抄録
A new theoretical consideration was made on the effect of various gases upon the microstructure of cast iron. A well-known phenomenon such as inverse chilling or abnormal crystallization, which has been never made clear by the so-called nucleus theory, may now be explained by the idea that oxygen dissolved in the molten cast iron disturbs graphite in its crystallization. In this case the oxygen is so apt to be adsorbed on the surface of graphite that its checking effect on the crystallization is extremely conspicuous, even though the concentration of oxygen is very small, only up to 0·001-0·01%. For instance, the "heredity" of pig iron, the complicated effect of superheating, slag, or atmosphere, or other unexpected phenomena which are often experienced by foundrymen, may probably be accounted by the fact that the oxygen has marked effect upon the separation of graphite from molten iron.
The solubility of oxygen in the molten cast iron is controlled by the concentration of silicon, being inversely proportional to √[Si]. From this relation we can calculate an approximate value of oxygen solubility. It is, further, noted that addition of small amount of Al or Ti has only a little effect on the oxygen solubility. On the other hand, with increasing temperature, the deoxidizing action of silicon becomes gradually weaker and hence the oxygen solubility in cast iron is favoured with the rise of temperature. It is due to this fact that superheating results in refinement of graphitic microstructure. Here too, it should be notcd that, as to the oxygen solubility, there exists a critical temperature above which the deoxidizing action of carbon becomes superior to that of silicon and the oxygen content begins again to decrease.
The oxygen content is, furthermore, governed by the composition of slag in such a way that the more acidic the slag, the smaller the oxygen concentration. In other words, cast iron molten under the strongly basic slag absorbs much oxygen and is liable to undergo abnormal crystallization. The "heredity" of pig iron is also attributed to the difference in oxygen content which results from the difference in the temperature and nature of slag in the blast furnace.
On the other hand hydrogen and nitrogen do not exert so much influence as oxygen, because, at the temperature of molten cast iron, they are hardly subjected to adsorption with graphite.
