Vol.2 No.3 2009

Research paper : A marked improvement in the reliability of the measurement of trace moisture in gases (H. Abe)−210−Synthesiology - English edition Vol.2 No.3 (2009) The strategy of developing the primary measurement standard by the diffusion tube method can be described using Eq. (1) as follows. The measurement method for each of the physical quantities on the right side of Eq. (1) is established. A generator is developed to fulfill the conditions xw >> xb and N >> Nb at xw ~ 14 nmol/mol and N ~ 780 nmol/h, and Eq. (1) can be approximated as N and F are stabilized (their variabilities are reduced). A technology is developed to measure N and F with small uncertainty, and the standard value xw can be determined using Eq. (2).However, the actual development was not performed in this order. There were several technologies, both major and minor, that had to be developed, and when advances in one technology were achieved, other technologies had to be revised. The development did not progress in one direction only.3.2.1 Measurement of evaporation rateN in Eq. (2) can be obtained by measuring the evaporation rate [g/h] and by dividing it by the molar mass of water (18.02 g/mol). As mentioned in section 3.1, we planned to use a magnetic suspension balance for the measurement of the evaporation rate. The magnetic suspension balance is structured so that the diffusion cell placed inside the generation chamber is suspended on an exterior electronic balance by a magnetic force. By using this balance, the mass-loss rate of the diffusion cell owing to water evaporation (evaporation rate) can be measured continuously without interrupting the generation of water vapor. The magnetic suspension balance at AIST was specially customized for the development of the trace moisture standard; the inner surface of the balance in contact with the gas is electropolished because it is part of the generation chamber; the balance can measure a change in mass of the diffusion cell of approximately 12 g at 1 µg resolution; the indication of the balance can be calibrated using external weights traceable to the SI, and so forth. Figure 7(a) shows an example of a measurement using the magnetic suspension balance. It can be seen that a small evaporation rate of 5.1 µg/h can be measured with sufficient accuracy. Figure 7(b) shows an example of a measurement without using the magnetic suspension balance. The generation chamber was exposed to the atmosphere at each mass measurement of the diffusion cell, and the diffusion cell inside the generation chamber was measured by suspending it using a wire connected to the hook of the electronic balance placed above the generation chamber. The diffusion cell used was the same as that shown in Fig. 7(a). For further details of the experiment, refer to Ref. [12]. In this case, an evaporation rate of 6.6 µg/h was measured, and it appeared that this method would suffice except for the fact that it was time-consuming. In fact, the diffusion tube method is often used as a simple trace moisture standard, after the evaporation rate has been determined in advance using gravimetry without using a magnetic suspension balance in a manner similar to that described above. This evaporation rate was used as a constant value under the same experimental conditions, in particular, temperature and pressure. However, from the comparison with values calculated theoretically, a problem was found in the experiment illustrated in Fig. 7(b). Assuming ideal diffusion in the diffusion tube method, the evaporation rate can be calculated from the inner diameter and length of the diffusion tube, the pressure in the generation chamber, and the temperature of the water inside the cell. The details of the calculation method are provided in Ref. [5]. In the case of the experiment illustrated in Fig. 7(b), the theoretical value was 4.1 µg/h, which differed slightly from the measured value. This difference could not be explained on the basis of the uncertainty of pressure, temperature, or other factors. In the experiment using the magnetic suspension balance (Fig. 7(a)), the theoretical value was 5.1 µg/h, which was consistent with the measured value. To confirm that this was not a coincidence, we performed experiments using diffusion tubes with different inner diameters and different pressures and temperatures in the generation chamber, and it was found that the theoretical and measured values were consistent Fig. 7 Data of mass loss measurement of diffusion cell.MSB stands for magnetic suspension balance.Fig. 6 Schematic diagram of mass loss measurement of diffusion cell using magnetic suspension balance.Diffusion cellTrace moisturein gasElectromagnetDry gas (N2)ControllerMagnetElectronic balancePosition sensor(a)With MSB(b)Without MSBTime / hMeasured value : 6.6 µg/hTheoretical value : 4.1 µg/hTime / hMeasured value : 5.1 µg/hTheoretical value : 5.1 µg/hMass loss / µgMass loss / µg000050100150200-1600-1200-800-400-300-200-100204060xwNF〜〜(2)


page 37