Energy-dependent scintillatior resolution

Physics Processes, Models and Cross Sections
1

Hi all, I’m working to build a G4 simulation of an X-ray telescope. It uses cerium bromide scintillators. I have parameters set up such as scintillation yield(s), indices of refraction, etc. and optical photons are produced and propagated correctly.

From what I can tell, in G4, the energy resolution "RESOLUTIONSCALE" of a scintillating material is a constant value. However, for cerium bromide, the energy resolution (FWHM) is not constant, but rather a linear function of energy—see, for instance, Quarati et al., figure 7..

Main question: Has anyone used or implemented energy-dependent scintillation resolution in the past?

Am I misinterpreting the energy resolution parameter "RESOLUTIONSCALE" available for scintillation? My understanding is that it is a fixed parameter that does not very with energy, so it is not a good model of CeBr3 resolution. At the moment I am thinking to modify the G4 scintillation source to incorporate energy-dependent resolution. I would be happy to share this modification with anyone in the future.

Thank you in advance!

~ William

2

Hi, In the scintillation process, the mean number of photons emitted is dictated by a Poisson or Gaussian distribution (see Book For Application Developers Release 10.7 page 244). The standard deviation about the mean is then the square root of the mean times a factor RESOLUTIONSCALE. So with RESOLUTIONSCALE=1.0, the standard deviation will be proportional to sqrt(particle energy), if SCINTILLATIONYIELD is a constant.

I usually treat energy resolution empirically, since calculating it can be complicated (there is optical photon transport to the PMT, PMT transfer functions, electronics effects, etc., …). I usually output a spectrum from Geant4 which will have resolution as described above and then apply an empirical curve from spec sheets or lab data in a MatLab program. I’m somewhat oversimplifying, but you get the idea.