RE: [fluka-discuss]: Simulation of coincidence summing effects in HPGe efficiency calibration

From: Joachim Vollaire <joachim.vollaire_at_cern.ch>
Date: Mon, 11 Apr 2016 09:16:54 +0000

Hi Andrea

One first comment which is not directly related to your problem concerns the definition of the geometry. Your first region does not need to be defined using 60 zones but can simply be done with the following three:

regAIR2 5 | +world -det_out -source
                 | +det_out +world +al_end
                 | +det_out +world -zero -source

Otherwise, your approach seems correct. I also checked with Co-60 and could reproduce with FLUKA the summing peak at 2.5 MeV. I think therefore that it should work for other radionuclides with more complicated decay scheme.

Otherwise my question is how do you compare the simulation with the experiment. I would say that in real life, the summing effect in the detector are not necessarily coming from the same radionuclide decay while in the simulation the response is the averaged behaviour of a single decay ?

Greetings
Joachim

-----Original Message-----
From: owner-fluka-discuss_at_mi.infn.it [mailto:owner-fluka-discuss_at_mi.infn.it] On Behalf Of Andrea Mattera
Sent: 30 March 2016 13:04
To: fluka-discuss_at_fluka.org
Subject: [fluka-discuss]: Simulation of coincidence summing effects in HPGe efficiency calibration

Dear Fluka experts,

I am trying to simulate the detection efficiency of a HPGe; specifically we are interested in the coincidence summing effects if a source with multiple gamma lines is placed in close proximity of the detector. In order to do this I am using a BEAM ISOTOPE and the HI-PROPE cards to define a radioisotope as a source and I am detecting the energy in the HPGe using a (series of) DETECT [input file attached].

We are comparing the Fluka simulation with measurements performed in very similar geometries and with the same sources (152Eu, 54Co, 133Ba):
we could verify (see pag. 1 in the attached pdf) that the two efficiency curves agree very well over a wide energy range if we are in 'ideal'
conditions (point-source placed at a relatively long distance to the source). This gave us some confidence that no big mistakes are there in the geometry and/or the scoring of particles.

However, once we reduce the source-to-detector distance - i.e. when we expect the first effects of coincidence summing to appear - problems start to arise. Our simulation seems to largely over-estimate the effects of coincidence summing for most points (sometimes even by a factor of 3 or more, compared to the measurement) - see pag. 2.
[OBS: another difference between the 'far point-source' vs 'close geometry' is that, in the latter case (both in measurement and simulations), the source is extended and mixed in concrete powder to account for self-absorption effects in the source]

My first question to you is: is Fluka able to simulate these effects with the settings above in the input file (ISOTOPE + DETECT)?
I would like to stress that we are not interested in (time-dependent) pile-up effects on the efficiency (the activity of the sources and hence the count-rate in our detector are very low). We only want to simulate coincidences coming from gamma cascades in the same decay event. From simple tests (e.g. with a 60Co source) it seems like Fluka shows a summing peak, which makes me hopeful that this is a good path to explore for our application [pag. 3].

But: am I missing something in the physics or in the implementation? If our approach is in principle correct, do you have any guesses or hints at where we should look for an error that could explain the differences between simulation and measurements?


Thanks a lot in advance for your help!


/Andrea

��������������������������������������������������������(�Ƨ��ځ맲��r���*'~�&������x%��Z�m����0��n���ߖ���i�'i���w�
Received on Mon Apr 11 2016 - 12:40:09 CEST