CHEMICAL INTERACTIONS AMONG DEFECTS IN Ge AND Si 597 



1.7 



1.6 



.5 - 



1.4 



1.3 



1.2 



1. 1 



1.0 



0.9 



0.8 



0.7 



10 



16 



10 



17 



5 6 



10 



18 



Nd in cm' 



Fig. 22 — Plots of correction factor ??, required to compensate for the depend- 

 ence of hole mobility on the density of scattering centers along a diffusion curve. 

 I? is plotted against the initial density of donor and is shown for the two extreme 

 cases of pairing and no pairing. 



either extreme, and therefore closer to unity in some intermediate situ- 

 ation. In any event the correct value of ?? can be read from Fig. 22 if the 

 experiments involve either extreme at the measurement temperature. 

 This has, in fact, been approximately the case in our experiments, in 

 which pairing is almost complete at the temperature where conduc- 

 tances have been measured. 



According to (11.11) a plot of S/2o against 's/l should be a straight 

 line of slope 



2.256^V^/2«>A^D°\ 



S = 



d 



\:eoNa 



(11.12) 



f Measurement of S therefore affords a measure oi D. Of course the ap- 



! parent D obtained in this manner can never represent anything beyond 



' some average quantity having the general significance of a diffusi\-ity. 



This follows from the previous discussion concerning the non-constancy 



