Scoring in amorphous vs crystal volume

Geometry, Fields and Transportation
1

Hi everyone.

I was wondering if you could help me with a doubt that I have.

I am trying to simulate an Al2O3 crystal with 1mm3 being irradiated with a proton beam.
Initially I considered it an amorphous cubic volume and scored the total deposited energy in the crystal (the goal is to scorer this quantity per particle step - I am working on the scorer now).
I then used the G4CrystalExtension where I put all the crystal parameters in G4CrystalUnitCell (below) and scored the same quantity.
I was expecting different results between the two simulation since now I was including the crystalline effects, however, I got the same data.
Is this suppose to happen or am I doing something wrong?

Plus, I am new to geant4 so I am still trying to understand everything.

Thanks in advance!
Cristiana

DetectorConstruction.cc

// define crystal
//

G4Box* solidCryst = new G4Box(“crystal”, cryst_dX/2, cryst_dZ/2, cryst_dZ/2);

G4ExtendedMaterial* CrystalMat = new G4ExtendedMaterial("crystal.material",cryst_mat);

CrystalMat->RegisterExtension(std::unique_ptr<G4CrystalExtension>(new G4CrystalExtension(CrystalMat)));
G4CrystalExtension* crystalExtension = (G4CrystalExtension*)CrystalMat->RetrieveExtension("crystal");

crystalExtension->SetUnitCell(new G4CrystalUnitCell(4.75 * CLHEP::angstrom,
                                                    4.75 * CLHEP::angstrom,
                                                    12.99 * CLHEP::angstrom,
                                                    CLHEP::halfpi,
                                                    CLHEP::halfpi,
                                                    2/3 * CLHEP::pi,
                                                    167));

CrystalMat->RegisterExtension(std::unique_ptr<MaterialExtensionData>(new MaterialExtensionData("ExtendedData")));
MaterialExtensionData* materialExtension = (MaterialExtensionData*)CrystalMat->RetrieveExtension("ExtendedData");
materialExtension->SetValue(57.);   

G4LogicalCrystalVolume* fBoxLogicCrystal = new G4LogicalCrystalVolume(solidCryst,
                                                                      CrystalMat,
                                                                      "crystal.logic");

new G4PVPlacement(0,
                  G4ThreeVector(),
                  fBoxLogicCrystal,
                  "crystal.physical",
                  logicWorld,
                  false,
                  0);
2

Hello Cristiana,

The crystal extension will only affect the results if the G4Channeling process is attached to the incident (charged) particle in question. This is the only physics process in Geant4 that makes use of this crystal extension. It is intended for crystal channelling (guiding of incident particles) and most useful when the particle is close to the channelling orientation (e.g. the plane). This channelling effect relies on some extra text files given in the channelling example and covers only a certain crystal axis for a certain crystal, e.g Si <100>.

Also, the channelling physics modulates the other physics processes through the generic biasing interface (i.e. more scattering close the nuclei, less between the planes) and a consequence of this is that the particle weight will quickly drop to zero, in which case the scoring of energy deposition or dose, that by default takes into account the weight, may not be correct.

Hope this helps,
Laurie

3

Hi Laurie.

Thank you for replying.

So I would need to have the text files specific for my crystal?
Do you know if this extension is usefull for straight crystals like the one I am simulating or only for bent crystals?

Again, thank you so much.
Cristiana

4

Hello Cristiana,

Yes, you would need the text files for a particular crystal lattice and also for a particularly orientation (e.g. <111> or <110> for Si). It absolutely works for straight crystals too where you will also see channelling effects. The files were originally produced using a separate code DYNECHARM++. See here:

Note though if you’re using a primitive scorer you should not include the weight when trying to accumulate deposited energy in the crystal as the weight will be wrong due to how the channelling works currently.

Cheers,
Laurie

5

Hi Laurie

The answer is helpful for me. I want to know how to count the energy deposition of each atom in the crystal and how to achieve it.

Thank you!
Kun ZHU

6

Hello Kun ZHU,

Energy deposition is not done on an atom-by-atom basis. The channelling process limits the step size to the channelling oscillation length / 100 for example. It is a continuum approach where we approximate the field of the individual atoms as a continuous potential well along a direction of travel. The regular Geant4 physics processes are modulated according to factors derived from the position in the potential well in the crystal.

A problem being fixed just now is that the weight is automatically updated with the biasing and eventually goes to zero (or a very small value), which is not correct. It is caused by using the biasing interface for it. If you do not use any other biasing you can safely ignore the weight when using any quantities and it should be correct.

You would have to histogram the energy deposition hits from the volume yourself along the direction of travel for example.

Cheers,
Laurie

7

Hello Laurie

I know that the energy deposition in the channeling example of Geant4 is not based on atoms, but if I want to know the energy deposition of each atom, how can I achieve it?

If I model each atom and get its energy deposition in this way, then what physical processes should be implemented or what should be called?

Thank you for your helping!
Best Regards!

Kun ZHU

8

By “energy deposition of each atom” what exactly are you picturing? Are you asking about the energy of each individual ionization event (because Geant4 assumes you’re ionizing atoms, not promoting valence electrons)? If that’s the case, you can set the electron production cut to zero and you’ll get all the secondaries created by ionization.

Or displacement events? G4 might treat these in the “microelectronics” processes, but I’m not certain.

Or non-ionizing energy loss? The latter can’t be attributed to an individual atom.

9

Hi Michael

I want to know how energy deposited on specific atoms within a crystal , and it can be used to better design crystals.

10

Hi Inevay:
I have another question. If the radionuclide inside the crystal decays, then the decayed particles have an initial direction.The channeling effect in this direction should be considered, but from the example in Geant4 only the cross section in a single direction is corrected.

How to solve it?