| Résumé | Impurities are believed to be one of the major issues in realizing the metal fixed-point temperatures of the ITS-90 with a low degree of uncertainty. This has raised interest in the individual effects of impurities on the phase-transition temperature of fixed-point metals. Surprisingly, impurities that do not affect a fixed-point temperature have been found experimentally. A possible explanation for this behavior is the formation of insoluble oxides of the added impurities consuming oxygen already present in the fixed-point cell (mostly as an oxide of the fixed-point metal). This is supported by several recent publications. However, all the results could be coincidental. This article presents more convincing proof for the formation of insoluble compounds born from impurities dissolved in the fixed-point metal. Based on refined doping experiments and using impurities that have not been investigated before, both the impurities' dissolution and the precipitation could be observed as an initial decrease (or increase) of the fixed-point temperature followed by a gradual return to its original value. The selected impurities (gallium and zinc in indium) were found to dissolve within a few days and precipitate out within no more than two weeks. The behavior of iron in indium was investigated as well, but the results are not conclusive. Finally, another series of doping experiments indicates that sulfur does not dissolve in indium in significant amounts, but forms insoluble compounds (probably sulfides) when added to the metal. This supports the general assumption that metal-non-metal compounds might be present in the cell without being noticed. © 2011 Her Majesty the Queen in Rights of Canada. |
|---|
