SELECTIONS FROM Astrophysical Techniques, Kitchin, C. R., (Bristol ; Philadelphia Institute of Physics Publishing, 1998) eBook ISBN: 0585295360 Printed ISBN: 0750304979 A Detector Breviary Two of the four most ubiquitous detectors used in astronomy, the eye and the photomultiplier, have now been discussed in some detail. The third major type of detector is the photographic emulsion and is dealt with in section 2.2 and the fourth is the CCD (see below). There are many other types of detectors but before going on to discuss these, they need placing in some sort of logical framework or the reader is likely to become more confused rather than more enlightened by this section! We may idealise any detector as simply a device wherein some measurable property changes in response to the effects of electromagnetic radiation. We may then classify the types of detector according to the property that is changing, and the scheme used in this work is shown in table 1.1.2. Other properties may be sensitive to radiation but fail to form the basis of a useful detector. For example forests can catch fire, and some human skins turn brown under the influence of solar radiation, but these are extravagant or slow ways of detecting the sun! Yet other properties may become the basis of detectors in the future. Such possibilities might include the initiation of stimulated radiation from excited atomic states as in the laser and the changeover from superconducting to normally conducting regimes in a material. Before resuming discussion of the detector types listed earlier, we need to establish the definitions of some of the criteria used to assess and compare detectors. The most important of these are listed in table 1.1.3. Table 1.1.3. Some criteria for assessment and comparison of detectors. QE (quantum efficiency) Ratio of the actual number of photons which are detected to the number of incident photons. DQE (detective quantum efficiency) Square of the ratio of the output signal/noise ratio to the input signal/noise ratio. τ (time constant) This has various precise definitions. Probably the most widespread is that τ is the time required for the detector output to approach to within (1 – e–1) of its final value after a change in the illumination; i.e. the time required for about 63% of the final change to have occurred. Dark noise The output from the detector when it is unilluminated. It is usually measured as a root-mean-square voltage or current. NEP (noise equivalent power) The radiative flux as an input, which gives an output signal-to-noise ratio of unity. It can be defined for monochromatic or black body radiation, and is usually measured in watts. D (detectivity) Reciprocal of NEP. The signal-to-noise ratio for incident radiation of unit intensity D* (normalised detectivity) The detectivity normalised by multiplying by the square root of the detector area, and by the electrical bandwidth. It is usually pronounced 'dee star'. The units cm Hz1/2 W–1 are commonly used and it then represents the signal-to-noise ratio when 1 W of radiation is incident on a detector with an area of 1 cm2, and the electrical bandwidth is 1 Hz. R (responsivity) Detector output for unit intensity input. Units are usually volts per watt or amps per watt. Dynamic range Ratio of the saturation output to the dark signal. Sometimes only defined over the region of linear response. Spectral response The change in output signal as a function of changes in the wavelength of the input signal. λm (peak wavelength) The wavelength for which the detectivity is a maximum. λc (cut-off wavelength) Wavelength(s) at which the detectivity falls to zero; wavelength(s) at which the detectivity falls to 1% of its peak value; wavelength(s) at which D* has fallen to half its peak value.Charge-Coupled Devices (CCD) Although only invented about three decades ago, these devices have already become as important a class of detector as the other three main types for astronomical work. Only their cost seems likely to restrict their use in the foreseeable future. Their popularity arises from their ability to integrate the detection over long intervals, their dynamic range (105), high quantum efficiency, and from the ease with which arrays can be formed to give two-dimensional imaging. In fact, CCDs can only be formed as an array, a single unit is of little use by itself. For this reason they are dealt with here rather than in Chapter 2. In astronomy, CCDs are almost synonymous with detectors of radiation. However, they do have other important applications, especially as computer memories. Their structure is somewhat different for these other applications and the reader should bear this in mind when reading about the devices from non-astronomical sources. The basic detection mechanism is related to the photoelectric effect. Light incident on a semiconductor (usually silicon) produces electron–hole pairs as we have already seen. These electrons are then trapped in potential wells produced by numerous small electrodes. There they accumulate until their total number is read out by charge coupling the detecting electrodes to a single read-out electrode. An individual unit of a CCD is shown in figure 1.1.11. The electrode is insulated from the semiconductor by a thin oxide layer. In other words, the device is related to the metal oxide–silicon or MOS transistors. It is held at a small positive voltage which is sufficient to drive the positive holes in the p-type silicon away from its vicinity and to attract the minority carriers, the electrons, into a thin layer immediately beneath it. The electron–hole pairs produced in this depletion region by the incident radiation are then separated out and these electrons also accumulate in the storage region. Thus an electron charge is formed whose magnitude is a function of the intensity of the illuminating radiation. In effect, the unit is a radiation-driven capacitor. Figure 1.1.12 Array of CCD basic units. Now if several such electrodes are formed on a single silicon chip and the depletion regions are insulated from each other by zones of very high p-type doping, then each will develop a charge which is proportional to its illuminating intensity. Thus we have a spatially digitised electric analogue of the original optical image (figure 1.1.12). All that remains is to be able to retrieve this electron image in some usable form. This is accomplished by charge coupling. Imagine