This conference covers a very broad range of topics; from IR array developments, through instrumentation for use on ground-based telescopes, to instrumentation for spacecraft, ranging from space projects about ready to be launched to some not yet in development. The past decade has been a very exciting time, driven largely by enormous advances in IR array technology.

The first prototype of a HgCdTe infrared detector array with 1024 X 1024 pixels developed by the Rockwell International Science Center has been tested in a new infrared camera at the UH 2.2 m telescope, the 0.6 m telescope, and the CFHT. At the 2.2 m tests were conducted both at f/31, where images of very high resolution were obtained using tip-tilt correction, and at f/10 for a wide field of view. At the CFHT both wide field imaging (f/8) and adaptive optics work was done. The HAWAII (HgCdTe astronomical wide area infrared imager) prototype device achieved very good performance. In the camera system, a double correlated readnoise of 15 e^- rms was achieved. The dark current at 1 V bias could be confirmed to be below 1 e^-, even though the device was operated above 77 K. The quantum efficiency is slightly below 50% and shows the wavy pattern characteristic of LPE-grown HgCdTe. The full well capacity is above 10⁵ e^- at 1 V bias, limited in our system by the dynamic range of the A/D converter. Data reduction is practically identical to what is used for NICMOS3 256 X 256 devices. Combined integration times of more than 1 hour have been used and demonstrate that the HAWAII devices are suitable for very deep imaging. The residual excess dark current problem known from NICMOS3 devices is not fully resolved. However, it appears less serious in our first HAWAII prototype device.

Photovoltaic detectors for ground based astronomical applications have experienced dramatic improvements during the last decade. Both the array format has been increased and the pixel performance has improved and is approaching fundamental limits. In view of this development a detection limit for the photon flux of the ideal detector will be derived, depending only on the temperature and the impedance of the detector. It is shown, that this limit is approximated by state of the art infrared arrays for long on chip integrations. In a multimode instrument covering the 1 to 5 micrometers spectral range a detector has to fulfill very different requirements. For high resolution spectroscopy low darkcurrent and read noise are required. For broad band thermal imaging a high well capacity is needed to reduce the speed required to read out the array before it saturates. Different possibilities to increase the well depth of infrared arrays have been investigated. First, an extra capacity can be added to the gate of the source follower in the unit cell of the multiplexer. Alternatively, the pixel capacity can be increased by increasing the doping concentration of the detector diode. The third possibility is to apply a large reverse bias voltage. This requires exceptionally good low doped InSb junctions which can be operated at a reverse bias voltage of 1 volt.

The ALADDIN 1024 X 1024 InSb array is now a fact rather than a concept, and the time has come to show test results. In this paper we present lab test data but we have also taken it to the telescope. The development program was a success and the array has met the design goals of the program. The few remaining problems involve hybridization and are expected to be solved soon. The ALADDIN program is a joint collaboration between the National Optical Astronomy Observatories (NOAO) and the U. S. Naval Observatory (USNO) with Santa Barbara Research Center (SBRC).

Hughes has designed a large-area staring Si:As impurity band conduction (IBC) focal plane array specifically for high-background longwave infrared (LWIR) astronomy applications. We derived the design parameters by surveying leading astronomers for their requirements. This paper describes summary results of these requirements and how they were implemented in the design. We discuss preliminary detector and readout data that confirm satisfactory operation. We define current status and plans for fabrication and test of detector/readout hybrids.

Large-format, very-long-wavelength infrared (VLWIR) hybrid focal plane arrays (HFPAs) based on doped-silicon blocked-impurity-band (BIB) detectors have been developed and demonstrated for a variety of astronomy applications. An HFPA consists of a BIB detector array interfaced via indium column interconnects to a matching cryogenic signal processor/multiplexer. Arsenic-doped silicon (Si:As) BIB detector arrays with useful photon response out to nearly 30 micrometers are the most fully developed embodiment of this technology. HFPAs with Si:As BIB arrays have been optimized for low, moderate, and high infrared backgrounds in 128 X 128-pixel formats, and a high-flux 256 X 256-pixel version is under development. For high-flux applications, both the detector array and multiplexer are optimized to handle incident flux densities > 10¹⁶ photons cm^-2s^-1, providing high spatial uniformity, high pixel operability, and background-limited performance down to low frequencies (< 10 Hz). Antimony-doped silicon (Si:Sb) arrays and 128 X 128-pixel Si:Sb HFPAs having response to wavelengths > 40 micrometers have also been demonstrated, primarily for use at low and moderate backgrounds. BIB technology offers producible, low-cost, high-performance focal planes for astronomy in the VLWIR.

New gallium-doped silicon 128 X 192 element arrays have been achieved at CEA-LETI- LIR (Infrared Laboratory) for imaging in the 8 - 14 micrometers spectral range. This program is in keeping with the previous detector developments for the ISOCAM camera (32 X 32 element arrays) and for ground-based observations (64 X 64 element arrays). The main features of the new detectors are: a pitch of 75 micrometers which leads to 10 X 15 mm² chip dimensions, two selectable storage capacities (respectively 0.1 and 0.5 pF), a DVR readout circuit achieved in an NMOS silicon line with 1.5 micrometers design rules. The main electro-optical performances are the following: a peak responsivity of 4.0 A/W, a noise of 58 fArms over the 0.1 - 128 Hz spectral band which is very close to the BLIP noise, and a corresponding noise equivalent power of 1.4 10^-14 W.

Single detectors and linear arrays of microbolometers utilizing the superconducting transition edge of YBa₂Cu₃O₇ have been fabricated by micromachining on silicon wafers. A D* of 8 +/- 2 X 10⁹ cm Hz^1/2/watt has been measured on a single detector. This is the highest D* reported on any superconducting microbolometer operating at temperatures higher than about 70 K. The NEP of this device was 1.5 X 10^-12 watts/Hz^HLF at 2 Hz, at a temperature of 80.7 K. The thermal time constant was 105 msec, and the detector area was 140 micrometers X 105 micrometers . The use of batch silicon processing makes fabrication of linear arrays of these detectors relatively straightforward. The measured responsivity of detectors in one such array varied by less than 20% over the 6 mm length of the 64-element linear array. This measurement shows that good uniformity can be achieved at a single operating temperature in a superconductor microbolometer array, even when the superconducting resistive transition is a sharp function of temperature. The thermal detection mechanism of these devices gives them broadband response. This makes them especially useful at long wavelengths (e.g. (lambda) > 20 micrometers ), where they provide very high sensitivity at relatively high operating temperatures.

We consider a 16 X 16 pixel GaAs photoconductive detector array as a new candidate for an ESA-sponsored detector development program. A photoconductor array covering the wavelength range from 150 to 300 microns can add to the capability of the far infrared imaging spectrometer in the model payload of FIRST (far infra-red and submillimeter space telescope). The GaAs detector array is a completely new development. The spectral response of GaAs has been known for years; gallium arsenide, however, was at that time not available in a quality to bring dark current and NEP at operating temperatures around 1 K down to levels of state of the art photoconductors. Recent progress in liquid phase epitaxy led to the production of very high purity GaAs. Contamination from the growth system can now be kept below the required donor concentration leading to n-type GaAs layers. Photoluminescence spectra indicate low compensation, a vital requirement for getting effective photoconductive detectors. Layers with a thickness of a few hundred microns were recently produced. The GaAs detector is likely to become available within the next few years. The paper discusses investigations of the material and first results of sampling detector development.

A novel homojunction interfacial workfunction internal photoemission (HIWIP) far-infrared (FIR) detector based on the interfacial workfunction (IWF) between a heavily doped absorber/emitter layer and a lightly doped (or intrinsic) layer is reported. The detector structures are classified according to their emitter layer doping concentrations (N_d). The threshold wavelength ((lambda) _t) is tunable in the IR wavelength range by changing N_d and bias voltage. This detector concept has been successfully demonstrated using forward biased commercial Si and Ge p-i-n diodes at 4.2 K. Threshold wavelengths ((lambda) _t) from around 40 - 220 micrometers for Si and up to 240 micrometers for Ge were experimentally obtained. A theoretical investigation including an estimation of workfunction dependence on N_d, quantum efficiency calculations, and dark current analysis is reported. Based on these results, the detector noise equivalent power (NEP) limited by thermal noise and background noise is calculated.

Extrinsic photodetectors are very sensitive to impacts by cosmic rays and charged particles from the Earth's trapped radiation belts. The residual effects of these impacts can, therefore, seriously affect the accuracy of astrophysical observations from spaceborne telescopes. Experimental and theoretical studies of the radiation effects in far-infrared detectors in low- temperature, low-background environments have been made. It is shown that these effects are associated with the generation of electron-hole pairs in the bulk of the detectors during the irradiation, and with the capture of the minority carriers by the compensating impurities. The results of the experiments, physical mechanisms of the observed effects and a new method for restoration of the pre-irradiation performance of the stressed Ge:Ga detector are discussed.

Low-background IR detectors have a very long-term transient response which, what is more, is highly dependent on the preceding irradiations. For these reasons to determine with a high precision the radiation flux of a cool space object, it has to be measured for a long time. The theory presented in this paper describes the detector operation under frequent flux changes and allows precise determination of the flux values from fast measurements, in spite of the strong memory effects. The theory is specialized for the ISOPHOT Si:Ga detector arrays and is confirmed by the experiments. Its use has to add to the productivity of the launched apparatus.

The NASA IRTF is building a multiple digital signal processor (DSP) based array electronics control system with the stated goal of having scaleable data processing power, readout, and throughput speed in order to support the latest generation 1024 X 1024 pixel arrays. This system is intended to satisfy the instrumentation requirements of SpeX, an NSF funded 1 to 5 micron medium resolution spectrograph (R approximately 2000 for 0.9 to 2.5 micrometers and separately approximately 3 to 5.5 micrometers , plus long slit single order coverage). Plans for SpeX include the use of a 1024 X 1024 InSb array for spectroscopy, and a portion of 1024 X 1024 HgCdTe or InSb array for slit field viewing and infrared guiding. This multiple array arrangement places unique requirements on the control electronics system. This paper details the design of the electronics system and its closely coupled hardware and software relationship which provides a high degree of flexibility in a simple modular framework. The design is an evolutionary upgrade of the current IRTF array control electronics system used in a 256 X 256 InSb based imager (NSFCAM), a 256 X 256 InSb, and 512 X 512 CCD in an echelle spectrograph (CSHELL) and a non-IRTF 128 X 128 Si:As BIB array based imager.

Wildfire is the array and instrument controller currently used in the infra-red instrumentation at National Optical Astronomy Observatories. Wildfire is a high performance, versatile transputer based controller which handles the clocking and readout of two-dimensional arrays along with all other aspects of instrument control. The system was originally designed to support the present generation of 256 X 256 infra-red arrays. This paper discusses the upgrade plan for Wildfire which is required to read out the newly developed 1 K X 1 K InSb arrays.

Rapid development of large format InSb and HgCdTe FPAs and the recent availability of Si:As impurity band FPAs to the scientific community have driven the need for flexible high performance drive and readout electronics. Simultaneously, increasing economic pressure throughout the global research community has emphasized the need for design efficiency. The development of a new third-generation scaleable architecture utilizing PGAs, fiber optics, on- camera CPU, and distributed frame processing with VME system compatibility is discussed.

The challenges associated with providing a commercial off-the-shelf infrared electronics system for reading out and controlling the next generation of highly parallel output focal plane arrays is examined. PolyCom's FLEXIRE^TM architecture is then shown to have the necessary flexibility and performance to meet the diverse needs of near and mid infrared imager and spectrometer type astronomical instruments. The FLEXIRE^TM architecture provides unique scalability features that allow the instrument designer to determine the compute power needed for their particular application and through software to reconfigure the system to interface to any number of off-the-shelf DSP boards that are based on industry standard bus architectures. The use of standard DSPs provides a well supported software environment for easy development of custom algorithms by the instrument development team. The open architecture and computer platform independence of the FLEXIRE^TM approach furthermore give the instrument designer enormous freedom in the design of his/her instrument.

We describe the shift-and-add performance of ALICE (array limited infrared control environment) with the IRCAM3 near-IR 256 X 256 camera on the UK Infrared Telescope on Mauna Kea, Hawaii. One-hundred-twenty-eight X one-hundred-twenty-eight sub-areas of the array can be read out in shift-and-add mode with exposure times of 20 ms, and observing efficiency of 100%. The system has produced images at K which show several diffraction rings and a central core of FWHM 0.15 arcseconds, and strehl ratios of 0.1 to 0.3 have been observed. At nbL (3.4 microns), strehl ratios of 0.4 have been achieved. Close binary systems (< 0.2 arcseconds) have been well resolved whilst the time-averaged seeing was approximately 0.7 arcseconds. We describe the techniques used to achieve this performance, and present some results of recent observations made using the system as well as giving an overview of current and planned applications of the ALICE system.

This paper presents an overview of the electronic imaging system design and performance for the long wavelength spectrometer that is to be deployed on the Keck Telescope. Because the LWS employs 96 parallel read-out channels and 24 parallel processing channels, the LWS serves as a forerunner of next generation infrared astronomical instruments that are to use highly parallel output focal plane arrays like the Santa Barbara Research Center ALADDIN Near-IR 32 output FPA or any of the Rockwell Mid-IR 16 output arrays. Several principles regarding noise and grounding issues are discussed with the hope that they may help the developers of the next generation of infrared astronomical instruments. It is concluded that low noise performance is not something that can be added after-the-fact in the lab unless the necessary efforts have been made in the early phases of an instrument's conceptual design. Low noise requirements will drive the highest level architectural issues of the electronics, and hence cannot be an afterthought.

3D, the next generation near-IR spectrometer developed at the MPE, offers, in a single integration, the opportunity to image an 8" x 8" field with a pixel scale of 0.5" or 0.3" across the entire K- or H-band simultaneously at a spectral resolution of R equals 1000 or R equals 2000 (K). Combining the advantages of imaging and spectroscopy increases the observing efficiency on small extended objects (e.g., galactic nuclei) by such a large factor over existing grating or Fabry-Perot spectrometers that subarcsecond near-IR spectroscopy on faint Seyferts, starbursts, quasars, or distant galaxies clusters becomes feasible for the first time on 4 m class telescopes. 3D, including a NICMOS III FPA at 25 e^-/single read, has been successfully operated at telescopes such as the 4.2 m WHT, 3.5 m Calar Alto, and 2.2 m La Silla. An additional tip-tilt seeing corrector for 3D called ROGUE correcting on up to 18th mag stars at 4 m-class-telescopes was successfully commissioned in the summer of 1994. The optical and electronic design of 3D as well as recent results are presented.

SIMON (`Spectrometre Infrarouge de Montreal') is a near-infrared (1.0 micrometers to 2.5 micrometers ) camera/spectrometer currently under development at the Universite de Montreal. The instrument will be used on the 3.6 m Canada-France-Hawaii telescope (CFHT) and the 1.6 m telescope of the Observatoire du Monte Megantic (OMM). It will house a 1024 MUL 1024 array with an image scale of 0.15" on the CHFT and 0.34" on the OMM. Two long-slit spectroscopic modes will provide resolving powers of 1300 and 5000 from 1.0 micrometers to 2.5 micrometers . The instrument could be interfaced with the adaptive optics system currently under development at the CFHT, providing diffraction-limited images at J, H, and K. This paper describes the general characteristics and the optical design of the instrument.

SHARP I and SHARP II are near infrared cameras for high-angular-resolution imaging. Both cameras are built around a 256 X 256 pixel NICMOS 3 HgCdTe array from Rockwell which is sensitive in the 1 - 2.5 micrometers range. With a 0.05"/pixel scale, they can produce diffraction limited K-band images at 4-m-class telescopes. For a 256 X 256 array, this pixel scale results in a field of view of 12.8" X 12.8" which is well suited for the observation of galactic and extragalactic near-infrared sources. Photometric and low resolution spectroscopic capabilities are added by photometric band filters (J, H, K), narrow band filters ((lambda) /(Delta) (lambda) approximately equals 100) for selected spectral lines, and a CVF ((lambda) /(Delta) (lambda) approximately equals 70). A cold shutter permits short exposure times down to about 10 ms. The data acquisition electronics permanently accepts the maximum frame rate of 8 Hz which is defined by the detector time constants (data rate 1 Mbyte/s). SHARP I has been especially designed for speckle observations at ESO's 3.5 m New Technology Telescope and is in operation since 1991. SHARP II is used at ESO's 3.6 m telescope together with the adaptive optics system COME-ON + since 1993. A new version of SHARP II is presently under test, which incorporates exchangeable camera optics for observations with scales of 0.035, 0.05, and 0.1"/pixel. The first scale extends diffraction limited observations down to the J-band, while the last one provides a larger field of view. To demonstrate the power of the cameras, images of the galactic center obtained with SHARP I, and images of the R136 region in 30 Doradus observed with SHARP II are presented.

We have just finished the first tests at the telescope of an infrared camera designed and developed at the Instituto de Astrofisica de Canarias (IAC). This camera, based on a 256 X 256 focal plane array, has been built to operate at the 1.5 m Carlos Sanchez IR telescope (CST) in the Observatorio del Teide (Canary Islands, Spain). In this paper we describe the final configuration and performance of the camera. Some images taken during two telescope commissioning periods are shown.

We describe the design, operation, and performance of a new high-speed infrared photometer using the solid-state photomultiplier (SSPM) detector. The SSPM was developed by Rockwell International Science Center and has single-photon counting capability over the 0.4 - 28 micron wavelength range, intrinsic time response of order 1 ns, and low detector noise (Petroff, et al., 1987). We have operated a 200 X 200-micron back-illuminated SSPM in a liquid-helium cooled dewar with a room-temperature transimpedance amplifier output. Single photon pulses can be easily distinguished above the amplifier noise. The individual photon pulses are binned at a selectable time resolution ranging from 5 microsecond(s) to 64 ms, and then written to Exabyte tape. In the first astronomical application of such a device, we have made observations of the Crab Nebula pulsar and Her X-1 at near-infrared wavelengths (J-, H- , and K-bands), and we present the instrument sensitivities established by these observations. We discuss other astronomical observations which are either planned or currently underway. Finally, we present design specifications and predicted performances for a second-generation SSPM high-speed infrared photometer.

The near infrared camera/spectrometer (or NICS, for brevity), that we are currently designing for first-light operations of the Italian Telescope Galileo (TNG), makes use of a set of optics for achieving both good quality imaging on equivalent fields of view on the sky as large as 4' X 4', and moderate resolution spectroscopy at a resolving power of 400 - 2200, with the possibility to reach 7500 in the future, over the near infrared bands from 0.95 micrometers up to 2.50 micrometers . We describe the details of the optical design, with the overall requests we imposed on the project in terms of performance and total size.

Near infrared imaging spectroscopy at spatial resolutions of 0.5 arc seconds will fundamentally change our understanding of active galactic nuclei. This long desired capability has been achieved for the first time by the latest generation of MPE instruments, ROGUE and 3D. ROGUE, the rapid off-axis guider experiment, is a low order adaptive optics system performing tip-tilt correction in the near infrared using natural guide stars. Three-dimensional is the MPE near infrared imaging spectrometer capable of simultaneous imaging and spectroscopy of the entire H and K atmospheric windows. ROGUE is capable of tip-tilt correction at 40 Hz in a 4 arc-minute diameter isokinetic patch using natural guide stars as faint as 18th magnitude. We discuss the design of the instrument, present the first astronomical results, and outline future efforts to incorporate variable image scales.

This paper describes installation of a long wavelength infrared (LWIR) acquisition camera on the Starfire Optical Range (SOR) 1.5 meter telescope, and reports initial performance results. This camera was designed for acquisition of satellites night or day, irrespective of target illumination. The camera may also have wide field-of-view astronomy applications. The optical design for this LWIR acquisition camera maps a 128 X 128 pixel Si:As impurity band conduction detector array onto the two milliradian telescope scene. A warm, aspheric germanium lens images the f/217 telescope beam onto a cold field stop, and the telescope pupil onto a cryogenic chopping mirror. The cryogenic chopping mirror has an opaque mask which serves as a radiation stop. A second, cooled, aspheric germanium lens reimages the field stop onto the detector array. Operation of the camera over the 7 - 24 micrometers detection band of the array is possible by replacing the interference filter and zinc selenide vacuum window. Problems and solutions pertaining to integration of the camera and cryocooler system on the telescope are described. Initial performance data reported include: optics/array radiometry, telescope background measurements, cryocooler induced telescope jitter measurements, and cryogenic chopping mirror characteristics.

A ground-based three-mirror zero-obscuration astronomical telescope is conceived, designed, and analyzed initially for use as a spectrographic camera covering near UV to near IR (320 - 900 nm) wavelength region in a fiber-fed instrument operating at McDonald Observatory of the University of Texas at Austin. The design is a modified configuration of UV-imager developed for auroral imaging by the POLAR spacecraft of the International Solar Terrestrial Physics (ISTP) Program. This paper presents the design and performance issues in the context of the additional system constraints and mechanical limitations imposed. Tolerance analysis in light of cost-effective fabrication and assembly is also presented.

Five infrared instruments are now under development or study for ESO's Very Large Telescope (VLT). ISAAC (infrared spectrometer and array camera) will be the first to be installed on the first of the 8 m unit telescopes where it will provide for imaging and long slit low and medium resolution spectroscopy in the 1 - 5 micrometers range. This instrument is being developed in-house at ESO and is now in the manufacturing and integration phase with installation scheduled for 1998. CONICA (high resolution near infrared camera) is intended primarily for 1 - 5 micrometers diffraction limited imaging and has been contracted to a consortium of institutes led by the Max Planck Institut fur Astronomie, Heidelberg, Germany with the Max Planck Institut fur Extraterrestrisches Physik, Garching, Germany as partner. Originally planed for coude, it has recently been redesigned for one of the Nasmyth foci of UT1 which will now be equipped with an adaptive optics system following a decision to delay installation of the coude foci. Following completion of a Phase A study, VISIR (mid-IR imager/spectrometer) has been selected as the next IR instrument and negotiation of a development contract is now in progress with the Service d'Astrophysique, Saclay, France who led the study consortium. This instrument is destined for the Cassegrain focus of UT2 and will provide both imaging and spectroscopic capabilities in the 10 micrometers and 20 micrometers windows. CRIRES (cryogenic IR echelle spectrometer), which aims for R approximately 100.000 in the 1 - 5 micrometers range, has been the subject of a concept definition study within ESO and has been rated highly enough scientifically to justify continuation of its associated immersion grating development program. NIRMOS (near IR multi-object spectrometer) has also been the subject of a concept definition study, led by the Observatoire de Meudon, and will soon enter a more detailed study phase of alternative concepts including the possibility of combining visible and near IR multi-object spectroscopy in a single instrument primarily for high redshift galaxy surveys.

An infrared camera called CONICA, for use at the first unit of the VLT (ESO), for diffraction limited observations in the spectral range from 1 - 5 micrometers is being built up by a consortium of the MPI fur Astronomie (Heidelberg, Germany) and the MPI fur Extraterrestrische Physik (Garching, Germany). The camera, which originally was planned for the Coude focus of the single telescope, now -- after postponement of the Coude trail -- is designed for application at the Nasmyth focus. Combined with an adaptive optics system this camera offers diffraction limited direct imaging capabilities. Imaging spectroscopy using a cold Fabry Perot etalon is offered as well as spectroscopy by means of grisms and imaging polarimetry. Optics and cryo- mechanics are designed for implementation of a 1024 X 1024 InSb IR-array. The imaging scale can be adapted for correct sampling of the wavelength dependent diffraction limited beam between 13.6 and 109 mas/pixel corresponding to a field of views between 13.9 and (phi) 73 arcsec. The spatial resolution is between 32 mas in J band (centered at 1.25 micrometers ) and 116 mas in M band (centered at 5 micrometers ). The instrument will be completely remote controlled.

A cryogenic echelle spectrometer is planned for the ESO Very Large Telescope to perform high-resolution (R >= 100,000) observations in the 1 - 5 micrometers range. Primary design goals are optical quality to exploit the 8 m telescope advantage and large spectral field to exploit the sensitivity advantage of a dispersive instrument with a large detector array. The basic instrument uses established technology; some components potentially leading to enhanced performance or additional capabilities are under development.

In 1992, the European Southern Observatory (ESO) committed a phase A study of a mid- infrared instrument for the 2nd unit VLT telescope, to a consortium of laboratories (SAp at Saclay, France; SRON at Groningen, Germany; and the Kapteyn Observatory at Roden, Netherlands). The results of the study are presented. One key scientific objective for this instrument is foreseen to be the study of dust. The required observing modes are (1) diffraction limited imaging both at 10 and 20 microns, and (2) spectroscopy at low resolution (R approximately equals 500) both at 10 and 20 microns. Another key domain is the study of atomic, molecular, and ionic lines observable in the atmospheric window at 10 and 20 microns. Given the various environments where the lines originate, medium (approximately equals 5000) to high (approximately equals 30,000) spectral resolution is needed. The optical design, as well as a mechanical layout, incorporating the various modes is described. The imaging and spectroscopic channels are separated. The spectrometer is based on a long slit all reflective design. Two optical configurations have been studied in detail. Because of the need for variable magnifications, the imager is based on refractive optics.

The Gemini Telescopes are being built to exploit the unique infrared sites of Mauna Kea in Hawaii and Cerro Pachon in Chile. Both telescopes are being designed to deliver 0.1 arcsec images at the focal plane at 2.2 micrometers which will include all tracking and enclosure affects. Beyond 2 micrometers , using fast tip/tilt secondaries these 8 m telescopes will be essentially diffraction limited. In addition the use of protected silver coatings for both the primary and secondary mirrors and efficient in-situ mirror cleaning means the Mauna Kea telescope should be capable of delivering focal plane emissivities of approximately 2%. The baseline design for the Mauna Kea telescope also includes an adaptive optics system capable of feeding a 1 - 2 arcminute corrected field to near infrared instruments mounted at the f/16 Cassegrain focus. Fully exploiting the superb characteristics of the Gemini Telescopes will require a new generation of instruments which will challenge both instrument designers and infrared array technologies. The baseline complement of infrared instruments includes a 1 - 5 micrometers imager, a 1 - 5 micrometers spectrometer, and a mid-infrared (8 - 25 micrometers ) imager. Several optical instruments will also be built under the baseline instrumentation plan.

The Gemini Infrared Imager is a 1 - 5.5 micrometers general purpose camera to be built by the Institute for Astronomy for the Gemini Telescope on Mauna Kea, Hawaii. The camera will provide both high spatial resolution and wide field modes, and support spectroscopic, coronographic, and polarimetric capabilities. The camera project is currently in its preliminary design phase. We present the results of the conceptual design study.

We present the conceptual design for a medium-resolution (R equals 2000, 6000) spectrometer for the near IR (0.9 micrometers - 5 micrometers ) to be used on the Gemini 8 m Telescopes. The design goal is to make optimum use of the unique characteristics of the telescopes: superb image quality and low near-IR background. This leads us to propose a mostly reflective design, with cold, pupil-reimaging fore-optics. We achieve a modest slit length of 100 arcsec. In addition to the basic high-spatial-resolution configuration, the design permits a number of important upgrades: a camera for use with a wider slit, a prism cross-disperser, an integral field module to spatially sample a small 2-D region of the focal plane. The spectrometer is designed around the next generation of InSb detectors, the 1024 X 1024 Aladdin arrays currently under development.

We present a design for a near-infrared (0.9 to 5.5 micrometers ) spectrograph for use on any large telescope. For example, the instrument meets all of the scientific and technical objectives requested by the Gemini Telescope Project for their facility infrared spectrograph. The features of the instrument include a wide range of rapidly selectable spectral and spatial resolutions, full-broad-band imaging, integral field spectroscopy, and several cross-dispersed modes. Much of the instrument is based on optical, mechanical, and electronic designs currently in use. The optical design has diffraction-limited performance and no vignetting over a 150" X 150" field of view. The mechanical design draws heavily on our extensive experience with cryogenic mechanisms and uses a cassette system for selection of the large number of possible configurations. The design is very modular and allows a staged implementation of the complete set of potential operational modes.

NIRSPEC is a recently funded, high-resolution, 1 - 5 micrometers cryogenic spectrograph for the Keck II telescope. The design of this new instrument is based on 1024 X 1024 InSb arrays and provides resolving powers of R equals 2,000 in non-cross-dispersed mode and R equals 25,000 in echelle mode with typically 5 to 6 orders on the array covering 60 - 90% of the selected waveband, J, H, K, or L, in a single observation. Later, even higher resolution can be achieved by using the proposed adaptive optics facility at Keck II and replacing some of the internal modules of NIRSPEC. This paper gives a brief description of the proposed design concepts, and a discussion of the detector and system constraints required to achieve the scientific goals of the instrument.

ISOCAM, the camera of the Infrared Space Observatory, will make images of the sky in the wavelength range 2.5 to 17 microns. The camera was integrated with the telescope and the three other scientific instruments in January 1994 and since then three sequences of tests have taken place: a set of tests meant to measure the straylight in the cryostat; a complete set of tests (with the integrated payload module alone) to measure the performances of the camera; and the same set of tests, but with the fully integrated satellite (payload plus service modules). In the last two cases, the results of the tests were nominal and within a few percents of those obtained in the ISOCAM calibration facility. But in all three cases, the ISOCAM pupil imaging lens was useful in detecting and identifying heat sources or leaks within the cryostat, thus taking its first images through the ISO telescope.

The near infrared camera and multi-object spectrometer (NICMOS) is a new instrument for the Hubble Space Telescope (HST). NICMOS provides opportunities for near infrared astrophysical investigations with HST in the 0.8 - 2.5 micrometers spectral region with 256 X 256 array detectors of HgCdTe manufactured by the Rockwell International Electro-Optical Center. This manuscript describes the characteristics and capabilities of the flight focal plane assemblies (FPAs) for all three NICMOS cameras as well as the three flight spares and three continuous evaluation FPAs. All of the FPAs exceed the original specifications by a significant amount and represent the first scientific space flight of infrared arrays for this wavelength region.

For aerospace applications a miniature, solid-state near infrared (NIR) spectrometer based on an acousto-optic tunable filter (AOTF) has been developed and built at Brimrose Corp. of America. In this spectrometer a light emitting diode (LED) array as light source, a set of optical fibers as the lightwave transmission route, and a miniature AOTF as a tunable filter were adopted. This approach makes the spectrometer very compact, light-weight, rugged and reliable, with low operating power and long lifetime.

A new low-cost absolute cryogenic radiometer of the electrical substitution (ES) type optimized for performing black-body models calibration is described. Nitrogen is used as a cryogenic liquid for cooling of a radiometer receiving cavity up to temperature 80 K. This absolute ESR has been developed for measuring the irradiance in the range of 10^-3 divided by 10^-6 W/cm² with the uncertainty of 0.1% for the upper level of irradiance range. The receiving cavity having 16 mm aperture, is fabricated from copper foil with thickness of 30 micrometers . A substitution winding of the receiving cavity is made of manganin wires. Measured heat conduction of the receiving cavity is 1.1 X 10^-3 W/K. A single time constant of the receiving cavity is 80 sec. Brief description of the design, operating principles, and measurements results of the new cryogenic radiometer at nitrogen temperature are given.

This paper is a report on the ongoing flight evaluation testing of Spectralon, a diffuse reflectance material which is slated for use as a calibration panel for several satellite-based earth-observing instruments. The present study focuses on tests of the stability of the material under exposure to levels of UV/VUV radiation which match those of the low-earth orbit environment. In earlier UV/VUV exposure tests, some degradation of the optical properties of the material were observed; this optical degradation has been linked to photochemical degradation of organic contaminants. A more stringent manufacturing protocol was designed to eliminate these contaminants. The second phase of UV/VUV exposure testing, reported here, was undertaken with the object of validating and optimizing this new manufacturing protocol. Results of this testing indicate that the new manufacturing protocol yields a significant improvement in the optical stability of Spectralon under UV/VUV exposure. These results also indicate that most of the observed degradation is caused by exposure to radiation in the 200 - 380 nm band. This finding suggests new avenues of investigation, as well as providing justification for a simplification of future test requirements.

A revised baseline mission concept for the Space Infrared Telescope Facility (SIRTF) has been developed by the SIRTF science and engineering teams over the past year. This mission, which would be carried into solar orbit by a Delta 7920 launch vehicle, retains the key scientific capabilities of earlier mission concepts at a fraction of the mass and cost. We review the scientific and technical innovations which have enabled this development.

SIRTF, and other infrared space astronomy projects, require detector arrays with extremely high sensitivity. It is a goal of SIRTF to achieve background limited performance at all wavelengths. Especially critical is the spectral region around 3 micrometers , where the zodiacal dust emission and scattering reaches a minimum, and where there are no other natural background sources, and where a cooled space telescope provides negligible background. This is a spectral region where many of the most interesting astrophysical sources can best be studied. The IRAC (infrared array camera) SIRTF experiment requires short wavelength detector arrays (2 - 5 micrometers ) that achieve background limited operation.

We present initial test results for far-infrared arrays built at the University of Arizona using Ge:Ga photoconductors and low temperature readouts operating at a temperature of 2 K. We also present separate test results for the Hughes CRC-696 multiplexer used in this program. The two array configurations considered are a 1 X 32 based on an older heated readout concept and a new 4 X 32 module that takes advantage of the benefits of having a true cryogenic readout. Based on these results, performance meeting the SIRTF goal of background-limited imaging can be expected for the 32 X 32 array under construction.

We report on the design, modeling, and construction of far-infrared focal plane array modules for the Space Infrared Telescope Facility (SIRTF). The modules consist of 4 X 32 detector elements, readout electronics, and interconnections. The modules, which are of Z- plane architecture, may be stacked to produce imaging arrays of at least 32 X 32 format. These arrays are the largest available operating in the wavelength range 50 - 120 micrometers . The design takes advantage of the Hughes CRC-696 readout which was specifically developed for use with far-infrared photoconductive detectors. Since the readouts have excellent performance at temperatures below 2 K, a simplified construction concept using the proven interconnection methods of wire boding and multilayer ceramic boards are used. We report on additional design considerations such as minimization of parasitic capacitance at the input node and accommodation of thermal annealing of the detectors.

We present the design for the multiband imaging photometer for SIRTF (MIPS). MIPS is a versatile instrument that provides diffraction-limited imaging at 30 micrometers , 70 micrometers , and 160 micrometers . MIPS also provides low resolution (5%) spectroscopy in the 50 - 100 micrometers wavelength range to allow the determination of far-infrared spectral energy distributions. The 70 micrometers array can also be used in a special high resolution mode that supports image reconstruction techniques for improved angular resolution. The one cryogenic mechanism on MIPS is a scanning mirror based on a space-qualified design used on the Infrared Space Observatory. We describe modifications to the scan mechanism to optimize it for use at very long wavelengths.

The infrared spectrograph (IRS) to be flown in the Space Infrared Telescope Facility (SIRTF) makes use of many recent technological advances and will enable numerous new scientific investigations. The IRS is a broad-band (5 to 40 micrometers ) low and medium resolution (R equals (lambda) /(Delta) (lambda) equals 80 and 600) spectrograph, designed to take advantage of the low background conditions provided by SIRTF; it has no moving parts, and it will be background limited over much of its wavelength range. The IRS is a joint project of Cornell University, the California Institute of Technology, the University of Rochester, Ball Aerospace Group, the NASA Ames Research Center, and the Jet Propulsion Laboratory.

The revised baseline mission concept of the Space Infrared Telescope Facility (SIRTF) has assumed that the infrared array camera (IRAC), one of three focal plane instruments designed for SIRTF, will be removed from SIRTF and carried out collaboratively on the Japanese- launched Infrared Imaging Surveyor (IRIS) mission. The IRIS mission is an approved program scheduled for launch in 2001. In addition to IRAC, which will cover the 3 - 8 micron wavelength region, the proposed IRIS mission will also contain a mid-infrared array camera and a far-infrared scanner, consisting of three linear array detectors. We review the scientific requirements and design of IRAC required for incorporation into IRIS.

The Clementine mission provided the first ever complete, systematic surface mapping of the moon from the ultra-violet to the near-infrared regions. More than 1.7 million images of the moon, earth, and space were returned from this mission. The near-infrared (NIR) multi- spectral camera, one of two workhorse lunar mapping cameras (the other being the UV/visible camera), provided approximately 200 m spatial resolution at 400 km periselene, and a 39 km across-track swath. This 1.9 kg infrared camera using a 256 X 256 InSb FPA viewed reflected solar illumination from the lunar surface and lunar horizon in the 1 to 3 micrometers wavelength region, extending lunar imagery and mineralogy studies into the near infrared. A description of this lightweight, low power NIR camera along with a summary of lessons learned is presented. Design goals and preliminary on-orbit performance estimates are addressed in terms of meeting the mission's primary objective for flight qualifying the sensors for future Department of Defense flights.

The Clementine mission provided the first ever complete, systematic surface mapping of the moon from the ultra-violet to the near-infrared regions. More than 1.7 million images of the moon, earth, and space were returned from this mission. The long-wave-infrared (LWIR) camera supplemented the UV/visible and near-infrared mapping cameras providing limited strip coverage of the moon, giving insight to the thermal properties of the soils. This camera provided approximately 100 m spatial resolution at 400 km periselene, and a 7 km across- track swath. This 2.1 kg camera using a 128 X 128 mercury-cadmium-telluride (MCT) FPA viewed thermal emission of the lunar surface and lunar horizon in the 8.0 to 9.5 micrometers wavelength region. A description of this lightweight, low power LWIR camera along with a summary of lessons learned is presented. Design goals and preliminary on-orbit performance estimates are addressed in terms of meeting the mission's primary objective for flight qualifying the sensors for future Department of Defense flights.

An IR camera based on a Rockwell 256 X 256 HgCdTe (NICMOS III) array has been recently acquired by the U.S. Naval Observatory as part of a program to evaluate the astrometric characteristics of current-generation IR arrays. Results of a pixel-by-pixel photometric evaluation of this HgCdTe device are presented. Overall, the array is of extremely high quality with less than 100 (0.2%) photometrically unusable pixels. With a fairly stringent photometric selection based on linearity and full well, the bad pixel count is about 420. The average pixel photometric response is linear to within about 1.6% rms over a range of 0 - 120,000 e^-; using a quadratic relation reduces the calibration errors to under 1% rms. For comparison, the results of a similar test with a Texas Instruments TI800 X 800 CCD (a prototype of the HST WF/PC 1) are shown.

A fiberoptic link with a transmission rate of 1 gigabit/s is used in the ESO IR data acquisition system. The link connects the detector front-end with a multiprocessor system, based on T9000 transputers. The link not only transmits data -- the architecture of the system allows distribution of data to the multiprocessor system in a flexible and simple way. The paper describes principles of link operation, why this speed is required, and briefly the components used.