Limitations on Metal Fixed Points Caused by Trace Impurities

Abstract

Freezing points of metals are used as reference temperatures, and a useful criterion for monitoring the realization of a point is that melt and freeze should provide the same temperature. This criterion is limited by the sensitivity of measurement and by the range of melting temperature caused by impurity. An alternative is to calculate the effect of impurities on the temperature from a knowledge of the phase diagrams and of the impurity concentrations. Experiments on tin points with additions of up to 56 ppm Fe and 36 ppm Sb are described and it is shown that such criteria are inadequate. The implications of this are discussed, and it is concluded that for high precision (better than 5 mK) it is necessary to use very pure samples from different sources and purified by different techniques, and to confirm experimentally that they provide the same temperature.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 16 citations exist upstream.