| Summary: | An unphysical peak in the momentum distribution of pi- after interacting with thin slices of beryllium | ||
|---|---|---|---|
| Product: | Geant4 | Reporter: | Sam Grant <sgrant> |
| Component: | physics_lists | Assignee: | Alberto.Ribon |
| Status: | ASSIGNED --- | ||
| Severity: | minor | CC: | Vladimir.Ivantchenko |
| Priority: | P4 | ||
| Version: | 11.0 | ||
| Hardware: | All | ||
| OS: | All | ||
| Attachments: | Three slides which briefly illustrate the problem. | ||
Thanks for reporting this problem. Which physics list are you using? If I have understood correctly, your primary particle is a 500 MeV pion-, on a 50 cm thick beryllium target, right? Hi Alberto, Thanks for your reply. Yes, that's correct. It's a 500 MeV pion- beam incident on a 50 cm thick beryllium target. Thanks, Sam (In reply to Alberto.Ribon from comment #1) > Thanks for reporting this problem. > Which physics list are you using? > If I have understood correctly, your primary particle is a 500 MeV pion-, on > a 50 cm thick beryllium target, right? And the physics list is QGSP_BERT, although the issue also appears with ShieldingM. (In reply to Sam Grant from comment #2) > Hi Alberto, > > Thanks for your reply. Yes, that's correct. It's a 500 MeV pion- beam > incident on a 50 cm thick beryllium target. > > Thanks, > Sam > > (In reply to Alberto.Ribon from comment #1) > > Thanks for reporting this problem. > > Which physics list are you using? > > If I have understood correctly, your primary particle is a 500 MeV pion-, on > > a 50 cm thick beryllium target, right? I have checked whether the problem is due to hadronic physics - in particular to the Bertini (BERT) model which is used in most physics lists (including the two mentioned - QGSP_BERT and ShieldingM) for dealing with pion inelastic interaction for projectile (pion) kinetic energy below a few GeV. Forcing inelastic interactions of 500 MeV pi- on Be target, with Geant4 versions 11.1.p02 and 10.1.p03, and using BERT hadronic model, I see a consistent spectrum of secondary pions, and similar between these two versions of Geant4. So, it appears very likely that the problem is not due to hadronic physics. As an extra check, I would suggest to try out (using Geant4 11.0.p02 or newer versions) the physics list QBBC or QGSP_INCLXX : in the first case, Binary Cascade model (BIC) is used for pions below 1.5 GeV, whereas in the second case, Liege intranuclear cascade (INCL) is used for pions. If you see the same unphysical peak in these two cases, then the problem is certainly not hadronics - giving that you see the same problem with 3 completely different hadronic models (BERT, BIC, INCL). |
Created attachment 817 [details] Three slides which briefly illustrate the problem. Hello, When simulating a beam of pi- passing through a beryllium target, using G4beamline-3.08 (geant4-11-00-patch-02), I observed an unphysical peak at ~50 MeV in the momentum distribution of pi- exiting that target. When running the same simulation using G4beamline-3.06 (geant4-10-02) I do not see the peak, indicating that this is likely the result of a bug in geant4-11. This peak is only noticeable when I set the maximum physics step through the material to <= 0.163 mm (the 'maxStep' parameter in G4beamline), so it may be being caused by an issue with particle tracking through thin slices of material. I used QGSP_BERT in this study. I have attached some slides which illustrate the problem. Thank you, Sam