Undergeneration of neutrons in DT fusion

Physics Processes, Models and Cross Sections
1

Hello, I am currently trying to simulate the neutron production of 100 keV deuterons into a thin (~1 um) thick target of tritium. My hand calculations predict ~23000 neutrons per 1E8 deuterons fired at the target, but I consistently get ~21000, and I was wondering if anyone knew why that might be.

My simulation is just a vacuum filled box with a single small slab of pure tritium, and my tallying mechanism is a command line scoring shell around the slab, and the energy of every neutron entering the shell is tallied. I have tried tallying the number of Inelastic collisions in the slab itself between the deuterons and the tritium and that number is approximately correct, however the number of neutrons generated is still wrong.

I am currently using the physics list from the Hadr03 example, with the G4EmLowEPPhysics module added, but the issue seems to resolve/slightly over predict the number of neutrons produced when the EM list is removed. Is there a way to both keep the EM physics and accurately predict the number of neutrons produced?

Thanks in advance for any help, below is my G4 version, operating system, and CMake version.

Geant4 Version: 11.01_patch-02
Operating System: Ubuntu 22
Compiler/Version: CMake
CMake Version: 3.22.1

2

Dear Arthur,

I have not the answer to your question (unfortunately) but I have a problem which I think you solved : I try to have Hardr03 hadron physics to produced D-T reaction (which I did) but then charge particles do not deposit energy in detectors (alphas/deuterons, see my recent post on this forum). If I use EM physics at the same time, then I do not have any neutrons and alpha production. You seem to have both. Could you give me a hint for that?
Thanks a lot
Olivier

3

Oops…my problem with EM physics was the statistics!!! Not enough to see fusion reactions!!!
I’m sorry.

4

Hi Oliver, looks like you’ve already solved the problem, but if you would like any more tips, I’ve had some luck making my target thinner, and artificially increasing the cross-section for inelastic reactions through cross-section biasing!

Hope this helps,
Arthur

5

Ok thanks, I will have a look to that!