APPLICATIONS OF ISOTOPIC INDICATORS 931 



solid lead, using ThB or RaD as indicators. The measurements of the 

 self-diffusion coefficient in liquid lead^^^ gave the result anticipated 

 from the known diffusion rates of lead in mercury and other related 

 elements. The diffusion rate in liquids is primarily determined by the 

 radius of the diffusing particle and the viscosity of the liquid : thus, 

 the replacement of a diffusing metal present in small concentration by 

 another related metal will not appreciably influence the rate of diffusion. 

 A very different behaviour was revealed, however, when the self-diffusion 

 in solid lead'^°^ was measured, using RaD as indicator. In the first 

 experiments carried out in collaboration with Groh, we soldered a 

 piece of radio-lead to the bottom of a rod of ordinary lead, whereafter 

 the system was kept at 280° for 140 days. After the lapse of that time, 

 we cut the system into four equal parts, rolled the four lead pieces 

 into thin plates, and placed them in an electroscope. No diffusion of 

 the radio-lead into the ordinary lead could be ascertained, showing 

 that the self-diffusion rate in lead must be at least several hundred 

 times smaller than that of gold in lead, as determined by Roberts- 

 Austin. 



This result necessitated the introduction of special methods of great 

 sensitivity for measuring diffusion. Since the rate of diffuson is inver- 

 sely proportional to the square of the thickness of the layer, we worked 

 out methods for the measurement of the penetration of radioactive lead 

 into ordinary lead layers to a depth of only a few microns. The a-par- 

 ticles emitted by ThB (more correctly, by its disintegration products, 

 ThC and ThC, which, however, attain radioactive equilibrium with the 

 former within a few hours) produce scintillations on a zinc sulphide 

 screen, the number of which is ascertained. The infinitesimal layer of 

 ThB, which is in intimate contact with a lead foil placed below it, is 

 then heated for a few hours to, say, 200°. If a diffusion of the ThB 

 atoms into the lead foil takes place, the count of the scintillations will 

 give a smaller value after the experiment than before. The range of 

 the a-particles in lead being only about 30 [jl, a shift of a small percentage 

 of the ThB atoms to depths less than 30 [i will suffice to reduce the 

 counts of scintillations to a noticeable extent. This method, worked out 

 in collaboration with Mrs. Obrutsheva^^^^ was later replaced by a 

 more sensitive and exact procedure applied in diffusion measurements 

 in solid metals, and also salts, in collaboration with Seith^^""'. ThB 

 was condensed on a foil or a single crystal of the metal, and the ionisation 

 produced by radiation emitted by the radioactive body was measured. 

 A slight diffusion of the ThB into the lead after heating sufficed to 

 diminish the ionizing effect registered by an electrometer. Instead of 

 the ionization produced by the a-rays, the ionization produced by recoil 

 particles accompanying the emission of a-rays was measured in some 

 cases. The range of the recoil particles in lead being only about 10~^ cm., 



59* 



