Focal-plane-array (FPA) hybrids of quantum-well infrared photodetectors (QWIPs) thinned to the membrane level are undergoing development. The developmental QWIPs in question are of the kind that exploit bound-to-quasi-bound absorption, as described, for example, in "Bound-to-Quasi-Bound Quantum-Well Infrared Photodetectors," which immediately follows this article. The present development encompasses both the thinned-array design and the method of fabrication.

The GaAs Substrate supports the QWIP FPA during initial fabrication. The substrate is removed by a sequence of processes, leaving only the QWIP FPA on the etch-stop layer.

Unthinned and partially thinned QWIP FPA hybrids developed previously have been characterized by crosstalk among pixels, thermal mismatches between the FPA hybrids and associated readout multiplexers, and poor light-coupling efficiency. The developmental thinned QWIP FPAs overcome these deficiencies and offer other improvements in performance, as follows:

  • The thermal masses of FPAs are so small that they adapt the thermal expansion and contraction coefficients of Si CMOS (complementary metal oxide/semiconductor) readout multiplexers.
  • The more favorable aspect ratios created by thinning maximize the efficiency of coupling of light from random reflectors.
  • Crosstalk among pixels is suppressed because after thinning, the remaining substrate thickness is too small to support appreciable crosstalk.

A QWIP FPA of the present type is fabricated on top of a 300-Å-thick layer of Al0.3Ga0.7As that, in turn, has been deposited on a relatively thick (several hundred µm) GaAs substrate. Then the thinned QWIP FPA is formed by removing the substrate from under the Al0.3Ga0.7As layer in the following sequence of processes:

  1. Formation of a QWIP FPA hybrid via indium bump-bonding process. This hybrid consists of QWIP FPA and Si CMOS readout multiplexer.
  2. The hybrid is backfilled with low-viscosity epoxy.
  3. By mechanical lapping and polishing with abrasives, most of the thickness is removed from the bottom side, leaving a substrate thickness of 100 µm.
  4. Chemical polishing with a solution of 1 part bromine and 100 parts methanol is used to remove the next 75 µm.
  5. The next 20 µm of thickness is removed in an 8-minute wet-chemical etch with a solution of 5 parts H2SO4, 40 parts H2O2, and 100 parts H2O.
  6. The remaining 5 µm of substrate thickness is removed by dry etching in a plasma formed from CCl2F2 gas. The Al0.3Ga0.7As layer acts as an etch-stop layer during this process, in that the plasma etches the GaAs substrate material much faster than it etches Al0.3Ga0.7As.
  7. An O2 plasma is used to remove a grayish film that remains after the CCl2F2 plasma etch.

This work was done by Sarath Gunapala, John K. Liu, and Mani Sundaram of Caltech for NASA's Jet Propulsion Laboratory. In accordance with Public Law 96-517, the contractor has elected to retain title to this invention. Inquiries concerning rights for its commercial use should be addressed to

Technology Reporting Office
Mail Stop 122-116
4800 Oak Grove Drive
Pasadena, CA 91109
(818) 354-2240

Refer to NPO-19833

This Brief includes a Technical Support Package (TSP).
Fabrications of thinned QWIP arrays for improved performance

(reference NPO19833) is currently available for download from the TSP library.

Don't have an account? Sign up here.

NASA Tech Briefs Magazine

This article first appeared in the September, 1998 issue of NASA Tech Briefs Magazine.

Read more articles from the archives here.