364 



BELL SYSTEM TECHNICAL JOURNAL 



deposition and are progressively wider and less prominent as the rate 

 of deposition is increased until for very rapid deposition the response 

 for all wave-lengths decreases rapidly from the beginning. In this 

 connection we may recall from the oxidation experiments that a slight 

 amount of diffusion or surface rearrangement may occur in such time 

 intervals as are available in the experiments involving slow deposition 

 of caesium. 



But the behavior of the different cells and even different positions 

 in the same cell may not be closely compared, for their initial states 

 were by no means identical. In each of these cells there was a differ- 

 ence in the colors of the two ends of the cathode due to a distortion 

 of the glow discharge in the quantitative oxidation step of their 

 preparation which was due to the special geometry of these cells 



.,4 6 8 10"2 2 



LAYERS OF CAESIUM DEPOSITED 

 4 6 8 10"' 2 4 6 8 10 2 



HOURS OF DEPOSITION 



50 70 



Fig. 18 — Spectral response changes in cell G, position 2. 



But certain gross conclusions may be drawn. In all cases the addition 

 of a few tenths of a layer of free caesium caused profound modification 

 in the spectral response characteristics. If we regard the results for 

 the slower rates of deposition as more typical of steady state conditions 

 of the cathode surface, it appears that we may increase the response 

 in the region about 7,500 A. at the expense of the response at and 

 beyond 10,000 A. by the addition of a few tenths of a layer of free 

 caesium. It should be remarked again, however, that in view of the 

 roughness of the surface these quantities should be reduced by a 

 factor of four, and the critical amount of caesium may well be less 

 than even a tenth of a layer on the actual surface. Further addition 

 of caesium causes a decrease in sensitivity at all wave-lengths, and 

 this effect is increasingly pronounced towards the infra-red end of 

 the spectrum. 



