Energy deposit and gamma rays due to neutrons

[Nynal95]

Hello everyone,

in my application, I simulate the propagation of neutrons through an absorber plate. The absorber materials are Aluminium, Copper, Iron, Lead, Paraffin or Concrete. I use Geant4 version 10.5. and a physics list including the high precision neutron package (attached). For the neutron data, I tried G4NDL4.5, ENDF B-VIII.0 and JEFF-3.3 libraries. I set the environment variable G4NEUTRONHP_DO_NOT_ADJUST_FINAL_STATE=1 .

I am most interested in the correct simulation of the neutron propagation. However, I looked at the energy deposit and the production of secondary gamma rays by neutron interactions (capture, (n,n’gamma)) in the absorber. Here, I get a strange behavior for almost all materials. Even for the (n,n’gamma) scattering, I get energy deposits larger than the initial neutron energy. When looking at the gamma energies, I get strange values. An example with ENDF-BVIII.0 library: For neutrons of 4.2 MeV interacting with aluminium, sometimes 3 gamma rays of approximately 2.99 MeV are created in 1 inelastic scattering reaction. Aluminium has an excited state of such an energy (ENSDF database), but then one would expect 1 gamma of this energy, not 3? Using JEFF-3.3, I haven’t seen this happen.

In this plot you can see the energy deposit spectrum of 4.2 MeV neutrons in 3cm aluminium:

To get the energy deposit, I do in my stepping action:
if(volume_prestep->GetName() == “Absorber” ){
fEDep+=step->GetTotalEnergyDeposit()/CLHEP::MeV;
}
, while fEDep is collected during one event. At the end of the event, it is resetted to 0.

I get the secondary particles in my stepping action via:
if(volume_prestep->GetName() == “Absorber” ){
const std::vector<const G4Track*>* secondary = step->GetSecondaryInCurrentStep();
}

Has anyone noticed similar problems? I am now not sure if I can trust the neutron energy after the reaction after all, does anyone know more about this?

Thanks very much in advance!

Nina

PhysicsList.cc (2.0 KB)

[civanch]

Hello,

your computations of energy deposition and access to secondary both are correct.

Concerning energy deposition itself one should mension, that in standard HEP simulation Geant4 conservs 4-momentum in all interaction on a very good level. When we apply precise neutron transport (HP models and ENDF data) results are not guranteed event by event but only in average. I am not an expert in HP, you need configure HP models using environment variables in order to have better energy balance. Have a look into other discussions.

VI

[Nynal95]

Hello,
thank you very much for your answer.
I have some problems understanding what “in average” means in this context. When I look for example at gamma rays produced in one single event, their number and energy will be wrong sometimes, but when I look at many events, the overall number of gammas of certain energies should be correctly predicted? And what does it mean for the neutron energy? Is it safe to look at a neutron energy loss spectrum (not energy deposit) calculated from the difference in kinetic energy, or if not what would be the correct quantity?

Thanks in advance,
Nina

[civanch]

Hello,

this means that if make energy balance: all secondary energy + energy deposited, in the case of HP models, this value will not be equal to the primary energy but inclusive spectra of low-energy particles will better fit to low-energy experimental data. In PhysicsLists without HP energy balance is guranteed but spectra of neutrons, protons, alpha will be less accurate. For gamma in HP I cannot say.

VI

[civanch]

Hello,

to be sure about energy balance you may try out QBBC or QSSP_BIC Physics Lists.

VI

[Nynal95]

Hello,

thank you very much for your answers. I will then try out the physics lists you mentioned and compare.

Best regards,
Nina

[huahua]

hi vicanch,
Is there any reference about the different between stander HEP and HP?
Thanks!