CESIUM-OX YGEN-SILVER PIIOrOELECTRIC CELL 340 



We have wished to obtain information regarding the amount of 

 caesium in the surface film, the amount of free caesium in the underlying 

 body of the matrix, to check up on the possibility of the diffusion of 

 free caesium between the surface and body of the matrix and to study 

 the relation between film thickness and the spectral response of the 

 cathode surface. To this end we conducted one series of experiments 

 in which known quantities of oxygen were admitted to cathode surfaces 

 prepared in the standard fashion, following the course of recovery in 

 response with time at various temperatures. In further experiments 

 caesium vapor was allowed to deposit continuously upon the active 

 surfaces of several cathodes of the same type, while observations were 

 made of the resulting changes in spectral response. 



Special cells were made and pumped under as nearly as possible 

 the optimum conditions. The first series of cells were similar in 

 structure to that shown in Fig. 1, except that to the upper end of the 

 bulb was sealed a short glass tube. Within this tube and attached 

 to its upper end was a thin-walled glass bulb containing oxygen at 

 reduced pressure. A steel ball was also included within the tube 

 which could be made to break the glass bulb of oxygen by agitation 

 of the cell. The second series of cells were constructed as shown in 

 Fig. 7. In each cell was supported a glass cylinder closed except for 

 a small orifice pointing obliquely towards the cathode surface. Within 

 this cylinder were a thin-walled bulb of soda-lime glass containing 

 free caesium and a steel ball with which the bulb could be broken 

 as before. 



Monochromatic light was obtained from a Bausch and Lomb No. 

 2700 glass spectrometer fitted with two slits and calibrated by the 

 makers to 10,000 A. The light source was a lamp with a helical 

 tungsten filament operated at a color temperature of 2,710° K. Read- 

 ings were corrected to an equal energy scale using the relative energy 

 curve for tungsten and the dispersion curve of the spectrometer. 

 At the blue end of the spectrum readings were very small due to the 

 low intensity of the light, the low response of the cell, and the wide 

 dispersion of the spectrometer. The values presented are from 

 6,000 A to 10,000 A. The measurements were made with 90 volts 

 across the cells and the photoelectric currents were amplified on a 

 vacuum tube bridge circuit so as to be rapidly readable on a micro- 

 ammeter. 



In Table 2 are presented data from four cells to which oxygen was 

 admitted, giving the amounts of oxygen, the equivalent number of 

 atomic layers of casium oxidized, the various aging and heat treat- 

 ments, and the integral sensitivities obtained after the several treat- 



