Fluctuation and error bars for Ion Energy deposition profile


Those questions might sound simple but I consider myself a beginner so please cope with me.
I have been using example TestEM11 for energy deposition profile (histogram 1) of ions.
Recently, I have been interested in understanding the fluctuation of those profiles between same run of particles.
I thought first that the error bars when running 10^4 particles was due to this fluctuation but then I tried to run only one particle and see fluctuation between runs. The error bars on 1 particle are so huge that I doubt it is due to physical parameters. (running run01.mac with protons 1MeV should show these)

So could someone help me understand the origin of the error bars? Why does a huge number of particles is required to get the expected energy deposition profile? And how can I obtain physical fluctuation?

All advise are appreciated.
Thank you in advance!


you do not attach plot which we may discuss. Of course, there are fluctuations , so each event should be slightly different.


Hi sorry for the mistakes.
So I slightly modified the mac file run01.mac from TestEM11. The particle is a proton 1 MeV and I use the empenelope physics. One plot is 1 proton only and second plot is 1000 protons. I would like to understand the causes of the error bars present in those plots. (What is caused by fluctuations? what is due to code (bining, cut treshold, stepMax… )?) I am interested only in observing fluctuations but I feel like the plots take more into account.

1proton1MEV.pdf (16.8 KB) 100proton1MEV.pdf (17.1 KB)


I am precising my question here.

I am interested into checking the fluctuation of Ion deposition in silicone for a 300 um detector. The energy deposition is computed with steps of 1um. The ions considered are from H to Fe, in the 1-100Mev/nucl range. I am using the empenelope physic list.
I am expecting to see straggling fluctuations in each step deposition depending on the energy of the particle. From my understanding, depending on the energy before the step’s deposition, the distribution of energy deposited in a step should follow either a Gaussian (lower energies), Vavilov (main part of our range of energies of interest) or Landau distribution.
Also these fluctuations should evolve (from Landau to Vavilov to Gaussian) along the steps as the particle is depositing part of its energy on its path.

I found in literature some mentions of this type of fluctuations included in Geant, I am just not sure how it is actually implemented, and if it is automatically used. I found G4IonFluctuations which seems to enforce this type of fluctuation but I am not sure if it is something that is required to add in the physic list at set up.

Could you help me figure out this? Or correct me if I understood something wrong.

Thanks in advance!

Hi @civanch ,
Should I post this new question as an independent topic?

Thanks in advance


sorry for the late answer. The topic is correct.

For ion ionisation we have G4IonIonisation as the main model of energy loss fluctuations. It is applied at each step of an ion. We have also an alternative new model for high energy ions derived from the ATIMA code but it is not avalable with production EM physics constructors.


Hi @civanch,

Thank you for answering. Thus I understand from your answer that G4IonIonisation includes the straggling fluctuations for H-Fe ions in the 1-100Mev/u range. Is it directly implemented in empenelope physic list (or another EM physics?) or should it be activated somehow?


each EM physics constructor has ion physics defined. Our advice of today is to use Opt4 (EMZ) physics as the most accurate for ion transport.



I managed to achieve the check I needed.

Thank you very much