I 



600 



THE BELL SYSTEM TECHNICAL JOURNAL, MAY 1956 



UJ 



D 

 O 



o 

 o 



tr 

 o 



4.5 



4.0 



3.5 



3.0 



2.5 



MM 



2.0 



1.5 



1.0 



1.14 



1.12 



1.10 



1.08 



1.06 



> 



a. 



D 

 O 



o 



LL 



WW 



1.04 



1.02 



1.00 



10 15 20 25 30 35 

 1(/t IN SECONDS 



40 45 



50 



55 



Fig. 23 — Curves illustrating the observed linear dependence of S/2o on the \/~t. 



X 10^^ cm~', shows a large reduction in diffusivity even at temperatures 

 as high as 200°C. 



The difficulties discussed in this section serve to emphasize the im- 

 portance of a direct transport experiment in which lithium atoms nni- 

 jormlij distributed throughout germanium or silicon, uniformly doped 

 with acceptor, are caused to migrate by an electric field, and their 

 mobilities measured. Because of the uniform dispersion of solutes the 

 mobility will be constant everywhere. Furthermore no diffusion poten- 

 tial will be involved, and also the refined formula (7.25) can be applied. 

 There are, however, many difficulties associated with the performance 

 of this type of measurement. 



In closing it may be mentioned that a few much less careful experi- 

 ments of the kind described here have been performed in boron-doped 

 silicon. The results indicate ion pairing in a qualitative way but more 

 definite experiments are needed. 



