USE OF FIELD EMISSION ELECTRON MICROSCOPE 911 



sisted of 8 turns, .3 cm diameter. The center of the coil was 2 cm from the 

 point, and the axis of the coil formed an angle of about 30° with the axis 

 of the tube. Since the composition of the source is essentially that of Batalum 

 getters which are known to evaporate Ba, it is assumed that nearly pure 

 Ba evaporated from it. There is, however, the possibiHty that some BaO 

 evaporated with the Ba. The tube has been in an operable condition for 

 about 12 years. 



The tube was baked at 400°C for one hour. Then all the parts were 

 glowed or heated to outgas them. It was rebaked at 410°C for three hrs. 

 The W loop, Ta filaments, and Ba coil were heated so hot that further 

 heating did not increase the pressure. The tube was sealed off at a pressure 

 of 2 X 10"^ mm with both Ta filaments at a high temperature. Soon after 

 the tube was sealed off the patterns for clean W and Ba on W were quite 

 unsteady: there were rapid variations in intensity of small bright spots or 

 flickering and there were more gradual changes in the pattern over large 

 areas. After glowing the W loop, Ta filaments, and Ba coil many times and 

 at successively higher temperatures, the flickering disappeared completely 

 and the slow large-area changes became less pronounced or required a longer 

 time to appear. Characteristic patterns could be reproduced at will for any 

 particular treatment. In the early stages the clean W pattern changed notice- 

 ably in one minute; later, the time required for a definite change to occur 

 increased to ten minutes, then one hour, then one day, and finally one week 

 or even one month. The effect of the residual gas was to enlarge the 211 

 planes and darken the 111 zone. The effect of this residual gas could be re- 

 moved at r = 8{X)°K in a minute. We suspect that the residual gas is mostly 

 CO. 



During the course of many experiments, the W loop was raised from 2200 

 to 2800°K. Gradually the voltage required to obtain a field emission of 10 

 microamps from clean W increased from 6000 to 9000 volts. In accordance 

 with Mueller's results^ we ascribe this to an increase in the radius of curva- 

 ture of the point from 2 to 3 X 10~^ cm. 



The lower portion of Fig. la shows an enlarged view of the W point and 

 indicates that the tip of the point consists of a single crystal. Hence all 

 possible crystal orientations should be represented on the surface. Figure 

 lb shows a still further enlargement of the tip of the point. We assume that 

 near the tip of the point the surface is a paraboloid. If the origin is taken at 

 the vertex, and y is measured along the axis and x perpendicular thereto, 

 the equation for the paraboloid is 



y = ^/2r (1) 



where r is the radius of curvature of the "point." The field near such a sur- 



