X-RAY MICROSCOPY 



i-i-i-'-i-w-L.jji 



w 



20 



SO\q\I 



Fig. 3. Spectral distribution form a thin alu- 

 minum target (accelerating voltage 34 kV) at 90° 

 to the electron beam (Amrehn and Kuhlenkampff, 

 1955). 



Fig. 4. Angular distribution from a thin alu- 

 minum target (accelerating voltage 34 kV) 

 (Kerscher and Kuhlenkampff, 1955). 



intensity (i.e., energy flux per unit solid an- 

 gle) in the forward direction is minimal and 

 that the intensity passes through a maxi- 

 mum at an angle dependent upon the elec- 

 tron energy and the quantum energy. Bohm 

 (1937-8) and Honerjager (1940) carried out 

 more detailed measurements (in the same 

 region of electron energy) on targets of alu- 

 minum and magnesium and confirmed the 

 essential aspects of the earlier investigations. 

 The maximum occurs at smaller angles as the 

 electron energy is increased; but for a fixed 

 electron energy, it is the softer components 

 of the spectrmn which have the smallest an- 

 gle of maximum intensity. Bohm and Honer- 

 jager observed additionally that, with de- 

 creasing target thickness, the forward mini- 

 mum became more marked, because of the 

 decreasing effect of electron scattering and 

 diffusion. 



It was noted that even at the high energy 

 limit there was some emission in the forward 



direction but this was attributed to the ef- 

 fect of (electron diffusion, even in the thin- 

 nest targets used. The theory of Scherzer 

 (1932) predicted that there would be no for- 

 ward emission, l)ut this involved the assump- 

 tion of small atomic numbers and high 

 electron energies, and the more general cal- 

 culations of Scheer and Zeitler (1955) show 

 that there is expected to be some forward 

 emission. 



The finite intensity observed by Honer- 

 jager in the forward direction using his thin- 

 nest (100 A) target affords evidence of this, 

 because this thickness is only one half of the 

 "Aufhellungsdicke" at this energy. Electron 

 scattering would not be expected to affect 

 the original distribution to any observable 

 extent. 



Kerscher and Kuhlenkampff (1955) have 

 used targets 250 A in thickness, and have 

 established accurately the shape of the angu- 

 lar distributions for several different photon 

 energies within the continuous spectrum ex- 

 cited in aluminum at 34 kV (Fig. 4). These 

 workers used a proportional counter and sin- 

 gle channel pulse anal^^zer to select particu- 

 lar bands of photon energies. This method 

 suffers from the disadvantage that the 

 boundaries of the selected channels are ill- 

 defined (because of the finite 'spread' in the 

 heights of pulses produced by photons of a 

 given energy) and Doffin and Kuhlenkampff 

 (1957) have recently avoided this difficulty 

 by using balanced filters, to select a well de- 

 fined energy band. 



Massey and Burhop (1952) have given a 

 comparison between the theoretical angle of 

 maximum emission (at the high energy limit) 

 and the experimental data of Kuhlenkampff 

 and of Bohm. The agreement is very good, 

 and the more recent data also fit well. 



X-ray Production in Electron -opaque 

 Targets 



Spectral Distribution. When the x-ray 

 target is thick enough to bring the electrons 

 to rest, the observed spectral distribution is 



656 



