We describe the on-orbit characterization of the HgCdTe detectors aboard NICMOS. The flat-field response is strongly wavelength dependent, and we show the effect of this on the photometric uncertainties in data, as well as the complications it introduces into calibration of slitless grism observations. We present the first rigorous treatment of the dark current as a function of exposure time for HgCdTe array detectors, and show that they consist of three independent components which we have fully characterized - a constant component which is the true dark current, an 'amplifier glow' component which results from operation of the four readout amplifiers situated near the detector corners and injects a spatially dependent signal each time the detector is non-destructively read out, and finally the 'shading', a component well known in HgCdTe detectors which we show is simply a pixel dependent bias change whose amplitude is a function of the time since the detector was last non-destructively read out. We show that with these three components fully characterized, we are able to generate 'synthetic' dark current images for calibration purposes which accurately predict the actual performance of the three flight detectors. In addition, we present linearity curves produced in ground testing before launch. Finally, we report a number of detector related anomalies which we have observed with NICMOS some of which have limited the observed sensitivity of the instrument, and which at the time of writing are still not fully understood.

This paper describes the results of a test program to evaluate four Rockwell HAWAII and two PICNIC near IR array detectors with a view to their application in imaging, spectroscopy and in fast telescope tracking and interferometer fringe detection. Results of the laboratory test of the arrays are presented, together with a guide for their general operation.

This paper is a review of current astronomy projects at Raytheon/SBRC in the near-IR band. Another paper in this same session (3354-11) covers astronomy projects in longer wavelengths. For ground-based astronomy, InSb arrays with formats of 256 X 256, 512 X 512, and 1024 X 1024 have been developed and tested. For space-based astronomy, four projects are discussed with array formats ranging from 256 X 256 to 2K X 2K. The space projects support instruments on the SIRTF, IRIS, NGST, and Rosetta missions. Representative data are presented from 1024 X 1024 and 256 X 256 arrays obtained by test facilities at NOAO and the University of Rochester.

Four ALADDIN Type II 1024 X 1024 InSb arrays have been tested in the NOAO IR Detector Lab for use in the Gemini IR instruments; the Gemini Near-IR Spectrometer, and the near IR imager. Santa Barbara Research Center has been able to successfully deliver science devices. The best of which will be selected for use at Mauna Kea and Cerro Pachon. These results are reported to show the progress in the development of these arrays.

A high performance, bias tunable, p-GaAs homojunction interfacial work function internal photoemission far-IR (FIR) detector has been demonstrated. A responsivity of 3.10 +/- 0.05 A/W, a quantum efficiency of 12.5 percent, and a detectivity D* of 5.9 X 10¹⁰ cm (root) Hz/W, were obtained at 4.2K, for cutoff wavelengths form 80 to 100 micrometers . The bias dependences of quantum efficiency, detectivity, and cutoff wavelength have been measured and are well explained by the theoretical models. The cutoff wavelength is modeled by a modified high density theory, and the quantum efficiency is predicted by scaling the free carrier absorption coefficient linearly with the doping concentration. The effect of the number of layers on detector performance and the uniformity of the detectors have been discussed. A comparison with Ge:GA photoconductive detectors suggests that a similar or even better performance may be obtainable.

Ge:Ga far-IR photoconductor 2D direct hybrid arrays are being developed for application in the focal-plane detectors of the far-IR surveyor, one of the two main instruments of the IR imaging surveyor satellite. The arrays are composed of Ge:Ga photoconductor arrays fabricated on one chip, Si- pMOS readout integrated circuits, and a hybridization of them done by using indium bump technology.

We describe the design, construction, and performance of the 32 X 32 Ge:Ga imaging array being built at the University of Arizona for the Multiband Imaging Photometer for SIRTF (MIPS). The array will support a number of operational modes in the MIPS instrument including natural background-limited mapping at 70 micrometers , super-resolution observations at 70 micrometers , and spectral energy distribution measurements between 50 and 100 micrometers . The array is constructed in a modular manner using eight 4 X 32 pixel building blocks. To meet the sensitivity and stability requirements, the array must have excellent photometric repeatability, low noise, and robustness to the effects of the ionizing radiation environment in space. Key elements in attaining this level of performance are the Ge:Ga detectors materials and the cryogenic CRC-696 readout electronics. We present laboratory data for a 16 X 32 prototype of the array, and describe the plans for the construction of the qualification and flight units.

The HAWAII-2 is an IR 2048² focal plane array (FPA) that is being developed for next-generation IR astronomy. It will supplant our HAWAII 1024² as the largest high- performance imaging array available for IR astronomy. As with our prior IR sensor, the flip-chip hybrid will consist of a low-capacitance HgCdTe detector array mated to a low- noise CMOS silicon multiplexer via indium interconnects. In order to accommodate reasonable telescope optics and fabrication of the large sophisticated readout using world- class submicron CMOS, the FPA has 18 micrometers pixel pitch. We anticipate > 5 percent yield of defect-free multiplexers using 0.8 micrometers CMOS. The HgCdTe detector arrays will be fabricated on large wafers including sapphire and silicon. Though the first FPAs will have 2.5 micrometers cut-off, the readout will be able to support longer wavelengths. Also reported are the latest 1024 X 1024 FPA results with 2.5 micrometers HgCdTe detectors.

Rockwell Science Center has developed a double layer planar heterostructure (DLPH) detector array fabrication process with significant advantages over the PACE-1 process now being used to produce 256 X 256 and 1024 X 1024 FPAs for low background IR astronomy. The DLPH detectors are p- on-n photodiodes fabricated in a double layer of wide and narrow bandgap HgCdTe grown by molecular beam epitaxy on CdZnTe substrates. The double layer structure provides superior surface passivation while the lattice matched CdZnTe substrate reduces the defect density. DLPH FPAs have been fabricated in array sizes up to 640 X 480 and with cutoff wavelengths as long as 15 micrometers . Quantum efficiencies are typically in the 0.5 to 0.8 range. For a 256 X 256 array DLPH detectors with 5.3 micrometers cutoff wavelength at 50K, the median dark current was 0.39 e-/sec at 0.5V reverse bias. For 7 of 17 individual DLPH detector with 10.6 micrometers cutoff at 30K, the dark current was less than 10⁴ e-/sec at 20 mV bias. For long cutoff wavelengths, the detector breakdown voltage is too low to permit signal integration directly on the reverse biased detector capacitance. Such detectors require a readout circuit that maintains the detector near zero bias and provides a separate capacitor to store the integrated signal.

Instrument platforms like the VLT represent a new challenge to IR focal plane technology. Since the large telescope diameter and the improved image quality provided by adaptive optics reduce the pixel scale, larger array formats are needed. To meet this challenge ESO is participating in development programs for both InSb and HgCdTe large format arrays. To cover the spectral region of 1 to 5 micron ESO has funded a foundry run at SBRC to produce 1024 X 1024 InSb arrays, which will be installed in ISAAC, the IR Spectrometer and Array Camera built for the VLT. Since the delivery of the 1K X 1K InSb array is delayed, the test results obtained with a 256 X 256 InSb array and the application of off chip cryogenic amplifiers to InSb detectors will be discussed. Results obtained with a (lambda) _c equals 2.5 micrometers Rockwell 1024 X 1024 HgCdTe array will be presented, where an off chip cryogenic operational amplifier was used yielding a rms read noise of 3 electrons. Sensitivity profiles of individual pixels have been measured with a single mode IR fiber. Limitations of PACE 1 technology, such as persistence, will be discussed. First results with the 1K X 1K array, which was installed in SOFI, an IR focal reducer providing 1-2.5 micron imaging and long slit grism spectroscopy at the NTT telescope, will be presented. Advanced techniques of real time image sharpening will also be included. An outlook to the development of (lambda) _c equals 2048 X 2048 HgCdTe array formats will be given. The optical layout of NIRMOS, a multi-object spectrograph for the VLT telescope, is base don the availability of 2K X 2K HgCdTe arrays.

Raytheon/SBRC has demonstrated high quality Si:As IBC IR FPAs for both ground-based and space-based Mid-IR astronomy applications. These arrays offer in-band quantum efficiencies of approximately 50 percent over a wavelength range from 6 micrometers to 26 micrometers and usable responses from 2 micrometers to 28 micrometers . For high background, ground-based applications the readout input circuit is a direct injection (DI) FET, while for low background, space-based applications a source follower per detector (SFD) is used. The SFD offers extremely low noise and power dissipation, and is implemented in a very small unit cell. The DI input circuit offers much larger bucket capacity and better linearity compared with the SFD, and is implemented in a 50 micrometers unit cell. SBRC's Si:As IBC detector process results in very low dark current sand our Raytheon/MED readout process is optimized for very low redout noise at low temperature operation. SBRC is committed to achieving still better performance to serve the future needs of the IR astronomy community.

High performance 128 X 128 Si:As and Si:Sb blocked- impurity-band hybrid arrays have been developed for ground- based and airborne astronomy. These devices cover the 5-25 (Si:As) and 15-40 micrometers (Si:Sb) portions of the spectrum. The peak detective quantum efficiencies quantum efficiencies exceed 50 percent for Si:As and 30 percent for Si:Sb. An anti-reflection coat is used to increase responsivity and to reduce internal reflections for the Si:As detectors. The multiplexer yields a linear output response vs. integrated charge. A special design feature of the multiplexer is a changeable nodal capacitance that allows dynamic switching of the well depth between 1.8 X 10⁶ and 1.8 X 10⁷ electrons. The single-sample read noises for the two states are approximately 75 and approximately 760 electrons respectively. These devices have been used successfully to perform astronomical observations in a number of instruments.

Manufacturing specifications of IR transmitting crystalline optical components for wavelengths >= 0.7 microns should be important to the IR astronomer when designing imaging systems especially where cost and delivery are also major program concerns.

The ESO IR detector high speed array control and processing electronic IRACE is designed as a modular system and supports readout and data processing of arrays with four as well as multiple output channels. In addition the system can handle multiple separate arrays and the data re routed to multiple processing chains. Detector front-ends are galvanically separated form data processing and system administration with fiberoptic links. Interfaces to different data processing systems for on-line data handling are implemented. The paper describes principles of system operation, and the achieved readout and on-line processing speeds.

The NASA IRTF is building a multiple digital signal processor (DSP) based array electronics control system for SpeX, an NSF funded 1 to 5 micron medium resolution spectrograph. SpeX will use a 1024 X 1024 InSb array for spectroscopy and one 512 X 512 quadrant of another 1024 X 1024 InSb array for slit field viewing and IR guiding. An additional system is also being produce at the Institute for Astronomy for the SUBARU IR camera and spectrograph (IRCS). Plans for IRCS include the use of a 1024 X 1024 InSb array for spectroscopy and one 1024 X 1024 InSb array for IR imaging. This document will provide the instrument derived requirements, an overall system description, and some of the tradeoffs and technical choices made. The design for both system is an evolutionary upgrade of the current IRTF array control electronics system used in a 256 X 256 InSb based imager, a 256 X 256 InSb and 512 X 512 CCD in an echelle spectrograph, an 800 X 800 CCD based tiptilt correction system and a non-IRTF 128 X 128 Si:As BIB array based imager.

An overview on the use of grisms from high refractive index optical materials is given. When using grisms manufactured from silicon or germanium two IR focal-reducers of the European Southern Observatory (ESO) can serve as medium resolution echelle spectrometers. A silicon echelle grism allowing for a spectral resolution of 5000 for a 1 arcsec slit is being developed for SOFI, a near IR instrument featuring a 1024 Rockwell HgCdTe detector at ESO's 3.5m New Technology Telescope. For TIMMI2, ESO's new 10/20 micrometers instrument for the 3.6m telescope a germanium echelle grism is being built. For TIMMI a 10 micrometers camera featuring a 64 X 64 detector a low resolution germanium grism yielding a spectral resolution (lambda) /(Delta) (lambda) equals 200 for a 1 arcsec slit has already been successfully commissioned. The manufacturing process, the status and performances will be presented. Moreover we show some astronomical results.

Linear variable filters have found increasing applicability in spectrally selective optical instruments. They serve as moderate resolution spectral discriminators in astronomical instruments and in reconnaissance equipment. They perform extremely well as 'sliding out-of-band blocking filters' when used in conjunction with grating spectrometers.

We describe a grism suitable for low-resolution, slitless spectroscopy in the IR region between 3.0 and 5.0 micrometers . The grism is fabricated in silicon using a three-mask, photolithographic process, resulting in an eight-step binary approximation to the normal sawtooth grating profile. Desirable features of this approach include the ability to incorporate aberration correction in the gratin and a gentle ruing relief profile permitting a conformal anti-reflection coating for improved efficiency. To demonstrate the performance of this grism in a practical applications, we have constructed a slitless spectrograph system using an off-the-shelf InSb camera and simple, uncooled, refractive optics. This system is well suited to observing compact, bright, transient phenomena without good a priori knowledge of their positions. We present examples of present instrument performance. An upgrade currently under construction will increase sensitivity by cooling more of the optical path and increasing the aperture of the collecting optics. We plan to use the improved instrument to observe the Leonid Meteor shower in November 1998.

In order to increase throughput and maximize sensitivity the next-generation of astronomical instrumentation is moving toward cryogenic, all-reflective, off-axis optical design solutions. These off-axis systems require mirrors which are produced with complex conic sections, demand a thermal optical performance at cryogenic temperatures, and must support lifetimes on the order of 5-10 years. SSG specializes in the design, development, fabrication and testing of off-axis, all-reflective optical systems, having produced > 40 such systems over the last 20 years. The majority of these system have been produced using nickel plated aluminum mirror substrates and aluminum metering structures in order to obtain a passively systems has long been a point of debate. In this paper we demonstrate the long term stability of nickel plated aluminum optics by presenting interferometric test data obtained on > 10 optical elements over a period of 10 years. Cryogenic stability is demonstrated by presenting system level wavefront data obtained over a wide thermal range down to 115K. In addition, we will present thermal test data obtained from a number of alternate metal optical materials: beryllium, bare aluminum, and aluminum/beryllium alloys.

The new design of image slicer developed at Durham University for 2D area spectroscopy is described. The unit acts as a coupler between the telescope and a spectrograph to reformat a square or rectangular field into a long slit. Its advantages over previous designs of image slicers and other methods using fibers, lenses or narrow-band filtering are discussed, mainly: large field, high spatial resolution, large number of spectral resolution elements, high transmission, and the small size of the instrument. The system is also easy to cool and is then well suited for IR spectroscopy. The proposed design is a new type of image slicer in which the original 2D image is sliced into narrow sub-images that are re-imaged side by side to form a long 1D image at the spectrograph input. The flexibility of the concept at the base of this new design is highlighted through the description of 5 different slicer designs. Three of these are for future instruments now at the design phase: the CGS4 slicer, the UIST slicer and the GNIRS slicer; the two others are studies for possible future slicers on GMOS and NGST. These designs show how easily our slicer can be added to an existing instrument, how it can be incorporated to the slit wheel of future instruments, and how multi-slit reformatting permits a much larger field of view.

Micromachined silicon gratings offer two great advantages to astronomical spectroscopy in the IR: (1) Photolithographic processing techniques permit the production of gratings with much larger groove constants than are possible with conventional wavelength coverage, despite the relatively small format of IR arrays. (2) One can use anisotropic etching to form gratings on dielectric wedges. By illuminating the grating through the dielectric, we can achieve higher spectral resolution for a given grating size or a smaller grating for a given desired resolution. We discuss the technical challenges involved in micromachining large grating grooves over large areas while holding positional accuracy to very tight tolerances. Manufacturing issues include material choices, surface preparation, and chemical and physical effects during processing. We also discuss our program for evaluation of the finished products, show result of measurements we have made on front-surface and immersion devices, and use these result to assess the potential of these devices for real-world astronomical applications.

SPIFFI is an integral field spectrograph with an HAWAII array that enables us to simultaneously take near IR spectra of 1024 spatial pixels in a hexagonal field of view on the sky. It can be used on 4 to 8 meter class telescopes with a maximum pixel scale of 0.5 arcsec and with adaptive optics pixel scales, Nyquist sampling the point spread function of the telescope. A fiber bundle of 1024 silica/silica fibers rearranges the 2D field of view into the 1D entrance slit of the spectrometer. A novel technique involving flared fibers is used to achieve a high filling factor and coupling efficiency. Each fiber tip in the bundle is flared to increase the fiber core diameter by a factor of 15. The tapered end is polished to form a spherical micro-lens with a hexagonal cross-section to couple light into the fiber core. Apart from yielding a high coupling efficiency and a high geometrical filling factor, the monolithic micro- lens/fiber system can be used at a working temperature of 77K without loosing positioning accuracy. The spectrometer optics is achromatic from 1.1 to 2.5 microns and use four reflection gratings on a wheel as dispersing elements with a resolving power from 2000 to 4500. The fore-optics includes the filter wheel, the cold pupil stop and a scale changing mechanism to switch between three different image scales according to observing and seeing conditions. The spectrometer collimator is a f/4.3 three lens achromat, the spectrometer camera is a f/1.2 folded Schmidt camera. The optical design of the spectrometer is distortion free to get straight, equidistant spectra that match the columns of the detector, thus minimizing cross-talk form adjacent spectra to less than 5 percent.

We are constructing a fully cryogenic near-IR multi-object spectrometer for use from 9000 angstrom to 2.4 micrometers . When completed in the summer of 1999, FLAMINGOS will be the first fully cryogenic near-IR multi-object spectrometer and will thus allow efficient background limited operation through the entire H and K windows. Due to its very fast, wide field optical design FLAMINGOS is also an excellent survey imager being more than an order of magnitude more efficient than current and planned near-IR cameras. When used for multi- slit spectroscopy FLAMINGOS will be a factor of 50 to 100 more efficient than current near-IR spectrometers. FLAMINGOS can accept nay input beam slower than f/6. This makes it extremely versatile allowing use on a large number of telescopes. FLAMINGOS will have a large collimated space with a grism wheel and a filter wheel, providing multiple object near-IR survey spectroscopy at resolutions of 600 up to 2000. It will have a small separate cryogenic dewar with a cycling time less than 6 hours, which will hold 11 cryogenic multi-slit plates that are fabricated outside the dewar. This will allow multi-slit spectroscopy 50-100 objects simultaneously.

The MDM/Ohio State/ALADDIN IR Camera (MOSAIC) is a general purpose near IR imaging camera and medium-resolution long- slit spectrometer in use on the MDM 1.3-m and 2.4-m telescopes and the Kitt Peak 2.1-m and 4-m telescopes. In cooperation with NOAO and USNO, MOSAIC is one of the first general-purpose near-IR instruments available to the astronomical community that uses a first-generation 1024 X 512 ALADDIN InSb array, with the capability to use a full 1024 X 1024 array once one becomes available. MOSAIC provides tow imaging plate scales, and a variety of long- slit grism spectroscopic modes. This paper describes the general instrument design and capabilities, and presents representative scientific results.

The SMIRFS prototype near IR fiber system has been designed to provide a multi-object capability in the J, H and K bands and an integral field capability from 1 to 2 micrometers in conjunction with the cooled grating spectrograph CGS4 at the UKIRT. In multi-object mode there are 14 fibers covering a 12 arcmin² field and in integral field mode there are 72 fibers covering a 6 X 4 arcsec field with spatial sampling of 0.63 arcsec. Both modes use fibers that are coupled to the telescope and spectrograph using lenslets. The information gleaned from these devices should benefit the next generation of IR spectrographs that will take full advantage of the larger chip formats that are now coming on line.

We have built a panoramic wide field near infrared imaging camera based on 4 Rockwell HAWAII 1024 X 1024 detectors. The baseline survey instrument operates in the region 0.8 to 1.8 micrometers on non-IR optimized telescopes with an upgrade at K band in 1999. The instrument was commission on the 2.5m INT and 4.2m WHT telescopes in December 1997 and January 1998. The main design goals in this project were to produce a highly productive astronomical instrument in a very short space of time and for low cost. Survey instruments are by their nature very versatile and CIRSI will support the wide range of astronomical interests at the Institute of Astronomy. Furthermore, since CIRSI is a traveling instrument and we are able to operate at a number of different telescopes to take opportunity of a range of image sizes and scales.

A new mid-IR spectrograph, the Texas Echelon Cross Echelle Spectrograph (TEXES) is under construction. The primary motivation for TEXES is to observe interstellar molecules at very high resolution. TEXES will operate at 7-25 micrometers wavelength with three spectrographic modes: a high resolution cross-dispersed mode, with R approximately equals 100,000, a mid-resolution long-slit mode, with R approximately equals 14,000, and a low resolution long-slit mode, with R approximately equals 2000. In hi-res mode, the primary disperser is a 36 inch long, R10 grating with a 7 mm groove spacing. The echelon is cross-dispersed with a 7 in long R2 echelle. In mid-res mode, the echelon is by-passed with an Offner relay, and the echelle is used by itself. In lo-res mode, a first-order grating is inserted over the echelle. For initial test, TEXES will use a Hughes Aircraft 20 X 64 pixel Si:As impurity-band array, which covers only two echelon orders. It will later be replaced with a 256 X 256 pixel array, which will Nyquist sample approximately 10 orders. The spectrograph has been assembled and tested with a partially complete echelon, demonstrating the soundness of the design. When we began this project, we were unable to find a vendor capable of machining or ruling a diffraction grating with the very coarse ruling required. Consequently, we attempted to hand-fabricate the echelon. We have not succeeded in assembling the echelon with the required precision, missing by about a factor of two. Fortunately, Hyperfine, Inc. is now capable of diamond machining the echelon. We are purchasing a machined echelon, and hope to complete the spectrograph by the end of summer 1998.

The Instituto de Astrofisica de Canarias (IAC) is undertaking the design and construction of a common-user near IR spectrograph (LIRIS) for the Cassegrain focus of the 4.2 m William Herschel Telescope sited at the Observatorio del Roque de Los Muchachos. LIRIS will be a near IR intermediate-resolution spectrograph designed to operate over a spectral resolution range between 1000 and 5000, with added capabilities for coronographic, multiproject and polarimetric observations. The instrument allows the combination of an adequate spatial resolution with a large useful field of view across the slit, thanks to the use of the new 1024 X 1024 pixel HgCdTe Hawaii detector manufactured by Rockwell. All the optics and mechanisms situated inside the cryostat will be cooled to below 100 K. The detector will operate at 77 K. Calibration and tracking will be made with the existing Cassegrain A and G Box, into which a near IR calibration system will be incorporated.

An imaging spectrometer is being designed to take advantage of recent improvements in the image quality achieved at the UK Infrared Telescope. The realization of near-diffraction limited imaging at two microns brings with it the possibility of significant improvements in sensitivity to IR observations. UIST will provide a versatile facility for high spatial resolution imaging and spectroscopy in the 1-5 micrometers wavelength range. We will present the opto-mechanical design of this new instrument, highlighting the innovative features. These include provision of multiple pixel scales within the camera and polarimetry via a Wollaston prism. One of the most challenging areas of the design is the inclusion of a cryogenic integral field unit for area spectroscopy over a 5 inch field. The spectroscopic modes include cross- dispersed spectroscopy over the complete 1-2.5 micrometers wavelength ranges and moderate resolution long slit or area spectroscopy over the complete 1-5 micrometers range. A higher resolution mode will also the included. This will allow USTI to take advantage of the very low backgrounds to be found between OH sky lines. The instruments will incorporate a 1024 X 1024 Indium Antimonide array from SBRC. The development of the IR array controller for UIST will also be discussed.

SpeX is a medium-resolution 0.8-5.5 micrometers cryogenic spectrograph being built at the Institute for Astronomy, University of Hawaii, for the NASA IR Telescope Facility on Mauna Kea. SpeX was funded by the National Science Foundation in July 1994. First-light is expected in 1999. The primary scientific driver of the instrument is to provide maximum simultaneous wavelength coverage at a spectral resolving power which is well-matched to many planetary, stellar and galactic features, and which adequately separates sky emission lines and disperses sky spectral resolutions of R approximately 1000-2000 simultaneously across 0.9-2.5 micrometers , 2.0-4.2 micrometers , or 2.4- 5.5 micrometers . SpeX will use an Aladdin II 1024 X 1024 InSb array in its spectrograph and an Aladdin II 512 X 512 InSb array in its IR slit-viewer.

The South Pole Imaging Fabry-Perot Interferometer (SPIFI) is a direct detection, imaging, submillimeter spectrometer. The spectral resolving elements are a pair of cryogenic, scanning Fabry-Perot interferometers which use a free- standard Ni mesh for the etalon mirrors. The detectors for SPIFI are a 5 X 5 array of bolometers coupled to the focal plane with Winston cones. An adiabatic demagnetization refrigerator cools the bolometers to approximately 60 mK while a ³He system operates simultaneously as a thermal guard. SPIFI is intended to operate on the ASO/RO submillimeter telescope at the South Pole and on the JCMT telescope on Mauna Kea and will be used to study the gas- phase reservoirs of carbon in star-forming regions in our own and near-by galaxies. SPIFI takes advantage of three things: (1) Advanced bolometers that achieve background limited performance at very high resolving powers. (2) The imaging capability and high spectral resolving power of Fabry-Perot interferometers. (3) The superb atmospheric transmission in submillimeter bands possible from the South Pole. The SPIFI uses state-of-the-art monolithic silicon bolometers fabricated at the NASA Goddard Space Flight Center. The cryogenic, scanning Fabry-Perots in SPIFI were designed and built at Cornell and are an evolution of the design used with great success for the Kuiper Wide Field Camera. The 1.7 m Antarctic Submillimeter Telescope/Remote Observatory exploits what is thought to be the best submillimeter observing site in the world.

Omega Cass is the new MPIA multi-mode camera for imaging and spectroscopy at near IR wavelengths between 1.0 and 2.5 micrometers . The Camera is equipped with an 1024 X 1024 HAWAII HgCdTe focal plane array from Rockwell. The cryogenic re- imaging optics are designed to cover a wide variety of observing conditions. The imaging scales can be changed during observations, allowing the observer to react to changing conditions. Three different lens sets provide scales of 0.3, 0.2 and 0.1 arcsec/pixel at the f/10 Cassegrain focus of the 3.5m telescope. In combination with a laser based adaptive optics system, available at the same telescope, these imaging scale correspond to 0.12, 0.08 and 0.04 arcsec/pixel, which double samples the diffraction limit at the shortest operation wavelength. A set of grisms allow low to medium resolution long slit spectroscopy up to R equals 1000. In addition, sensitive polarimetry can be done with Wollaston prisms and wire grid analyzers. Omega-Cass is mainly designed for the 3.5m telescope on Calar Alto, although it may be used at any other telescopes with a focal ratio slower than f/8, including the MPIA's 2.2m telescopes on Calar Alto and La Silla.

Two wide field, multiple beam, near IR cameras are being developed for the MMT and Whipple Observatories. For the MMT a triple beamed, 1024 X 1024 pixel camera for the 6.5-m f/5 is being designed. As a prototype for the MMT and for use on the Whipple Observatory 1.2-m telescope, a dual beamed, 256 X 256 InSb array camera with three selectable magnifications has been built and in use for the past year.

A 1-5 micrometers IR camera and spectrograph (IRCS) is described. The IRCS will be a facility instrument for the 8.2 m Subaru Telescope at Mauna Kea. It consists of two sections, a spectrograph and a camera section. The spectrograph is a cross-dispersed echelle that will provide a resolving power of 20,000 with a slit width of 0.15 arcsec and two-pixel sampling. The camera section serves as a slit viewer and as a camera with two pixel scales, 0.022 arcsec/pixel and 0.060 arcsec/pixel. Grisms providing 400-1400 resolving power will be available. Each section will utilize an ALADDIN II 1024 X 1024 InSb array. The instrument specifications are optimized for 2.2 micrometers using the adaptive optics and the tip-tilt secondary systems of the Subaru Telescope.

The following capabilities have been identified as high priority for future Gemini instrumentation. (1) A natural guide star/laser beacon adaptive optics (AO) system at Cerro Pachon, and laser beacon capability for the Mauna Kea AO system. (2) A near IR coronagraph/imager for Cerro Pachon. (3) 1-2.5 micrometers multi-object spectroscopic capability, including IFU and multi-slit capability for use with AO corrected images, and wide field multi-object capability over at least 5 arcmin field of view. (4) Polarization modulators for optical and IR wavelengths at both Mauna Kea and Cerro Pachon. (5) A high stability lab optical spectrometer for Cerro Pachon, with resolutions around 120K and >= 300K.

The University of Florida is developing the mid-IR imager, called GatirCam, to be used primarily, but not solely, at the southern hemisphere Gemini telescope at Cerro Pachon, Chile. Key features of GatirCam are its fully reflective optics, its very high mechanical rigidity, and the fact that the associated electronics are very similar to those is in use successfully on similar instrumentation. Design studies for GatirCam indicate that it will meet or exceed all critical requirements of image quality and performance. A low-resolution spectroscopic mode is also currently under consideration for implementation in GatirCam.

We discuss the main design features of the Gemini Near-IR Imager (NIRI) and its scientific capabilities. NIRI is designed to fully exploit the excellent image quality and low telescope emissivity expected from the Gemini telescope on Mauna Kea. It offers a range of pixel scales matched to different scientific objectives and has spectroscopic as well as polarimetric capabilities. One of its main design features is the use of a near-IR 2 X 2 Shack-Hartmann wavefront sensor for tip-tilt and focus control.

The design of a near-IR spectrometer for the Gemini 8m telescopes is described. This instrument, GNIRS, provides coverage from 0.9 to 5.5 micrometers at several spectral resolutions and two pixel scales. Capabilities include an imaging mode intended primarily for acquisition, a cross- dispersed mode covering wavelengths from 0.9 to 2.5 micrometers , and provisions for an integral field unit. The design of the GNIRS is conservative, as it must meet tight schedule and resource constraints; it nonetheless provides high throughput and operational efficiency, minimal flexure, and the flexibility needed to support queue observing. The optics are a combination of diamond-turned metal optics for the fore-optics and collimator, and refractive optics for the cameras. The mechanism include a two-axis grating turret; all mechanism are deposited by means of internal detents. The instrument achieves low flexure within its weight budget by the use of a modular structure composed of cylindrical light-weighted sections into which individual mechanisms and optics modules are mounted. Extensive analyses of mechanical and optical performance have been performed. The GNIRS has passed its critical design review, and fabrication is now underway.

The design and development of NIRSPEC, a near-IR echelle spectrograph for the Keck II 10-meter telescope is described. This instrument is a large, facility-class vacuum-cryogenic spectrometer with a resolving power of R equals 25,000 for a 0.4 inch slit. It employs diamond-machined metal optics and state-of-the-art IR array detectors for high throughput, together with powerful user-friendly software for ease of use.

We present a technical description, and up-to-date status report, of the mid-IR imaging Fabry-Perot interferometer system (MIRFI) currently under development by the IR Astronomy Group at the University of California, Irvine. MIRFI, which is designed primarily for operation with the new Keck 10-m telescopes, will facilitate diffraction- limited imaging at extremely high spectral resolution in a variety of important mid-IR ionic and molecular spectral lines. The unique design goals of MIRFI, namely its high spectral resolving power and diffraction-limited imaging capability, are ideally suited for studying the dynamics of extended, optically obscured regions of ionic or warm molecular gas. Such regions include, but are not restricted to, (1) HII regions associated with young OB star clusters, (2) molecular cloud shocks associated with young stellar outflows, (3) photodissociation regions, (4) the Galactic Center, and (5) galactic nuclei and extragalactic HII regions.

The long wavelength IR camera is a facility instrument for the Keck Observatory designed to operate at the f/25 forward Cassegrain focus of the Keck I telescope. The camera operates over the wavelength band 7-13 micrometers using ZnSe transmissive optics. A set of filters, a circular variable filter, and a mid-IR polarizer are available, as are three plate scales: 0.05 inch, 0.10 inch, 0.12 inch per pixel. The camera focal plane array and optics are cooled using liquid helium. The system has been refurbished with a 128 X 128 pixel Si:As detector array. The electronics readout system used to clock the array is compatible wit both the hardware and software of the other Keck IR instruments NIRC and LWS. A new pre-amplifier/A-D converter has been designed and constructed which decreases greatly the system susceptibility to noise.

The instrument ensemble at ESO's Very Large Telescope will be extended by a high-resolution echelle spectrograph for the 1-5 micrometers wavelength range. Primary design goals are high spectral resolution, wide coverage and high sensitivity. Several novel features are introduced with this instrument. At the same time, development time and risk will be minimized through the use of concepts available from earlier ESO VLT instruments.

A high resolution near IR camera (CONICA) for the firs VLT unit is under development, which will provide diffraction limited spatial resolution being combined with the adaptive optics system NAOS. CONICA serves as a multi-mode instrument for the wavelength region between 1.0 and 5.0 micrometers , offering broad band, narrow band or Fabry Perot direct imaging capabilities, polarimetric modes using Wollaston prism or wire grid analyzers and long slit spectroscopy up to a spectral resolution of about 1000 per two pixel. We presented a first concept of CONICA in 1995. In the mean time, large parts of the instrument have been manufactured, the cryostat and the adapter have been finished and first cryogenic test have been performed. This paper describes the actual design and status of development of CONICA focusing on those aspects which have not been described in detail before or the design of which have been changed in the mean time.

In this paper, we present VISIR, the mid-IR instrument to be installed in 2001 on the telescope unit number 3 or the European Very Large Telescope program. The instrument combines imaging capabilities over a field up to about 1 arcmin at the diffraction limit of the telescope, and long- slit grating spectroscopy capabilities with various spectral resolutions. The contract to design and build VISIR was signed in November 1996 between ESO and a French-Dutch consortium of institutes. One year after the signature of the contract, VISIR has successfully passed the preliminary design review. The results of the first year of studies are presented here.Emphasis is put on the optical design which is in its final form.

The imaging photopolarimeter ISOPHOT on-board the European satellite ISO houses 144 background detectors of Si:Ga, Si:P, Ge:Ga and stressed Ge:Ga, all sampled by newly developed cold read-out electronics. There is large temporal radiation damage to most of these detectors on the daily passage through the earth's radiation belts. In addition the Ge:Ga detectors exhibit a continuous responsivity increase caused by the cosmic radiation far off the earth. Effective curing procedure shave been developed to heat out these effects. The in-flight sensitivities achieved are close to the pre-flight predictions for most channels. At 100-200 micrometers cirrus confusion is a serious limit for the detection of faint objects on large parts of the sky. The cold filter wheel carrying 56 optical elements, such as filters, apertures and polarizers, as well as the focal plane chopper, operate with high precision and very low power consumption. Due to an effective cold internal baffle system the measured near-field straylight was close to the pre- flight theoretical prediction based on APART simulations. THe sun and moon straylight at 25 and 175 micrometers was measured during several solar eclipses. Drift and transients of the detectors, non-linearities of the preamplifiers, ionizing radiation effects and a complex optical path make the photometric calibration of this instrument challenging. Because most of these effects are reproducible, a calibration accuracy of < 30 percent is already available for most photometric modes. Examples of observations, including the 175 micrometers Serendipitous Sky Survey, will highlight the capabilities of the instrument.

The long wavelength spectrometer on-board the European Space Agency IR Space Observatory (ISO) uses a grating and one of two Fabry-Perot interferometers to make medium and high resolution spectroscopic observations in the 43-196.9 micrometers wavelength range. The instrument has been in continuous use since the launch of ISO in November 1995. In this paper we describe the calibration of the instrument and its performance, both spectroscopic and photometric, over the two years of instrument operations.

The telescope system of a Japanese IR Astronomical Space Mission, 'IR Imaging Surveyor (IRIS)', is described. The IRIS is a cryogenically-cooled telescope, being planned to be launched in 2003. It will make astronomical observations from near-IR to far-IR regions. The IRIS telescope system is a Ritchey-Chretien type, whose primary mirror size ins 700mm in diameter and whose system F ratio is 6. In order to share the focal plane with two scientific instruments and a focal- plane star sensor, it has a clear field of view of 38 arcminutes in radius. It is being designed to achieve the diffraction-limited performance at 5 micrometers for temperatures below 10K. The IRIS telescope will use light-weight silicon carbide (SiC) mirrors. The current estimate of the primary mirror weight is 9 kg and the goal of total weight of the telescope system is less than 27 kg. Preliminary tests of small size SiC mirrors at 4.2K suggest that slight distortion of the surface figure detected at low temperatures can be reduced by improved CVD processes. The telescope system is designed to meet the launch conditions of the M-V rocket and to have the fundamental frequencies above 100 Hz.

The far-IR Surveyor (IRS) is one of the two focal plane instruments of the IR Imaging Surveyor, IRIS, which is a Japanese IR astronomical satellite. FIS is designed primarily to perform an all-sky survey with several photometric bands like IRAS. Advantages of FIS to IRAS are its high detectivity of point sources and its longer wavelength capability. These features are gained by remarkable improvement in detector technology. FIS adopts currently developed unstressed and stressed Ge:Ga array detectors to cover 50 to 200 micrometers in wavelength. Due to highly sensitive detector system, it is expected to detect over 10 million objects by the all-sky, including a lot of high-z objects. FIS also has spectroscopic capability by a Fourier spectrometer covering 50 to 200 cm^-1 in wave number with spectral resolution of 0.5 cm^-1. The same detector arrays of the scanner are used and these two functions are switched. As a result of combining a spectroscopic function with the scanner, FIS becomes a unique instrument. The basic observation mode of the FIS is an all-sky survey using the scanner. The spectroscopic function is operated in the pointing mode in which it can take longer integration time. Spectral information can be used to estimate the redshifts of strange objects detected by the all-sky survey. The spectrometer is also a powerful instrument to reveal the physical properties of galactic and nearby sources.

Basic design and current development status of IRC: an IR camera on-board the IRIS is presented. IRC is one of the focal-plane instruments of the 70cm cooled telescope of the IRIS. IRC utilizes recently developed large-format IR arrays for imaging and low-resolution spectroscopy at wavelength 1.8-26 micrometers . IRC consist of 3 camera channels: NIR, MIR-S, and MIR-L. These 3 channels simultaneously observe different fields of the sky, with diffraction-limited spatial resolution. One critical component limiting the performance of the IRC is the performance of large-format arrays: 512 X 412 format InSb and 256 X 256 format Si:As IBC arrays operation at very low temperature. Performance test of the Si:As array manufactured by the Hughes/SBRC is under way, and the preliminary results is presented. Design of the camera optics and the optical components is also presented. IRC is operated under the pointing observation of about 500 sec exposure time, and the development goal is to achieve high point-source detectivity limited nearly by the confusion due to faint astronomical sources.

The joint US and German SOFIA project to develop and operate a 2.5 meter IR airborne telescope in a Boeing 747-SP is now in its second year. The Universities Space Research Association , teamed with Raytheon E-Systems and United Airlines, is developing and will operate SOFIA. The 2.5 meter telescope will be designed and built by a consortium of German companies led by MAN. Work on the aircraft and the preliminary mirror has started. First science flights will begin in 2001 with 20 percent of the observing time assigned to German investigators. The observatory is expected to operate for over 20 years. The sensitivity, characteristics, US science instrument complement, and operations concept for the SOFIA observatory, with an emphasis on the science community's participation are discussed.

SOFIA will permit observations not possible from ground based telescopes, while retaining a number of their major advantages for observers. These include the opportunity to change focal plane instruments frequently, and continuous access to the instrument while observing. SOFIA is being designed to maximize the benefits of these features and to assure optimum performance of the instruments, within the constraints of available resources. This paper describes the top level optical, mechanical, and electronic interface parameters and configuration issues foreseen for Science Instruments on SOFIA.

NASA's Stratospheric Observatory for IR Astronomy (SOFIA) will enable unprecedented IR acuity at wavelengths obscured from the ground. To help open this new chapter in the exploration of the IR universe, we are developing the Airborne IR Echelle Spectrometer (AIRES) as a facility science instrument. Full funding was awarded for a four year development in October, 1997. The instrument is scheduled to come on-line with the observatory in the Fall of 2001. It will be used to investigate a broad range of phenomena that occur in the interstellar medium. AIRES will use a 1200 mm long, 76 degree blaze angle echelle to combine high resolution spectroscopy with diffraction-limited imaging in the cross-dispersion direction. Its three 2D detector arrays will prove good sensitivity over a decade in wavelength. An additional array will be used as a slit viewer for (lambda) <EQ 28 micrometers to image source morphology and to verify telescope pointing. Our scientific motivation, preliminary optical design and packaging, focal plane configuration, echelle prototyping, and cryostat layout are described.

The spectral-photometric IR camera SPICA is proposed as one of the German science instruments of the Stratospheric Observatory for IR Astronomy (SOFIA). It will cover a wavelength range of 20-220 micrometers with three large area detector arrays. With the 2.5 m SOFIA telescope, SPICA will provide unprecedented diffraction limited spatial resolution in the far-IR. In addition, low resolution 3D-imaging spectroscopy is planned. While the silicon array will be commercially available, the germanium arrays are being developed, including their cryogenic multiplexers. The overall instrument concept, its camera optics and the status of the detector development will be presented. The instrument is being developed by the DLR Institute of Space Sensor Technology in Berlin with support of several German and US partners.

To study narrow features in quiescent molecular clouds, a high spectral resolution, high sensitivity instrument is required. We present the design and capabilities of a mid-IR spectrograph being built as a PI instrument for the Stratospheric Observatory for IR Astronomy. The Echelon- Cross-Echelle Spectrograph (EXES) will operate from 5.5-28.5 micrometers in three spectroscopic modes: R approximately 10⁵, 10⁴, and 1500. EXES is similar in concept to our ground based instrument, TEXES. EXES consists of three chambers. The first chamber contains focal-reducing optics. The second chamber houses the high resolution grating, an echelon. Discussion of the echelon can be found elsewhere in this volume. The third chamber contains an echelle grating and a first order grating mounted back-to-back. A flip mirror selects operating mode by either directing light into the echelon chamber or allowing the light to pass through to the grating chamber. Re-imaging optics upstream of the detector provide two plate scales. The detector, a 256 by 256 pixel Si:As IBC, is in a baffled subsection of the third chamber. Finally, the low resolution grating serves as a slit positioning camera when it is rotated face on.

We describe our design for an imaging far-IR spectrometer for NASA/DARA's SOFIA observatory. The design of the instrument is driven by the goal of maximizing observing efficiency. Since the sensitivity of well designed FIR instruments is limited by the thermal background of telescope and atmosphere, observing efficiency can only be increased by increasing the throughput of the spectrometer and the number of simultaneous data channels. Our instrument will feature two separate medium resolution grating spectrometers with common fore-optics feeding two large Ge:Ga arrays. The two Littrow spectrometers operate between 45-110 micrometers , and 110-210 micrometers , resp., in 1st and 2nd order. Multiplexing takes place both spectrally and spatially. An image slicer redistributes 5 by 5 pixel fields-of-view along the 1 by 25 pixel entrance slits of the spectrometers. Anamorphic collimator mirrors help keep the spectrometer compact in the cross-dispersion direction. The spectrally dispersed images of the flits are anamorphically projected onto the detector arrays, to independently match spectral and spatial resolution to detector size. We will thus be able to instantaneously cover a velocity range of approximately 1500 km/s around a selected FIR spectral line, for each of the 25 spatial pixels. For calibration and flatfielding we use blackbody calibrators internal to the instrument, at signal levels comparable to the thermal background of the telescope. An image rotator compensates field rotation during long integrations. Estimated sensitivity of the spectrometer is approximately 2 by 10^-15W/(root)Hz/pixel.

FIRST and SOFIA are both future IR observatories with 3m class main mirrors having sophisticated instrumentation aboard. The present design of the FIRST imaging spectrometer PACS requires two large far-IR photoconductor arrays of 25 X 16 pixels each, the baseline material is stressed and unstressed Ge:Ga. A gallium arsenide photoconductive detector which is sensitive in the far IR (FIR) wavelength range from about 60 micrometers to 300 micrometers might offer the advantage of extending considerably the long wavelength cut- off of presently available photodetectors. FIRGA is an ESA sponsored detector development program on this matter involving international partners. The aim is a monolithic 4 X 32 demonstrator array module with associated cryogenic read-out electronics. Recent progress in material research has led to the production of Te-doped n-type GaAs layers using liquid phase epitaxy. We prepared sample detectors from those material and investigated their electrical and IR characteristics. First measurements indicate that GaAs has in principle considerable potential as a FIR photon detector. Theoretical modeling of GaAs detectors can help with the detector design and allows the prediction of response transients as a function of detector parameters. Present development activities are mainly concentration on material research, i.e. the production of GaAs:Te with improved FIR characteristics. Results of the current test and measurements are reported. The FIRGA study is intended to prepare the technology for large 2D GaAs detector arrays for far IR astronomy.

The powerful tools of integral field spectroscopy and adaptive optics have made great contributions to the progress in astronomy in recent years. The combined use of these techniques now enables spectroscopy in the near IR close to the telescope diffraction limit. This will provide new and interesting insights into a variety of objects such as AGNs, QSOs, circumstellar disks around highly extincted YSOs, etc. Spectroscopy at or close to the telescope diffraction limit has some caveats which one has to be aware of when designing the instruments so as to maintain the maximum possible throughput and to optimize spectral resolution. Astronomical campaigns with our H- and K-band integral field array spectrograph 3D in combination with the laser guide star adaptive optics system ALFA at the 3.5-m telescope at Calar Alto require special observational techniques in order to make the most efficient use of the observing time available. Chopping by moving the telescope to do background subtraction makes it necessary to relock the A.O. system on the guide star after moving the telescope back to source. This procedure is usually rather time consuming. The aperture interchange module (AIM), which we present here, enables us to perform chopping between source and black sky while keeping the telescope fixed at a certain point in the sky. For this purpose AIM uses two different optical channels. The ON channel always points to the center of the 3'ALFA FOV, picking off a FOV of roughly 4 inch by 4 inch. With the OFF channel one can choose any off-center position within the ALFA FOV except a central obscuration of 36 inch diameter. The AIM optics are designed in such a way that the optical pathlengths for the on- and off-axis positions are kept equal. AIM also includes a scale changer which magnifies the scale from 0.25 inch/pix to 0.07 inch pix. The 3D spectrometer itself is equipped with two interchangeable grisms, so that one can choose between H- and K-bands and between spectral resolutions of 1100 and 2100. The commissioning run of AIM together with 3D and ALFA took place in July 1997 at the 3.5m Calar Alto telescope.

A concept of a sensitive cryogenically cooled far IR detector is presented. The device consists of a photosensitive Ge:GA element with one blocking and one injecting contact. Operation of the structure as a photodetector involves two steps. The measurements of the signal occurs when the reverse polarity of the bias voltage applied to the device and the registered photocurrent is proportional to the photon influx. Since the blocking contact does not supply majority carriers into the device, the photocurrent consists only of the carriers generated by IR photons from the acceptor impurities. When the space charge accumulated on the acceptors reaches the value sufficient for significant variation of the electric field in the bulk the device is reset by applying the forward polarity of the bias, which restores the initial charge state of the impurities. Device model and experimental data illustrating properties of the structure at moderately low photon fluxes are also discussed.

The inherent non-linearity in the photometric response of NICMOS-3 HgCdTe arrays can lead to photometric calibration errors of 1-2 percent or more when using standard flattening and calibration techniques involving single flat-field exposures and mid-level exposures of standard sources for calibration. Furthermore, the useful dynamic range of the device must be restricted to achieve even this precision, since deviations from linearity are greatest at higher exposure levels. We describe a technique to digitally linearize NICMOS-3 images by interpolating within the characteristic response curve for each pixel as determined from a series of exposures of a flat screen. Using astronomical observations, we quantify the improvements that can be achieved in final photometric precision and useful dynamic range.

Low-noise and low-power cryogenic readout electronics are developed for a focal plane instrument of the IR Imaging Surveyor. We measured the static characteristics and the noise spectra of several types of silicon MOSFETs at the cryogenic temperature where silicon JFETs do not work well due to the carrier freeze-out. The 'kink' behavior of n- channel MOSFETs was observed below the carrier freeze-out temperature, but it was not obvious for the p-channel MOSFET. It was demonstrated the p-channel MOSFETs can be used for the cryogenic readout electronics of the IRIS's far-IR array with an acceptable performance. The amplifier integrated with these MOSFETs showed low-noise at 2K under a low power consumption of 1 (mu) W per MOSFET. We now design and evaluate several circuits that are fabricated by the CMOS process for cryogenic readout.

GaAs JFET with various gate sizes, which ranges from 0.5 to 200 micrometers in gate length and from 2 to 200 micrometers in gate width, are fabricated and their DC characteristics and low- frequency noise spectra are measured at low temperatures in order to develop cryogenic electronic circuits for a far-IR detector array on board a satellite such as the Japanese IR satellite. We obtained the following results from the noise measurements: (1) Noise spectra of GaAs JFETs are dominated by a 1/f noise and include some generation-recombination noises in low-frequencies. (2) The 1/f noise voltage is found to remarkably depend on both the gate length and the drain-source voltage, but the gate width and the gate-source voltage have not almost concern with the 1/f noise voltage. Therefore, we suppose that the electric field in the channel of the GaAs JFET mainly contributes to the 1/f noise. By using these characteristics, the GaAs JFET having very low power dissipation and very low noise will be designed for cryogenic readout circuits at low temperatures.

We are developing a stressed Ge:Ga 2D array detector that will be used for balloon-borne and satellite-born astronomical observations at wavelength between 100 and 200 micrometers . We have succeeded in making a 4 X 8 element stressed array detector with a stress of 600 N/mm² and responsivity peak wavelength moved to about 165 micrometers . This has the largest number of pixels at the present time. The responsivities of the detector are high enough as well as those currently in use. This detector has a compact structure and a small total pixel size, and thus, it can be used for satellite-born instruments that have severe space limitation.

SubWavelength Structured Surfaces (SWS), by synthesizing effective index of refraction, offer an attractive way to mimic antireflective coating effects. It is of particular interest for some IR materials of high index of refraction such as CdTe or KRS-5. These material are often used for entrance window in cryogenic IR instrument in the 20 microns band. For these materials, multilayer antireflective coatings provide limited performances in transmission, while expected performances of SWS can be very high even for a wavelength range covering both the N and Q atmospheric windows, from 7 microns to 28 microns. The SWS simulates a gradient index layer. Its main parameters are its pitch and its depth. The pitches required depend on the IR material index. For CdTe and KRS5, they are around 3 microns to work in N-band and Q-band and around 6 microns to work only on Q- band, and the depth required is around 10 microns to work till 28 microns. We have tried a new approach to realize these structures by using excimer laser ablation technique. We describe the used technique and our results for different materials such as CdTe, KRS5, CsBr and CsI. Antireflection structured surfaces on CdTe could offer an increase in transmission better than 25 percent at 24 microns. We measured a transmission efficiency of near 96 percent between 23 micrometers and 35 micrometers on KRS-5, and more than 95 percent between 18.5 micrometers and 35.5 micrometers on CsI.

Deep cryogenic field effect transistors (FETs) which are able to operate under liquid helium temperatures have significant advantages over conventional cryogenic Silicon- Junction-FETs or Si-metal-oxide-semiconductor-FETs as readout circuits of a far-IR focal plane array detector: simple operation, simple system structures, and large transconductance. We report the testing of an InGaAs-channel heterojunction field effect transistor (HJFET) operating at 4.2 K designed for a readout circuit of a cryogenically cooled far-IR detector. In this report, we present current- voltage characteristics, transconductance, low-frequency noise (LFN) characteristics, and the influence of the gate leakage current on the LFN characteristics of the HJFET. Input-referred noise voltage as low as a few hundred nanovolts at 1 Hz was measured for the HJFET with a 100 X 100 micrometers ² gate area. We discuss further possibilities for the fabrication of HJFETs with an extremely small input current of less than 10^-15 A.

We discuss the design and performance of a solid KRS-5 grism used in NSFCAM, a 1-5.5 micrometers facility IR camera at the NASA IR telescope Facility on Mauna Kea. The grism was built by Carl Zeiss, Jena-Germany, and cost 13,800 dollars. It is used with order-sorting filters in the L-, K- and H-bands, and provides a spectral resolution of R equals 280. We also discus the advantages and disadvantages of solid grisms over replica resin grisms, and illustrate the performance of the grism with some astronomical observations.

We have designed a near IR spectrograph, sensitive in the 1.5-5 micrometers range, that uses a silicon immersion echelle grating. The cross-dispersed design demonstrates that immersion echelles allow compact spectrographs which have excellent spectral coverage and very high resolving power. Our instrument will have continuous spectral coverage over a 5.7 percent passband at 2.3 micrometers or a 7.6 percent passband at 4.6 micrometers and resolving power ranging from R equals 87,000 at 4.6 micrometers to 109,000 at 2.3 micrometers . We discuss design issues that are unique to spectrographs using silicon immersion echelle gratings in the near IR such as grating parameters, geometry, and the mechanical and thermal properties of large pieces of single crystal silicon.

The optimal operating temperature range for Indium Antimonide detectors is typically near 35 Kelvin. Commercially available miniature split-Stirling cycle cryocoolers present an attractive approach to detector cooling. These units offer stand-alone operation, small size, light weight, low power input, low vibration, moderate cost, and reasonable lifetime. However, currently available units have inadequate cooling capacity at 35 Kelvin when operated in a normal manner. We have substantially increased the low temperature cooling capacity of commercial cryocoolers by utilizing the 77 Kelvin intermediate temperature available in liquid nitrogen cooled instruments. We thermally connect the liquid nitrogen cold sink to the middle of the cryocooler coldfinger, shunting heat from the coldfinger to the LN2. The resulting performance improvements and careful thermal design of the detector mount to minimize parasitic heat loads a low miniature split-Stirling cycle cryocoolers to provide adequate cooling of large format Indium Antimonide focal plane arrays.

To built a 3K X 3K pixel near-IR FPA, we have made a package and a multi-chip module for Mitsubishi 1040 X 1040 PtSi CSD, which is one of the largest SWIR FPAs. Mosaicing demands smallest gaps between chips to achieve a large fill-factor and controlled flatness to fit a camera focal plane. The package of 52-pin half-pitch PGA has been designed to be smaller than the bear chip. After the chip is glued on the package and wire-bonded, nine packages with the chip are arrayed in three by three on a multi chip module (MCM) of 6 cm X 6 cm area. The fill-factor of the imaging area is 89 percent. The package and MCM are made of AlN ceramic of high thermal conductivity. MCM, therefore, plays a role of an efficient heat sink. The surface of the package, with which the chip is in contact, has been polished with accurate flatness as well as MCM. As the result, the height of nine chips built on MCM are uniform within approximately 20 micrometers in 6 cm X 6 cm area. The mosaic array will be equipped in a near-IR camera for astronomical observations of a wide field view.

The IR camera and spectrograph (IRCS) a facility instrument for the 8.2m Subaru Telescope is being built at the University of Hawaii, Institute for Astronomy. IRCS will use a 1024 X 1024 InSb array for spectroscopy and another 1024 X 1024 InSb array for IR imaging. In a collaborative effort with the team members of SpX², a test system has been fabricated for joint testing of 1024 X 1024 InSb ALADDIN II based arrays. This document is a preliminary report on the test results of the science grade array provided by the Subaru Telescope Project. It is a possible candidate for inclusion in IRCS.

The IR Camera and Spectrograph for the Subaru telescope uses a series of reflective and transmissive slits. The width of the slits ranges from 48 micrometers to 440 micrometers . The requirements for both types of slits include sharp edge definition, good surface figure at cryogenic temperature and high reflectivity. Several different substrate materials and fabrication methods were investigated. The substrate materials considered include aluminum, copper, tungsten carbide, chromium carbide, and sapphire. The fabrication methods investigated include photo-etching micro machining using UV laser, electroforming, diamond turning, conventional polishing and electrical-discharge-machining. The pros and cons of each material and fabrication method will be described.

Stressed Ge:Ga is currently the most suitable detector type for very low background operation in the 115 to 200 micrometers range. Nonetheless, substantial advances have been required to develop stressed Ge:Ga detectors that work at the background limit in SIRTF. Both dark current and read noise have been improved significantly for the SIRTF devices. The design also takes account of space flight requirements such as the necessity to anneal the focal plane thermally using a minimum of cryogenic power dissipation, and the desire that any failures not propagate through an entire focal plane. The SIRTF 2 X 20 pixel focal plane will have dark current of about 200 e/s, read noise of 100 e rms, and responsivity > 7 A/W. As a result, even in the darkest parts of the sky, it will reach the background limit in less than 4 seconds of integration.

Despite the advances in digital signal processing, the analog circuits which amplify and define the bandwidth of the low level signals from CCD and IR array detectors are a critical element in obtaining the best possible signal-to- noise ratio. The choice of components and topology for these front-end circuits is discussed, including the effects of input voltage and current noise density, coupling strategies, shield driving, physical layout, and grounding. The signal chain use in Imaging Sciences Laboratory instruments is presented as an interpretation of these considerations.

A low cost tip-tilt wavefront stabilization system has been put into operation on the Blanco 4-m telescope on Cerro Tololo. A light-weighted f/15 secondary mirror is driven by three commercial piezoelectric actuators. A dichroic at the Cassegrain focus separates optical reference and IR science beams. A steerable high-speed optical CCD sensor, coupled to a dedicated PC for control and image processing, provides positional feedback to the secondary. The IR field is reflected to one of several science sensors. We present a system description, initial performance measures at the telescope, and directions for future improvements.

The long wavelength spectrometer on board the ESA IR Space Observatory employs doped geranium photoconductors to perform spectroscopy in the 43 to 197 micrometers waveband. The instrument has been in continuous on orbit operation for over two years - longer than any other experiment in this waveband. Invaluable data have been gathered on the long term performance of a beryllium doped germanium detector and both stressed and unstressed gallium doped germanium detectors in the presence of ionizing radiation in the form of cosmic rays and charged particles trapped in the Earth's magnetic fields. In this paper we report on the in-orbit performance of the detectors and in particular on the long term behavior of the dark current and responsivity.

Silicon diffraction gratings fabricated by photolithography and anisotropic chemical etching offer unique advantages over conventional ruled gratings. Previous work showed a high amount of scatter from etched gratings and several waves of wavefront distortion due to an overetch problem. We recently began to study ways to improve the optical performance of etched silicon gratings. Defect etching was used to verify that our starting material was of high quality and that it could be polished to the required optical flatness without producing subsurface damage. We compared the effect of different etch mask materials and etching variables on the scatter and wavefront aberration of test gratings. Gratings masked with silicon nitride achieved sufficiently low wavefront distortion to be useful in high resolution spectroscopy. The scatter was improved by a factor of tow over the grating fabricated several years earlier, but is still a factor of 3 worse thana ruled echelle. A better lithography mask should reduce the scatter. The major issues in fabricating large etched gratings for astronomy are achieving good uniformity across the surface and the compatibility of thick silicon disks with standard semiconductor processing hardware.

This paper presents an overview of the electronics multi- array control system design that can control several different types of focal plane arrays simultaneously. This system is used with the SUBARU standard data acquisition system MESIA. MACS2 and MESSIA materialize a wide wavelength observation at visible and near IR wavelengths that requires different types of arrays. MACS2 consists of four types of cards. An isolation card is required for one imaging system. A clock driver card and an ADC card are required for an array, and a preamp card is required for an ADC card. Each card is daisy-chained through differential signals. Every array does not have to be placed closely, and no more signal lines are required even when controlling more than one array and type. The bias voltages to operate arrays and the offset voltage at the analog input can be controlled and monitored form a host workstation. We can arrange various environments to evaluate focal plane arrays without any modifications of printed circuit boards or any wiring. MACS2 is very useful and powerful for evaluating different types of arrays. Also we could save time to swap a spare card when a card for whichever detector is broken, and the maintenance of a recent complex imaging system becomes easier. MACS2 will be installed in TRISPEC, two InSb, WFCT, SIRIUS.

Abstract not available.

ABEL is currently at the beginning of the design phase at the Instituto de Astrofisica de Canaris. The instrument will be equipped with the 256 X 256 Santa Barbara Research Corporation InSb FPA which will provide a working spectral range from 1 to 5 microns. For image mode three different platescales are envisaged: 0.2 inch/pixel, to be used in the thermal IR to avoid detector saturation; 0.4 inch/pixel, which will allow for sufficient sampling of the median seeing limited images below 2.5 microns; and 1.0 inch/pixel, which will be the standard in spectroscopic operations and during wide field imaging. For spectroscopy, a standard moderate spectral resolution of about 400 will be available in the JHKLM windows, which will be all fully covered in a single exposure. Additional higher spectral resolution is under consideration, which at least double. ABEL will offer a wide variety of slit widths and shapes, ranging from 1 inch to 3 inches, and including dog-leg shape. The thermal design is based on a two stages closed cycle cooler, the first stage being used for the passive optics while the second will cool directly the detector to about 30 to 40 K. The instrument is planned for the late 99 and a major cooperation with the Osservatorio de Arcetri is underway. ABEL will be installed in the f/13.8 Cassegrain focus of the 1.5m Telescopio Carlos Sanchez, at the Spanish Observatorio de El Teide, in the canarian island of Tenerife.

MIRAC2 was built for ground-based astronomy at Steward Observatory, University of Arizona and Harvard-Smithsonian Center for Astrophysics. It utilizes a Rockwell HF-16 128 X 128 arsenic-doped silicon blocked-impurity-band hybrid array with a wavelength range of 2 to 28 micrometers operating in a liquid helium-cooled cryostat at 5K. Reflective optics, and externally actuated detector and pupil slides provide a variety of magnification and focal ratio settings without opening the cryostat. Nominal settings at the NASA IRTF and UKIRT give diffraction-limited imaging with .34 and .27 arcsec/pixel, respectively. The sensitivity on the IRTF at 11.7 micrometers , 10 percent bandwidth filter, chop-nod, source in one beam, 1 sigma, one minute total time is 25 mJ/arcsec surface brightness and 43 mJy point source.

CISCO is an IR camera and spectrograph based on a single 1024 X 1024 HgCdTe array detector, which has been developed as a back-end spectrograph of OHS. It is also designed to be mounted on the Cassegrain or Nasmyth focus directly as an independent instrument. In addition to the normal imaging and spectroscopy modes, CISCO has a slitless prism spectroscopy mode at resolving power of approximately 30. This mode is primarily aimed at detecting the H(alpha) emission line of forming galaxy at z equals 2.05-2.65. The development of CISCO is in near completion, showing results of test observations carried out using a 1.5m telescope.

COHSI was successfully commissioned at the United Kingdom IR Telescope on Mauna Kea during a seven night observing run which coincided with this conference. Here we briefly describe the instrument and give a preliminary report on its performance at this time. The suppression optics and masks worked extremely well and the instrument background was found to be very low.

With the increasing number of adaptive optics (AO) systems being developed for major observatories, it is important to understand the requirements and the capabilities of the IR instrumentation that will utilize these systems. During the past two years, we have constructed and upgraded an IR camera system based on the NICMOS 3 IR array for the Chicago Adaptive Optics System (ChAOS). ChAOS is a high order adaptive optics system built by Edward Kibblewhite and his research group for use at the ARC 3.5 Meter Telescope. The IR camera has been used with the AO system during several observing runs and achieved near diffraction limited images form 1 to 2.3 microns. We present these early imaging result and discuss several of the unique problems faced by this camera. We will also discuss the potential capabilities of AO instruments and discuss our observing plans for the near future. Finally we describe our ongoing development of an IR spectrograph for the system.

We describe the configuration and operation modes of the IR camera/spectrograph: TEQUILA based on a 1024 X 1024 HgCdTe FPA. The optical system will allow three possible modes of operation: direct imaging, low and medium resolution spectroscopy and polarimetry. The basic system is being designed to consist of the following: 1) A LN₂ dewar that allocates the FPA together with the preamplifiers and a 24 filter position cylinder. 2) Control and readout electronics based on DSP modules linked to a workstation through fiber optics. 3) An opto-mechanical assembly cooled to -30 degrees that provides an efficient operation of the instrument in its various modes. 4) A control module for the moving parts of the instrument. The opto-mechanical assembly will have the necessary provision to install a scanning Fabry-Perot interferometer and an adaptive optics correction system. The final image acquisition and control of the whole instrument is carried out in a workstation to provide the observer with a friendly environment. The system will operate at the 2.1 m telescope at the Observatorio Astronomico Nacional in San Pedro Martir, B.C. (Mexico), and is intended to be a first-light instrument for the new 7.8m Mexican IR-Optical Telescope.

LEWIS is an IR spectrograph designed primarily for spectroscopy in the 3 micrometers region. It is an echelle type spectrograph using a coarse groove grating together with a prism as a cross-disperser. Using LEWIS, we can observe the whole L-band in one exposure with a resolving power over 1250, which makes observations very efficient. A Santa Barbara Research Center 256 X 256 InSb array is employed as a detector. The grating used is characterized by large groove spacing of 125 micrometers and is utilized at very high orders, 25th-37th order in the L-band. A closed-cycle cooler is employed to keep the optics at 90K, and to maintain the detector at 30K. So far, scientific observations have been made at the Steward Observatory 60 inch telescope on Mt. Lemmon, the Steward Observatory 61 inch telescope on Mt. Bigelow, and the Wyoming IR Observatory 88 inch telescope on Mt. Jelm. The achieved throughput of the spectrograph including the quantum efficiency of the detector is about 20 percent. With the present detector control system, observations are background limited at 3.5 micrometers using multiple correlated sampling, and a limiting magnitude of 8.2 mag is achieved for S/N equals 20 in 30 min integration time with 1.5m class telescopes.

Two imaging systems for 1 meter telescope with f/ratios of f/10 and f/16 capable of simultaneous operation in any one of the U, B, V or R passbands together with any one of the J, H or K passbands are described. The visible bands are separated from the IR bands by a dichroic beam-splitter. Identical fields are imaged on a 2048 by 2048 CCD with 15 micron pixels, and an InSb 1024 by 1024 array with 27 micron pixels at a scale of 0.3 arc-seconds per pixel. The system uses reflective optics prior to the dichroic followed by refractive optics to produce the final f/ratios. The astigmatism introduce into the IR transmitted beam by the dichroic substrate is corrected using a cylindrical mirror. An additional prismatic correction necessitated by the dispersion in the beam splitter is also addressed.

CAMIRAS, the Saclay ground-based mid-IR camera, was built in 1990. At the time it was equipped with a 64 X 64 pixels Si:Ga array, developed by LETI/LIR, as a by-product of the detectors for ISOCAM. CAMIRAS has been intensively used at CFHT and at the Nordic Optical Telescope since 1991. In 1995, LETI/LIR delivered 192 X 192 Si:Ga arrays. Such a detector has been integrated in CAMIRAS in March 1995, and its acquisition system has been upgraded. The instrument is very versatile; several configurations allow to optimize it either at 10 microns or at 17 microns. CAMIRAS was mounted with the new 192 X 128 detector array at CFHT in July 1996. THis paper gives a full instrumental description of CAMIRAS. A particular attention is given to the spectro- imaging capability of the instrument achieved with a circular variable filter. Results from the last 3 observing campaigns carried out at the TIRGO observatory and at the Nordic Optical Telescope illustrate this paper. Spectra of comet Hale-Bopp showing the silicates features in the 8-13 microns wavelength range are presented.

Operating in the wavelength range 5-28 micrometers , the mid-IR array eXpandable (MAX) camera contains a 128 X 128 Si:As Blocked Impurity Band detector tailored for high background, low noise performance. The term 'expandable' refers to our ability to substitute one of the newer generation 256 X 256 arrays, when they become available. The all-reflective optics provide diffraction limited performance, and a large, interchangeable filter set allows both broad and narrow band imagery. MAX was built to our specifications by IR Laboratories of Tucson, and it achieved first light on UKIRT in November 1995. In this paper, we present the design and real-world performance of MAX, together with a sample of the scientific results already published.

This paper describes a near IR imaging spectrograph designed for very high throughput, many spatial pixels, and a wide field of view. Each spectrum contains approximately twenty spectral channels. This balance spatial and spectral coverage is driven by limited detector size and represents an excellent match to many astrophysical problems. We achieve the high efficiency and wide field of view by placing an array of microlenses in the focal plane wheel of a pre-existing camera/spectrograph. A narrow band filter prevents overlap of the individual traces. Ray tracing and preliminary test at the telescope demonstrate the workability of the concept. Initial observations with the imaging spectrograph should take place son after this conference.

Spectroscopic studies of degree-scale IR line emission from telescopes on the Earth's surface are very challenging. Not only does the emission from atmospheric OH molecules fluctuate in time and vary spatially, but also the extent of interstellar clouds makes it impossible to establish a zero point for flux measurements close to the emission regions. Beam-switching can solve both problems, but traditional telescopes cannot switch between fields that are several degrees apart before the OH emission changes significantly. We present here the design for a telescope we have constructed in order to obtain quantitative measures of the extended, UV-excited, near-IR line flux from the Galactic Center and from nearby star-forming clouds. It uses a coelostat to beam-switch across angels as large as 10 degrees in less than 1 second. Its 20 arc-minute beamsize, combined with its 150 mm aperture, gives it a surface brightness sensitivity comparable to much larger telescopes, while maintaining portability. This telescope, integrated with a Fabry-Perot spectrometer, has been used successfully at both McDonald Observatory and the Cerro Tololo Interamerican Observatory and has mapped near-IR molecular hydrogen emission over the inner 10 degrees of the Galaxy for the first time.

Large astronomical spectrographs designed for use in the visible for use in the visible can operate efficiently well beyond the long wavelength cutoff of CCD detectors. Given the expense and complexity of constructing IR-optimized high resolution or multi-object spectrographs, it is prudent to explore the range of scientific programs possible utilizing modern near-IR arrays at the focal plane of historically visible wavelength instruments. For the past three years, we have used the NICMASS camera, a 256 by 256 HgCdTe imager developed at the University of Massachusetts, at the camera 5 focus of the Coude Feed Spectrograph on Kitt Peak for moderate and high resolution IR spectroscopy in the 1-1.8 micrometers range. This configuration has been used at a spectral resolution 7200 using a 316 1/mm grating an extremely stable platform permitting radial velocity determinations to better than 1 km-s ^-1. We will discuss some scientific results obtained with this novel configuration and the performance limitations imposed by the ambient temperature spectrograph beyond a wavelength of 1 micrometers . We also discuss plans to evaluate the suitability of NICMASS for multi- object near-IR spectroscopy on the Hydra Bench Spectrograph at the WIYN telescope on Kitt Peak.

ARIES, a new 1-5 micrometers camera/spectrograph, is designed to capitalize on the exceptionally low thermal background and high optical throughput offered by the f/15 adaptive secondary system being built for the upgraded 6.5m MMT. With two state-of-the-art IR arrays, ARIES will provide diffraction-limited imaging in the JHKLM atmospheric windows and also echelle, long-slit spectroscopy at resolutions of 2,000 and 30,000. ARIES will also supply global wavefront tip/tilt information to the adaptive system using cryogenic pick-off mirrors to access field stars over a 50 arcsec diameter field at wavelengths from 1-2 micrometers .

LISA is the low resolution imaging spectrometer channel for SPICA. the proposed spectral-photometric-IR camera for the joint US and German stratospheric far IR observatory SOFIA. LISA will cover a spectral range from 40 micrometers to 220 micrometers at a spectral resolution (lambda) /(Delta) (lambda) approximately 22. It will record simultaneously individual diffraction limited spectra from several tens to several hundreds spatial positions of a rectangular field of view on the sky. A novel technical and optical solution will be presented for such an approach together with a brief discussion on the expected performance and possible astronomical applications.

KIR is a 1024 by 1024 near-IR camera used with the adaptive optics Bonnette (PUEO) of the Canada-France-Hawaii Telescope. The camera houses a 1024 by 1024 HgCdTe and simple refractive optics providing diffraction-limited images with an image scale of 0.035 inch/pixel. First light was obtained in December 1997. The throughput of the camera, from the top of the atmosphere down to the atmosphere down to the detector including PUEO, is 19 percent, 20 percent and 21 percent at J, H and K, respectively. This project is a collaboration between the Universite de Montreal, the Observatoire Midi Pyrenees and the Canada-France-Hawaii Telescope. The design and performance of the instrument are presented in this paper.

We present the optical, mechanical and electronic design of MAGNUM-MIP. The MAGNUM project plans to carry out multi color monitoring observations for hundreds of AGNs over several years under remote and automated operation. MAGNUM- MIP has two channels that offer optical and IR broad-band imaging observations at the same time. The IR channel has a SBRC InSb 256 by 256 array which covers a wavelength range from 1 to 4 microns, and the optical channel uses a 1024 by 1024 SITE CCD which covers 0.35 micron to 1 micron. The two channels use the same optics and a beam splitter. We adopted a reflecting optical system in order to get good imaging quality over the wide wavelength range. Because the monitoring is expected to be carried out remotely for several years with minimum manual support and maintenance, the camera is designed to work with only semi-annual maintenance. It has a mechanical cooler, a low outgas design, and an automated vacuum system.

The ISL is a successful astronomical instrumentation program that has completed three major instruments and many smaller projects since 1987. We have developed the capabilities to perform all aspects of instrument design and construction and a range of unique skills and methods. We maintain a permanent staff that currently consists of two scientists specializing in optical design and detector systems, a seniors mechanical engineer, a programmer, an electronic engineer, a mechanical designer, two machinists, and a lab assistant. Instrumentation projects also draw upon faculty and graduate student effort.

Imaging planets, brown dwarfs and disks around nearby stars is a challenging endeavor due to the required scene contrast. Success requires imaging down to m equals 20-25 within arcseconds of stars that are 4th-6th magnitude. Light scattered and diffracted from a variety of sources increases the background flux in the area of interest by orders of magnitude masking the target objects. As first shown by M. B. Lyot in 1939 masks can be placed in the focal pane and pupil planes of a camera to occult the bright central source making it possible to image the faint extensions around it. CoCo is an experiment in using a coronagraphic camera, for IR observations, on a large telescope in an effort to understand how a coronagraph can help and how to properly design one of the new generation of large telescopes. Recent result with CoCo show a factor of 5-10 reduction in background levels in the area from 2-7 arcseconds from the central object. This paper will describe those result and summarize what has been learned towards building coronagraphic cameras for today's large telescopes.

We are developing a high-resolution cross-dispersed echelle spectrograph for installation at one of the coude foci of the new AEOS 3.67 meter telescope, operated by the Air Force Space Command on Haleakala, Maui, Hawaii. The spectrograph will consist of two major subsystems, an optical arm for the wavelength range 0.5-1.0 microns and a SWIR arm for the range 1.0-2.5 microns. The optical arm will include a mosaic 4096 by 4096 thinned CCD array, providing coverage of the wavelength range in two settings at a resolving power of 50,000. The CCD camera will be operated in frame-transfer mode. The IR arm will consist of a compact, folded cross- dispersed cryogenic echelle spectrography. The SWIR detector will be a 2048 by 2048 HgCdTe array, based on the existing HAWAII 1024 by 1024 devices. The large-format detector will permit coverage of the entire J or H band in a single grating setting with a resolving power of 60,000, and the K band in two settings. The high resolution, coupled with careful attention to scattering and stray light in the optical system, will permit exploitation of the low sky background between the strong OH airflow lines. Adequate order separation will be maintained to permit work on moderately extended objects while still retaining sky subtraction capability. The spectrography is expected to be available for use in early 2000.

SCORE is a cross-dispersed echelle spectrograph, built as a prototype for the Short-High module of SIRTF's IRS instrument. It operates over the 7.5-15 micrometers N-band atmospheric window, and has ben used on Palomar's Hale telescope several times since November, 1996. Since the initial run, a number of improvements have ben undertaken or are in the process being undertaken which enhance SCORE's performance and simplify its operation. One such addition, now completed, is a second detector array which serves as a slit-viewer with 12 inch diameter field of view around the slit. This viewer allows easy acquisition and guidance for sources with dim or absent optical counterparts, and accurately registers the position of the slit on the source with the recorded spectrum. Software written in the IDL environment optimizes the extraction of spectra form SCORE's mid-IR crossed-echelle data. The echelle, while providing the advantage of increased pixel utilization, introduces several difficulties, including curved orders, order cross- talk, and differentially slanted lines. These and other instrumental artifacts must be removed to achieve the highest spectral signal-to-noise. The pixel efficiency will be further increased by the use of a grism predisperser. The grism will provide approximately even spacing between orders of the echelle, in contrast with the decreasing spacing towards shorter wavelength orders generated by the current grating. SCORE is already one of the most powerful short- slit spectrographs operating in this wavelength band, and, with the implementation of these improvements, will deliver even greater capability.

We describe a cryogenic, high-resolution spectrograph (Phoenix) for the 1-5 micrometers region. Phoenix is an echelle spectrograph of the near-Littrow over-under configuration without cross dispersion. The foreoptics include Lyot re- imaging, discrete and circular variable order sorting filters, a selection of slits, and optics for post-slit and Lyot imaging. The entire instrument is cooled to 50 K using two closed cycle coolers. The detector is a Hughes-Santa Barbara 512 X 1024 InSb array. Resolution of 65,000 has been obtained. Throughput without slit losses is 13 percent at 2.3 micrometers . Recent results are discussed. Phoenix is a facility instrument of the National Optical Astronomy Observatories and will be available at CTIO, KPNO, and Gemini.

The design concepts of a mid-IR camera/spectrograph, based on a HUGHES/SBRC 320 by 240 Si:As IBC sensor chip assembly (SCA), are presented, The system will operate in the 2 to 28 micrometers wavelength interval and will be optimized in the 10 micrometers regime. This SCA is divided into 32 regions, each with an independent output. The outputs, after being amplified and sampled, are multiplexed into 8 high speed 16 bit A/D converters. The initial configuration allows readout rates of up to 60 frame/s. A higher speed frame readout configuration is foreseen. A 32 bit deep memory and a high speed ALU, synchronized to the detector, will co- add/subtract the frames, making a real time visualization during the integration process possible. The detector, reflective optics, low resolution gratings, several cold stops, baffles, up to 12 filters and a CVF will be allocated in a 10 inch diameter working surface LN₂/LHe dewar. The system will be linked to a workstation, providing a user friendly environment. The system is planned to operate at the Observatorio Astronomic Nacional in San Pedro Martir, B.C., Mexico.

Omega Prime is a wide-field near-IR camera for the prime focus of the Calar Alto 3.5 m telescope in Spain. The detector is a 1024 X 1024 pixel HAWAII array made by Rockwell. The image scale is 0.4 arcsec/pixel, giving a field of view of 6.8 by 6.8 arcmin. In order to maximize the throughput, the optics were designed as a prime focus corrector with only three lenses. This simple design without a cold pupil provides an excellent image quality over the entire field of view. To reduce thermal background at wavelengths longer than 2.2 micrometers , Omega Prime has a series of cold internal baffles and an additional torodial mirror outside the dewar. This annular reflector causes detector pixels to 'see' mostly the cold interior of the camera. The camera has been in operation since May 1996 and has been used for a variety of scientific programs. Including a very deep K survey covering 1000 square arcmin to a 5 (sigma) limit for point-sources of 20.5 magnitude.

Polarimeters at optical and near-IR wavelengths are increasingly available as part of facility instruments at major observatories, and are used for a large number of astronomical programs, ranging form nearby star-forming regions to high-redshift galaxies. Polarimetry is used in both imaging and spectroscopic modes and at both low and high spectral resolutions. As degrees of polarization are usually low a large collecting area is needed to get the high signal to noise required for accurate polarimetry. Thus polarimetry can take particular advantage of the new generation 8m telescopes such as Gemini. Techniques for obtaining high precision measurements used for IR polarimeters on UKIRT and on the AAT, together with the performance achieved for both imagers and spectrometers are presented. The implementation of the same techniques proposed for Gemini instruments is described.

Measurements were performed to verify the straylight suppression in the IR Space Observatory telescope using the ISOPHOT instrument. These test comprised the near-field straylight by bright stars and planets as well as the far- field straylight by the Sun, the Earth, and the Moon. No significant straylight above the specifications reflecting the astronomical needs for low surface brightness absolute sky measurements could be detected at 25 and 170 micrometers . In some cases comparison to preflight straylight simultaneous were possible. The consistency of the predictions from the models with the measurement results confirms the reasonable assumptions made for the simulations. This will allow to further optimize the telescope design for future low background IR space telescopes.

We present the wavefront error budget and optomechanical tolerance analysis for NIRSPEC, a high-resolution near-IR echelle spectrograph for the Keck II telescope. The error budget accounts for aberrations induced by optical design residual, manufacturing error, cryogenic degradation, mounting effects, and misalignments. The allowed errors due to misalignments are used with boresighting and vignetting requirements in an optomechanical tolerance analysis.

We describe the optical alignment and image quality testing of Michelle, the all-reflective mid-IR astronomical spectrometer and imager being built at the Royal Observatory Edinburgh for the UKIRT and GEMINI telescopes. The design strategy called for optical alignment by manufacture, with the only means for adjustment being the machining of sacrificial pads under key optical components. The success of this approach in meeting the alignment error budget is discussed, including the description of a method for identifying the optical axis of the optical train using field rotation. We present the result of image spot and wavefront error measurements and compare them with the instruments opto-mechanical specification.

The Space IR Telescope Facility (SIRTF) contains three focal plane instruments, one of which is the IR Array Camera (IRAC). IRAC is a four-channel camera that provides simultaneous 5.12 X 5.12 arcmin images at 3.6, 4.5, 5.8 and 8 microns. The pixel size is 1.2 arcsec in all bands. Two adjacent fields of view in the SIRTF focal plane are viewed by the four channels in pairs. All four detector arrays in the camera are 256 by 256 pixels in size, with the two short wavelength channels using InSb and the two longer wavelength channels using Si:As IBS detectors. The IRAC sensitivities at 3.6, 4.5, 5.8, and 8.0 microns are 6, 7, 36, and 54 microJanskys, respectively. Two of the most important scientific objectives of IRAC will be to carry out surveys to study galaxy formation and evolution during the early stage of the Universe, and to search for brown dwarfs and superplanets.

The Photoconductor Array Camera and Spectrometer is one of the three proposed instruments for ESA's Far IR and Submillimeter Telescope. It employs two 16 by 25 pixels Ge:Ga photoconductor arrays to perform imaging photometry and imaging line spectroscopy in the 80-210 micrometers wavelength band, with optional extensions to 60 micrometers and approximately 300 micrometers . In photometry mode, it will simultaneously image two bands, 80-130 micrometers and 130-210 micrometers , over fields of view of approximately 1 foot by 1.5 feet and approximately 2 feet by 3 feet, respectively, with full beam sampling in each band. In spectroscopy mode, it will image a field of approximately 50 inches by 50 inches, resolved into 5 by 5 pixels, with an instantaneous spectral coverage of approximately 1500 km/s and a spectral resolution of approximately 175 km/s. In both modes background-noise limited performance is expected, with sensitivities of 3.5-5 mJy or 2.5 by 10^-18/m², respectively.

The anomalous motion of the near IR camera and multi-object spectrometer (NICMOS) detector arrays was originally discovered and characterized during ground optical testing, in a large, high fidelity Hubble Space Telescope (HST) simulator. To monitor the state of the cryo-mechanical system, as NICMOS traveled among several testing sties, a portable stimulus was needed. The cold-well displacement monitor (CDM) was quickly assembled from a very simple design. The 'cheaper, better, faster' approach proved to be a winner here. Off-the-shelf optics, a simplified interface to the instrument, and a limited set of requirements were used. After calibration against the large refractive aberration simulator/Hubble opto-mechanical simulator (RAS/HOMS), the CDM gave results of similar accuracy to RAS/HOMS. It became the primary tool for the difficult job of managing the NICMOS cryogen system up through launch.

The grating turret is one of many parts of the Gemini near IR spectrograph being designed for the Gemini 8m telescope project. It provides four gratings and one flat mirror mount, with each one able to tilt in discrete 1/4 degree steps over a full +/- 7.5 degree range while maintaining a 1/10 pixel or 2.7 micrometers position stability. There are distinct design features satisfying space, weight, and performance limitations. We use 2 motors, one to control the grating selection, and one for grating tilt. These motors are locate outside the vacuum/cryogenic environment for easy access. We employ a proven technique using rubbing plates to provide thermal contact for effective cooling. 3D modeling was used extensively to optimize weight and structural stiffness for the mechanism.

Reviewed is a focus stage designed to accommodate the positioning and stability requirements of the detector arrays in the Gemini North Telescope's Near IR imager (NIRI). Focus axis translation of the two detector arrays is required, while sub-micron deflection stiffness about all other axes is of paramount importance to the successful operation of NIRI. The stiffness requirement coupled with a cryogenic vacuum environment led to a flexure design. Testing of the prototype stage mechanism to date has shown transverse deflections of < 1 micrometers , positioning repeatability of 1 micrometers , and satisfactory cryogenic performance.

The cryogenic optical performance of an all aluminum three mirror anastigmat re-imager developed for the NIRSPEC instrument is reported. Details pertaining to the optical and mechanical design, structural/thermal modeling, initial performance projections, optical/mechanical fabrication, optical alignment, and optical testing procedures are all presented. The operational performance of the optical system at ambient and at cryogenic temperatures is presented and compared with initial performance projections. The optical subsystem has been delivered to UCLA for integration into the NIRSPEC instrument, final installation will be done at the Keck Observatory.

We present a high-resolution gimbal mirror mechanism which will perform the beam steering for the on-instrument wavefront sensor section of the Gemini near-IR imager. In turn, the wavefront sensor will generate correction signals for the tip-tilt and fast-focus secondary mirror. Preliminary testing of the current version of the gimbal assembly has revealed positive result when operated at room temperature, but demonstrated hysterisis problems at cryogenic temperatures. Described in this paper are the specifications, design and performance characteristics, and integration of the gimbal mechanism with the rest of the wavefront sensor system.

The Gemini Near IR Imager (NIRI) is a cryogenic instrument cooled by two closed-cycle cryo-coolers. The vacuum jacket is a hexagon shaped vacuum vessel made of three sections. Each section is forged out of aluminum 6061. All the internal structural components are made of aluminum 6061T6 except the supporting trusses, which are made of titanium. All the internal structural members are stress relieved to maintain dimensional stability and good optical alignment. The thermal insulation includes floating shields and cold shields. Two closed-cycle coolers are mounted opposite to each other and electronically synchronized in order to cancel the vibration caused by the oscillating expansion valve. Several different fabrication methods and stress relief methods are discussed.

A review of the capabilities of the near-IR camera and multi-object spectrometer (NICMOS) on the Hubble Space Telescope and of the environment in which the instrument operates is presented, together with the observing strategies which can be adopted to maximize the scientific return from the instrument. The wavelength dependence and intensity of the background detected by NICMOS, cosmic-ray impacts, image quality, and photometric calibrations are discussed. Examples of actual observations performed with NICMOS in wavebands which are difficult to access from the ground are shown.

The IR camera and spectrograph (IRCS) for SUBARU and Gemini near-IR imager (NIRI) instruments have a common design for all wheels, based on a modified geneva mechanisms with a locking cam actuated detent pin. The geneva design, in combination with the spring loaded detent mechanism, allows the stepper motor/spur gear drive to decouple from the wheel at each aperture position. The detent mechanism positions the wheel precisely. The need for precise motor control and wheel position encoding is reduced because of the detent mechanism. Six of these mechanism are filters wheels requiring repeatable aperture positing. The other seven mechanisms include of a slit wheel, grism wheel, pupil mask wheel, 2 beam steerers, a focal p;lane mask wheel, and a beamsplitter wheel. These mechanisms require repeatable, stable and accurate positioning. The number of aperture positions for the 13 wheels range from 2 to 16. The mechanisms are aligned and tested at room temperature and operated at 60 K, requiring an athermal design, for which the modified geneva mechanism is ideally suited. This paper will discuss the prototype development and final mechanical design of specific wheel mechanisms completed for the IRCS and NIRI instruments at the Institute for Astronomy.

We are developing an instrument to study the morphology and kinematics of the molecular gas and its interrelationship with the ionized gas in star forming regions, planetary nebulae and supernova remnants in our Galaxy and other galaxies, as well as the kinematics of the IR emitting gas in starburst and interacting galaxies. This instrument consists of a water-free fused silica scanning Fabry-Perot interferometer optimized in the spectral range from 1.5 to 2.4 micrometers with high spectral resolution. It will be installed in the collimated beam of a nearly 2:1 focal reducer, designed for the Cassegrain focus of the 2.1 m telescope of the San Pedro Martir National Astronomical Observatory. Mexico, in its f/7.5 configuration, yielding a field of view of 11.6 arc-min. It will provide direct images as well as interferograms to be focused on a 1024 X 1024 HAWAII array, covering a spectral range from 0.9 to 2.5 micrometers .

The internal support structure of the Gemini Near IR Spectrograph (GNIRS) comprises a series of substructures which are interconnected to support the optical components and their mechanisms.A very stable support structure is required in the GNIRS to exploit the high image quality rings; this type of structural did not provide sufficient stiffness. This concept was replaced by a novel type of structure employing lightweight cylindrical models, with each module produced by numerically controlled machining from a solid. Finite element analysis is combined with 3D layout techniques to develop an optimized structural configuration for each module. A parametric process was performed for the design optimization to produce the highest fundamental frequency for a given weight, as well as to deal with the normal concerns about global deformation and stress.

NASA Kuiper telescope personnel identified computer-aided telescope balancing as a needed capability which will save significant amount of time and will improve SOFIA operational effectiveness. Automated telescope balancing is considered a 'critical technology need' for SOFIA. It is necessary to balance the telescope to accommodate the different science instruments. The telescope must be adequately balanced to enable the pointing and tracking system to operate properly. In-flight balancing may be necessary because the mass properties of the science instruments can change during its operation and because of disturbance torques. In the past, a trial and error procedure was used to balance the Kuiper telescope, which can take up to several hours and requires a highly skilled, experienced technician to perform the task. Various approaches for balancing the telescope are reported in this paper. Potential benefits are: 1) enable balance compensation to be performed quickly, thereby reducing operations costs, 2) enable a much wider range of balance compensation to be performed in-flight, which may increase science observation time and science return, 3) enable technicians of any skill level to balance the telescope with the aid of a computerized system, and 4) improve the performance of the telescope by increasing the precision of the balance, which may increase science return.

We describe the design and performance of the near IR telescope experiment (NITE), a rocket-borne instrument designed to search for IR emission from baryonic dark matter in the halos of nearby edge-on spiral galaxies. A 256 X 256 InSb array at the focus of a 16.5 cm liquid-helium- cooled telescope achieves near-background-limited sensitivity in a 3.5-5.5 micrometers waveband where the local foreground from zodiacal emission is at a minimum. This experiment represents the first scientific application of a low-background IR InSb array, a precursor to the InSb arrays intended for SIRTF, in a space-borne observation. We describe the flight performance of the instrument and preliminary scientific result from an observation of NGC 4565.

We are constructing Mid-IR Test Observation System (MIRTOS) as an IR imaging system to evaluate and monitor the performance of the Japanese National Large Telescope, SUBARU. The system combines two cameras. One of the camera is for near-IR (NIR) and the other for mid-IR (MIR). They capture images simultaneously at the rate fast enough to freeze the seeing. Simultaneous NIR images are useful not only for evaluation of the image quality of the telescope but for two-wavelength shift-and-add that preserves diffraction-limited angular resolution of MIR images for a long integration. The system also has telescope emissivity mapping mode that images the telescope entrance pupil in MIR. For the MIR camera, a Santa Barbara Research Center (SBRC) Si:As IBS array with 320 by 240 pixels is used with pixel scale of 0.067 arcsec/pixel that takes enough samples to make diffraction limited images at 8 microns. For the NIR camera, an SBRC InSb array with 256 by 256 is used. Pixel scale is 0.028 arcsec/pixel. It is optimized to detect position of the brightest speckle in images at the wavelength of 2.2 microns with wide enough field of view to image a reference point source. In the emissivity mapping mode, a temperature controlled black body is inserted just outside the dewar window as an absolute calibration source for the telescope emissivity. In this paper we will present detailed design of MIRTOS.

We discuss the site conditions for astronomy at the South Pole and over the Antarctic plateau. We find that these conditions are the most favorable on Earth for sensitive observations at thermal IR and sub-millimeter wavelengths. We further discuss plans to develop IR facilities to exploit this potential.

We describe an imaging Fabry-Perot instrument, and give examples of its astronomical applications. The salient features of the instrument are: wide wavelength coverage with a single etalon, operation near the focal plane in a converging beam, resolving power R approximately 4000, relatively easy portability to other telescopes.

ABU is a NOAO IR imaging camera designed for evaluating the performance of the 1024x1024 Aladdth InSb array. For this experiment, it was outfitted with five filters (see Figure 9) m the 3-5 micron range to exploit the low water vapor and lower air temperatures at the South Pole. At the South Pole it was integrated with the CARA SPIREX (South Pole Infrared Explorer) telescope. Figure 1 is a picture of the telescope showing the environmental box (the white box by the author). which protected ABU and its electronics from ambient environmental conditions.

We describe the requirements, constraints, and goals for FLITECAM, the first light IR test experiment camera being built at UCLA for SOFIA. The camera must allow testing of the testing of the telescope/observatory and provide first- light images for public outreach and publicity. In addition, the camera should become a facility-class instrument for use by the general SOFIA user community. The camera is relatively simple and inherits many of the designs from previous instruments built in the IR Imaging Detector Laboratory at UCLA. It will offer wide-field imaging, high- resolution imaging for observing diffraction-limited images at >= 3 micrometers , low-resolution grism spectroscopy, and pupil-viewing. FLITECAM will be delivered for observatory tests in early 2001. The project will not formally start until NIRSPEC is delivered and commissioned at the Keck Observatory.

We have developed a new readout system for bolometers optimized for the Planck Surveyor experiment, a satellite mission dedicated to survey the Cosmological Microwave Background. The bolometer resistance is measured in a bridge with a capacitance load, using a periodic square wave bias current in order to remove the 1/f noises of the electronics. The use of a capacitance allows to reduce the transient signal and to get rid of the Johnson noise. The bias voltages are fully controlled by computer, and the lock-in detection is digital. This system has been implemented and successfully tested on the Diabolo ground- based astronomical experiment. Using the advantages of our readout system, we have built and fully tested an engineering model of the space qualifiable electronics which fulfills the scientific and technical requirements of the Planck Surveyor bolometric instrument: low noise system down to 0.1 Hz, electrical power consumption lower than 40 Watts and volume lower than 15 liters. Our presentation will consist in a full description of this readout system and a review of the current test results. Our system could also be adapted, with some modifications, to other space born instruments which use bolometers.

We investigate the use of non-cryogenic instrumentation for near IR spectroscopy. With this technique, it is possible to apply in the J and H bands some instrument concepts and observing techniques used in the visible. We present observations of the thermal background in H. We derive some instrument requirements for minimizing and handling it. We recommend the use of short wavelength cutoff wavelengths or linear filters in H. We present observations of the sky emission, and do confirm previous upper limits of the continuum emission between the OH lines. We discuss some applications of non-cooled near IR spectroscopy.

PACS covers the wavelength range 80-210 micrometers in spectrometric and photometric imaging modes. The long wavelength camera is a 16 X 25 pixel array of stressed Ge:Ga detectors. In order to demonstrate the feasibility of such a large array, one of the 25 linear arrays was manufactured. It consists of 16 elements of 1.5 mm³ each separated by ceramic plungers and stressed by one single mechanism. As preamplifier a dedicated CMOS circuit was developed, based on similar circuits successfully operating in ISO's photometer. In particular, it was intended to increase the gain of the CRE in order to minimize the debiasing effects on the low bias operated detectors. Two complete linear demonstrator arrays were manufactured and independently tested under various low background conditions in a 1.7 K environment at MPIA and MPE. The feasibility of the concept chosen was demonstrated in several functional tests. Valuable experience was gained to guide the development of the next generation of CREs and arrays.

The IR Spectrograph (IRS) will provide the Space IR Telescope Facility (SIRTF) with low and moderate-spectral resolution spectroscopic capabilities from 4 to 40 microns. The IRS is composed of four separate modules, with two of the modules providing R approximately 50 spectral resolution over 4 to 40 microns and two modules providing R approximately 600 spectral resolution over 10 to 37 microns. Each module has its own entrance slit in the focal plane and the IRS instrument has no moving parts. The low-resolution modules employ long slit designs that allow both spectral and 1D spatial information to be acquired simultaneously on the same detector array. Two small imaging sub-arrays in one of the low-resolution modules will also allow IR objects with poorly known positions to be accurately placed into any of the IRS modules' entrance slits. The high-resolution modules use a cross-dispersed echelle design that gives both spectral and limited spatial measurements on the same detector array. The one-sigma continuum sensitivity requirements for the IRS low-resolution modules' one-sigma line sensitivity requirements are 6.0 by 10^-23 W- cm^-2 in the same integration time. Internal calibration sources allow the IRS to perform self monitoring of detector sensitivity changes. Careful thermal design allows all four modules to be powered up simultaneously and still input less than 4 mW total power into the SIRTF liquid helium bath. The optical, mechanical, thermal, and electrical design of the IRS is discussed, as is the IRS on- orbit operational concept.

The short wavelength spectrometer (SWS) is one of the four instruments on-board of ESA's IR SPace Observatory (ISO), launched on 15 November 1995. It covers the wavelength range of 2.38-45.2 microns with a spectral resolution ranging from 1000-2000. By inserting Fabry-Perot filters the resolution can be enhanced by a factor 20 for the wavelength range from 11.4-44.5 microns. After the successful launch the instrument was tested and calibrated during a period of spacecraft checkout and performance verification. The opto- mechanical construction of the instrument appears to behave extremely well. The instrument performance is on all aspects as expected, except for the detector sensitivity where the noise is dominated by effects of particle radiation. We given here an overview of the in-orbit performance, discuss the calibration and present some result from trend analysis of the most important instrument and detector parameters.

Many IR sources are dusty; embedded stars are obscured, often completely, and their light is absorbed. The starlight heats the dust, typically to temperatures of tens or hundreds of Kelvin, and the heated dust radiates in the far IR, at wavelengths for which the Stratospheric Observatory for IR Astronomy (SOFIA) is optimized. These dusty targets radiate most or all of their energy in the far IR: broadband imaging with the highest possible spatial resolution is the natural starting point form which to develop an understanding of their morphology and energetics. Because SOFIA is the largest far IR telescope, it delivers the best spatial resolution. The wealth of detail revealed when resolution improves often result in startling insights, as new pictures of old favorites from the Hubble Space Telescope so regularly remind us. We therefore believe that most SOFIA studies will begin with high spatial resolution broadband imaging, and that a facility science instrument is required to serve this heavy and continuing workload.

We present design considerations for a mid-IR imaging polarimeter, TNTCAM II. Using a 256 by 256 Si:As BIB array, the camera will be unparalleled as an polarimeter/imager by any instrument currently in use at these wavelengths. Thanks to NSF support, access by the general astronomical community will be arranged. In polarimetry mode, TNTCAM II will be sensitive to linear polarizations as small as 0.2 percent. Polarized emission from cosmic sources will be modulated at a frequency high enough to remove atmospheric and system noise fluctuations. Dewar design and the optical system are discussed, including CCD assisted digital shift and ad tip- tilt correction and use of a rotating entrance window assembly allowing on-the-fly f-ratio adjustment and optimal throughput across the entire 5-25 micrometers band. The camera can contribute to the understanding of YSOs and evolved stars, obtaining high resolution mid-IR observations of dusty environments immediately surrounding these objects. In imaging mode mosaics of extended objects can be made in 2' by 2' sub-fields. In polarimetry mode, B-fields in YSOs can be probed by dust emission from hot cores, incidentally constraining grain alignment scenarios in young stellar environments.

We present here briefly the scientific objectives, the optical and mechanical design as well as the main components and observing modes of the new thermal IR multimode instrument for the 3.6-m telescope of European Southern Observatory at La Silla, Chile. Also we describe shortly the software package consisting of instrument control, data acquisition, quick look data reduction, calibration control and off-line data reduction that is necessary to operate the instrument and reduce the data. New values of refractive index of KRS-5 at low temperatures are given.

We present an iterative inversion method for the restoration of chopped and nodded images, typical of thermal IR astronomy with ground based telescopes. The method computes the smallest solution subjected to the constraint of positivity. The restored images exhibit artifacts, related to the chopping amplitude, which can be predicted by looking at the mathematical structure of the problem. However these effects can be strongly reduced by combing a few images taken with different chopping/nodding throws of small amplitude. Preliminary results on synthetic data are very promising. Restored images show high cosmetic quality and a typical restoration error smaller than 10 percent. We also present restorations of real images taken at the UKIRT telescope with the MAX camera. The availability of an image restoration method for mid-IR images would have a major impact on the telescope design and observing strategy at these wavelengths.