Zn concentration determination in CdZnTe by NIR spectroscopy

Growth of epitaxial layers of CdxHg1-xTe (CMT) onto Cd1-yZnyTe (CZT) substr ates is widely reported as being performed in order to minimize misfit disl ocations at the growth interface. The lattice parameters of CdZnTe alloys w ith y similar to 0.04 are chosen to match that of x = 0.22 CdHgTe. However, the rapid rate of change of lattice parameter, as a function y, means that the uniformity and definition of the required Zn concentration is potentia lly very critical. Segregation effects in the bulk growth process means tha t commercial CZT suppliers are unwilling to accept tight specifications on 'y' and, in addition, different internal techniques and calibrations are be ing used to determine 'y' Previously published studies of the nondestructiv e, near IR assessment of the CdZnTe band edge, required extrapolation techn iques to define a cut-on point. Empirically determined correction factors w ere then applied to account for the effect of substrate thickness variation s but no allowance was made for potential variations in the transmission va lue of the substrate. We report a study of the NIR band edge cut-on, define d by the wavelength corresponding to an absorption coefficient alpha = 10 c m(-1), which automatically corrects for thickness or transmission variation s. Data have been correlated against XRD lattice parameter data to determin e the required Zn concentration. (C) 1999 Elsevier Science B.V. All rights reserved.