Strange peak at 2.27 MeV in Co60?

Physics Processes, Models and Cross Sections
1

Hi everyone,
I am trying to get Co60 spectra from a 7-physical volumes germanium detector. The source is kept at around 3 cm from the front surface.
image

For ion I am using

/gps/ion 28 60 0 2505.8

I am getting a strange peak at Co60 around 2.7 MeV.

image
image1129×847 17.1 KB

Can someone please help me attribute the reason for it?

2

Hi,

I have two questions:

  1. The last of the 4 numbers after the /gps/ion command corresponds to the energy of the ion. Why do you need to setup non-zero energy? I think if the source is at rest, that number shall be zero, no?
  2. in real life, there is always a non zero probability of having the so-called “summing”, in this case, two gammas (1.2MeV, 1.3 MeV) entering the sensitive volume and leaving all the energy at once (2.5MeV). In real life sometimes it is not possible to separate the components of the summing, in Geant4 simulation you may want to separate them. So, question: does your scoring/SD separate energy deposition from different particles? If the answer is no, I would say the extra peak that you see corresponds to a summing peak. If this is correct, you will see the same effect independently of the decay.

Please, do not mix summing effect (which depends on the physics case and detector efficiency) with the pile-up effect (which is caused by random independent events happening within a very small time window, and therefore it depends only on the rate). Both effects may happen at the same time.

I hope that may explain the situation.

Best,
Alvaro

3

Thanks for your reply.

I am trying to simulate only the gamma rays (not the beta electrons) that is why I am starting from the excited state of Ni60 instead of Co60.

Please correct me if I am wrong here, I think the energy in the four parameters of gps/ion is the excitation energy as opposed to the kinetic energy.

4

Actually the 'strange peak’was talking about the sharp bump just before the sum peak.
Sorry for the incorrect terminology.

5

You’ve done it correctly. The fourth argument is the excitation energy. You would specify the kinetic energy separately, which you don’t need to do here (in real life, there would be a small nuclear recoil that gives a Doppler broadening to the gamma lines, but that’s much smaller than your detector resolution).

Your simulation is fine to see the “pure” gamma spectrum, but be aware that electrons can also deliver energy in germanium detectors. So if you plan to compare your simulation to a real, built device, you should go back to letting the Co-60 itself decay.

6

The ‘strange’ peak is at ~2.25 MeV, not 2.7 MeV. There are actually three such peaks; each one corresponding to a main full energy peak but ~0.2 MeV lower. Their shape on the log plot is very reminiscent of the triangle peaks in HPGe detectors due to inelastic fast neutron scattering in the germanium. Unfortunately, there are no neutrons involved here. So, it’s a little puzzling.

7

Thanks Mike,
Ya this was just a very toy model sort of spectra so I kind of supressed beta decay electrons.

8

Hi John,

Thank You for your reply.

Ya sorry the peak I was referring to was 2.25 (not 2.7) Mev

Actually I am not sure of it’s purely by chance or has something substantial physically but
the Sharp edge of the first two peaks seem to occur roughly (3-4 keV) of the Compton edges for the two gammas of Co60.

Yes, indeed there is no neutron involved as far as I am aware.

9

I think the sharp high energy edges of those ‘triangles’ are the Compton edges for the respective gammas. They are, as best I can tell with a ruler, at exactly the correct energy for the Compton edge. The shape in between each full energy peak and its Compton edge are just due to dissipative losses from the full energy peak.

1 Like
10

I did not follow the discussion in detail. So, I am sorry if this mail is out of game.
With example rdecay01, I noticed a somewhat different energy spectrum if we start with an excitation energy of 2505.753 keV (red) or 2505.8 keV (blue)

Ni60.mac.txt (398 Bytes)

Ni60
Ni60998×883 31.3 KB

1 Like
11

I remember this, Michel, with other elements! The nuclear excited states, with particular decays, are tabulated for a specific energy value. If you set the excitation to exactly what is in the table, then G4 will do the decays to the tabulated states.

But if you “round off” (like 2505.8 instead of 2505.753, in this example), then G4 will do a “generic” de-excitation. I think it tries to throw off a gammas to get to one of the known states, but it can also throw off the whole excitation energy at once, or jump to some lower state than you intended, etc. Hence the blue vs. red differences that Michel has shown.

1 Like
12

Thanks John, for pointing this out.

But I was wondering that the 2.5 Mev peak is not a true photoelectric peak. So can we still get Compton edge and continuum for that.

Sorry this is not directly related to Geant but can you please share some comments here.

Thank You

13

Thanks Michel for bringing this up.

Yes I am aslo puzzled at this, but I thinks Mike has cleared it.

14

Thank You Mike for clarifying this, it really explains it nice

I see ! so does this mean that the 2.5 Mev peak that I am getting is overestimated as compared to real scenario and has two contributions

1.) True coincidence due to the two (1.17 and 1.33) gammas being depositing their energies simultaneously.
2.) Direct energy transition from whole excitation energies as you pointed out?

Thanks a lot for your replies.

15

Maybe. If you’re really wanting to model the Ni-60 de-excitation, you should try re-rerunning with the “exact” energy used in G4:

/gps/ion 28 60 0 2505.753

(as Michel noted). If you do this, may be post the resulting spectrum and see how it differs from what you have now?

16

That makes sense. The red spectrum (using the exact excited level energy) agrees with the 60Ni decay scheme following 60Co beta decay. The true decay scheme also has two weak lines at 2.5057 MeV (intensity 0.00005) and 0.3469 MeV (intensity 0.008) that are too weak to be seen in the red spectrum. The blue spectrum seems to be out to lunch.

1 Like
17

The 2.5 MeV peak is due to either a single gamma ray from the deexcitation of the 2.50573 MeV level (probability ~ 0.00001) or the deposit of energy from 1.33250 MeV and 1.17321 MeV gammas from the cascade dexcitation of that level (probability 1). For the latter, the cascade time scale is effectively instantaneous, that is, too short for the detector to tell the difference between two gammas with summed energy or a single of the same energy. So both are effectively normal full energy peaks. There may be ways using sophisticated electronics to distinguish the two effects, but the simulation doesn’t include it.

1 Like
18

Thanks, Mike, I’ll try doing this.

19

Ah, I see. So, the detector “thinks” 2.5MeV to be an independent gamma ray.
Thanks, John, for the super helpful explanation.