The photomultiplier is a vacuum photoemissive device and is based on a photocathode, a series of electron multiplying electrodes called dynodes and an anode, as shown in Figure 1 :
The photocathode is a photosensitive surface consisting of a support (usually transparent to the incident radiation) coated with an opaque material whose ‘photoelectric threshold value’ (Le. the electron escape energy) is sufficiently low to allow a portion of the incident radiation energy to be used to ‘free’ electrons from the photocathode material (i.e. enable the external photoeffect to take place). The nature of the photocathode coating material therefore depends on the energy range of the expected incident radiation, that is on the incident light wavelength.
The free electrons thus produced by the photocathode are accelerated by the applied bias voltage towards the dynodes which are coated by a material with a low work function, so as to enable secondary emission and therefore current multiplication. The output current collected at the anode Ia is therefore related to the initial phototocathode current Ipk by a current amplification factor G, as shown in eqn (3.6):this latter depends on the number of dynodes n (i.e. the number of stages of amplification), the dynode bias voltage Vd and the dynode material/structure factor α (Le. the secondary emission efficiency), as shown in eqn (3.7):where A is a constant and α is usually between 0.7 and 0.8 (Hamamatsu, 1983). Equations (3.6) and (3.7) show that the output anode current Ia is a linear function of the incident radiation Ein if the gain G is kept constant. This, according to eqn (3.7), can be achieved by keeping the dynode voltage Vd constant. The value of the supply voltage and the design of the dynode bias network is therefore critical both to the current gain actual value and its linearity. The power supply needs to be a stable, low ripple voltage source and the dynode bias needs to be insensitive to the voltage variations produced by the dynode photo-current Ipd (i.e. the dynode current needed to replace the electrons used in the photocurrent amplification). A simple potential divider network between the photocathode, the dynodes and the anode is sufficient but, for most applications, a voltage stabilizing zener diode and pulse decoupling capacitors also need to be used (Hamamatsu, 1983; Ferranti, 1982).
Like all vacuum components the photomultiplier is not suitable for applications where large mechanical forces are likely to be exerted on its envelope, furthermore its gain is affected by magnetic fields thus making the photomultiplier unsuitable for applications near (or in) strong magnetic fields without proper shielding (note that both drawbacks can be eliminated by the use of high permeability metal casings). However, in view of its wide dynamic range of operation (typically 4 to 5 decades) and its high speed characteristic (typical value for the anode current rise time is 2 nS, limited largely by the electrons’ transit time and the electrodes stray capacitances) the photomultiplier is ideal for high sensitivity, high speed light amplifica tion, such as in the case of Scanning Laser Rangefinders.