抄録
A new copper mould devised by Derge to take a wedge-shaped and therefore rapidly chilled sample for rapid vacuum fusion analysis of molten iron and steel for total oxygen content during the steel making process was made and tested by the authors, and the following throughout defects were found : (1) the sharp bottom part of the cavity of the mould often could not be completely filled with molten iron and therefore the available chilled and bright part of sample taken was often too small to effectuate accurate analysis ; (2) the heat capacity of the mould was so small that the mould became conaiberadly hot,not permitting to touch with bare hands even after it was used only once, and this made it impossible to continue sampling with the same mould in spite of Derge’s pretension. Samples taken with Derge’s mould were analysed for oxygen by the Gakushin hydrogen reduction method which is expected to give nearly the same value as the vacuum fusion method when oxygen is contained mainly in the form of iron oxide in samples. The values so obta- ined almost agree with those obtained by the Gakushin rapid method for determining oxygen in molten iron and steel, orthe modified Herty’s aluminum method.In using the Gakushin rapid method it was found. (1) blank values were sometimes nearly equal to or even greater than true values and therefore they could not be ignored ; (2) the use of not only ashless but finer filter paper su h as T\={o}y\={o} filter paper No. 5c or No. 6 is preferable, because ordinary ashless filter papers are not so fine and fail to catch finer alumina particles.
