110 The Electron Microscope 



sary even for a moderate reduction of the focal length. The 

 asymptomatic values of the potentials, measured in units of the 

 object potential, are marked on the trajectories. Only very weak 

 diverging lenses can be therefore realized, at a prohibitive cost 

 in voltage. 



We must now remember the clause which accompanied the 

 theorems on the impossibility of diverging lenses and corrected 

 lenses : without space charge. With negative space charges it 

 would be obviously possible to produce diverging lenses, as the 

 electrons nearer the axis would repel those outside. The prob- 

 lem is only how to realize negative space charges of considerable 

 magnitude. An elementary estimate shows that we require 

 charge densities of the order of a hundred electrostatic units per 

 cm^, which could be realized with electron concentrations of the 

 order lO^^-lO^^/cm^. Concentrations of this order produce by 

 their space charge voltages of the order of 100-1,000 volts in 

 devices of a few cm, and if they are in motion with velocities of 

 this order, they produce currents of the order of hundreds of 

 amp, which, if they were allowed to flow to the electrodes would 

 soon convert the device into a vacuum furnace. Therefore, it is 

 imperative to keep these huge space currents away from the 

 electrodes, and to make them circulate inside the vacuum space. 

 A device which realizes this to some extent is known for a long 

 time. It is the static magnetron of A. W. Hull.^^ Figures 41-44 

 illustrate a suggestion by which the principle of the static mag- 

 netron could be adapted for the solution of the problem of 

 corrected objectives with very short focus. 



Figure 41 is a longitudinal section of the projected objective, 

 with no claim as regards correct dimensions. The movement 

 of the electrons which image the object, and which will be 

 called for short beam electrons, is from left to right. The objec- 

 tive consists of two lenses, a condensing lens at the left, and a 

 diverging, space charge lens at the right. The condensing lens 

 is a combination of a magnetic lens, and of a unipotential electro- 

 static lens. The reason why such a combination has to be used 

 will be explained later. The magnetic and the electrostatic field 

 are shown separately. The middle electrode ''E" of the uni- 



