Charge exchange model (G4ChargeExchange) for 7Li(p,n)7Be reaction at lower energy region

Physics Processes, Models and Cross Sections
1

Hello Everyone,

I have been working on simulating the energy spectra of p,n reaction on 7Li target (thin) at lower energies up to region of 20 MeV. The physics models which I have implemented in my code for this purpose are shown below:

                       Hadronic Processes for proton

Process: hadElastic

    Model:              hElasticLHEP: 0 eV  ---> 100 TeV

 Cr_sctns:            GheishaElastic: 0 eV  ---> 100 TeV

Process: chargeExchange

    Model:           Charge Exchange: 0 eV  ---> 100 TeV

 Cr_sctns: Glauber-Gribov nucleus nucleus: 0 eV  ---> 2.88022e+295 J  

 Cr_sctns:            GheishaElastic: 0 eV  ---> 100 TeV

Process: inelastic

    Model:                      FTFP: 4 GeV ---> 100 TeV

    Model:            BertiniCascade: 0 eV  ---> 5 GeV

 Cr_sctns: Glauber-Gribov nucleus nucleus: 0 eV  ---> 2.88022e+295 J  

 Cr_sctns:          GheishaInelastic: 0 eV  ---> 100 TeV

===========================================================

As far as I understand, BertiniCascade Model produces break up continuum neutrons whereas the Charge Exchange Model (G4ChargeExchange) covers only the resonance peak of neutrons in this case.

However, the energy spectrum in the lab frame that is obtained from simulation is different from the experimental result. Comparison plot for a particular angle is attached (in Fig. blue: experimental result, red: simulation result) herewith. The peak positions and peak width of resonance neutrons are not fully matched with that of the experimental results. A discrepancy has also been observed in energy spectra of continuum neutrons at less than 5 MeV region. Note that similar type of result is observed using geant4.10.02 and geant4.10.05.

It would be helpful if someone could suggest how I could improve this simulation result. Do I need to use some other physics model or cross section?

Is the G4ChargeExchange model capable of reproducing resonance peaks of 7Li(p,n)7Be reaction and/or this kind of reaction at the lower energy region (upto 20 MeV)?

I greatly appreciate any kind of suggestions/opinion/information.

Thank you.Neutron_E_Spec_LabFrame

2

Dear S. Seth,

I cannot tell you about the chargeExchange model, I have no experience with it. What I can say is that I would say that you have maybe not defined your physics list properly (the GheusaElastic cross section is included twice), but I am not sure about this.

Other possiblity is to use the G4ParticleHP physics list with the G4TENDL data libraries. Then you will be using ENDF/B-VII.1 to model the (p,n) reactions. Maybe you will get better results. The only issue is that you should use G4ParticleHP only below 10 MeV (have a look to the README file inside the G4TENDL data library), since the cross section for 7Li(p,*) in ENDF/B-VII.1 extends only up to 10 MeV.

To use G4ParticleHP for protons, you can use the QGSP_BIC_AllHP physics list. You should also download the G4TENDL data libraries from the geant4 web and set the environmental variable “G4PARTICLEHPDATA” pointing to the G4TENDL data library folder.

Regards,

Emilio

1 Like
3

Dear Emilio,

Thank you for your suggestions and pointing out the mistake in implementation of cross-section.

When I implemented the following physics models (Bertini Cascade + Charge Exchange), I obtained the similar result which I attached in my previous post. (Please note that, in this work only for the inelastic process of proton I added a data set for the cross section : G4BGGNucleonInelasticXS). I have attached the simulation result for 8 MeV protons (Red: simulation result, Blue: experimental result) using the physics model shown below.

                       Hadronic Processes for proton

Process: hadElastic

    Model:              hElasticLHEP: 0 eV  ---> 100 TeV

   Cr_sctns:            GheishaElastic: 0 eV  ---> 100 TeV

Process: chargeExchange

    Model:           Charge Exchange: 0 eV  ---> 100 TeV
 Cr_sctns:            GheishaElastic: 0 eV  ---> 100 TeV

Process: inelastic

    Model:                      FTFP: 4 GeV ---> 100 TeV

    Model:            BertiniCascade: 0 eV  ---> 5 GeV

 Cr_sctns:       Barashenkov-Glauber: 0 eV  ---> 100 TeV

Cr_sctns:          GheishaInelastic: 0 eV  ---> 100 TeV

In addition, according to your suggestion we have used QGSP_BIC_AllHP physics list and set the environmental variable “G4PARTICLEHPDATA” pointing to the G4TENDL data library folder. I have shown below the physics list for protons we obtained after this modification. The neutron spectrum in the lab frame at a particular angle for 8 MeV incident proton beam is attached (Red: simulation result, Blue: experimental result). This result does not match well with the experimental result whereas the peak position looks okay.

 Hadronic Processes for proton ( from QGSP_BIC_AllHP physics list)

Process: hadElastic
Model: hElasticCHIPS: 0 eV —> 100 TeV
Cr_sctns: Barashenkov-Glauber: 0 eV —> 100 TeV

Process: protonInelastic
Model: QGSP: 12 GeV —> 100 TeV
Model: FTFP: 9.5 GeV —> 25 GeV
Model: Binary Cascade: 199 MeV —> 9.9 GeV
Model: ParticleHPInelastic: 0 eV —> 200 MeV
Cr_sctns: ProtonHPInelasticXS: 0 eV —> 200 MeV
Cr_sctns: Barashenkov-Glauber: 0 eV —> 100 TeV

I made a comparison of simulation-results for these two sets of physics models. It looks Bertini Cascade + Charge Exchange model provides better results than that obtained using QGSP_BIC_AllHP physics list.

Please let me know whether we need to add any other model along with the QGSP_BIC_AllHP physics list or I made any mistake in implementation of this physics list.

I appreciate any suggestions/opinions.

NeutronSpecta__LabFrame_diff_PhList.pdf (74.3 KB)

With best regards and thanks
Susnata

4

I think that what you have done is correct and, I am afraid, this is the best you can do right now with Geant4!
Regarding the Charge-Exchange process that you are using, I think it is a simplistic model, not yet validated and used extensively, and by default it is supposed to be used above 20 MeV. This means to me that we should not expect too much precision when using it below this energy, as in your case.
Anyhow, the author of this model has been informed about your question and I hope he will comment.
Another try that you might consider is to use INCL intra-nuclear cascade model, for instance via the physics list QGSP_INCLXX.

1 Like
5

@ribon Thank you for providing this useful information and also for your suggestion. According to your suggestion, I have executed the code with QGSP_INCLXX physics list. The physics models for proton that I obtained using this physics list are listed below:

    Hadronic Processes for proton

Process: hadElastic

    Model:             hElasticCHIPS: 0 eV  ---> 100 TeV

 Cr_sctns:      ChipsProtonElasticXS: 0 eV  ---> 100 TeV

 Cr_sctns:            GheishaElastic: 0 eV  ---> 100 TeV

Process: protonInelastic

    Model:                      QGSP: 15 GeV ---> 100 TeV

    Model:                     PRECO: 0 eV  ---> 2 MeV

    Model:           INCL++ v5.2.9.5: 1 MeV ---> 20 GeV

 Cr_sctns:       Barashenkov-Glauber: 0 eV  ---> 100 TeV

 Cr_sctns:       Barashenkov-Glauber: 0 eV  ---> 100 TeV

 Cr_sctns:          GheishaInelastic: 0 eV  ---> 100 TeV

As I found that with this physics list interaction rate decreased. I ran the code for about 112 M protons of energy 13 MeV.

Result is attached herewith (Red Line: simulation, Blue circle : experiment). Simulation result shown here is for 253 neutrons.

nspectra_Ep13MeV_QGSP_INCLXX

Please let me know if you have any other suggestions.

Thanks and best regards

Susnata

6

Hello,

I would suggest to use reference Physics Lists with the Binary cascade and/or HP (high precision) neutrons. For example, QGSP_BIC, QGSP_BIC_HP, QGSP_BIC_AllHP. It is the best what we can propose for today.

VI

7

Please look here. Probably it will help.