Hi, I've created a simple example that tried to replicate the results shown in the "Physics Reference Manual" for the HP Capture Process using 15MeV neutrons and U-238. If I plot only the individual energies of each one of the gammas, the results are in perfect accordance with the ones of the manual. But then I tried to sum up the energies for all the gammas that come out in each event, and in the result I have events that have up to 40MeV of energy released in gammas and others with practicly no energy at all! If I have 15MeV neutrons being captured and U-239 has a binding energy of ~4.9MeV, shouldn't the sum of the energy be arround 19.9MeV? Following is a copy os part of my Physics List: theParticleIterator->reset(); while ( (*theParticleIterator)() ){ G4ParticleDefinition* particle = theParticleIterator->value(); G4ProcessManager* pmanager = particle->GetProcessManager(); pmanager->SetVerboseLevel(2); G4String particleName = particle->GetParticleName(); if (particleName == "neutron"){ // neutron // construct processes for neutron theNeutronCapture = new G4HadronCaptureProcess("hp_capture"); //register HP model for neutrons G4NeutronHPCapture* theHPCaptureModel = new G4NeutronHPCapture; theHPCaptureModel->SetMaxEnergy(20*MeV); theNeutronCapture->RegisterMe(theHPCaptureModel); // add processes pmanager->AddDiscreteProcess(theNeutronCapture); } //if (particleName == "neutron") } //while ( (*theParticleIterator)() ) Thank you for your time. Best regards Alexandre Lindote
Dear Alexandre, for some materials, energy conservation is only on avarage. This in particular the case for those materials, where for capture gammas no data on corellations are available. Here a decision was to be made. We can either have the spectra and mean multiplicities correct, or we can bias the spectra in an attempt of conserving the energy event by event. Most applications of this code at present are of dosimatric nature. They care very much about the correct energy spectra and mean multiplicities and not so much about particle corellations within individual events. Hence the current implementation. Many greetings, Hans-Peter.
