The ideal detector for application in a variety of astronomical studies at ultraviolet, optical, and infrared wavelengths is one which can simultaneously over a wide waveband provide the wavelength, time of arrival and location of each photon falling upon the detector from the widest possible field. Intrinsic energy discrimination, although typical of modern X-ray detectors such as X-ray CCDs, does not occur in any existing ultraviolet or optical detector systems. Any detector development which provides these key features will make a major contribution to astrophysical research by improving the sensitivity limits for faint object spectroscopy, and by simultaneously providing the photon arrival time.
Perryman et al. (1993) proposed, largely on theoretical grounds, that optical detectors based on superconductors could make a major advance in the development of instrumentation for application in astronomy. Since this proposal experimental demonstration of the basic principles using niobium-based superconductors by Peacock et al. (1996a,b) has shown the validity of this approach. In this paper we extend these initial ideas to the more general case of an arbitrary superconductor, consider the effects of multiple tunnelling, and indicate the wavelength resolution that may ultimately be expected under various conditions.
Despite the considerable strengths which have made the CCD the pre-eminent detector for optical astronomy, three performance weaknesses still exist: the lack of an inherent spectral resolution, the limited quantum efficiency at short wavelengths (in the blue/ultraviolet), and the inability to photon count--and thereby provide the arrival time of the incident photon (Delamere 1992). Current photon counting detectors cannot compete with CCDs except with respect to time resolution, and for applications involving very low signal levels. None provides a measure of the photon wavelength, all have a limited quantum efficiency due to the available photocathode materials, and all are limited by their photon count rates (Timothy 1988; Petroff & Stapelbroeck 1989). Photon-counting ability has potential applications in many fields of astronomy, e.g. the study of the optical properties of variable sources, as well as compensation for the effects of atmospheric seeing. It is however the lack of any intrinsic wavelength discrimination which is a serious limitation of all available photon detectors, and which the superconducting detector overcomes for the first time.
Detectors based on semiconductors have a bandgap (1.1 eV for silicon) of order of the (optical) photon energy. Photoabsorption results in typically only one electron being excited from the valence to the conduction band, irrespective of photon wavelength. This precludes, without some form of amplification, the possibilities of photon counting, and/or wavelength discrimination.
This is not however the case for superconductors. As broad-band, noiseless, high speed photon counting detectors, with inherent spectroscopic ability--and thus the ability to cover efficiently in a single detector the waveband from the ultraviolet to the near infrared with medium spectroscopic resolution--they could allow developments in a number of different fields of astronomy. For example the observation of emission-line complexes or continuum absorption features such as the Lyman edge in very faint extragalactic objects may allow the direct determination of their red shifts. Applications requiring time resolution, either to measure intrinsic variability characteristics (e.g. Dravins 1994) or to overcome atmospheric effects, will also become more accessible.