More detailed calculations of the photocurrent

The treatments of the surface recombination above assume that there is only one trap site. In practice, this will rarely be the case indeed, as the analysis of the a.c. data for p-GaP above illustrated, there may be a near constant distribution of surface states throughout a substantial fraction of the bandgap. For an n-type semiconductor, the total recombination current may then be written, for Gerischer s case (a) [136] above 

In his treatment of the photoresponse in the semiconductor, Wilson assumed, as had Memming, that quasi-equilibrium conditions obtained across the depletion layer, i.e. the product np is a constant. 

At the depletion layer boundary, np c WDpw and at the surface, nP — psAD exp (vs), where we assume, even in the depletion layer, n An, Ap p, and pw and ps are the hole concentrations at x = W and 0, respectively. Hence 

Inserting this into eqn. (363) gives an explicit expression for the total flux and hence, from eqn. (392) the observed photocurrent 

It is as well, at this point, to reiterate the approximations underlying Wilson s approach. They are 

---