﻿550 Photoelectric Properties of Thin Metal Films. 



could be photographed through the thickest film used. The 

 wave-length o£ this line was 2302. The spectrum was a line 

 spectrum. 



Currents. — The results obtained from the measurements 

 of currents are : — 



1. That for thin films the maximum emergent current is 

 harper than the maximum incident current, and vice versa 

 for thick films. 



2. For emergent light, the current rises rapidly with 

 potential difference at first, and approaches its maximum 

 value for a small value of this potential. The turning-point 

 of the curves between the rapid increase and the slow in- 

 crease to the maximum value, is at about 2 volts for the 

 18-minute and the 55-minute films, and at about 4 volts for 

 the 8-minute film. For incident light, the rise of the current 

 with the potential difference is not so great at first as for 

 emergent light, and the turning-points to the saturation- 

 values are at much higher potentials, of the order 8-12 volts. 



These results show that light assists electrons to leave 

 a metal more in the direction of the light than in the opposite 

 direction for thin films, but when the films are thick the 

 reverse is the case. 



Without attempting to explain the whole phenomena, 

 attention might be called to the difference in the form of the 

 curves for incident and emergent currents (figs. 3-0). This 

 difference cannot, I think, be explained by a dissymmetry in 

 the tube. 



An explanation of the form of the curve for thick metals 

 was put forward by Ladenburg and Markau *. They supposed 

 that the same number of electrons leave an electrode no 

 matter what the potential difference between it and the 

 receiving electrode is. The fact that a certain potential 

 difference is necessary in order to obtain the maximum 

 current, they explain as being due to electrons being reflected 

 from the receiving electrode and getting back to the emitting 

 electrode when the potential difference is small. If this is 

 the case, then the form of the curve ought to be the same for 

 incident and emergent light. Again, in the present experi- 

 ments the receiving electrode is the w T alls of the vessel, and 

 so is perpendicular to the emitting electrode, so that any 

 reflected electrons ought to go to other parts of the walls and 

 not get back to the film. 



The original explanation of the form of the curve given 

 by Lenard t, that a certain potential difference is necessary 



* Verk. d. deutsch, Ffrijs. Ges. 1908, n. 14, p. 562. 

 t Ann. der Phys. viii. 1902, p. 149. 



