Vol.2 No.3 2009

Research paper : A marked improvement in the reliability of the measurement of trace moisture in gases (H. Abe)−219−Synthesiology - English edition Vol.2 No.3 (2009) the water concentration, but there is variation in the values. For example, a difference of approximately 6 % was reported between a value in one study and another value cited in the same study for comparison (JQSRT 2005, vol. 94, pages 51-107). Similarly, the maximum difference is 20 % or more in another study (JQSRT 2002, vol. 75, pages 493-505). Therefore, rather than already having a sufficiently accurate value, I have an impression that the CRDS trace moisture analyzer used a value which happen to agree relatively well with the standard values. I hope that, by feeding back the results obtained using a method traceable to the SI, the uncertainty of the absorption cross section will be reduced, and the reliability of trace moisture measurement using cavity ring-down spectroscopy will increase further.The matrix gas dependence of the absorption cross section is not small, for example, there is a twofold difference in values for the absorption line between nitrogen and oxygen (Appl. Opt. vol. 44, pages 611-619). However, this relationship can be considered to be almost constant in a small temperature range, and therefore, once the conversion factor is accurately determined at a temperature around which measurement is performed, only the calibration of the instrument for nitrogen will be necessary, and the measurement of trace moisture in oxygen can be performed using the conversion factor. By this method, for matrix gases for which the conversion factor can be obtained with small uncertainty, the reliability of the measurement can be maintained even if there is no trace moisture standard for those matrix gases. Furthermore, the users can skip the calibration for those matrix gases and reduce the management cost. Therefore, I think this is an effective method of trace moisture measurement in gases other than nitrogen. However, when the absorption line of a matrix gas lies in the same wavelength range as the absorption line of water (such as in the case of NH3), we need to develop an analysis technology to compensate for this effect and to extract information on the moisture concentration. Depending on the type of gas, measurement cannot be performed at all because the absorption of the matrix gas is too strong, and this method may not be effective for all gas types. It is necessary to investigate for which type of gas this method is applicable by considering the uncertainty and the experimental difficulties of determining the conversion factor.4 National measurement standards and commercially available measuring instrumentsQuestion and comment (Akira Ono)By using the results of this research, it was clarified that many of the trace moisture analyzers that are already commercially available do not meet the specifications. I think this is an example where a significant social loss may occur owing to the unreliability of commercially available measuring instruments whose measurement capabilities are not adequately demonstrated because of the lack of the corresponding national measurement standard. What lesson should we learn from this?Answer (Hisashi Abe)Regarding the conventional commercially available measuring instruments, my understanding is that this situation arose because the instrument manufacturers and retailers endeavored to meet the demands of users who were actually seeking a measurement method of trace moisture, and this situation continued without the national standard for a long time. I think this must be considered as a serious matter, and we must strongly recognize the social importance and impact of standards. Considering the situation of the trace moisture standard of other countries and the period when the CRDS trace moisture analyzer, which was essential for realizing the standard using the diffusion tube method, became commercially available, I think it was technologically difficult to establish the trace moisture standard before 2001. Ideally, even in such a situation, there should have been a shared awareness among the users, instrument manufacturers and retailers, and researchers of the national metrology institute concerning the technological reasons that prevented the development of the standard and the reliability of the trace moisture measurement without the standard. This is not just the case for trace moisture measurement but can be said to apply to other measurements in a similar situation. How can the three parties share a similar awareness? In particular, this is very difficult to achieve in a situation wherein the researchers have negative opinions (although they are not necessarily correct) about the users’ demands. How do we obtain opportunities to publish and discuss such negative matters? It is not easy to bring up negative matters as topics at academic societies and in scientific papers, and even if such a matter is brought up, the probability that this information reaches the users, instrument manufacturers, and retailers is not high. If a technological issue is clarified, it can be published in Synthesiology as a research scenario, but even this is difficult without accompanying successful scenarios. One solution is to deepen the understanding of measurement among the users, instrument companies and retailers, and researchers by exchanging opinions using the occasions of visiting the laboratory and by attending the meetings of the Metrology Club organized by the NMIJ. I cannot think of any other particularly effective solution. I am still not fully able to understand what lesson can be learned from this case.However, concerning the past “social loss” related to trace moisture measurements, I believe the following. Before the trace moisture standard was established, there was no way of objectively investigating the measurement capabilities of trace moisture analyzers, and this fact itself was discovered only after the standard was established. Considering this situation, I think it is a bit too much to say that there was a social loss caused by the unreliable measurements. At the time, no major problems occurred because of the measurements (at least, on the surface), and rather than calling it a social loss, I think we should express it more positively: recently, it has been discovered that there is plenty of room for improvement in trace moisture control in industry.5 Application of diffusion tube method to organic standard gasesQuestion and comment (Akira Ono)In this research, the diffusion tube method was applied to water to generate a standard gas of trace moisture. I think that this method is also applicable to organic substances for the generation of standard gases with a known concentration value. What do you think about this possibility?Answer (Hisashi Abe)AIST is the first to succeed in using a diffusion tube method as a primary measurement standard, but this method was originally used as a method for producing a low-concentration standard gas for various substances including organic substances (Crit. Rev. Anal. Chem. 2005, vol. 35, pages 31-55). Although there are some limitations, such as it can be used only for substances with a certain degree of vapor pressure at the experimental temperature and for those that do not suffer from the dissolution of the matrix gas, it is generally applicable to substances other than water. In particular, for highly adsorbent substances and those that are unstable and for which long-term storage is difficult, it is difficult to maintain their reliability as a standard gas supplied in a high-pressure container, and the method of continuously generating the standard gas (dynamic method) such as the diffusion tube method is more feasible. The diffusion tube method is introduced in the International Organization for Standardization (ISO 6145-8) as well as in the Japan Industrial Standard (JIS K 0225) as a


page 46