I am studying the activation of air around a D-T neutron generator by comparing FLUKA and MCNP.
The results of the two codes are very similar considering the radionuclides that come from neutron reactions.
However, Fluka gives addiation radionuclides (C-11, O-15 and N-13) coming from reactions of secondary protons. These radionuclides are not present in MCNP results (or are present only partially, as in the case of N-13) because in MCNP I did not initially consider protons transport.
Trying to activate protons transport in MCNP too, I obtained very different results in term of protons energy and fluence. As default settings, MCNP considers data tables for protons but when these data are missing or when protons energy is outside data energy range, models are used instead.
In my problem, protons energies are fully covered by the range of data tables (so the models are called only in case of missing data) and this is the reason why I didn’t change any default parameters in MCNP.
The proton fluence of FLUKA is the follow (very short run):
According to MCNP, secondary protons with energies more than 4 MeV are not possible.
My question is:
How Fluka handle protons transport? Does it use some models or cross section data? Which type of cross sections or models Fluka use? Will these information allow me to reproduce the same transport condition in MCNP?
14 MeV neutrons can definitely generate protons of more than 4 MeV in your geometry/materials (e.g., in silicon, calcium, nitrogen, …).
Note that this has nothing to do with the proton treatment in FLUKA, since they are the product of neutron reactions (whose treatment in FLUKA, below 20 MeV, is based on data libraries).
As for the further proton transport and interaction, different physics ingredients come into play: stopping power, multiple Coulomb scattering, nuclear inelastic/elastic scattering length, and nuclear interaction mechanism. In FLUKA, these are modelled and independently benchmarked.
Also, a sharp opposition data vs models is quite misleading, since data are discrete and their use on a continuous range implies already some assumptions, while models are shaped by data.
In conclusion, this information certainly does not allow you to reproduce the same conditions (as) in MCNP, since you are dealing with two different codes.
Dear Francesco,
thank you for your suggestions.
I am using JEFF3.3 in FLUKA and ENDF/B-VII.1 in MCNP and in terms of neutron transport they give me very similar results (neutron fluence, neutron dose and radionuclides that come from neutron reactions.)
Probably, some materials in the two codes are not the same because MCNP suggests to use isotopic tables instead of FLUKA that works by default with natural composition of the elements. So, in some case in MCNP I considered only the isotope that is the main constituent of an element.
Could this be the reason?
I’d say not for the generation of protons above 4 MeV, since I see them produced on 28Si, 40Ca, 14N. which are the dominant isotope of the respective element.
