Previous Up Next

Doping

The concept of substitutional doping sounds bizzare for an amorphous compound. We would expect that the disorder of the material could easily incorporate atoms with higher or lower coordination. Nevertheless, the preference for tetrahedral bonding in amorphous silicon seems strong enough to allow substitutional doping, at least to some extent. Chittick's paper already includes some preliminary results on doping by using mixtures of silane and phosphine (PH₃). They found the typical signatures of doped semiconductors, which are reduced activation energies and increased conductivities, but they note that the effect is quite weak compared to crystalline material. A substantial progress was later reported by Spear et.al. who also succeeded in p-type doping by adding diborane (B₂H₄) to the plasma [Spear-1975ssc, Spear-1976pm].

Spear's data on room temperature conductivty (upper panel) and activation energy (lower panel) of amorphous silicon films prepared from mixtures of silane with diborane (left) or phosphine (right). Chittick's earlier data is shown by the open symbols.

The dashed line in the center of the graphs represents undoped material, but the measured data suggests that there is still slight n-type conductivity in these samples. Intrinsic material with low conductivity and activation energies close to mid-gap could be obtained only by compensation of the n-type behaviour using p-type doping. Today, a higher purity of the source gases and better background pressures in the vacuum systems generally yield quite intrinsic material.

Previous Up Next