Vol.2 No.1 2009

26/88

Research paper : Expansion of organic reference materials for the analysis of hazardous substances in food and the environment (T. Ihara et al.)−23−Synthesiology - English edition Vol.2 No.1 (2009) 5.2. We therefore determined that it is the ideal RM among the national RMs currently available for quantitative NMR. Certain national RMs at AIST, such as potassium hydrogen phthalate (NMIJ CRM 3001-a) and 1,4-dichlorobenzene (NMIJ CRM 4039-a), qualify for condition 1), but potassium hydrogen phthalate does not dissolve easily in organic solvents, and therefore, fails to satisfy condition 2) in our view. Similarly, 1,4-dichlorobenzene is highly sublimable and does not meet condition 3). At present, no national RMs have been developed specifically for quantitative NMR. We are currently in the process of developing the AIST national RMs that satisfy condition 4) as well as 1) to 3).5 Final status of primary standardsQuestion and comment (Akira Ono)You assert that, in principle, the ideal outcome of the application of quantitative NMR would be the development of a single primary standard that serves as the national RM for all organic compounds. Realistically, how many national RMs do you expect are required when this future traceability system is completed? Do you have any specific candidates in mind as organic compounds for the national RMs?Answer (Toshihide Ihara)In this study, our priority was to minimize the number of national RMs required, thus reducing development time and expense. That is why we proposed the use of transfer materials in the multi-stage calibration process. Benzoic acid has served as the primary standard for all organic compounds we have measured so far. This success gives us confidence that a traceability system based on a single national RM can be constructed for all organic compounds for which 1H NMR measurement can be performed.On the other hand, such a traceability system has its disadvantages. Multi-stage calibration is time-consuming and increases uncertainty. If the accuracy or swiftness of analysis becomes more important for users, it is necessary to develop multiple national RMs with different polarities and chemical shifts. We are looking at ways of restricting calibration to single stage. To handle organic compounds that do not have protons, it is necessary to develop quantitative NMR for other nuclei, such as phosphorus and fluorine, along with the corresponding national RMs.6 Preparation and use of transfer materialsQuestion and comment (Akira Ono)I ask about how the transfer materials are used. When this new, efficient traceability system is completed in the future, will AIST produce, store, and disseminate these transfer materials as needed? Or can the reagent manufacturers that produce working RMs make the transfer material when needed, and dispose it when they are done? Answer (Toshihide Ihara)In our paper, we envisioned the transfer material to be prepared by the developers or suppliers of the working RMs (RM producers) according to their objectives. To ensure appropriate evaluations, the transfer materials will not be prepared for each batch, but the RM producers will be responsible for producing and storing them for a certain period.Also, as described in chapter 7, if quantitative NMR becomes widely used as a quantitative analytical method for organic compounds, prepared transfer materials can be used. Moreover, AIST or RM producers may supply easy-to-use transfer materials as RMs.7 Comparison of quantitative NMR and freezing point depression methodQuestion and comment (Akira Ono)My question concerns the analytical results in Table 2. In the freezing point depression method, uncertainty for purity determinations rarely exceeds the upper limit of 100 %, whereas in many cases using quantitative NMR, the upper limit for analytical result exceeds 100 %. Such results are unreasonable. Since the freezing point depression method directly measures impurities in pure substance, the upper limit for analytical result over 100 % is rare. Using quantitative NMR, on the other hand, measurement of the main components is performed when the concentration of the pure substance is diluted to about 1000 mg/L. Isn’t this one reason why the upper limit for analytical result can rise above 100 %? Isn’t this the case where quantitative NMR is fine for measuring components in a solution but is inappropriate for measuring the purity of pure substances? If so, quantitative NMR seems to be most promising for Product Realization Research surrounded by the dotted line in Fig. 6. I’d like to hear the authors’ views on this.Answer (Toshihide Ihara)Although the factors contributing to the uncertainty of quantitative NMR are not separated in Fig. 5 between preparation uncertainty and measurement uncertainty, preparation uncertainty is not relatively small. Thus, when applied to purity determination, quantitative NMR is undeniably inferior to the freezing point depression method in terms of uncertainty for preparation of solutions, and purity determination higher than the upper limit for analytical result exceeding 100 % is obtained as a result, as you pointed out (however, this does not indicate any bias in the purity determinations).The freezing point depression method cannot be applied to measure concentrations of components in solution, but there are many examples where the characteristics of quantitative NMR can be applied, as you also pointed out. Because many organic solvents contain hydrogen, we must find ways of reducing these effects so NMR can be applied to protons. In Product Realization Research, including the development of quantitative NMR equipment, solving the issue of protons in solution and enabling measurement of concentrations of components in solution are keys to establishing the use of quantitative NMR.8 Other candidates for universal calibration technologiesQuestion and comment (Akira Ono)In chapter 7, “Future Directions,” you raised the possibility that universal calibration technologies other than quantitative NMR may be found in the future. Are there any candidate calibration technologies at this time?Answer (Toshihide Ihara)In section 4.1, we stated that a universal calibration technology should theoretically be an analytical method qualified as a primary ratio method (measures the value of a ratio of an unknown to a standard of the same quantity; its operation must be completely described by a measurement equation).Although not yet established as an analytical technique, one candidate the Authors are examining is a combination of chromatography and atomic emission spectrometry. In this process, the analytes are separated from the sample by chromatography. Then each analyte is introduced into high-temperature plasma and atomized into constituent carbon, hydrogen, oxygen, and other atoms. These atoms can then be measured to find the emission of spectrally separated (for example) carbon atoms. By adding a primary standard containing a known quantity of carbon to the sample, the quantity can be combined with the emission of carbon to find the quantity of analyte, as the primary standard itself is also atomized. The point here is that the efficiency of atomization is not dependent on the molecular species. Currently, the combination of gas chromatography and helium-plasma atomic emission spectrometry can obtain uncertainty of 5 % (95 % confidence interval). Further

※このページを正しく表示するにはFlashPlayer9以上が必要です