| Summary: | Neutron elastic scattering does not conserve quantum numbers & 4mom | ||
|---|---|---|---|
| Product: | Geant4 | Reporter: | Mikhail.Kossov |
| Component: | processes/hadronic/models | Assignee: | dennis.herbert.wright |
| Status: | RESOLVED FIXED | ||
| Severity: | critical | ||
| Priority: | P2 | ||
| Version: | 7.1 | ||
| Hardware: | PC | ||
| OS: | Linux | ||
User defined isotope abundances are now handled in Geant4 cross section classes. Tags hadr-cross-V08-02-00 (hadronic/cross_section directory) and hadr-man-V08-02-00 (hadronic/management directory) contain the new code, which has been tested for most data set classes in the cross section directory. |
G4LElastic model (np case at present) sometimes has only a leading neutron in the alive projectile with the practically the same energy (no scattering, no recoil protons). In other cases all energy is transfered to the proton (?!) and the neutron stops and disappears (a fake charge exchange?). Only in about thirty pecent of events both a leading neutron (-1) and a secondary proton (0) appear: #-1,PDG=2112,C=0,B=1,Tkin=2.224731e+00 4M=(5.224152e+01,-2.362195e+01,0.299723e+02;9.417904e+02)M=9.395656e+02 # 0,PDG=2212,C=1,B=1,Tkin=8.202987e+00 4M=(-5.235558e+01,2.367352e+01,1.102679e+02;9.464753e+02)M=9.382723e+02 dP_z ---1.402402e+02 instead of 1.40e+02 --- (MeV/c) dE --- 10.4277 instead of 10.3731 --- (MeV) Nonconservation is dP_x=1.14 MeV/c, dP_y=0.52 MeV/c, dP_z=0.24 MeV/c, dE=50 keV The G4LElasticB which aims to substitute the the G4LElastic model provides two particles, but mostly two protons (!), which is absolutely unacceptable. In addition the G4ElasticProcess hadronic process, which uses G4LElastic or G4LElasticB models is not sensitive to the user defined isotop content of the element, so e.g. the deuterium target can not be properly simulated. The user defined isotop content is always substituted internally by natural abundance of isotopes.