Thermionic Currents with Potentials. 55T 



considered that for a long time the ions are formed as fast 

 as they are removed. The rate of removal depends on the 

 rate of drift across the film due to thermal energy and to 

 the applied field. Since the rate of escape is a function of 

 the rate of drift to the surface, and since the rate of drift 

 depends on the applied potentials, the emission clearly depends- 

 on the applied potentials. For potentials high enough for 

 the drift under the applied field to be relatively more im- 

 portant than the drift due to the thermal energy there 

 would be a nearly linear relationship between currents and 

 potentials. This agrees with observation for potentials above- 

 40 volts. For the current below 40 volts emission is a 

 function principally of the escape due to thermal energy, and 

 the current-potential curve is analogous to the unsaturated 

 portion of an electronic current-potential curve. This, again, 

 agrees with experiment. The initial condition where ions- 

 escape without having to cross a region of neutral material 



Az 

 should give a small value of — . The later stage should give 



a large value whose magnitude depends upon the thickness- 



Ai 

 of the film. Hence the observed initial growth in -r- is- 



° i 



explicable also. It is known that wires may be " ag^d out." 

 That is, after sufficient heating no more positive ionic currents- 

 are obtainable. This decay of the positive current corre- 

 sponds to the using up of ionizable material. Presumably 

 the film gradually evaporates. The last traces of film, no. 

 doubt, persist for a long time, but in many cases at least the 

 film becomes very thin. Now observation showed that in all 



cases there was a value of -.- as long as the currents were- 



i ° 



obtainable. In most cases the final value was relatively 

 small, something of the order -1. In some cases the value of 



— was large even to the end. 



The over-shooting is also consistent with the theory of 

 ionization within the surface-film. The escape of ions at, say,. 

 40 volts is a function of the rate of drift to the surface of the 

 film. Owing to the small velocity of drift with 40 volts, 

 there is a relatively large number of ions near the surface 

 of the film. If now the rate of drift is accelerated by 

 changing the potential to 400 volts, there occurs a relatively 

 large current during the time in which a new equilibrium ion. 



