Photomultiplier Sensitivity Enhancement
Optical Devices that Increase Photocathode Quantum Efficiency
Photomultiplier photocathodes are something of a compromise, as they have to capture light effectively but offer little impediment to the escape of the liberated photoelectrons. These requirements conflict to some extent, so in order to obtain best overall efficiency the photocathode has to be somewhat translucent. It has been known for a long time that if the incoming light is directed at an angle, such that the light that passes through the photocathode unabsorbed is subjected to a total internal reflection, it will bounce back through the photocathode for another chance. In principle this process can be repeated any number of times, although probably the best practical compromise is a system that gives a double bounce, since multiple-bounce systems have a much smaller entrance aperture. Previously published data suggest that the improvement is greater at longer wavelengths, and greater with S20 than with bialkali tubes, but we're just about to try this for ourselves. In principle, an improvement of 50-100% looks possible, but we'll have to see for ourselves. If it works well enough, we're going to make a prism assembly (that we've already been rash enough to design on paper) which will allow the pmt to be used at its normal orientation instead of at an angle. In principle it could be fitted as an accessory to any fluorescence system that we've ever sold. This is a potentially a free lunch of sorts, but not, alas, in the financial sense, since it would probably cost £200-£300, but we think it could well be worth the cost.
