A proposed focal-plane array (FPA) of quantum-well infrared photodetectors (QWIPs) would sense images in three different infrared wavelength bands simultaneously. These and other QWIP infrared image sensors are undergoing development for use in measuring temperature distributions in images; potential applications include scientific (e.g., remote sensing of the Earth and other planets) and military (e.g., discriminating between targets and other objects in terms of temperature).
A one-color QWIP FPA can produce data for an absolute-temperature map of the scene only if the emissivities of the objects in the scene are known. A two-color QWIP FPA can produce data for an absolute-temperature map of the scene even when the emissivities are unknown, provided that the sensed radiation consists solely of thermal radiation from the objects in the scene. The proposed three-color QWIP FPA could produce data for a temperature map of the scene even when the sensed radiation from the object included an unknown component of infrared light, reflected from objects in the scene, that originated at an external source.
The proposed device would be fabricated from a wafer that would comprise multiple layers of AlxGa1-xAs formed by molecular-beam epitaxy on a semi-insulating GaAs substrate. Each pixel of the device would contain a stack of three multiple-quantum-well (MQW) AlxGa1-xAs structures. Each MQW structure would be designed for peak photosensitivity in one of the three wavelength bands (see figure): The top MQW structure would be designed for bound-to-bound photoexcitation of electrons in the wavelength band of 14 to 15µm; the middle MQW structure would be designed for bound-to-quasi-bound excitation in the wavelength band of 10.5 to 11.5µm; and the bottom MQW structure would be designed for bound-to-continuum excitation in the wavelength band of 7 to 8µm.
Each MQW structure would consist of approximately 30 periods, each period comprising (1) an AlxGa1-xAs barrier layer 500 Å thick and (2) a GaAs well layer. The mole fraction of aluminum (x) in the barrier layers and the geometric depth of the wells would be chosen to obtain the required spectral response; and the foregoing parameters would be chosen, along with the doping densities and the precise number of periods, to optimize the device for a specific application. The top and middle MQW structures would be separated by a 0.5-µm-thick n+-doped GaAs contact layer. The middle and bottom MQW structures would be separated by a 0.5-µm-thick undoped (and therefore highly electrically resistive) GaAs isolation layer. Two n+-doped GaAs contact layers would be grown on both sides of the isolation layer to serve as independent electrical contacts for the middle and bottom MQW stacks.
The three-color QWIP structure as described thus far would be grown on top of a 0.5-µm-thick n+-doped GaAs bottom contact layer on top of an AlxGa1-xAs etch-stop layer on the GaAs substrate. A 300-Å-thick Al0.3Ga0.7As stop-etch layer would be grown on top of the topmost n+-doped GaAs contact layer. On top of the stop-etch layer, a 1.3-µm-thick n+-doped GaAs contact and cap layer would be grown, and a light coupler would be fabricated in this layer as described subsequently. The contact layer between the top and middle MQW structures would be short-circuited to the contact layer between the bottom MQW and the substrate, to establish a common (ground) bus. Only three indium bumps per pixel would be needed for bonding the device to a silicon readout integrated circuit that would provide independent readout for each wavelength in each pixel.
As explained in more detail in several prior NASA Tech Briefs articles, in order to make photoexcitation possible, a light coupler is needed to alter the polarization of normally incident light so that the light propagating within the device is polarized at least partly along the through-the-thickness direction. In the proposed device, the light coupler would be an achromatic array of reflectors comprising a square pattern of cells, each cell comprising a three-by-three array of subcells. Within each cell, the depths of eight subcells would be chosen to obtain destructive interference at the middle wavelengths of all three wavelength bands; the depth of the ninth subcell would be chosen at random. The cell surfaces would be coated with Au/Ge and Au for ohmic contact and reflection.
This work was done by Sumith Bandara, Sarath Gunapala, John K. Liu, Daniel Wilson, and William Parrish 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
Refer to NPO-20683