This paper compares the photon detectors used to date in high energy physics and astrophysics experiments, with particular emphasis on design features, problems and mistakes. The paper also describes a new direction in the area of photon detection.
Invited talk presented at the International Workshop on RICH Detectors (RICH 95)Uppsala, Sweden, June 12-16, 1995 * Work supported by the Royal Swedish Academy of Sciences through its Nobel Institute for Physics, and Department of Energy contract DE-AC03-765SF00515. J. Seguinot and T. Ypsilantis have recently described the theory and history of Ring Imaging Cherenkov (RICH) detectors [1,2]. In this paper, I will expand on these excellent review papers, by covering the various photon detector designs in greater detail, and by including discussion of mistakes made, and detector problems encountered, along the way. However, I will not consider overall system geometries, radiators, optics, gas systems, etc. As a result, there will be no discussion of angular resolution or the value of N o achieved in the various experiments.Photon detectors are among the most difficult devices used in physics experiments, because they must achieve high efficiency for photon transport and for the detection of single photo-electrons. For gaseous devices, this requires the correct choice of gas gain in order to prevent breakdown and wire aging, together with the use of low noise electronics having the maximum possible amplification. In addition, the detector must be constructed of materials which resist corrosion due to photosensitive materials such as TMAE, the detector enclosure must be tightly sealed in order to prevent oxygen leaks, etc.The most critical step is the selection of the photocathode material. Typically, a choice must be made between a solid (CsI) or gaseous photocathode (TMAE, TEA). A conservative approach favors a gaseous photocathode, since it is continuously being replaced by flushing, and permits the photon detectors to be easily serviced (the air sensitive photocathode can be removed at any time). In addition, it can be argued that we now know how to handle TMAE, which, as is generally accepted, is the best photocathode material available as far as quantum efficiency is concerned (see Fig.1 and ref.3).However, it is a very fragile molecule, and therefore its use may result in relatively fast wire aging. A possible alternative is TEA, which, in the early days, was rejected because it requires expensive CaF 2 windows, which could be contaminated easily in the region of 8.3 eV and thus lose their UV transmission; there was also a suspicion that it had low quantum efficiency. Although the CaF 2 windows are still expensive to use, the contamination problem appears to be manageable, and the results of several tests presented in this paper remove the second concern by confirming the high quantum efficiency reported in ref. 1. The TEA molecule is also not as fragile as the TMAE molecule, so that its wire aging effects should be less severe. This would certainly be helped by cond...