COMBINED MEASUREMENTS ON ETCHED GERMANIUM SURFACES 1021 



change in surface potential produced by illumination of the surface to 

 the change in the quasi-Fermi level for minority carriers. By measuring 

 dY/db rather than dY/dL, discussed in Reference 2, the surface re- 

 combination velocity is eliminated from the surface photo-voltage data: 

 the limiting values of dY/dd, after correction for the Dember effect, 

 ought to be (po/ni) and — (ni/po) , no matter what the surface recombina- 

 tion velocity may be. 



By combining this information with the field-effect data, one can de- 

 duce the quantity (5Ss/55)r • This and the previous differential, deduced 

 directly from the field-effect data, completely define the dependence of 

 charge in surface traps on the two independent parameters Y and 

 8 — that is, the dependence on chemical environment and on the bulk 

 non-equilibrium carrier level. 



The further interpretation of the cjuantities (dI,s/dY)s=o , (52s/55)r 

 and s in terms of the distribution of surface traps is postponed to the 

 succeeding paper. Here it is sufficient to say that the results are con- 

 sistent with the assumption that the traps responsible for surface re- 

 combination are also those pertinent to the field effect and surface 

 photovoltage experiments. Then the ciuantity (d2s/^F)a=o depends only 

 on an integral over the distribution in energy of traps; (31,^/88) y depends 

 also on the ratios of cross-sections for transitions to the valence and 

 conduction bands; and s depends in addition on the geometric mean 

 cross-sections. 



II. OUTLINE OF THE EXPERIMENT 



The experiment is carried out with a slice of germanium, 0.025 cm 

 thick, which is supported in such a way that there is a gap 0.025 cm wide 

 between the slice and a metal plate. Substantially ohmic contacts are 

 attached to the ends of the slice. Three kinds of experiment are now 

 carried out: 



(i) The conductance of the slice is modulated by illuminating it 

 with a short flash of light; the subsequent decaj^ of photoconductivity 

 with time is studied, and the time-constant of the exponential tail 

 measured. 



(ii) A sinusoidally varying potential difference of about 500 volts 

 peak-to-peak is applied between the metal plate and the germanium. 

 Facilities are available for measuring the changes in conductance pro- 

 duced by the field. The sample is also illuminated with light chopped 

 at a frequency different from that of the applied field. One measures: 



(a) the magnitude of the peak-to-peak conductance change in the dark; 



(b) the same in the presence of the light; and (c) the change in con- 



