Assessment of thermal immersion effects in the melting and freezing points defined by the International Temperature Scale of 1990 is one of the vital issues of modern thermometry. In documents of the Consultative Committee for Thermometry, the deviation of the experimental immersion profile from the theoretical value of the hydrostatic effect at a height of about 3 cm to 5 cm from the thermometer well bottom is used for the estimation of the uncertainty due to unwanted thermal effects. This estimation assumes the occurrence of solely the hydrostatic effect all along the height of the well inner wall. Real distortions of the temperature gradient at the bottom and at the top part of the well caused by the change of heat-exchange conditions are not taken into account. To define more precisely the temperature gradient along the height of the well, a miniature PRT with a 30 mm sensitive element and a sheath length and diameter of about 60 mm and 6 mm, respectively, were used. Also, the measurements of fixed-points temperature at noticeably different slopes of immersion profiles due to variations of the thermometer heat exchange and phase transition realization conditions were produced by means of a standard platinum resistance thermometer (SPRT). The measurements were carried out at the tin and zinc freezing points. The immersion curves measured with a miniature thermometer demonstrated an increase of the temperature during its lifting in the first 1 cm to 3 cm above the bottom of the well. The measurement results at the zinc freezing point by means of the SPRT have not confirmed the correlation between the immersion curves, the received value of the Zn freezing temperature, and the estimation of its uncertainty.

In CCT documents it is stated that "…for the freezing curves of the metallic fixed points, the maximum observed temperature on the plateau should be taken as the best approximation of the liquidus temperature. The fixed points should be realized with the inner and outer liquid-solid interfaces and extend past the maximum by 10 % to 20 % of the fraction frozen, to clearly establish the value of the maximum and the resolution of its determination." Also, it is accepted that "…the inner interface is essentially static. It is the temperature of the inner liquid/solid interface that is measured by the thermometer." The analysis of freezing curves obtained by the standard method of fixed-point realization shows that the parameters of the initial part of the freezing curve, the mean temperature value of which is usually taken as the liquidus temperature, depend on how the inner interface is initiated. Variations in the duration and intensity of initiation cause changes in the initial part of the freezing curve and in the resulting SPRT measurement. Moreover, the relation between the duration of the initial section of the plateau with a minor temperature change and the duration of its final section with a significant slope also depend on the initiation method used and on the furnace temperature. The effect of freezing initiation conditions on the measurement result is individual for each fixed point because of the differences in thermophysical properties of metals and in conditions of the heat transfer from the liquid-solid interface to the thermometer. Aluminum has a maximum value of the melting specific heat in comparison with other metals used in ITS-90 fixed points; in the present study, the effect of the intensity and duration of the inner liquid-solid interface initiation was investigated both experimentally and through calculation.

No abstract

No abstract