Physics List

[hapol]

[NOTE: this is a copy of the original issue from Piotr, sent 25/11/14 and lost when moving the group]

I've made some tests of the physics libraries and compare the longitudinal range, straggling and radial range. As I was using ionGasModel combine to emstandardphysics_opt3 because it resolved the problem we had on the energy fluctuations in the elastic reaction 13Cpp13C, I wanted to see why this model works fine. Here are the physics libraries that I was comparing :

emstandard : the standard electromagnetic library in Geant4 emstandard_opt3 : G4IonParametrisedLossModel used for ion ionisation based on ICRU73 ion stopping data and G4NuclearStopping process added. emLivermore : based on the Livermore low-energy electromagnetic model. ionGasModel : only used in TestEm7 exemple. It was created specialy for low-density gas volume to disable the effective charge approach of the ion. In the example it used on top of emstandard library. Here are the range values of a 60MeV 12C beam :

in H2_STP (0.084 mg/cm3)

in D2_STP (0.167 mg/cm3)

in iC4H10STP (2.67 mg/cm3)

As the ActarSim physics libraries are based on Gamma therapy example, I noticed that in addition to the electromagnetic library one can add a specific ion library (ionstandard which load the HadrontherapyIonStandard library) and a stopping library (G4StoppingPhysics).

I tested these extentions with ionGasModel in iC4H10 and it seems that ionStandard library doesn't work fine.

[hapol]

[NOTE: this is a copy of the original issue response from Yassid, sent 26/11/14 and lost when moving the group]

Hi Piotr

Thank you very much for the nice report. I have a question about these plots. When you use IonGasModel, do you use it together with any other library? I understood from the email that the good results are obtained when you combine ionGasModel+emstandardphysics_opt3 but from the plots I infer that only one library at a time was used for each calculation.

Best regards Yassid

[hapol]

[NOTE: this is a copy of the original issue response from Piotr, sent 26/11/14 and lost when moving the group]

Hi Yassid, In TestEm7, ionGasModel is set as an emPhysiclibrary, combined with emstandardphysics(_opt0), that's why I used this way here. I am looking if there is no incompatibility with other emlibraries but I think it is ok. In the actual version of ActarSim, ionGasModel is set as a add-on to the emlibrary and maybe it is better to use it like this. I will have a look at the range of ionGasModel + emstandardphysics_opt3 how are the ranges of ions.

[hapol]

[NOTE: this is a copy of the original issue response from Piotr, sent 26/11/14 and lost when moving the group]

Ok so I updated the plots with the combination emstandard_opt3 + ionGasModel

For H2STP :

D2STP :

and iC4H10:

[hapol]

[NOTE: this is a copy of the original issue response from Piotr, sent 26/11/14 and lost when moving the group]

Since ionGasModel set the ionisation model to BraggIonGas and BetheBlochGas Models, I will set it as an add-on to the emphysicslibrary, with a flag to be sure that we use it after defining the emlibrary.

emstandardphysics_opt3 seems more accurate since it use the nuclear stopping process but I noticed that there is no electron coulombScattering process included as in emstandard.

Now there is another library I didn't mentioned, it is emstandardphysics_opt4. Indeed I have a problem when I use it for H2 and D2_STP, it stop working after loading the msc process for electron. But it works for iC4H10 gas so maybe there is a problem for low density gas. Anyway opt3 should be enough (opt4 differs mainly for electron process).

[hapol]

[NOTE: this is a copy of the original issue response from Piotr, sent 27/11/14 and lost when moving the group]

I found a problem using the ionGasModel : the electron cut gives weird results. I wanted to cut off the delta electrons by putting a huge cut on them (/ActarSim/phys/setECut 10 km) but the track become very short. In fact above 2 mm cut the ion track is changing and there are less steps. You can see the track of a 80MeV Carbon12 in iC4H10 using ionGasModel + opt3 :

for a 1mm cut :

2mm :

3mm :

10 km :

When I use only emstandardphysics_opt3, here are the tracks :

for 1 mm :

3 mm :

10 km :

I don't know why it behave like that but if we can't turn off the delta electrons (or at least reduce there number) without changing the track length, maybe ionGasModel willl be not suitable for our simulations, despite it has the best agreement with TRIM. I will try to see if we can fix that and also we should compare ActarSim to the real datas to see which physics library we should use.