Infrared arrays with HgCdTe as the light-sensitive layer, such as have been developed up to sizes 2048x2048 pixels for the James Webb Space Telescope, are near-ideal detectors for imaging and spectroscopy in the region ~1-5 microns. However current construction requires fabrication on CdZnTe substrates, which are expensive and limited in availability. The key to making larger (up to 14,000x14,000 pixels) and less expensive infrared detectors lies in using silicon wafer substrates, since large silicon wafers are common in the high volume semiconductor industry and their coefficient of thermal expansion is well-matched to that of the silicon readout circuits.
While the use of silicon substrates has been a major goal in the field of developing infrared detectors, the main limitation over the past 15 years has been the large lattice spacing mismatch between silicon and commonly-used infrared light-sensitive materials. The mismatch causes defects that can result in higher dark current, or valence holes that lead to reduced quantum efficiency and image persistence.
Enlisting the expertise and fabrication capabilities of Raytheon Vision Systems, detector expert Dr. D. Figer of the Rochester Institute of Technology plans to deposit the HgCdTe light-sensitive layer on silicon using the very promising technique of Molecular Beam Epitaxy (MBE). By maintaining vacuum during MBE processing, defect density has been shown to be reduced and the resulting prototype devices have achieved the anticipated performance. Very large, affordable infrared arrays will be essential for making optimum use of the proposed ~30m class ground-based telescopes and their availability has clear implications for fields beyond astronomy, including medical imaging and remote sensing.
Funding for the development of large infrared arrays on silicon is being provided by NSF's Division of Astronomical Sciences through its Advanced Technologies and Instrumentation program.
