Methods of dislocation distribution analysis and inclusion identification with application to CdTe and (Cd,Zn)Te

Although the density of dislocations is used as an indicator of the quality of semiconductor wafers, it is also necessary to provide a description of the dislocation distribution for those materials prone to polygonization (namely the formation of sub-grain boundaries). In this work it is shown that visual inspection of etch pit distributions is insufficient and that quantitative comparison with random distributions is more informative. This was achieved in two ways: (i) by comparison with a Poisson distribution and (ii) by using normalized radial dislocation density plots, the former yielding the type of distribution and the latter the characteristic sub-grain tilting angle for the sample. Although these methods are general, examples for melt-and vapour-grown CdTe and (Cd, Zn)Te are given. Also reported upon in this work is the use of etching solutions to reveal inclusions in CdTe and (Cd, Zn)Te and hence permit their chemical identification by SEM/EDAX.