Overcoming the Limitations of the SIE and OME Methods in Assessing the Effects of Impurities in Temperature Fixed Points.

Impurities are still among the primary concerns regarding the realization of many fixed points of the International Temperature Scale (ITS-90). Several methods have been suggested to correct for these effects. The most promising strategy, with regard to the achievable uncertainty level, is the 'sum of the individual estimates' (SIE) method. It involves a chemical analysis of the material and a calculation of each of the detected chemical species' effect on the phase-transition temperature of the fixed-point substance. This correction can be accurate only if all the detected impurities are completely dissolved. Given the recent evidence for insoluble impurities in metal fixed points, this strategy needs to be modified; otherwise, it may lead to an inaccurate estimation of the impurity-related effect on the fixed-point temperature. In this article, a correction method is set out that reflects the crucial distinction between soluble, insoluble, and partially soluble impurities. This 'sum of the individual estimates for the dissolved species' (SIEDS) method starts from a chemical analysis but takes into account only the dissolved particles. For this purpose, different types of substances are considered as possible dissolved impurities and are discussed from a chemical point of view. For those impurities where data are insufficient, only an uncertainty estimation is possible. For this purpose, the 'overall maximum estimate of the dissolved species' (OMEDS) method is derived from the SIEDS method as the new counterpart to the well-known 'overall maximum estimate' (OME) method. [ABSTRACT FROM AUTHOR]