The Effect of Water Vapour on the Diffusion Coefficients of Ions in Nitrogen and Oxygen
Proceedings of the Royal Society A
In a previous paper* an account has been given of an investigation of the effect of water vapour on the diffusion coefficients and mobilities of ions in air. I t was found th at as the concentration of water vapour in the air varied, the values of the diffusion coefficients for positive and negative ions showed rather wide oscillations. These oscillations appeared to be of an irregular character, but a long series of experiments showed th at the values of the coefficients found for any vapour
... nd for any vapour pressure could be reproduced with a con siderable degree of accuracy. In a subsidiary investigation, it was found that the mean mobilities of the ions varied in the same way with vapour pressure, if the fields to which the ions were exposed were less than about 1 • 4 volts per centimetre. In stronger fields the oscillations were suppressed, the mobility values plotted against vapour pressure lying on a smooth curve. The method used for the determination of the diffusion coefficients was practically the same as th at originally used by Townsend. The chief difference was the employment of only one long and one short tube for the capture of the diffusing ions, instead of the sets of 24 tubes in parallel used in Townsend's apparatus. The ionization was produced by a-particles from polonium. A rather small volume of air, enclosed and kept at a definite humidity, was passed alternately through the long and the short tube, after exposure to the a-radiation, and the diffusion coefficient was calculated from the ratio of the concentrations of the ions in the issuing air by the use of Townsend's formula. W ith this method of working, there will be an accumulation in the air of ozone and oxides of nitrogen formed by the action of the a-particles.t Since the influence of these bodies on the ionization has been established, ^ seemed possible that the periodic variation in the diffusion coefficients might be caused by their presence. The examination of nitrogen and oxygen separately was * J .