Fusor X-ray Spectra

[Jon Rosenstiel]

Here's what happened (see pic) when I pointed my Teledyne Isotopes U/5 FIDLER scintillation probe at my operating fusor.

In case you’re not familiar with the FIDLER… (Field Instrument for Detection of Low Energy Radiation). It is made specifically for the detection of low energy x-ray and gamma ray radiation. It uses a 5” diameter by 0.010” thick NaI(Tl) crystal coupled to a 5” pmt. (Ebay purchase. Had to replace the pmt and an open divider resistor to bring it back to life)

Ok, what’s with the photopeaks? I was under the impression that the fusor’s bremsstrahlung radiation would consist of a broad range of energies up to the acceleration voltage. And I don’t believe it’s characteristic radiation because the energy peaks change with input voltage. (38kV input, 34.3keV peak. 43kV input, 39.5keV peak)

I’m stumped here. What am I seeing?

Jon Rosenstiel

[DaveC]

Jon - Interesting results, to be sure. My expectation would also be that the upper end of the Xray spectra is approximately the top excitation energy.

A possible answer lies in the fusor geometry, which provides colliding ion(electron) beams at approxmately the energy of the input KeV. But colliding beams at 43 kV, have an actual energy about equal to 43 + 43 KeV or ~ 86 KeV and one would expect Xray bremstrahlung up to that energy.

I am not sure exactly what one gets in the way of equivalent collision energies, in the fusor, since that depends on where the ions(or electrons) begin their paths. So the KeV received could be somewhat less than the applied voltage, per beam, making the the Xray energy less that twice the voltage.

Your data indicates the energies are about 71.67% of twice the applied kV.

Very nice set of curves.

Dave Cooper

[Richard Hull]

interesting! Could this be some sort of re-emission (secondary) peak caused by the primary X-rays working inside the fusor shell metal? Just a wild stab here. Some sort of XRF "L" shell stuff on a heavy metal?

Richard Hull

[Jon Rosenstiel]

A few of things I'm going to try this weekend, maybe will help shed light on things.

1. Pb and maybe Al absorbers, various thicknesses.
2. Position the detector closer to the fusor, and try various positions.
3. Aim the detector into the veiwport. (From a distance)!

Suggestions welcome.

Jon Rosenstiel

[ChrisSmolinski]

How well calibrated is the probe? Something looks amiss... the
highest x-ray energy should be equal to the applied voltage, 38 and 43
kV respectively in your two runs. I see values in the graph of about 45
and 53 kV (eyeballing them).

Lead has some L edges in the 15 kv range, which could explain the
second, lower energy, peaks. As the photon energies get lower,
attenuation gets higher, until you dip below the edge. The L shell
fluorescence energies are in the 10 kV range, which also helps to
explain the peak.

It does look (to me) that you have a gain error (kv vs bin number)
between the two runs.

Otherwise it has the correct general "look" to it. The 'ideal' spectrum
shape would be a straight line, intersecting the X axis at the applied
voltage, going at a diagonal to the Y axis. This then is modified by the
various materials in the beam (plus effects of the target itself) giving
the characteristic peak at about three quarters maximum energy (give
or take depending on filtering).

If you had a thin piece of silver, you might see fluorescence at about
22 keV, or with some tin around 25 keV.

[Richard Hull]

Chris,

Remember that if Jon is doing fusion during the run he has a viable multi-mev proton generator inside as well!!! Those have gotta' do something crashing around inside the fusor as all their energy is disappated inside the device. There would be as many protons, recoiling tritium and He 3 atoms as neutrons.

Lotsa' stuff over the accelerator voltage in there to do all manner of gamma/x-ray related stuff. With, say, a half million exoergic fusions/second (only a quarter million neuts/sec) the inside of the fusor is a nasty place to be.

Richard Hull

[Carl Willis]

Hi Jon,

Interesting spectrum. I tried collecting some gamma spectra from my fusor when I was looking for prompt capture gamma emissions. I found the peaks I was looking for, but as in your case there were confusing peaks as well. There are so many things going on in a fusor that it is very hard to figure out what you might be seeing.

My first guess is that the higher-energy peak is a "real" peak corresponding to bremsstrahlung, since it is voltage-dependent. Also, there shouldn't be any other radiations of those energies that compare in intensity to the bremsstrahlung. Certainly the shape of the peak is a bit "off" if that guess is correct, but shielding is going to be good at removing the lower-energy constituents of the bremsstrahlung spectrum.

The lower "peak" might contain a prominent iodine escape peak superimposed on compton and backscattered photon counts. This peak is always ~ 25 keV below the photopeak and corresponds to escape of a characteristic I x-ray of this energy from the NaI detector volume, a process which is enhanced in thin detectors like yours.

Just guesses! Another experiment would be to look at an ordinary x-ray tube with a NaI detector and see if a similar spectrum is obtained. Also at these low energies, NaI(Tl) has a pretty poor energy resolution, which makes it hard to properly ID what is seen.

I'd be interested to see what energy you get for the Am-241 gamma. This is an easy test to see if you are calibrated correctly.

In my experiments, I saw the effects of having activated the sodium and iodine in the crystal by neutrons. The result was a lot of noise (a monotonic spectrum decreasing with increasing energy) in the lower-energy part of the spectrum that decayed away exponentially over time after the fusor was off. Beta (and to a lesser extent, gamma) emissions from the activation products were responsible.

Cool stuff! Please post more spectra when you get them.

-Carl

[Jon Rosenstiel]

Chris, Carl

The curves I posted are still in my mca's memory and I'm recalibrating now. (mca and hvps have been warming up 2+ hours)

Originally I used the 32keV K-shell x-ray of Ba (Cs-137 source) and the 59.5keV gamma peak of Am-241. I’m doing the re-cal using the same sources.

The re-cal changed things just slightly. With a fusor input 43kV input (red curve) the low peak is 11.1keV and the high peak is 39.7keV. (My posted graph shows 10.9keV and 39.5keV).

Applying the re-cal to the 38kV input curve (blue) gives a peak of 34.3keV. (Was 34.5keV)

Checking an x-ray tube spectrum is on my list of things to do.

Gotta’ go, getting past my bedtime!

Jon Rosenstiel

[Jon Rosenstiel]

Two more graphs....

The upper graph (fusor shell x-ray spectra) is similar in content to the graph attached to my original post, but my setup has changed considerably.

Instead of the 5" FIDLER detector I used a 2" Bicron unit. I also paid more attention to shielding in order to keep the mca's dead time under 30%. (Was over 80% with the FIDLER)! Not good for linearity.

Two things happened when using the FIDLER detector in a large x-ray flux.
One, the crystal seemed to develop what I call an “after-glow”. And two, my preamp was saturating. (I’m sure the 2” Bicron would suffer the same problems in a high flux situation, but its small size made it much easier to shield)

I first noticed the “after-glow” phenomenon while taking a spectrum of an x-ray tube. (More on the x-ray tube later) I was astonished when the spectrum on my mca kept “growing” after I had cut the x-ray tube’s power. The “after-glow” decays quickly (about 5 minutes) but it does a number on the spectrum.

About the x-ray tube…. It was powered up, the mca was taking a nice spectrum, and all seemed good…. until I heard some popping noises! I peered over the shield to observe the anode glowing bright red! The electron beam had burned a hole in the anode disc (platinum perhaps?) and the vacuum was ruined. Nuts!

The lower graph (fusor viewport x-ray spectra) shows the fusor x-ray spectra at various input voltages with the detector aimed into the viewport. In an effort to attenuate the viewport x-ray barrage I placed the Bicron detector as far away from the viewport as space would permit (53”) and mounted a sheet of lead with a 1/8” orifice onto the viewport.

Thoughts on what all this means are welcome.

Jon Rosenstiel

[Frank Sanns]

Nice work Jon. The "afterglow" that you are seeing may be due to x-ray activation of the NaI.

To analyse a sample by x-ray diffraction, you grind it up with potassium iodide or chloride then put it into the x-ray beam. You start with a pure white powder but when you open up to remove the sample after it has been in the x-ray beam, you get this lavender color that persists for a few minutes. After that it is white again. I do not know for sure if the same activation can happen with the sodium salt but it would not surprise me. The energy from the activation has to be lost as heat or light. If the latter then you will have many photons to detect long after the x-rays have stopped.

Frank S.

[Richard Hull]

There is neutron activation in NaI for sure and it is a big problem in the real world. I don't know it you are making neutrons or not.

Richard Hull

[Jon Rosenstiel]

During the x-ray tests I was making neutrons, about 1.7E+06n/s worth.

Even with an x-ray tube I could light up the NaI crystal, no problem.

Jon Rosenstiel

[3l]

Hi John:

You may be seeing the Emmisive decays of thallium in your detector. Thallium is a large emitter of gamma and xrays
when activated,so large in fact that thallium is used for hand held x-ray devices. Portable battle field xray that uses a 10 microcurie ingot of thallium as the xray source. Also Sodium when activated creates a gamma ray that can make a positron electron pair in the detector. The two gammas can cause many levels of the Iodine to shift up and down including the s p k & l
Shells. Oh yeah Iodine is used for cancer therapy when activated
It is an intense gamma emitter.
Another detector may be in order since all the elements in the detector can be activated.

Happy Fusoring!
Larry Leins
Fusor Tech

[Steven Sesselmann]

Interesting thread from 2004, but as always one of Jon's carefully prepared experiments are going to have educational value.

I found this after searching for "bremsstrahlung" and only two posts came up ?

Probably the only two spelled that way

I have a question....

In the D+D reaction the fusion product T or He3 are both positively charged particles with approximately 1 Mev energy, these charged particles are bound to hit the fusor shell, either once where they become embedded or possibly they could bounce a couple of times, but where are the x-rays or bremsstrahlung from these?

Where on the spectra would you expect to find them?

Steven

[Carl Willis]

Hi Steven,

There will be some characteristic x-rays (i.e. derived from electronic transitions in the target atoms) from fusion products striking the chamber, but they will be overwhelmed by characteristic x-rays and bremsstrahlung from electrons emitted by the cathode. In theory, bremsstrahlung are also emitted by the abrupt stopping of massive charged particles, but because of their great mass they are vastly less efficient at producing bremsstrahlung compared with electrons.

Energetic fusion products will induce nuclear energy-level changes by inelastic scattering, notably on Fe-56, as will neutrons. The Fe-56 prompt gamma radiation from this process is easily detected with a scintillation detector at about 840 keV, as seen in some spectra made by Jon and myself from the vicinity of operating fusors. It's hard to tell how much of this radiation comes from (n,n') and how much from scattering of charged fusion products.

-Carl

[Steven Sesselmann]

Carl,

Those were my thoughts, interesting that you and Jon have measured it.

One would imagine that an atomic nucleus with a mass of 3 amu traveling with an energy of 1 mev and stopping abruptly, would have to emit a gamma of exactly 1 mev.

Of course it will never stop abruptly, but will undergo a number of inelastic or elastic collisions before becoming thermal.

I wonder if we could estimate the number of collisions by carefully measuring the gamma spectra.

I am interested in this, because I want to find out if a secondary fusion reaction is probable inside the S.T.A.R. cathode.

If I line the cathode with beryllium, the He3 and T collisions with the cathode lining would be slightly elastic and energy would be absorbed quicker, possibly slowing the He3 or T to levels where a secondary fusion reaction might take place.

I estimate that an He3 or T might have to bounce off beryllium 18 times, before reaching energies less than 100 kev. If this was the case, we should be able to measure a spectra characteristic of these bounces.

Your thoughts..

Steven

[Richard Hull]

I think we can view the He3 and T reactions much like D-D. That means that there will be wasted opportunities on almost every collision. Only a few will do what we want them to do.

This is true of the alpha, Be - n reaction where about 10,000 alphas make only 1 neutron.

Nuke stuff via bombardment and collision to achieve a specific result is always akin to trying the herd 20,000 cats. They never do what you want them to do.

Richard Hull

[Steven Sesselmann]

Richard,

Like cats, the He3 and T nuclei have 9 lives or more, you can smash them against the wall 9 times or more before they die...

Inside the S.T.A.R. cathode they are mechanically confined, because they are charged particles, they can not penetrate the wall.

The question is can we measure the growls as they are smashed against the cathode wall, using gamma spectrometry and then work out how many lives they have?

Steven