Abstract
An X-ray investigation was carried out for the crystal structure of cobalt heated at temperatures above the transformation point, and it was found that the β phase of face-centred cubic lattice is retained more in the surface layer of about 0.02mm than in the interior, and also that the grain size is smaller in the former than in the latter, as is shown in Fig. 1 (Pl.), Fig. 2 and Table 1. In the case of annealed file powder, which should be markedly affected by the above surface phenomenon, the greater part of the powder was, as expected, of cubic structure.
In order to inquire into the cause of the above surface phenomenon, the same experiment as above was also made with very pure cobalt, but since the result was almost the same as before, the phenomenon does not seem likely to be attributed to the impurity in the specimen. By testing a specimen, which was etched to remove the surface layer affected by the cold work of machining and polishing, the above phenomenon was found to have been very much reduced. Therefore this phenomenon may be explained by the reason that the grain refinement in the surface layer due to a severe cold work before heat treatment slackens the trans-formation of cubic phase into hexagonal.
A specimen, the surface of which was polished after annealing was also examined by X-rays. The result was that the polished surface was entirely of hexagonal phase, while in the interior more or less quantity of the cubic was retained. This difference is evidently attributable to the stress due to polishing.
The surface phenomena as found in the present investigation call crystal analyst' attention to the fact that the surface layer does not necessarily have the same structure as the interior, especially for metals having a low transformation point or for those which easily super-cools. The uncertainty can of course be eliminated to a large extent by dissolving off of the doubtful surface.
