Investigations on the mechanism of the errosion of nozzle, guide and runner bricks by flowing molten steel, scums and non-metallic oxide inclusions in the process of bottem casting for producing conventional carbon steel with a basic open-hearth furnace as well as with a converter were carried out.
This paper concerns chiefly with the observation of external appearance and the microstructure of eroded brick after pouring the rimming and killed steels, and with the study of various scums adhering to the surface of residual steel rod in runner opening and ingot, of the float on the upper surface of molten steel in mould, and of the non-metallic inclusions. Also the chemical and mineralogical investigations were carried out to know the relation between the chemical composition of scum and non-metallic oxide inclusions.
From the experimental results the author concluded that the erosin of the bricks by flowing molten steel was closely connected to the floating minute particles of oxide slag, and that the origination of high aluminous sand would probably be not directly concerned in the erosion of bricks by molten steel.

View full abstract

It is generally accepted that when portland cement is kneaded with water or the solutions of various chemicals the change in all stages, extending from setting, hardning, and finally to the development of the strength, will be observed, and that the change varies with the nature of the chemicals.
And there are already some long period studies of the hydration products and the mechanism of hydration by measuring the rate of heat evolution using ice- and other similar calorimeters.
These investigations, however, are limitted to the chemicals familier to the chemistry of cements, namely, gypsume and lime, and also there are only a few papers on the investigations on the initial stage of hydration with relatively large amount of water.
Concentrating to the attention to the initial stage of hydration the authors followed up continuously the changes by measuring the heat of hydration of cement past using an adiavatic calorimeter of own construction, and the approach was made to the mechanism of the reaction through the analysis of the kinetics of the reaction.
This paper concerns the hydration of 3CaO⋅Al₂O₃ whose rate could be represented by Jander's equation of diffusion in membrane, and which suggested that the rate determining step was the diffusion of the component ions of C₃A through the membrane. Under electron microscope the membrane was observed as the pile of hexagonal thin plates of 300-500 Å and 2-10μ thick.
The trend of the appearance of the hydration products of all agents, i.e. CaSO₄, CaCl₂, NaOH, suger, and alcohol, single and mixture, was the same except the difference of minor importance, although some of them accelerate and the others retered the hydration. As the reaction mechanism was shown to be the same through out all aditional agents it follows that another physical factors should be taken into consideration in order to interpret the change of the consistency during the setting.
On the other hand, the coexistence of CaO, CaSO₄, Na₂PO₄, and Borax showed the linear rate of hydration, suggesting that the reaction proceded at the liquid solid interface.
The authors assumed that these phenomena came from whether the reaction proceded without the formation of membrane or the solution was comming in direct contact through the cracks in the membrane.
Moreover, in some cases of the addition of sugar and alcohol the hydration stopped completely after the period of violent reaction.
In short, the analysis showed that the chemicals gave the very complicated effects from the initial to the final stage of setting.

View full abstract

The shape and the size of carbide grains in nickel bonded titanium carbides were studied by metallographical methods to confirm the influences of sintering atmosphere, temperature, and of the amount of metal binder on the nature of carbide grains.
The carbide and nickel powders were mixed and ball-milled with methyl alcohol for 25 hours. Thus five titanium-carbide nickel compositions containing 10, 20, 30, 40 and 50 percent Ni by weight were prepared. Besides, another five ones, including 5 percent TiO₂ were also prepared. Compacts of 12mmφ×10mm were pressed with the presure 1t/cm², which were sintered at 1500°C in a graphite boat in either dry hydrogen or argon, and were quenched into water. The specimens were cut in vertical direction into two halves and polished with diamond paste. From the visual observations of microphotographs the statistical measure of grain size as well as the factor describing the grain shape were worked out.
It was found that the size and shape of carbide grains varied with the amount of metal binder, i.e. with the amount of liquid phase existing during the sintering process. function
The specimens containing more than 30 percent Ni showed that the grain sizes was the of the sintering temperature only, being quite inde endent of the composition. Those of less than 20 percent Ni, however, proved that the grain size decreased with decreasing amount of metal binder.
In the absence of oxygen the fraction of angular carbide grains increased with the amount of metal binder, while in the presence of oxygen the increase of spheriform grains was observed. The sintering at lower temperature, 1400°C, showed that the grain growth as well asthe influences of atmosphere and of the amount of metal binder on the grain size were greatly reduced.

View full abstract