The densities of air-free water have been measured with an accuracy of about 3 ppm between 5°and 80°C. Since the relative measurements were more precise than the absolute measurements, the densities were adjusted to bring them into agreement with the defined density at 4°C. This was done by calculating the weight of the water in the dilatometers from the densities of Chappuis as expressed by the Tilton and Taylor equation up to 40°C. Comparisons are made with other literature values, particularly above 40°C, including a close examination of the water density data of Owen et al. Finally, a modified form of the Tilton and Taylor equation was derived to fit the water density data from 0-80°C with a mean absolute deviation of 0.7 X 10-e g/ml.Since we believe these values represent the best data available at this time, a table of water density values calculated from this equation at 0.1°intervals over the entire range of 0-80°C is also presented. This laboratory has been investigating the thermodynamic properties of aqueous rare earth salt solutions for some years, and as part of this study, precise thermal expansion data were needed. The dilatometric method of measurement chosen to obtain the thermal expansion data required an accurate measurement of thermal expansions of the dilatometers themselves. To this end, water was initially chosen as the calibrating liquid as it was readily available in high purity, and accurate, consistent density data over the required temperature range of 5-80°C seemed to be available.The water density data of Chappuis (3) covering the temperature range of 0-42°C as represented by the Tilton and Taylor equation (12) are generally accepted as a standard in this country as well as by the International Bureau of Weights and Measures. Furthermore, these data have been extended to 80°C by Jones et al. ( 6) and Steckel and Szapiro (11), and to 85°C by Owen et al. (9). Unfortunately these sets of data differ by as much as 27 X 10~6 g/ml at 80°C which is approximately 10 times what we believe to be our allowable error for calibration purposes. Owen et al. attributed this difference to the use by Jones et al. of only a linear equation as opposed to a quadratic equation to represent the volumes of their dilatometers as a function of temperature. This criticism appeared reasonable, and thus the density data of Owen et al. which extended the Chappuis data to 85°C was assumed to be correct.Owen et al. had found that the expression, AV/At = a + b£av (1) in which iav is the mean temperature of the interval At, represented the volume expansions of their dilatometers over the temperature range of 5-85°C. As the borosilicate glass dilatometers used were similar in size and shape to the ones used in this research, it was expected Equation 1 would be obeyed in this research as well. However, when the densities of water determined by Chappuis and by Owen et al. were used as the standards for our data, the AV/At vs. £av plots obtained were not quite linear. This fact plus the disagreement between the data of ...

No abstract

No abstract

No abstract

No abstract