THERMIONIC ELECTRON EMISSION 



457 



An examination of a number of thoriated tungsten filaments with a 

 microscope showed that the diameter of the tungsten crystals was 

 of the same order as the values of h given above, namely 10~^ to 10^' cm. 

 These filaments had been given the customary heat treatment in a 

 vacuum at temperatures near 2800° K. for times measured in minutes 

 and at temperatures near 2100° K. for many hours. It was natural, 

 therefore, to form the hypothesis that different crystals have different 

 adsorptive properties and that consequently different crystals in the 

 same filament should be covered with varying amounts of thorium. 

 Different crystals will then have different work functions. The values 

 of ju found by the above analysis are consistent with this hypothesis 

 since 2ju, which is the difference in work function between a hill checker 

 and a valley checker, is always considerably less than 2.0 volts which 

 is the maximum difference in work function between clean tungsten and 

 thorium on tungsten. 



5 10 15 20 25 30 35 40 45 50 55 60 65 



"\/~F~ (V VOLTS PER cm) 



Fig. 20 — Experimental log i vs. V F curves for thoriated tungsten for various values of 

 /at 1270° K./ = fraction of surface covered with thorium. 



If this hypothesis is true, then as a thoriated tungsten filament is 

 activated, the curves for the various stages of activation should all 

 correspond to approximately the same value of b. Figure 20 shows 

 such a family of experimental curves taken by W. H. Brattain of these 

 laboratories. Curves 3, 4, 5 and 6 were taken at T — 1270° K.; curve 



