[ 162 ] 



XIV. The Effect of Gases on the Contact Difference of 

 Potential between Metals at Different Temperatures. By 

 0. W. Richardson, F.R.S., Wheatstone Professor of 

 Physics, and F. S. Robertson, M.Z.E.E., Lecturer in 

 Electrical Engineering, University of London,. King's 

 College *. 



THE present paper deals with some observations of the 

 curves connecting thermionic electron current and 

 applied potential difference (characteristic carves) for the 

 case of a straight hot wire and a coaxal metal cylinder when 

 small known pressures of gases, principally hydrogen, are 

 present. The currents dealt with were rather small, being of the 

 order 7 x 10~ 8 amp., and approximate saturation was attained 

 under the experimental conditions with-accelerating potentials 

 of the order of a volt. Attention has been directed chiefly 

 to the region in which the curves approach the voltage axis. 

 In this region the shapes of the curves are determined by the 

 distribution of velocity among the emitted electrons, by the 

 geometry of the electrodes and the full of potential down the 

 hot wire. Whilst these factors were not exactly identical in 

 all the comparative experiments, it is believed that the effects 

 of variation in them are practically negligible, with the 

 result that the shapes of the different curves under comparison 

 are invariable. This is true, as may be seen from the 

 diagrams, figs. 2, 4, and 5, not only for the curves taken 

 under high vacuum conditions, but also for those taken when 

 small quantities of hydrogen were admitted. 



Whilst in general the curves are of the same shape they 

 are not in identical positions, but those obtained in the 

 presence of hydrogen are displaced along the voltage axis as 

 compared with those obtained in a vacuum. Inasmuch as 

 for a given maximum or saturation current the shapes of the 

 curves are the same, it is reasonable to infer that the infra- 

 saturation currents will be sensibly the same function of the 

 true voltage between the electrodes. Now the voltage 

 plotted in the diagrams is the voltage between the central 

 point of the hotwire and the surrounding cylinder as read on 

 a voltmeter which in effect is inserted between these points. 

 This differs from the true voltage by an amount equal to the 

 contact difference of potential between the surfaces plus an 

 effect arising from the drop of potential due to the heating- 

 current down the wire. With the arrangements used the 

 last effect is believed to have been sensibly the same in all 

 the experiments ; so that the displacement of the curves along 

 the voltage axis should be a measure of the change which 

 took place in the contact potential difference between the hot 

 * Communicated by the Authors. 



