An Experimental Investigation of the Long-Term Stability of Triple-Point-of-Water Cells

Abstract

Contamination of triple-point-of-water (TPW) cells by the chemical components of the borosilicate glass that contains the water is now widely recognized as the principal contributor to long-term drift of the cell temperature. To add to the available experimental data, a comparison of 24 TPW cells of various ages (from 10 years to 59 years), manufacturers (NRC, Jarrett, Isotech), and materials (borosilicate glass and fused quartz) was undertaken in 2013. Twelve cells from this group were compared to one another in 1997. By comparing the current inter-cell temperature differences to those determined 16 years earlier, it was found that some cells have remained stable, others have become colder (as might be expected from ongoing dissolution of the glass), and one or two show an apparent increase in temperature that seems anomalous. Also included among the 24 cells are five cells of borosilicate glass and five of fused quartz that were purchased 10 years ago. By comparing the relative temperature differences among this group of borosilcate and fused-quartz-encapsulated cells to the values obtained when they were last compared 6 years ago, it was found that the average temperature of the borosilcate group of cells decreases by \(-6\,\upmu \mathrm{K}\,{\cdot }\,\mathrm{year}^{-1}\,({\pm }2\,\upmu \mathrm{K}\,{\cdot }\,\mathrm{year}^{-1})\), in reasonable agreement with an average drift of \(-4\,\upmu \mathrm{K}\,{\cdot }\,\mathrm{year}^{-1}\) suggested 12 years ago. It was concluded that fused quartz is the superior container for TPW cells.