Single heterojunction photodiodes suitable for the 1.0 to 1.3 µm spectral region of interest for fiber optics communication have been prepared from lattice-matched p-type epilayers of the quaternary III--V alloy semiconductor InGaAsP on n-type InP substrates. The characteristics of these heterojunctions are demonstrated by mesa photodiodes made from Zn-doped epilayers of composition In0.84Ga0.16As0.34P0.66 on Sn-doped InP. Their room temperature spectral response extends from the q, 0.96 µm self-filtering cutoff of the InP substrate to Ëœ 1.13 µm determined by the bandgap of this particular quaternary composition. Responsivities of 0.46 A/W and external quantum efficiencies of 0.54 are measured at 1.05 µm.

The use of charge transfer devices on the focal plane allows signals from a large number of detectors to be multiplexed onto a few video lines, thereby reducing the leadout complexity and making possible focal planes with many thousands of individual detectors, with a resultant increase in performance over present systems. Hybrid technology, which interfaces silicon CCDs with optimum infrared detector materials, offers a high performance, near term solution for advanced focal planes. The electrical and mechanical aspects of hybrid technology are addressed, and reasonable solutions are presented for both PC and PV detectors multiplexed by a CCD with minimum loss in S/N. The systems implications of this approach to large focal planes are described. Finally, experimental results obtained at Honeywell on detector and detector/CCD performance are presented. Progress in this area has been rapid, and the large hybrid focal planes will begin appearing in infrared systems in the next few years.

The performance of infrared sensitive photodiodes may be improved at higher operating temperatures by a reduction of excess volume in which minority carriers are thermally gen-erated within a diffusion length of the junction. This technique requires surface recombi-nation velocities typically less than 100 cm/sec. Calculations for HgCdTe indicate a peak D* above 1.5 x 10¹¹ cm Hz½/w may be obtained at 190 K for a 5.0 µm cut-off wavelength.

The figures of merit which are most commonly used to describe the performance characteristics of infrared detectors are briefly summarized. An indication of the applications in which the various figures of merit are most useful is also included, where feasible.

This paper describes work performed for the Ballistic Missile Defense Advanced Technology Center to assess the current state of optical sensor calibration and test facilities and to establish a recommended program for meeting projected BMDATC calibration and testing needs. Included in the paper are a summary of existing capabilities and calibration approaches, typical data produced during the assessment, and the recommended approach to satisfying the expected ground test requirements of astronomical infrared sensors.

A new long-wavelength infrared test facility is being developed at Ames Research Center to conduct proof-of-performance tests of focal plane arrays for the Infrared Astronomical Satellite (IRAS) telescope. This facility is believed to be unique in that it will cali-brate full-sized arrays designed for the 5-120 µm region of the IR spectrum under simulated spaceflight conditions and at background levels as low as 8 x 10⁶ photons/cm² second. Most of the tests will be performed in a main test chamber which contains a liquid helium shroud, blackbody sources, an arrangement to simulate zodiacal background, and spot scanning optics. A noncontaminating high-vacuum pumping system will be used to evacuate the chamber prior to cool-down to prevent molecular deposition. All housekeeping and focal plane signal outputs are multiplexed, digitized, and transmitted on high-speed data lines to a PDP-11/70 computer. The computer provides real-time displays on a color CRT terminal and stores data for subsequent reduction and hard copy output. An overview of the conceptual design and performance specifications of the test facility are given.

A system to evaluate photodetectors sensitive in the 0.5 to 2.0 micrometer spectral band has been designed, fabricated and calibrated. The system consists of a spot scanner which is controlled by a desk top calculator. Responsivity data is sampled and processed by the calculator. The system is used to evaluate photodetectors for uniformity of responsivity, quantum efficiency, optical area and crosstalk.

Infrared observations are germane to nearly every problem of contemporary astrophysics. Within the next decade, NASA expects to carry out at least two major cooled space telescope projects for infrared astronomy. This paper discusses a complementary pair of telescopes, the Infrared Astronomical Satellite (IRAS) and the Shuttle Infrared Telescope Facility (SIRTF), which are being developed to study the infrared sky. Brief descriptions, with emphasis on the focal plane designs, and specific objectives of both telescope systems are given.

The objective of the Far Infrared Sky Survey Experiment (FIRSSE) is to measure the spatial and brightness distribution of the celestial background and astronomical sources in the wavelength region between 8 µm and 120 µm. A superfluid helium (2.2K) cooled sensor with background photon noise limited detector arrays will be flown on an Aries sounding rocket. This paper outlines the current design concept of the sensor and focal plane array layout of the experiment.

The Shuttle Infrared Telescope Facility (SIRTF) will impose conditions on the infrared detectors in its focal plane instruments which are significantly different from the operat-ing conditions characteristic of ground-based systems. The tenuous atmosphere above the 400 km nominal flight altitude has no measurable absorption so that detectors operating at all wavelengths from 5µm to 1000µm can achieve high radiometric sensitivity. Emission of the atmosphere will be significant only in a few strong spectral lines. Emission from interplanetary dust is expected to be the limiting radiometric background in the region between 10µ and 30µ. Shuttle ejections, especially of water vapor and carbon dioxide must be strictly controlled in order to prevent the emission of these materials from limiting the telescope's sensitivity, and ejection of particles must be controlled to prevent their appearance as false targets. The telescope and detectors will be cooled to cryogenic temperatures in order to reduce telescope self-emission and to obtain preamplifier-noise-limited sensitivity. Charged particle radiation will generate false pulses which in the horns of the Van Allen belts and in the South Atlantic Anomaly may interfere with observations. Using an infrared camera as an example instrument, predictions of these detector operating conditions and resulting performance are presented for semiconductors (HgCdTe, Si:As, Ge:Ga) appropriate to the shorter wavelength bands (to 120µm), and for a Ge bolometer for the longer wavelength region.

Problems exist in calibrating a cryogenically cooled infrared sensor system in the laboratory. Although there exists limited laboratory facilities which can simulate the actual operational environment, the sensor-tester interface has, in the past, created difficulties in calibrating LWIR sensors. The procedures used to calibrate the AFGL infrared celestial survey experiments are described in detail. It was found that the stars are reliable calibration sources which offer several advantages over laboratory references. The long term stability of the sensor systems as well as linearity of the system are discussed. Anomalous behavior in the parallel biased detectors was observed and is described. Suggestions are offered to improve calibration of future cryogenically cooled infrared celestial sensors.

This paper presents the major results of a NASA supported development effort on Ge:Ga and Ge:Be photoconductive detectors for use in far infrared astronomical observations from a space platform such as IRAS. Ge:Be is useful in the wavelength range from 30- to 55-µm, and Ge:Ga from 50- to 120-µm. The theory of operation and some typical test data which represent the current state of the art are given. Under a background photon flux density on the order of 1 X 109 ph/sec/cm², an operating temperature of 3K and a measuring frequency of 1 Hz, Ge:Be detectors achieved an NEP (40 µm) of 2 to 3 x 10^-16 watt/Hz½, and Ge:Ga detectors achieved an NEP (100 µm) of about 1 X 10-16 watt/Hz 1/2. At a lower backa,round flux calculated to be 4 X 108ph/sec/cm², one Ge:Ga detector attained an NEP (100 µm) of 5 X 10-17 watt/Hz 1/2. The reproducibility and yield of these detectors are adequate to consider their use in large focal plane arrays for far infrared astronomical observations.

Pyroelectric detectors offer the advantages of room temperature operation and wide spectral response. High performance requires good thermal design of the detector and its mounting structure to maximize low frequency responsivity and to minimize temperature noise. The intrinsic detector noise is caused by dielectric losses. In the crystalline ferroelectrics, such as Sr _0.5 Ba _0.5 Nb₂O₆ (SBN) and LiTaO3, dielectric loss tangents on the order of 0.0003 have been obtained and a intrinsic detector noise may be limited by material impurities. A low frequency D* of 2 x 10⁹ cm Hz /W has been measured for these materials. At these loss tangent values, temperature noise and preamplifier noise for both JFET and CCD amplifiers presently limit performance. Despite its lower pyroelectric coefficient, the polymer PVF₂ is a competitive pyroelectric material for detector arrays using CCD signal processors.

Thermoelectrically cooled detector packages offer rugged, lightweight, relatively inex-pensive devices to the designer and user of infrared systems operating at intermediate (150°K - 250°K) temperatures. This paper explores some of the more important features of thermoelectrically cooled detector packages, their strengths and limitations in comparison to other cooling modes, and their reliability. A brief survey of some of the many applications of these devices is included. Additionally discussed are some of the factors which the designer must consider when choosing a cooling technique for an IR system.

The attractions of using thermoelectric (TE) coolers in IR imagers, as compared to cooling with either a Joule-Thomson cryostat or mechanical tyre refrigerator, are their relative light weight, low life cycle costs, and excellent reliability. The vitality of TE technology derives from a coincident maturation in three "sub-technologies": TF coolers, Intermediate-Temperature-Operation (ITO) detectors, and integrated focal plane electronics. TE coolers are available which provide 3n-50 milliwatts cooling nower at 195 V using less than 3 watts of input power. Small-geometry detectors sensitive to 3-5 micron radiation are thermal noise limited when operated above 170 K but still exhibit respectable detectivities of near 1. 0x1011cm-Hz ^1/2 /W at 193 K. Large detector arrays with integrated focal plane signal processing are now being developed which will improve system sensitivity beyond that of our first-generation devices. This paper briefly reviews the status of TE coolers, ITO detectors, and focal plane electronics, and presents the problems and trends in TE technology.

The optical properties of materials used in LWIR spectral filters are strongly temperature dependent. Spectral transmittance curves are presented for several interference filters and filter substrates at room temperature and at cryogenic temperatures. The dependence of out-of-band rejection upon optical geometry is also discussed.

A high performance 8 to 12 µm 72-element (Hg,Cd)Te detector array has been built for an advanced difffaction-limited serial-scanned FLIR system. The avelzage peak detect it for the array is 1.7 x 1011 cm Hz ^1/2 /W which results in a cumulative detectivity (~ rn D avg.) of 1.4 x 10¹² cm Hz ^1/2 /W . The cumulative detectivity controls system sensitivity. The cumulative D* for this array is over 2.3 times higher than a modular FLIR array with 180 elements and a better than average peak detectivity of 4.5 x 1010 cm Hz ^1/2 /W or a cumulative detectivity of 0.6 x 10¹⁰ cm Hz ½/W.