Since plastic scintillator neutron detectors are being discussed, I thought I might add a little about my detector setup.
A piece of standard, undoped scintillation plastic will only provide effective detection of fast neutrons greater than several hundred keV, at lesser energies the recoiling proton has very little energy and you'll eventually lose it in the system noise. Further, as long neutron energies are less than about 10 MeV, scattering is isotropic and the energy spectrum of the recoil protons is a simple uniform distribution from an energy of zero to that of the incident proton.
The probability of neutron scattering in a detector which is an inch or two thick is quite high, and as the thickness increases multiple scattering events become possible. Multiple scattering events don't really present any problems for simple counting; since the scattering events occur in such a short period of time compared to the decay time of the scintillator the light from events sum and create a higher amplitude pulse than would have been produced by a single event. A real problem with plastic is sensitivity to nearly all types of radiation. Luckily, an operating fusor shouldn't produce anything but neutrons which can pass through a detector housing. There is still the cosmic ray background to deal with, and a good bit of this is high amplitude stuff you can't cut out by increasing your discriminator level.
All of the above issues should be considered if you want to use scintillation plastic to detect neutrons. I built my detector in a thick stainless steel housing and lined it with 0.25" of lead. This will completely shield it from any x-rays produced by the fusor and help a little bit in reducing background. The PMT is also magnetically and electrostatically shielded, and the removable back of the housing is sealed from light with a black gasket. Some thought should be given to the setting of the discriminator setting. You want to cut out all the noise you can, but as the level is increased, the neutron detection effeciency drops since the recoil proton spectrum extends all the way to zero. For reference, my SCA allows me to select a window of energies to monitor anywhere within its 0-10 volt range. I set my lower level discriminator at 0.5V and and a window width larger enough to count everything above this. With my 2" diameter by 2" deep detector I get a background count of about 240 cpm. The system has a detection efficiency of over 30% when calibrated on an AmBe source. This agrees well with theoretical predictions.
Joe Zambelli
Created on Tuesday, January 16, 2001 3:30 PM EDT by Joe Zambelli