Vol.2 No.3 2009
Research paper : A marked improvement in the reliability of the measurement of trace moisture in gases (H. Abe)−214−Synthesiology - English edition Vol.2 No.3 (2009) sensitivity coefficient was determined from the experiments.The process of evaluating uncertainty involves completing Eqs. (3) and (4) by similarly evaluating the uncertainties of other physical quantities in the two equations by an experimental or theoretical consideration. In this process, a table of uncertainty evaluation is normally created to organize and visualize the results. The uncertainty evaluation is complete only when this table is filled. The table of uncertainty evaluation that we completed in this research is shown in Table 2 (including uncertainty components in Eq. (3) only). Generally, in establishing a standard, uncertainty evaluation is as difficult as or more difficult than the development of the instruments used in the standard.3.4 Summary of establishment of the trace moisture standardFrom the results shown in sections 3.2.1-3.2.3, we finally succeeded in developing a trace moisture generator that can generate trace moisture in the range between 12 nmol/mol (ppb) and 1200 nmol/mol (ppb) (corresponding to a frost point between -100 ºC and -75 ºC at atmospheric pressure), and we achieved the goal for the generation range. An uncertainty of about 3 % was achieved, as seen in Table 2, surpassing the target of 11 % for the relative standard uncertainty at 12 nmol/mol (ppb). Thus, the trace moisture standard was established as intended.In this research, we consider that the selection of measuring instruments was extremely important. Without the magnetic suspension balance, CRDS trace moisture analyzer, and critical-flow Venturi nozzle flow meter, we believe that the establishment of the trace moisture standard would have been extremely difficult. One key to success was that, although the CRDS trace moisture analyzer and critical-flow Venturi nozzle flow meter were initially not considered, we were able to notice the problems in the instruments used initially in the relatively early stage of the research and were able to introduce new instruments relatively soon after that. Another key factor was attributable to the effort spent on customizing the instruments used in this research. Although not hitherto mentioned in this paper, we adequately discussed our special requirements with the manufacturers before installing their instruments and also the improvement as needed after installation.4 Preparation of the calibration service systemOn the basis of the trace moisture standard established in this research, AIST initiated the calibration service for measuring instruments in the range between 12 nmol/mol (ppb) and 240 nmol/mol (ppb) (a frost point in the range between -100 ºC and -85 ºC at atmospheric pressure) in May 2007. The calibration range was expanded in May 2009, and we now provide a calibration service for the range from 12 nmol/mol (ppb) to 1200 nmol/mol (ppb) (a frost point in the range between -100 ºC and -75 ºC at atmospheric pressure). The Chemicals Evaluation and Research Institute (CERI) developed a trace moisture generator using a diffusion tube method without a magnetic suspension balance, and using the CRDS trace moisture analyzer as a reference standard for obtaining the traceability to AIST, they initiated a calibration service for the range between 12 nmol/mol (ppb) and 1200 nmol/mol (ppb) in July 2009.5 High-performance trace moisture analyzerEven if a trace moisture standard and calibration service are available, highly reliable measurement cannot be realized without a high-performance trace moisture measuring instrument, and the task of finding such an instrument is necessary. Here, the results of a preliminary performance test using the AIST trace moisture standard on some commercially available trace moisture analyzers are described.5.1 Performance test on CRDS trace moisture analyzerAs mentioned in section 3.2.2, in the process of establishing the standard, we found that the CRDS trace moisture analyzer has an excellent performance in measuring trace moisture. Figure 13 shows an example of the measurement of trace moisture using the CRDS trace moisture analyzer. It is demonstrated that the CRDS trace moisture analyzer has good response and stability. The result also demonstrates the good response and stability of trace moisture generation using the AIST standard generator. It was confirmed that the CRDS trace moisture analyzer is capable of measuring trace moisture as low as about 1 nmol/mol (ppb) with excellent linearity in the range between 12 nmol/mol (ppb) and 1400 nmol/mol (ppb) from the comparison with the standard.5.2 Performance tests on other commercial measuring instrumentsA similar performance test was carried out on some conventional chilled-mirror hygrometers. The tested hygrometers were the products of two companies, and the lower limit of the measurement range specified in the catalogs was a frost point of -100 ºC (14 nmol/mol (ppb)) or less for both products. The results are shown in Fig. 14. In this test, the product of Company A presented problems 0.435.20.3002.3760.2494.6021200.00.442.20.3000.9900.1041.918500.00.530.530.3000.1980.0210.384100.001.60.780.3000.0990.0590.71550.002.10.420.3000.0400.0240.28620.002.90.350.3000.0240.0140.17212.00 Standard valueUncertainty component ・Evaporation rate・Adsorbed/desorbed moisture・Flow rate of dry gas・Residual moisture in zero gasCombined standard uncertaintyRelative combined standard uncertainty [%]*All units are in nmol/mol (except relative combined standard uncertainty)Table 2 Evaluation table for uncertainty of trace moisture generation*.