Srinivas,

The FLP at the surface of semiconductors due to the presence of high density surface states is an observed phenomenon. The complexity of this effect is that it depends on the surface properties of the material both chemically and geometrically. So, the findings may change from a laboratory to the other according to the surface treatment and the contacted environment. However there is studies of the germanium surface under well controlled conditions. Paul Handler and William M. Portnoy found that the surface of germanium has a strong p-type conductivity which is consistent with your observation of fermi level pinning at the surface near the valence band edge. They attributed this to : there is a two-dimensional surface state band at the surface which overlaps in energy a two-dimensional valence band just beneath the surface according to their expression in their paper at the link: http://journals.aps.org/pr/abstract/10.1103/PhysRev.116.516

I think this model can account for your experimental findings. An old paper but very useful.

Best wishes

In a recent, very nice PES study Kuzmin and coworkers show data that they interpret as such that defects are not responsible for the Fermi level pinning in n-Ge.

http://dx.doi.org/10.1103/PhysRevB.94.035421

They further argue that epitaxial BaO layer can alleviate FLP; this is in agreement with our own data, where we were able to demonstrate well-behaved C-V behavior (Fig. 3d, attached) with epi-BTO, albeit on p-type Ge.

http://onlinelibrary.wiley.com/doi/10.1002/admi.201500193/full

In addition to Kuzmin's work referenced above, there are quite a few works on this topic, with various competing conclusions. A couple of weeks ago, for example, Li and Robertson used DFT to argue that Y passivates an oxygen-related point defect in GeO2 which they suggest causes FLP. 

http://dx.doi.org/10.1063/1.4974291

There are quite a few interesting works from the last decade, I'd be happy to refer you to others as well, but my conclusion is that the manner doesn't seem to be fully settled just yet. Some of the best (non-epi) FLP alleviation/'unpinning' works I encountered utilized ultrathin oxides, e.g. http://iopscience.iop.org/article/10.1143/APEX.1.051406 

and additional important contributions by Nishimuri, Toriumi and coworkers.

Thanks for your reply Lior, I have seen these works in the past. It doesn't answer my question unfortunately. As you say, it is still an open discussion. I was hoping for an explanation for the origin of donor and acceptor type surface states, as they are frequently invoked in literature to explain FLP in Ge.

Srinivas,

The FLP at the surface of semiconductors due to the presence of high density surface states is an observed phenomenon. The complexity of this effect is that it depends on the surface properties of the material both chemically and geometrically. So, the findings may change from a laboratory to the other according to the surface treatment and the contacted environment. However there is studies of the germanium surface under well controlled conditions. Paul Handler and William M. Portnoy found that the surface of germanium has a strong p-type conductivity which is consistent with your observation of fermi level pinning at the surface near the valence band edge. They attributed this to : there is a two-dimensional surface state band at the surface which overlaps in energy a two-dimensional valence band just beneath the surface according to their expression in their paper at the link: http://journals.aps.org/pr/abstract/10.1103/PhysRev.116.516

I think this model can account for your experimental findings. An old paper but very useful.

Best wishes