DISTRIBUTION AND CROSS-SECTIONS OF GERMANIUM SURFACES 1051 



10 

 W 

 to 



10 



to 



1.0 



0.8 



0.6 



0.4 



0.2 



-0.2 



-0.4 



-0.6 



-0.8 



1.0 



-6 



-2 2 



Y-Ln\ 



Fig. 2 



Experiment and theory for 



95 



)K^- 



X + X- 



Solid lines theory; circles and dots, with smooth curves through the points, repre- 

 sent experimental results for n and p-type samples, respectively. 



we see maxima at F — ^w X = 2.0 for the p-type sample, and 3.5 for the 

 n-type sample. Both these values are within the limits to (n x given in 

 the previous paragraph, thus confirming the estimate made there. Fig. 

 3 shows a comparison between the experimental results and (24). The 

 graph has been fitted horizontally, by setting (n x = 2.5, as found above ; 

 vertically, to agree with the mean value at that point. The agreement 

 with experiment is reasonable, although again, just as in Fig. 2, the ex- 

 perimental variation of s with ( Y — in X) is rather slower than one would 

 expect. 



The fact that the experimental values, both of surface photo-voltage 

 and of surface recombination velocity, vary more slowly than expected, 

 is susceptible of a number of interpretations: (i) The deduced distribu- 

 tion of fast states might be wrong. However, the most likely alternative 

 distributions — isolated levels, or a completely uniform distribution — 



