210 ANNUAL, REPORT SMITHSONIAN INSTITUTION, 1939 



lines. This indicates that no more than 10-" gm. of hafnium were 

 required for spectrographic detection, but actually the efficiency of 

 excitation of the hafnium atoms is much greater under these condi- 

 tions than under conditions in which only the minimum detectable 

 amount of material is introduced into the arc. 10"^ gm. can usually 

 be detected without difficulty, however. 



Emission methods were thus shown to be insufficiently sensitive for 

 the problem at hand. Since the crystals were transparent, the absorp- 

 tion method could be tried, and this method was found satisfactory. 

 Ultimately, amounts down to 10"^^ gm. of hafnium were successfully 

 located in the crystals by absorption methods. 



The first step in this new attack was to determine the absorption 

 spectrum of hafnium salts of the type present in the crystal. The 

 hafnium-containing crystalline material was dissolved in water, and 

 a quartz vessel filled with the resulting solution was placed between 

 the light source (a hydrogen discharge tube, which gives a good con- 

 tinuous spectrum in the ultraviolet) and the slit of the spectrograph. 

 On the same plate an absorption spectrogram was taken of similar 

 crystal material which contained no hafnium, as shown in plate 2, 

 figure 2. The differences in absorption were obviously due to the 

 hafnium salt, which was found to absorb strongly in the ultraviolet, 

 a little in the violet, and practically not at all in the visible region. 



The final step was to locate spots in the crystals which showed this 

 characteristic absorption. For this purpose a spectrograph and an 

 ultraviolet-transmitting microscope were combined. Sections of the 

 crystal were thus photographed in each individual line of the mercury 

 spectrum from a quartz mercury arc. Thus the crystal was photo- 

 graphed under high magnification in transmitted monochromatic 

 green light, in blue light, in violet light, and in the light of several 

 ultraviolet lines. 



Opaque materials in the crystal showed dark in all these photo- 

 graphs. Transparent materials which were not hafnium showed 

 light in all, as in plate 3, figure 1, where each bow contains a separate 

 photogi-aph. The hafnium salts showed transparent in green and 

 blue light, darker in the violet light, and opaque in the ultraviolet, 

 as in plate 3, figure 2- Thus any particle which showed this pro- 

 gressive absorption could definitely be called a hafnium carrier, and 

 from its opacity its hafnium concentration could be determined. 



This method was then further simplified by using a microscope 

 adapted to either visible or ultraviolet light, and taking duplicate 

 photographs of each crystalline particle, immersed in an oil of the 

 proper index of refraction, first with green light as in plate 4, figure 

 1 (left), and then with ultraviolet light of the desired wave length, as 



