Problem 2149 - Boron10 in G4NDL4.5
Summary: Boron10 in G4NDL4.5
Status: RESOLVED FIXED
Alias: None
Product: Geant4
Classification: Unclassified
Component: processes/hadronic/models/neutron_hp (show other problems)
Version: 10.4
Hardware: All All
: P4 minor
Assignee: Emilio
URL:
Depends on:
Blocks:
 
Reported: 2019-03-17 11:28 CET by michel.maire
Modified: 2019-06-17 13:14 CEST (History)
3 users (show)

See Also:

Attachments
output of Hadr03 (4.29 KB, text/plain)
2019-03-17 11:28 CET, michel.maire 		Details
plot of Q (35.16 KB, image/gif)
2019-03-17 11:30 CET, michel.maire 		Details
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this problem.
Description michel.maire 2019-03-17 11:28:21 CET 
Created attachment 553 [details]
output of Hadr03

See hypernews hadronic #1687 from Dongwon Lee.

Neutron (5 MeV) on Boron10. In attachement, output of Hadr03 for 2 physics constructors: BinaryCascade and neutronHP (with G4NDL4.5).

The second attachement is plot of Q: BIC in top, neutronHP in bottom.

From which we can conclude that channel 1 of neutronHP (line 88 of output) is wrong.
Comment 1 michel.maire 2019-03-17 11:30:16 CET 
Created attachment 554 [details]
plot of Q
Comment 2 Emilio 2019-06-14 10:40:01 CEST 
Dear Michel,

I have been investigating this problem, and here are the conclusions.

The neutron + B10 --> 2 alpha + triton reaction do not have final state data in G4NDL4.5. Thus, ParticleHP uses a model to generate the two alphas and the triton. The model do not try to conservate energy and momentum event-by-event, but on average. On top of this, ParticleHP tries to conservate energy and momentum event-by-event in some situations, by default. This option can be disabled by defining the environmental variable G4NEUTRONHP_DO_NOT_ADJUST_FINAL_STATE (highly recommended, in my opinion).

Now, when running Geant4 with the two options you get:
  default --> Q = 1.85 MeV
  G4NEUTRONHP_DO_NOT_ADJUST_FINAL_STATE --> Q = 0

And both are wrong. I have been investigating the problem and I found a bug. When fixed the results are:
  default --> Q = 2.09 MeV
  G4NEUTRONHP_DO_NOT_ADJUST_FINAL_STATE --> Q = 327 keV

So now the Q-value is OK when running when the G4NEUTRONHP_DO_NOT_ADJUST_FINAL_STATE option. In the other case Geant4 tries to force energy conservation event-by-event, unsuccessfully.

For the moment, the issue is fixed when using the G4NEUTRONHP_DO_NOT_ADJUST_FINAL_STATE option. Fixing the other option can be done for this specific case but will require a much greater effort. This can be done in a future. However, it is not possible, in all the cases, to force energy conservation when using G4NDL data libraries. Only in some specific cases.

Regards,

Emilio
Comment 3 michel.maire 2019-06-17 13:14:58 CEST 
HI Emilio,

Thanks for your investigation.
Some time ago (more than one year ...) there were discussions within hadronic group about the necessity to simplify, reduce the number of environmental variables in ParticleHP, and do not adjust final state by default.
I have never seen any progress in this direction ...