Problem 1738

Summary: photons emitted in inelastic (n,n')-reaction
Product: Geant4 Reporter: Artem Zontikov <zontikov.a>
Component: processes/hadronic/models/neutron_hpAssignee: tkoi
Status: RESOLVED WONTFIX    
Severity: trivial    
Priority: P5    
Version: 10.1   
Hardware: PC   
OS: Linux   
Attachments: photon spectrum

Description Artem Zontikov 2015-04-27 12:25:19 CEST
Hi.

I encountered with a strange result simulating photons emitted in neutron inelastic scattering. The problem geometry was a "G4_C" cube with dimensions 1mm x 1mm x 1mm placed inside "World" volume filled with "G4_AIR". The cube was irradiated by pointlike neutron beam.

Attached you can find some plots with spectrum of photons generated in cube in energy interval [4.4375-4.4395] MeV. With energy of incident neutron increasing the second peak (~4.438 MeV) appears. It feels like the left peak originates when 1st excited state is a result from de-excitation of 2nd and higher levels, while the right peak (~4.439 MeV) is practically equals to F*n*sig, where F - flux/cm^2, n - number of nucleus, sig - cross-section of 1st excited state production.

All the simulations were performed with "FTF_BERT_HP" physlist. Flag for not adjusting final state was set to 1.

Thank you in advance.
Comment 1 Artem Zontikov 2015-04-27 12:26:09 CEST
Created attachment 339 [details]
photon spectrum
Comment 2 Artem Zontikov 2015-04-27 18:07:05 CEST
It looks like somewhere the energy of transition from the 1st level to the ground state is confused with the level energy itself.

Level energy: 4.438910e+03 MeV
Energy of transition: 4.438030e+03 MeV
Comment 3 tkoi 2015-04-28 14:15:29 CEST
Hi 

Thank you for reporting.
I successfully reproduced the phenomena you see.
I will do further investigation later.

Tatsumi
Comment 4 tkoi 2015-04-30 00:36:54 CEST
The origins of gammas are identified.
There are two sources of gammas. 
One is defined in ENDF data and the other is Gamma cascade code in NeutronHP.
The energy of gamma from the former source is 4.439000MeV and the later is 4.438030MeV.
Energy dependent behavior comes from, 
(n,n’_1st) reaction gives  4.439000MeV gammas because the  energy of gamma from the first excitation is defined in the ENDF data. But that kind of definitions do not exist in higher states, thus Geant4 use internal gamma cascade codes to generate final states so  we will have 4.438030MeV gammas from (n,n’_2nd and higher). Increasing energy of projectile neutrons also increases the gammas from higher excitation states. 
We respect ENDF data as much as possible and also you are setting “not adjusting final state” flag, so I close this report as “WONTFIX”.

Anyway, thank you very much for reporting this, the behavior is new for me. 
Tatsumi
Comment 5 Artem Zontikov 2015-05-01 12:15:22 CEST
Thank you for detailed explanation. Would you be so kind to clarify some issues regarding the topic? Please, correct me if I wrong.

1. I suspect that Geant4 generates gammas from (n,n’_1st) reaction using multiplicities defined in MF=12 section of ENDF data. So, if such section is missing (unfortunately, it happens sometimes) Geant4 will use internal gamma cascade code for the 1st excited state as well as for higher levels.

2. Is it safe to set energy of transition for the 1st state equal to the 1st level energy in text files under $G4NEUTRONHPDATA/Inelastic/Gammas/ to achieve consistency with ENDF data?
Comment 6 tkoi 2015-05-01 16:23:04 CEST
1) Correct
Probably you are already aware that is only given for (n,n’_1st) and not  for higher. 

2)I think so, (but this is my personal impression)
The value is derived from ENSDF library, so it will not be modified by us. 

Tatsumi