X-RAY MICROSCOPY 



Fig. 7. Angular distributions in the forward hemisphere from an electron-opaque aluminum target, 

 for electron energies of 10.05 and 12.05 keV respectively. The curves in the upper quadrant represent 

 observed data. Those in the lower quadrant have been corrected for self -absorption within the target. 

 (Cosslett and Dyson, 1957). 



in the region of 6-12 kV. The data resemble 

 very closely that obtained by Kuhlenkampff 

 at 90°. Fig. 6 shows the spectral distribution 

 in the forward direction from a gold target 

 for four different accelerating voltages. 



Angular Distribution. The angular dis- 

 tribution of the total radiation from an elec- 

 tron-opaque target of gold was measured at 

 10 and 25 kV by Oosterkamp and Proper 

 (reported by Botden et al. (1952)). At the 

 higher electron energy there was a slight rise 

 in intensity at increasing angles to a maxi- 

 mimi at 15° from the forward direction, 

 showing that electron scattering does not 

 quite obliterate the type of distribution ob- 

 served with electron-transparent films. At 10 

 kV there was no rise (i.e., no minimmn in 

 the forward direction) , and it was considered 

 that this was due to absorption in the target, 

 the effect of which w^ould increase with in- 

 creasing angle. Their measurements were 

 made with an ionization chamber, for dosi- 

 metric purposes; this would in fact accentu- 

 ate the lower photon energies which show 

 little anisotropy even when an electron-trans- 

 parent target is used. 



In the work just referred to no attempt 

 was made to discriminate between the dif- 

 ferent energies within the continuous spec- 

 triun, but a series of measurements using 

 targets of aluminum, copper and gold has 

 been described by Cosslett and the writer 

 (1957, 1959a), in which a proportional 

 counter was used for energy discrimination. 

 Near the high energy limit the radiation is 

 pronouncedly anistropic, although much less 

 so than in the case of ideally thin targets. 

 For example, for aluminmn at 10 kV, at a 

 quantum energy of 9 keV, the anisotropy 

 (defined as the intensity per unit solid angle 

 in the direction of maximum emission di- 

 vided by that in the forward direction) is 

 calculated to be 4.1, for an ideally thin tar- 

 get, whereas the "thick target" value ob- 

 served in these measurements was 1.6 (see 

 Fig. 7). At lower quantimi energies the 

 anisotropy is even less and below about 0.5 

 of the high energy limit the radiation is vir- 

 tually isotropic. 



Comparison with theory is difficult, but 

 in principle it is possible to deduce the ex- 

 tent of electron scattering by comparing the 



658 



