I am trying to do some very simple photofission simulations: a small 235U sphere is bombarded by a 10MeV photon beam. I have been able to define everything I need and I do get results. Nevertheless, the results I get differ significqntly from results I have obtained with other codes. I wonder if I have correctly set up the scoring and the physics that I need for this simulation.
Physics: I have activated photonuclear reactions with the PHOTONUC card for all energies, I have also unabled other processes which I do not need (such as radioactive decay) to reduce compouting time.
Scoring: in order to estimate the number of photofissions induced in the target per emitted primary photon I have used a USRBIN c ard with region point binning and FISSIONS for the quantity scored. I also used a AUXSCORE card to filter only the photon induced fissions. I belive that this cards give out the number of photon induced fissions in the target per primary photon, is this correct?
The results I get differ from other simulations I have made, here are my results: 235U_10MeV_21.bnn.lis (517 Bytes)
Thank you for your question.
It would be easier if you point me to the numbers that you are trying to achieve (from other simulation software or references).
To be precise, I think that your scoring should give you fissions per primary per cm^3. So if you want to have fissions per primary you would have to multiply the result by the volume of your TARGET region.
Small remark: if you don’t want radioactive decay then simply don’t put the RADDECAY card.
The number of photofissions/cm^3/primary I get in my target with FLUKA is 1.88E-3 while the value I am expecting is closer to 2.23E-3photofissions/primary. This gets even worse if the FLUKA result has to be multiplied by the target’s volume which is 0.0654cm^3 (sphere of radius=0.25cm).
I must have defined something wrong then, either the physics or the scoring. Any tip would help.
In fact I was wrong! The number you are getting is fissions/primary so the discrepancy is not that big. Let me look into your code again and come back to you with a more detailed answer.
Ok, cool. I thought I had seen that for fission scoring with USRBIN in region binning then it is only normalized per primary and not per unit volume. So yes, then the difference is not too big, it may be fine.
If you do see anything else that I could improove I’d apreciate it. Thanks!
I tested your input file and everything seems fine to me from the point of view of the simulation setup. My question is, where is the number 2.23E-3 photofissions/primary coming from?
Please find below the FLUKA U235 photofission cross section compared with available datasets. 235Uphotofiss.pdf (20.8 KB)
Thanks for testing it, that helps me a lot. The 2.23E-3 comes from simulations done with other codes.
However, what I do not get is: to my understanding FLUKA uses cross section data libraries only for low energy neutron interactions. Thus, the photofission interactions are simulated using physical models (particularly the giant dipole resonance for the energies in the order of 10MeV). Do those “FLUKA photofission cross sections” are the ones obtained by said models? or are those tabulated data that FLUKA uses for simulating? Would you have a source for that data?
The 2.23E-3 comes from simulations done with other codes.
without actually specifying which codes are we talking about. FLUKA is a well-established software in the field and if you’re comparing its results to some undefined Monte Carlo codes then it’s justified for me to ask why are you taking the other result as your benchmark or the correct answer?
We can try to dig deeper into the origin of this discrepancy, but I would have to know if that even makes sense - meaning why your number has more credibility then the FLUKA result.
Tabulated data embedded in FLUKA determine the GDR reaction cross section adopted by the code, giving the photon interaction probability. FLUKA models determine then the reaction channel selection, giving - as a resulting fraction of the above reaction cross section - the photofission cross section represented by the yellow curve in the file earlier posted, where it is compared to photofission datasets extracted from EXFOR.
The other code (MCNP) is one I have more experience with, while I am knew with FLUKA so I thought I might have done something wrong. I am just trying to learn FLUKA and in order to safely say I did the simulation correctly I compare the results to other codes, thats all. I dont have any reason to belive that it does have more credibility, it is not that I question what FLUKA does, what I question is my capabilities to correctly use FLUKA .
I think the difference makes sense given the cross sections seen in the image you attached.
You find it nowhere since it’s calculated runtime. Moreover, as already pointed out on several occasions in this forum, FLUKA is not designed to extract microscopic cross sections, rather to calculate resulting macroscopic quantities.
.