Low energy Cu(p,x) cross section

Dear Jesutofunmi,

Thanks for your post.

The LOW-MAT card is unrelated to (p,n); it sets the correspondence between FLUKA materials and low-energy neutron library materials for (n,*) reactions and elastic scattering below 20 MeV.

Looking back at the RELEASE-NOTES, the difference between FLUKA 4.1.1 and 4.2.1 impacting Cu(p,n) is the near-threshold refinement in the proton reaction cross section for {63,65}Cu. Incidentally, there was a typo in the RELEASE-NOTES (Co<->Cu) which we have spotted and fixed thanks to your post.

The 63Cu(p,n)63Zn yield is unaffected by this improvement, since the threshold for this particular channel is at around 4.2 MeV, well above where the aforementioned change applies.

Instead, for 65Cu(p,n)65Zn the improvement is visible in the excitation function at energies immediately above threshold (2.17 MeV), as displayed in the figure below. Note the cyan curve (FLUKA 4.2.1) is in better agreement with experimental data extracted from EXFOR (black symbols) in the near-threshold region than the red curve (FLUKA 4.1.1).

65Cu_pn

However, the improvement is rather between 2.17 and 4 MeV. In the 5-10 MeV domain that you quote I do not see a difference, neither in 63Cu(p,n) nor in 65Cu(p,n). Could you provide evidence? Specifically:

  • a plot displaying the alleged new peak and any necessary input files for your case study.
  • Where do you see the “new peak”? Comparing different FLUKA versions? Comparing against experimental data? Please clarify.

With kind regards,

Cesc