
** The crucibles were taken out as soon as they attained up to 1200° or 1400°.

View full abstract

The oxides such as CaO, MgO, BaO, SrO, CdO, ZnO, NiO, CeO₂, Co₃O₄, PbO, Ag₂O, SnO, Cu₂O, CuO, MnO₂, Al₂O₃, Cr₂O₃, CrO₃, TiO₂, (Nd, Pr)₂O₃, As₂O₃, Na₂O, K₂O, ThO₂, UO₂, SiO₂, Fe₂O₃ and B₂O₃ were added to zirconia and the heating effects of these additional reagents were observed by the difference of thermal expansion curves.
Four types were distinguished in the zirconia thermal expansion-contraction curves under our experimental conditions. CdO, ZnO, SnO, Ag₂O, Al₂O₃, As₂O₃, Na₂O, and K₂O mixtures were found to belong to Type I; SrO, BaO, NiO, Co₃O₄, Cr₂O₃, and CrO₃ mixtures to Type II; CaO and MgO mixtures to Type III; and CeO₂, ThO₂, MnO₂, UO₂, TiO₂, (Nd, Pr)₂O₃, Cu₂O and CuO mixtures to Type IV.
Arranged in order of their efficiency for the stabilization of zirconia, Type III comes first, Type IV second, Type II third and Type I fourth.
TiO₂, MnO₂, CeO₂, and CaO were the effective mineralizers for sintering of zirconia. These sintered zirconia had high compressive strength: for example, TiO₂ had 3551 kg/cm², CeO₂, 2653 and MnO₂, 2570.
MnO₂ was the most effective reagent for crystal growth of zirconia and TiO₂ the least effective.
The minimum amount of CaO required for the complete stabilization of ZrO₂ was as follows: about 3% at 1740°C firing, 2.5-5.0% at 1700°C, 5.0-7.5% at 1600°C and 5.0-7.5% at 1500°C. In the case of MgO, 3-4% at 1740°C, less than 5% at about 1700-1600°C and 5-10% at 1500°C.
ZrO₂ stabilized with CaO showed better results than MgO-stabilized zirconia in their chemical, physical and mechanical properties.
In the CaO 5% and ZrO₂ 95% mixtures, added CaO as salt combined with various anions, the heating effects due to anion were different from one another.
The order of these effects on the properties of the fired products were as follows: Crystal growth: CO₃^-->SO₄^-->OH^->F^->Cl^->PO₄^---, firing shrinkage: F^->SO₄^-->OH^->PO₄^--->CO₃^-- and compressive strength: F^-(4967kg/cm²)>SO₄^--(3934kg/cm²)>OH^-(3340kg/cm²)>CO₃^-->(3106kg/cm²)>PO₄^---(2160kg/cm²).

View full abstract

Effect of firing operation on physical properties of expanded vermiculite from Fukushima, Japan, was studied from the viewpoints bulk density, brittleness and thermal conductivity. As the firing temperature was raised from 20° up to 1100°C, the bulk density of the vermiculite was diminished from 1.170 down to 0.130. It was desirable the comparatively thin flakes of nearly uniform grain-size were introduced into a heated expansion furnace to receive equal heat treatment. The brittleness of the expanded vermiculite was increased as the firing temperature was raised. But the prolonged duration of firing had no effect on the bulk density and the brittleness in the range of temperature below 1000°C. On the other hand, when fired at 1100°C, the product became blackish and increased the brittleness with prolonged duration of firing over about 1/4min, though there was no appreciable change in less duration of firing. This tendency was strengthen at the temperature above 1100°C. To obtain the maximum expansion of golden material, it was seen from the data presented that Fukushima vermiculites should be subjected suddenly to the temperature from 1000° to 1100°C for a few seconds. Thermal conductivity was decreased with a decrease in bulk density due to an increase in the temperature of firing. The values of thermal conductivity of expanded products, obtained from the unsieved grains of vermiculite, with bulk density of 0.175 were 0.045 and 0.073kcal/m. hr °C at the mean temperatures of 100° and 300°C, respectively. Those values with bulk density of 0.150 were 0.042kcal/m hr °C at 100°C. and 0.075keal/m. hr °C at 300°C. The measurements of thermal conductivity were made with loose-filled samples pressed by 20 per cent.

View full abstract

In continuing the previous studies (This Journal 62, 785-789 (1954)), the results of further studies were reported by the same authors, as the following abridgement. (1) The strength depression by heating the hardened prismatic test pieces in air at 100 and 60°C for 60, 200 and 300 minutes were systematically tested and the results were fully discussed by comparing with those in the previous paper. (2) Fine powder of broken scrap of electric porcelain insulator, which is called as Scherben and was used already in the previous paper, was also added to the Portland cement sample in the proportion of Cement 1.0 to 1.0, 0.5, 0.4, 0.3, 0.2 or 0.1 of Scherben, and these mixed cement samples were tested on the strength depression by the rapid treatment of heating to cooling from 100, 90, 80, or 50°C to 0, 20, 50, 80 or 90°C. The addition of small amount (about 10-30%) of Scherben to the Portland cement was proved to give good effect to the mixed cement as the cementing material of electric porcelain insulator.

View full abstract