234 



RADIATION BIOLOGY 



1948; Rodda, 1949). Figure 3-31 presents diagrammatically the elec- 

 trode arrangement of a typical nine-stage secondary-emission photo- 

 multiplier tube, the 931-A, which employs electrostatic-field focusing of 

 the electrons. A photoelectron emitted by the photocathode is acceler- 

 ated by a potential of 40-100 v to dynode 1, where several secondary 

 electrons are released. The secondary electrons are again accelerated to 

 dynode 2, and each one again releases several new secondary electrons. 

 The process is repeated from dynode to dynode until finally the avalanche 



^z 



MICA SHIELD 



Fig. 3-31. Diagram of the electrode 

 arrangement in a nine-stage electrostatic- 

 focusing photomultiplier tube of 931-A 

 type. The paths of the photo- and 

 secondary electrons are indicated for the 

 first two stages. {From Engstrom, 1947.) 



25 50 75 100 125 



POTENTIAL PER STAGE, v 



150 



Fig. 3-32. Effect of dynode voltage on 

 the current amplification in a nine-stage 

 931-A-type photomultiplier tube. {From 

 Engstrom, 1947.) 



of electrons is collected by electrode 10, the anode. If the secondary- 

 electron yield is R per primary electron striking the dynode, the number 

 of dynodes is n, and the photocurrent is u, then the anode current is 

 RHo. The secondary-electron yield varies from 1 to 10 depending upon 

 the nature of the dynode surfaces and the accelerating potential between 

 them. For the 931-A multipher with a potential of 100 v per dynode, 

 the yield R is over 4, which for nine stages gives an amplification of 

 about a million. 



The accelerating potentials may be obtained from a bank of small 

 hearing-aid-type 90-v B batteries or from a voltage divider across a regu- 

 lated power supply. The power supply must be very well regulated, 

 however, because, unlike that of the vacuum photoelectric cell of two 

 electrodes, the response of the multiplier is extremely sensitive to applied 

 voltage. The current amplification, and thus the over-all sensitivity, is 

 nearly proportional to the logarithm of the applied voltage, as shown in 

 Fig. 3-32. Varying the voltage per stage from 75 to 100 v produces 

 nearly a tenfold increase in sensitivity. 



At constant flux intensity and applied voltage, photomultipliers exhibit 



