Vol.5 No.1 2012

Research paper : An analysis method for oxygen impurity in magnesium and its alloys (A. Tsuge et al.)−29−Synthesiology - English edition Vol.5 No.1 (2012) formed in the atmosphere, even though they were formed in the nitrogen-purged glove box.It is inferred that such oxidation of the chip is caused by a small amount of remaining oxygen in the glove box, resulting from the purging by blowing off the oxygen with nitrogen, without vacuuming.We reached a conclusion that the chip sampling is not appropriate for on-site analyzing with a simple procedure. To avoid the oxidation, extremely careful operation is necessary. The operation of fabricating the chips and collecting them with a tablet frame placed in the glove box was not only difficult and required skills, but also handling a lathe, which is a rotating machinery, with thick gloves is quite dangerous. Therefore, we considered that the chip sampling is inappropriate for an analysis method to be done easily on site and should not be adopted as a sampling method for a standard.From the experimental results about chip sampling, we obtained two findings which are helpful for subsequent research. First, as seen in the measurement result of the core of the rod shown in Fig. 6, it is inferred that the effect of oxygen on the bulk surface was almost negligible in the developed analysis method. It implies that the oxidation is limited to the very thin layer on the surface despite high oxygen affinity and susceptibility to oxidation of Mg. It also implies that the effect of surface oxidation could be kept at the same level as the detection limit of the IGF-IRA method if the surface area of the sample was relatively small compared to the volume. This led to the development of the core drill method described below. Second, another finding is that oxygen content could be kept within a certain range if the chip thickness remained roughly constant. This is important for preparing samples for verification of the analysis value and for the joint analysis tests.2.2.2 Core drill methodAs mentioned in 2.2.1, the effect of surface oxidation can be suppressed by using a bulk specimen with a small surface area against its volume. Meanwhile, the sampling method is required to be able to sample the material without bias. In addition, the specimens of a shape suitable for the analysis, have to be sampled from any part of objects which may have various configurations. Another unique requirement of the IGF-IRA method is that the specimen must be placed most certainly in the graphite crucible. In the commercially available IGF-IRA device, the graphite crucible is degassed after conducting the inert gas purge of the furnace, and the specimen is dropped by its own weight through the sample drop path of about 8 mm in diameter. That does not matter for heavy samples such as iron due to its large specific gravity. In contrast, there is the possibility that Mg, which has low specific gravity, will be lodged in the path for some reason. To fulfill these two requirements, the specimen should be made small and spherical as much as possible. However, there is a trade-off between the quantity of the specimen used and the precision of the measurement value. In the case of a spherical sample, the more time will be needed for processing, the more profound influence of oxidation is exerted on the analysis result. In our preliminary study, a cylinder specimen, which enables easy sampling, was adopted. After much trial-and-error, we chose a cylinder with a diameter of 7 mm and a length of 4 mm as the smallest size specimen while maintaining the measurement precision.To efficiently collect the sample of this shape from the object to be measured, we prepared a custom-made core drill to core a rod of 7 mm in diameter. Using the core drill, the rods were sampled from any area of any depth in the object, and the specimens were cut out at arbitrary depth from the rod. The appearance of the core drill is shown in Fig. 7. Figure 8 shows the case where 130 specimens were cut out systematically from the extruded billet of 170 mm in diameter and 500 mm in length. It should be noted that collecting specimens for distributional analysis from such a large object could be done within two days.3 Investigation of reliability of the method for international standardizationFig. 7 Core drill for collecting sample (inner diameter 7 mm)Fig. 8 Example of sample collection by core drill for the measurement of oxygen distribution in the extrusion billet170 mm No.1 -1 No.1 -13 500 mm No.1 -5 No.1 -6 No.1 -11 No.1 -12 No.1 -4 No.1 -8 No.1 -2 No.1 -3 No.1 -9 No.1 -7 No.1 -10 130 samples in totalEach sample is 7 x 50 (mm)Casting directionCut to 50 mm x 10No.10No.9No.8No.2No.1Sampling by core drill


page 32