A short introduction to the subject of the meeting, IAU Symposium No. 167, New Developments in Array Technology and Applications is given. CCD and Array detectors have become the detectors of choice at optical observatories all over the world. Direct imaging, photometry and spectroscopy are all vastly improved as a result. Thirteen IAU Commissions joined in sponsoring this meeting which indicates the wide interest in this subject. In the five days of the symposium the following topics were discussed: New Developments in CCD Technology, New Developments in IR Detector Arrays, Direct Imaging with CCDs and Other Arrays, Spectroscopy with CCDs and Other Arrays and Large Field Imaging with Array Mosaics. A few papers concerning Astrometry with CCDs were given in the poster sessions. Scientific results were also presented in the poster sessions.

Two-D array detectors are one of the key components of the instruments being built for the four ESO 8-m telescopes, the VLT project. Three optical and two infrared instruments are under construction. One faint object imager and multi-slit spectrograph to operate in the spectral region 350–1000 nm (acronym FORS) is based on a 20482 high-efficiency CCD. A two-channel echelle spectrograph to work in the 300–1000 nm region (UVRS) is built around a mosaic of 20482, 15 μm CCDs. A multifiber, new spectrograph is based on a 4096 × 2048, 16 μm CCD. The two infrared high-resolution instruments, one imager/long-slit spectrometer (ISAAC) and one high-resolution camera are designed to work with adaptive optics and in the speckle mode (CONICA). Both are designed to incorporate IR arrays up to a format of 10242 and will work in the spectral region of 1–5 μm. The main properties of the instruments and of detectors are presented and their close interdependence is illustrated.

The Keck II Deep Imaging Multi-Object Spectrograph (DEIMOS) is a general purpose, faint object, multi-slit, double-beam spectrograph which offers wide spectral coverage, high spectral resolution, high throughput, and long slit length on the sky. This powerful instrument will be the principal optical spectrograph on the Keck II telescope. DEIMOS is optimized for faint-object spectroscopy of individual point sources, low-surface-brightness extended objects, or widely distributed samples of faint objects on the sky. To obtain high resolution (∼ 1 å) and wide spectral coverage (up to 5000 å) the spectrograph uses wide angle cameras and large CCD detectors with many pixels.

This paper describes some of the work being carried out to obtain the CCD detectors required for the DEIMOS spectrograph. In addition, results are presented on the fabrication and characterization of a 4k × 2k three-side buttable CCD produced by Orbit Semiconductor, a silicon foundry in San Jose, California. This CCD was fabricated to test the ability of Orbit to produce high quality scientific CCDs with the characteristics required for detectors to be used in DEIMOS and other optical instruments of the Keck Observatory.

We review the present status of CCD use at major observatories, discuss the various developments that are leading to improved sensors. The choice of devices is quite extensive, but far from stable! A look at progress over the last few years helps to put present and future developments in perspective. Several areas of active CCD development are presented and discussed.

Lately, Charge Coupled Device (CCD) detectors have had great advances both in the visible and in the X-ray spectral range. However, the technology applied to these devices in the ultraviolet (UV) spectral region has not developed as well, because of some problems connected with the interaction between UV radiation and the materials typically used in semiconductor technology. In our laboratory the ultraviolet response of some UV-enhanced CCDs has been investigated. In particular, the quantum efficiency of coronene and lumigen coated and of back-illuminated ion implanted CCDs have been measured in the 304–11000 å spectral range. Very interesting results have been found, mainly for a one ion implanted CCD with quantum efficiency values of more than 60% at 304 å. Some measurements of the response uniformity of this spectral region have also been made. The results obtained encourage the possible use of these detectors in ultraviolet astronomy with very good performance.

The requirements of current and next generation CCD controllers in the areas of CCD device and system architectures, readout noise, number and speed of readouts are reviewed together with such operational requirements as system flexibility, power consumption, cost and weight. The basic components of a CCD controller are described, including the timing sequencer, clock drivers, video processor and computer interface. The capabilities and implementation of the CCD controller developed at San Diego State are reviewed. An upgraded controller is described to overcome limitations in the area of readout speed and efficient support of multiple readout capability.

The use of CCD detectors as elements of an active acquisition system is reviewed. In such systems, the CCD image acquisition, data-analysis and the instrument and telescope controls are no longer separated elements whose actions are coordinated by an astronomer and/or operator, but are parts of a global system. The interaction between incoming data (nature of the object, registered flux, current PSF and atmospheric transmission), observer's requirement (S/N, spatial and temporal resolution) and forthcoming CCD exposure characteristics (CCD preparation, exposure time, read-out parameters) is examined. The requirements for the CCD electronics, data acquisition system are evaluated and examples of recent application in imaging, spectroscopy, photometry and auxiliary equipment are given. An attempt is made to analyze future technological trends and possible bottlenecks in such systems and to propose simple rules to adopt when designing CCD hardware and software.

Advanced CCD detectors are the backbone of modern astronomical instrumentation working in the 300–1000 nm spectral range. The status and results of the different projects to design, manufacture and implement new CCD detectors for the ESO VLT instrumentation are presented. Emphasis is put into ESO's development activities together with industry's, and into first measurement results of thinned arrays in large formats obtained in 1994. In addition some recent cryostat developments are shown to illustrate the different requirements of VLT instruments in this area.

Infrared array detectors, like silicon CCDs a decade before, have revolutionized infrared astronomy. The quality and performance of the current generation of devices has already allowed astronomers to obtain infrared images at wavelengths out to 2.2 microns which are as deep as the best CCD images. High resolution infrared spectroscopy is now a reality and ground-based imaging to 35 microns has been achieved. Several classes of low-noise infrared array detectors with formats of 256 × 256 pixels are now in routine use, and developments are under way which will produce detectors of 1024 × 1024 pixels (for the near IR) within the next year. This paper will briefly review the state-of-the-art and compare and contrast the properties of available arrays. Progress in the field is illustrated with recent near infrared photometry obtained with a new two-channel imaging system developed at UCLA.

Lick Observatory has recently developed a near infrared camera for astronomical imaging. The new camera has been built around a Rockwell NICMOSII 256 × 256 HgCdTe focal plane array. The dewar and optics were manufactured by Infrared Laboratories in Tucson, Arizona while the electronics and data system were designed and fabricated at Lick Observatory.

The new instrument was designed to image faint galaxies and other astronomical objects as well as embedded infrared sources in star formation regions. The camera has cold focal plane scales of 0.2, 0.4 and 0.6 arcseconds per pixel when used at the telescope. Standard H, J and K as well as narrow band special purpose filters are contained in two independent motorized filter wheels.

The camera performance with different plate scales and filters as well as the mechanical, optical and electrical lay-out will be shown. M 42 will be used for characterization of the camera's performance.

This paper gives an overview of infrared array detectors which have been tested and used at ESO. The performance of arrays using Reticon type readouts, CCD readouts and switched FET multiplexers have been evaluated for both InSb and Hg1−xCdxTe detectors. Performance limitations specific to the NICMOS3 256 × 256 Hg1−xCdxTe detector installed in the ESO infrared array camera IRAC2 are addressed. The first test results with a high well capacity SBRC 256 × 256 InSb array are also presented.

Advanced readout techniques for image sharpening tested on a 2.2-m telescope are discussed briefly. A new generation of instruments being built for the VLT, the very large telescope project of ESO, is designed to house large format 1024 × 1024 IR arrays. A fast data acquisition system is currently being developed at ESO. The system is capable of handling the high data rates generated in the thermal infrared by large format low well capacity arrays. It can also cope with the low read noise required for flux levels of ≤ one photon/sec. It will first be installed in ISAAC, the Infrared Array Camera and Spectrometer built for the VLT (Moorwood 1993). The present status of both the detector developments and the data acquisition system is reviewed.

Over the past ten years a technological revolution has occurred in the development of two-dimensional infrared array detectors for astronomical observations. The wide application of these arrays for both ground-based and space observations has resulted in a profound change in the capabilities and perspective for infrared astronomy, resulting in new views of the infrared sky. A review will be presented describing these detectors, the numerous advantages they provide for astronomical observations, the present state of array technology, and the potential for future growth.

The improvement of photoconductors and photovoltaic detectors for λ < 200 μm has displaced bolometers as detectors of choice for many applications requiring high sensitivity. Continued development of bolometers for operation at low temperatures (T < 0.1 K) has resulted in significant improvements in their sensitivity, making them excellent choices for many broad band applications at λ > 200 μm, cryogenic spectrometer applications in the submillimeter, and applications requiring extreme stability and ease of calibration.

I will describe the development of bolometers over the past decade, with particular emphasis on the detectors we have developed at GSFC. Detectors have been constructed in 36 element arrays with NEP ∼ 5 × 10−18 W/√Hz and a response time of ten ms. Such detectors are very useful for many current problems in Cosmic Microwave Background studies and submillimeter spectroscopy. We will describe the current state of development of the detectors and the improvements we are pursuing.

There has been significant progress in the development of superconducting tunnel junction detectors for operation in the far infrared and submillimeter spectral range. Though at an early stage, these detectors have great promise of excellent sensitivity and ease of array fabrication. I will discuss the current developments in this area.

Broadband observations at 5–20 micron with large ground-based telescopes are dominated by thermal background radiation from the telescope optics, instrument and sky. Astronomical sources typically contribute less than one percent of the total detected flux. In most cases the direct, unprocessed image of a bright source (before background subtraction) is indistinguishable from an image of blank sky (Fig. 1). This paper gives an overview of the fundamental concepts, conditions and limitations of high-background mid-infrared imaging. The techniques used to acquire and process the image data are summarized, illustrated with examples from our 58 × 62 pixel mid-infrared camera astronomy program.

A large-format PtSi array (effectively 1k × 0.5k pixels) has been incorporated into an infrared camera intended for survey work using the 0.75-m telescope at Sutherland. The array is very uniform and almost free of cosmetic defects. The camera, its electronics and the operational procedures that we use are described, together with the methods of data reduction. The limiting magnitudes that can be observed in the JHK broad bands are given.

The rapid progress in the focal plane technology enables us to use large format infrared sensors, such as 256 × 256 InSb/HgCdTe and 1040 × 1040 PtSi arrays. Infrared two-dimensional sensors make possible not only the imaging observations but a deep detection limit. The development of a large format infrared array is one of the most important breakthroughs in observational astronomy.

We propose to build a mosaic infrared camera for the SUBARU 8-m telescope. The SUBARU telescope is designed to reach a diffraction limited image at infrared wavelengths with a wide field of view (six arcsec at the Cassegrain focus). The camera is designed to cover the entire field of view with PtSi infrared sensors and to employ a weighted shift-and-add operation and a real-time image processing. The efficiency of the mosaic infrared camera and power of the 8-m telescope have a strong potential to meet challenging problems. Most of the regions of the near infrared sky are not covered with enough sensitivity. It is essential to conduct infrared deep and wide surveys.

I will briefly treat some of the developments relating to CCD photometry which have taken place over the last two to three years, and speculate on those which can be anticipated to take place in the near future. CCDs have been in widespread use as astronomical detectors for slightly more than a decade. For the majority of visual-light projects they have displaced the photographic plate and the photomultiplier as the detector-of-choice, except in applications requiring large area coverage or high time resolution, but even here the development of arrays and the use of more sophisticated electronics has permitted encroachment into these domains. The use of large, low-noise detectors for CCD photometry places stringent demands on telescope optics, instruments, controllers, calibration procedures, data reduction methods and data storage, and forces a holistic approach to managing the data flow from detector to final storage medium.

For an earlier review on this field see Walker (1993); other papers presented at IAU Colloquium No. 136 and IAU Symposium 167 (these proceedings) should be consulted for different topics, alternative viewpoints, and more in-depth discussions of specific subjects.

We review the status of CCD photometry with emphasis on applications in projects that require considerable amounts of observing time at medium-aperture telescopes and at tasks complementing projects carried out at large telescopes. Associated problems of CCDs are discussed, together with unsuspected difficulties affecting millimagnitude accuracy mainly caused by pixel-size mismatch, cooling, data acquisition, non-linearity of response and improper standardization.