An experimental investigation of thermal conduction through vapours

Both the viscosity and the thermal conductivity of a gas are, according to Maxwell’s Law, independent of pressure over a wide range. This law is deduced by the application of the kinetic theory to a gas that obeys the gas laws, and there is, for real gases under normal conditions, abundant experimental evidence of its truth. In the case of vapours, however, the position is very different. The viscosities of various vapours have been made the subject of several investigations, but there is little direct evidence that the viscosity of an almost saturated vapour does not vary with its pressure. For instance, Rankine, in his work on the viscosity of bromine vapour, found that, although similar values could be obtained by different methods—using widely different pressure conditions—when the vapour at these pressures was at a temperature several degrees higher than its condensing point, definite irregularities were obtained when it was only slightly superheated. The same irregularties were noticed by Braune, Basch, and Wentzel for bromine vapour and, to a less extent, are apparent in their work on mercury vapour. Similar effects appear in the determination of vapour viscosities by Braune and Linke. Again, in C. J. Smith’s account of his experiments on the viscosity of water vapour it is admitted that, although the results, which were obtained under conditions of considerable superheating, agree fairly well with those of Speyerer for almost saturated steam, the methods of calculation depend, in both cases, on the validity of the gas laws when applied to the vapour. On the other hand, it has been shown by Boyd in work on the viscosities of nitrogen and hydrogen under pressure that the respective viscosities are increased by as much as 25 and 10% by an increase of pressure of about 100 atmospheres.