An experimental study has been carried out on the performance of n-type x =
0.31 HgCdTe photoconductive detectors in order to evaluate two different e
tching techniques; dry plasma etching, in the form of H-2/CH4 reactive ion
etching (RIE), and wet chemical etching using bromine in hydrobromic acid.
Two-dimensional laser beam-induced current (LBIC) imaging was employed as a
n in-line process monitoring tool to evaluate the lateral extent of reactiv
e ion etching (RIE) induced doping changes in the HgCdTe epilayer following
mesa delineation. Responsivity and noise measurements were performed on fa
bricated mid-wavelength infrared (MWIR) photoconductive devices to evaluate
the influence dry plasma etching has on material properties. For a signal
wavelength of 3 mu m, 60 degrees field of view, and a temperature of 80 K,
background limited D-lambda* performance was recorded for wet chemical proc
essed devices but not for the dry plasma processed devices. The D-lambda* v
alues obtained for wet chemical and dry plasma etched photoconductive detec
tors were 2.5 x 10(11) cmHz(1/2)W(-1) and 1.0 x 10(10) cmHz(1/2)W(-1), resp
ectively, Mercury annealing, which has been shown to restore the electrical
properties of dry plasma processed HgCdTe, could be used to lessen the inf
luence that RIE dry plasma etching has on photoconductor detector performan
ce.