Do incident particles see less material when atoms are scattered?


Say I have a perpendicular beam of neutrons striking an aluminum slab with a perfect crystal lattice. If the first neutron in the beam is energetic enough, it can scatter an Al atom away from its lattice site and the second neutron in the beam will “see” one less Al atom in its path (I’m considering a very ideal case for simplicity).

However, from my understanding, Geant4 doesn’t account for crystal structure or atom positions when using the Geant4 Material database, NIST compounds, atom mixtures, etc. (I found some relevant links stating this at and

And so, my question is how does Geant4 account for atoms that have already been scattered away from the path of an incoming particle, if at all? Does Geant4 keep track of how much material each incident particle “sees” for every run despite how it doesn’t account for the crystal lattice? Or is the likelihood of scattering the same for each particle in the beam such that each particle sees the same material for every run?

It doesn’t. Each event is independent, and sees exactly the same geometry as was configured at the beginning. Geant4 does not model the atomic-level structure of the material, nor does it keep track of that structure from event to event.

The best way to think about it is this: G4 treats volumes and materials as perfect continua. As the track is thrown, G4 picks a position for the track to stop on a step. The assumption is that no matter what that position might be, there will be an atom there to interact with. If another position were chosen nearby, there would be an atom at that other position. There’s no lattice structure, no grain structure, no variation in density with position.


Thank you for the detailed response! I thought this was the case, but I’m glad to have someone confirm this.

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.