Anomalous primary ionisation for Ne / 5% CH4 using HEED

[SSJ]

Hello All,

I have been using the example fe55.C which is available in the garfieldpp/Examples/Heed directory for checking the amount of primary ionisation that a particular photon can produce for different gas mixtures. I have used 2 photon energies: 5900 eV and 8000 eV , and two gas mixtures: Ar / 5% CH4, and Ne / 5% CH4 at 1 atm and 293.15 K. Further, I have reduced the dimensions of the cylinder inside which the x-rays can ionise the gas mixture.

Original dimensions in fe55.C Example:
diameter = 7.8
length = 10.
The dimensions I am using:
diameter = 0.1
length = 0.45

Hence, because of the low dimensions that I am using, it is expected for most of the photons to escape without interacting at all. Hence, while filling the histogram, an if condition of if (ne > 0) hElectrons.Fill(ne); is used by me.

The reason for using low dimensions is to observe escape peaks, that is when K-alpha X-rays of Ar or Ne can escape from the medium if an inner electron is extracted from their atoms because of the incident radiation.
The W-values of Ne / 5% CH4 and Ar / 5% CH4 given by HEED are: 34.995 eV and 26.4256 eV respectively.

Hence, now that we know the W-values, we can expect beforehand what should be the difference in the number of primary electrons in the primary peak and secondary peak for the mixtures, knowing the K-alpha X-ray energy values of Neon and Argon. From hyperphysics, the values for Neon and Argon taken by me are 848.61 eV and 2957.682 eV.

Hence,
Expected difference for Argon mixture: K-alpha energy / W-value = 2957.682 / 26.4256 which is around 110
Expected difference for Neon mixture: 848.61 / 34.995 which is around 25

The results from HEED for Ar / 5% CH4 seem perfectly as expected, that is there is a difference of about 110 primary electrons between the primary peak and the secondary (escape) peak, but I can’t understand the results for Ne / 5% CH4. For 5.9 KeV photons, the secondary peak is somewhere I wouldn’t expect based on my calculations, and for 8 KeV photons, the expected primary peak becomes the secondary peak which is a bigger mystery to me. So my question is that are there some physics processes that I am missing here?

Here are the attached 1D histograms showing number of Primary electrons obtained across multiple number of independent events:

1. Ne / 5% CH4, 5.9 KeV:
   It is seen that difference between primary electrons for both peaks is more than 25. I can’t understand why.
   
   
2. Ne / 5% CH4, 8 KeV:
   Here, the expected primary peak becomes secondary. Again, I can’t understand the physics behind this.
   
   
3. Ar / 5% CH4, 5.9 KeV:
   Perfect match with calculations: Difference of about 110 primary electrons between both peaks.
   
   
4. Ar / 5% CH4, 8 KeV:
   Again a perfect match with calculations: Difference of about 110 primary electrons between both peaks.
   
   

Attached code:
fe55.C (2.7 KB)

In summary, my question is- why do the number of primary electrons produced by the X-rays in Ne / 5% CH4 show a sort of behaviour which I cannot predict, when the results for Ar / 5% CH4 seem to be perfectly as per my predictions. Are there some physical processes that I am not taking into account in case of Ne / 5% CH4 mixture?

Thank you.

[couet]

I guess @hschindl can help.

[hschindl]

Yes, I’ll look into it…

[hschindl]

Hi,
the structure in the Ne/CH₄ spectrum in the plot you sent is due to the fact that a significant fraction of the electrons (photo-electrons and Auger electrons) leave the geometry before depositing all of their energy. With the diameter of the tube increased to 3 mm and the half-length increased to 5 cm, I get the spectrum attached which looks more reasonable. The escape peak is visible but much less pronounced than in Argon since Neon has a smaller fluorescence yield.
larger_tube.pdf (16.9 KB)

[SSJ]

Hello,

The attached spectrum indeed looks reasonable, with an escape peak which is more or less 30 primary electrons away from the main peak. The less pronounced escape peak makes sense by taking the concept of fluorescence yield into account.

So what can be the reason for a significant fraction of the electrons to leave the geometry in Neon, but not in Argon? Is it simply because of lower density of Neon, as compared to Argon? Further, both of my Ne / 5% CH4 spectrums show the “anomalous” peaks at about 100 electrons. Is there something special about this number? Shouldn’t one expect the peaks to shift rightwards as a more energetic photon is used.

Thank you.

[hschindl]

@SSJ

So what can be the reason for a significant fraction of the electrons to leave the geometry in Neon, but not in Argon? Is it simply because of lower density of Neon, as compared to Argon?

I think it’s simply because the electron mean free path and, consequently, the practical range are larger in neon than in argon.

I don’t think there’s anything special about the peak but I can’t say off-hand why the spectrum has this exact shape. If you want to investigate a bit more, you could plot the distribution of the secondary electrons produced by the photo-electrons and Auger electrons (loop over the ne electrons and retrieve their coordinates using GetElectron).

[SSJ]

Alright! Thanks for the direction.