Vol.5 No.1 2012

Research paper : An analysis method for oxygen impurity in magnesium and its alloys (A. Tsuge et al.)−35−Synthesiology - English edition Vol.5 No.1 (2012) Discussion with Reviewers1 OverallComment (Masahiro Okaji, AIST, currently at Chino Corporation)This research clearly aims for industrial application from the beginning and engages in R&D toward the standardization as an outlet, and I think it is appropriate as a Synthesiology paper.2 International situationComment (Yasuo Hasegawa, Energy Technology Research Institute, AIST)You discuss the situation of Japan, China, and Korea in “1 Introduction,” but I think it will become easier to understand the situation of the industry if you add a graph of the share of raw material production and goods production by countries.Answer (Wataru Kanematsu)I added the pie chart of the production of magnesium metal by countries based on the report by the US Geological Survey, as Fig. 1 in chapter 1. I also corrected the figures for the world share based on this (95 % 85 %). For goods production, the market scale is small compared to other non-iron metal products, and the statistics of the industrial organizations do not have magnesium as an independent survey subject.3 Process of selecting the IGF-IRA methodComment (Masahiro Okaji)Please describe the process of choosing the IGF-IRA method among several oxygen analysis methods, from the scenario-driven standpoint.You start with the inert gas fusion infrared absorptiometry (IGF-IRA) in 2.1. I think it will be logically easier to understand if you start with the comparison of the phenol dissolution method and CPAA method described in 3.1, and show why the IGF-IRA is more appropriate for standardization. Please reconsider the line of reasoning.Answer (Akira Tsuge)For the process of choosing the IGF-IRA method, I added the sentences describing the problems of the analyses using activation or wet operation, and the fact that IGF-IRA method is used widely in industry and therefore appropriate for standardization at the beginning of “2.1 Development of the basic technology” to clarify the reason for the appropriateness of standardizing IGF-IRA. 4 Process of the selection of sampling methodComment (Masahiro Okaji)As is the case with the preceding discussion, for the sampling method, I think the reasoning is natural and easier to understand if you describe the comparison of the two methods (chip sampling and core drill methods) in the beginning.Answer (Akira Tsuge)I added the description at the beginning of “2.2 On the sampling method” and at the beginning of “2.2.1 Chip sampling.” I clarified the process by explaining the fact that the effect of surface oxidation in the material with high oxygen affinity like magnesium cannot be avoided in general chip sampling method, and therefore the core drill sampling that has small effect on oxidation was chosen although homogenizing process cannot be done.5 Justification of agreement with CPAA methodComment 1 (Masahiro Okaji)In the last paragraph of 3.1, you explained the justification of agreement between this method and the charged particle method. However, there are assumptions in the description of the cause of discrepancy of the data, and I think a more objective explanation is required here. For example, in the later paragraph, it is written, “Sample no.3 … probably contained moisture and hydroxides.” Is this the result of measuring the concentration of moisture and oxides? Also, for the degree of agreement, can you say it is satisfactory in the comparison with the requirement level of uncertainty?Answer 1 (Akira Tsuge)The moisture and hydroxides were not measured. These are error factors caused by the assumption that the sample no.3 used in this experiment is a sample oxidized by increasing the humidity of the chips.For the requirement level of uncertainty, the analysis value of the oxygen in metal by the IGF-IRA method is allowed to have about 2 %~10 % fluctuation by relative standard deviation (this depends on the amount of sample used). This relatively low precision is attributed to the fact that there are many cases where the oxygen analysis value in metal is used for the detection of failures associated with the increase of an order of magnitude in the analysis value in daily product process management. In this sense, this method is considered to fulfill the required standard for uncertainty for use at production sites.Comment 2 (Yasuo Hasegawa)You compared the two different analysis methods in chapter 3, and discussed the cause of discrepancy with this analysis method. If you can assume a definite reason, shouldn’t you verify whether the degree of agreement improves when the cause of discrepancy is eliminated? If not, I think you can just say that the results agree within the margin of error.Answer 2 (Akira Tsuge)Since analysis results for the sample no.3 did not agree within the margin of error, we investigated the cause of discrepancy. In the comparison among the analysis methods with different analysis principle, investigating the differences originating from the analysis principle is necessary even if the result agrees within the error range; in inert gas fusion method, oxygen contained as oxide in the sample is analyzed, whereas oxygen of any form in the sample is done so in CPAA method. In this case, the difference exceeded the error range in sample no.3.6 Validity of the result of the joint analysis experimentComment (Masahiro Okaji)In 3.3, you mentioned that as the result of the joint analysis experiment by AIST and three other institutions, AIST and one institution were able to have measured values. However, you did not show the data so the reader cannot examine the validity. I think you have to demonstrate how the consistency is improved by knowing the relationship of the “applied power – graphite crucible temperature” using experimental data.Answer (Akira Tsuge)The results of the joint test in Japan were summarized in Table 3. I also added the explanation of each analysis value and the reason for ND (not detected).7 International standardizationComment 1 (Yasuo Hasegawa)I think the process for international standardization is informative for people involved in such a project. I think it will help such efforts if you show the process in a chronological table.Answer 1 (Wataru Kanematsu)The efforts for international standardization are described in chapter 4 in chronological order. For a better understanding, I organized the contents in Table 4.Comment 2 (Masahiro Okaji)In 4.2, the position of Korea is unclear in the ISO standard proposal. Since international cooperation is important,


page 38