Created attachment 438 [details] Histograms of the simulation. Details are explained in the text. Hi, I am Chanyoung L1agi Lee and here to ask whether my result is physically possible or just a bug. Let me explain the setup. Based on Hadr02, one of the hadronic cases in Geant4 extended examples, I simulated the target-fixed experiment. Proton beam is set with varying beam energy. Uranium target also set but its target length is fixed to 6 mm. Physics list is set to FTFP_INCLXX. I scored ions, which are created by the inelastic collision, except Hydrogen. The 1st slide of the attachment shows the result of it. The histogram shows the momentum distribution of ions. (Scaled with the total number of ions, the cross section is scored instead of the event number.) The strange point is the peak at the 4 GeV. Independent from the beam energy, it always has the peak at ~4 GeV. In 2nd slide, I chose some ions and also drew momentum histogram. As shown, it seems like ions are separated by momentum. So I plotted the histogram of atomic mass of total ions with momentum cut of 2 GeV. 3rd slide shows the histogram of momentum of total ions with the cut. As the beam energy goes down, ions are separated by the cut clearly. Ions in the middle zone has large momentum, and the edge is small. Summary. 1. Collision always have the peak at the 4 GeV in the histogram of momentum. (It makes 4 GeV momentum with just 100 MeV proton beam!) 2. Ions created from the collision are separated by the momentum cut. Ions in the middle zone of atomic mass (about 70-160) make the high-momentum peak. I am confused with those results and want to know how physics works under them. Would you help me? Related research, theory, or paper... any of them will be great help. Thanks.
Thank you for your feedback! Two questions: 1. Which version of Geant4 have you used (e.g. to produce the plots)? 2. Have you try to repeat the same exercise with another physics list, for instance FTFP_BERT?
Hello, Can you please clarify if you are plotting total momentum or momentum per nucleon? A 4 GeV/c recoiling uranium nucleus only carries a few tens of MeV energy. Thanks, Davide Mancusi
(In reply to Alberto.Ribon from comment #1) > Thank you for your feedback! > Two questions: > 1. Which version of Geant4 have you used (e.g. to produce the plots)? > 2. Have you try to repeat the same exercise with another physics list, for > instance FTFP_BERT? 1. Tried with all versions from 10.1 to 10.3. 2. No. I will try with FTFP_BERT then check FTFP_BERT also reproduce the result.
(In reply to Davide Mancusi from comment #2) > Hello, > > Can you please clarify if you are plotting total momentum or momentum per > nucleon? A 4 GeV/c recoiling uranium nucleus only carries a few tens of MeV > energy. > > Thanks, > Davide Mancusi Oh, should have noticed that. It is total momentum. I got the value from the track with GetMomentum() function an a G4ThreeVector, and calculated its magnitude.
(In reply to Chanyoung L1agi Lee from comment #4) > (In reply to Davide Mancusi from comment #2) > > Hello, > > > > Can you please clarify if you are plotting total momentum or momentum per > > nucleon? A 4 GeV/c recoiling uranium nucleus only carries a few tens of MeV > > energy. > > > > Thanks, > > Davide Mancusi > > Oh, should have noticed that. It is total momentum. I got the value from the > track with GetMomentum() function an a G4ThreeVector, and calculated its > magnitude. got value from the track with GetMomentum() function as a G4ThreeVector. Sorry for typo.
(In reply to Davide Mancusi from comment #2) > Hello, > > Can you please clarify if you are plotting total momentum or momentum per > nucleon? A 4 GeV/c recoiling uranium nucleus only carries a few tens of MeV > energy. > > Thanks, > David(In reply to Alberto.Ribon from comment #1) > Thank you for your feedback! > Two questions: > 1. Which version of Geant4 have you used (e.g. to produce the plots)? > 2. Have you try to repeat the same exercise with another physics list, for > instance FTFP_BERT? I tried with FTFP_BERT, and it showed similar result. FTFP_BERT showed the peak at 3 GeV/c. One more, the cross sections in low momentum part is quite less than FTFP_INCLXX.
Hello, I have verified that for 100 MeV proton on Uranium, with the latest version of Geant4 10.4 (released 10 days ago), I get very similar results for all of the following hadronic models: Bertini intranuclear cascade (BERT), Binary cascade (BIC), INCLXX, Precompound/de-execitation (PRECO), and Bertini interfaced with native Precompound/de-excitation (so-called "BERP") instead of using its own internal nuclear de-excitation. Given that all of these models conserve energy-momentum, and are based on quite different assumptions, it is natural for me to conclude that the peak around 4 GeV/c in the total momentum of fragments is indeed a physical effect, and not a bug or an unphysical artifact of the simulation. Unfortunately, I am not able to explain quantitatively this result and I am not aware of any paper where this is discussed. However, I have a rough, intuitive explanation, which relies on energy-momentum conservation. Given the relative small momentum (100 MeV/c) of the proton projectile on the Uranium nucleus at rest, the fragments (of any kind: light, medium or heavy) must have a momentum and energy which is allowed by energy-momentum conservation. In particular, their momenta must roughly compensate each other, and their kinetic energy must be the Q-value of that reaction. So, if there are 2 medium fragments, it is very likely that these carry the two highest momenta, and should be roughly "back-to-back" to conserve the momentum of the reaction. Given that available kinetic energy from these reactions (with 2 medium fragment) cannot vary too much, it means that the momentum of these two fragments cannot fluctuate too much from one reaction to another, always considering two medium fragments in the final state. It turns out that this value is around 4 GeV/c , which I could not predict, but it makes sense because it corresponds to an average momentum-per-nucleon of about ~40 MeV, with a corresponding average kinetic-energy-per-nucleon of about ~1 MeV. In the much less frequent case that the reaction (100 MeV/c proton on Uranium) produces one heavy fragment plus a few very light fragments, then the momentum conservation and the kinetic energy available in the reaction forbids the possibility that the heavy fragment carries a large momentum, otherwise it could not be balanced by the remaining few light fragments. I hope that this simple and rough explanation could be acceptable to you: in any case I cannot do anything else and therefore I am going to close this ticket. Regards, Alberto
