Vapor pressure measurements over supercooled water in the temperature range from −10 1 °C to +10 −2 °C

Abstract

An accurate measurement of saturation vapor pressure of supercooled water is a strong challenge in metrology, mainly due to difficulties concerning keeping water at a liquid state at temperatures well below the melting point; thus few experimental data covering limited temperature ranges (down to about 253 K) are reported in literature. For this reason , an investigation of the water vapor -supercooled water equilibrium along the saturation line is carried out at Istituto Nazionale di Ricerca Metrologica (INRIM). Measurements cover the temperature range from 261.26 K to 273.25 K, corresponding to a saturation vapor pressure from about 244 Pa to 611 Pa. The experimental apparatus includes a pyrex sample cell, kept in a liquid bath at a constant temperature with a millikelvin stability and connected to a manifold where the pressure is measured using one capacitive diaphragm pressure gauge. In this work, the water sample preparation, the measuring method and measurement corrections are reported; moreover, a comparison between experimental and literature data is conducted along with the most used vapor pressure formulations. Measurement results are discussed and uncertainty sources estimated. The resulting expanded relative uncertainty (k = 2) varies from 0.058 % at 273.25 K to 0.160 % at 255.26 K.

Accurate vapor pressure measurements of supercooled water in the temperature range between 252 K and 273 K

“…In order to make the data given by Bottomley [16] comparable with all the others, they have been re-adjusted as differences between the saturation vapour pressure of supercooled water and that of a corresponding stable ice phase at the same temperature calculated using the IAPWS 2011 formulation [31]. As Figure 3 highlights, the results of this work and measurements previously carried out by Beltramino et al [19] compare favourably; they are consistent within their combined uncertainty.…”

“…For the sake of clarity, the uncertainty bars for the first and the fourth run are shown. Measured pressures are also compared with the measurements available in literature [15][16][17][18][19]26]. All the data are plotted as differences, ∆p, with respect to the IAPWS G12-15 equation.…”

“…The latter were able to measure saturation vapour pressure at temperatures down to 254 K. Even wider was the temperature range explored by Fukuta and Gramada in 2003 [18], from the melting point down to 243 K, by using a dew-point hygrometer method. Recently, Beltramino et al [19] was able to carry out accurate saturation vapour pressure measurements down to 261 K, maintaining water in supercooled state by adopting a degassing procedure based on several cycles of water freezing, high-vacuum pumping and thawing under low pressure. Keeping water in a supercooled state becomes increasingly demanding as the temperature decreases from the melting point, especially for the long time intervals (several hours) usually required for accurate measurements.…”

“…This state is limited to nanoscopic volumes of water (droplets with diameters of some micrometres confined in micropores) lasting at most some milliseconds. New measurements of the saturation vapour pressure over supercooled liquid water along the saturation line have been carried out at INRIM in the temperature range between 252 K and 273 K. The aim of the work was to investigate a range where few data were previously reported [15][16][17][18][19]26]. Although results of notable experiments, saturation vapour pressure values reported in literature are often lacking of a clear uncertainty assessment and few information about their traceability can be deduced.…”

“…Moreover details on the instruments calibration against INRIM reference standards are explicitly mentioned. The significance of the comparison with literature data [15][16][17][18][19]26] and known formulations [2][3][4][5][6][7][8][9][10][11][12][13] is also commented.…”