Inelastic Neutron Scattering Experiments

[Jon Rosenstiel]

Inelastic neutron scattering (fast neutrons) is a process in which the incident neutron (provided its energy is greater than that of an excited state of the target nucleus) is scattered from the target nucleus at a lower energy, leaving the target nucleus in an excited state. The target nucleus then promptly decays back to its ground state, usually by the emission of one or more γ- rays. The reaction is written: (N,N’G)

This report documents recent inelastic scattering experiments I ran on titanium, Ti-48, and magnesium, Mg-24.

A little background on the setup:

On my fusor’s specimen shelf I placed a 3.0 cm stack of lead plates to attenuate the inelastic scattering gammas (849 keV from Fe-56 and 1434 keV from Cr-52) from the fusor’s stainless steel shell. The NaI detector was positioned 3.0 cm above the topmost lead plate. The fusor was run at high power (close to 1 kW) for these experiments. Each 20 minute run consisted of two 10 minute runs separated by a 15 minute period to allow the fusor to cool down.

Photo: (Showing setup)

Magnesium specimen is in place. Not shown is the lead cylinder (1.0 mm thick by 9 cm diameter by 8 cm high) that I use to shield the detector from scattered x-rays.

Upper chart: (Titanium)

The prominent peak labeled 988 is the 983.5 keV gamma from Ti-48’s 983.6 keV energy level.

Lower chart: (Magnesium)

Magnesium is a bit of a mystery. There’s no doubt it’s got a prominent gamma (peak labeled 1370 in the chart below), but I cannot find any (N,N’G) data on magnesium. I did find that Mg-24’s first (lowest) energy level is 1368.675 keV, so the 1370 keV gamma is most assuredly from this energy level. But, is this gamma from the (N,N’G) reaction in magnesium, or am I just overlooking something?

I’ve been using this site: http://www.nndc.bnl.gov/ensdf/browse_top.jsp for data on inelastic scattering reactions. To see data on Ti-48 look under “Browse by mass” and click on “48”. Then go down the list to 48Ti and click on “ENSDF”. Place a check mark in the box (last entry) next to the “TI(N,N'),(N,N'G), 48TI(N,N'G)” category. Then next to "Get selected ENSDF datasets:" click on “HTML” to see Ti-48’s (N,N’G) reaction data.

If anyone has trouble opening the two *.gif images please let me know and I’ll fix it.

Jon Rosenstiel

[attachment=2]Magnesium.gif[/attachment][attachment=1]Titanium.gif[/attachment][attachment=0]IE experiments 033a.jpg[/attachment]

[DaveC]

More, nice work, Jon. Interesting observation on Mg. Your explanation seems in good agreement with the experimental data.

Just a simple question on the physical setup, which may originate in the perspective of the photo. Is your sample platform just outside the fusor flanges, or is it outside the thinner shell area of the fusor? I guess in either case, the lead shielding makes that aspect of position more or less irrelevant.

Thanks again for sharing your results.

Dave Cooper

[Jon Rosenstiel]

Dave,
As you surmised, it's all in the perspective... there is about 4 mm between the underside of the sample platform and the outside surface of the conflats.

Jon Rosenstiel

[Carl Willis]

Jon,

This is a fascinating report as usual. The strange energy from Mg seems to point to an impurity, in my opinion. Another possibility is that you are exciting some other kind of fast neutron capture reaction, of which magnesium does participate in a few:

Mg-24 (n,p) Na-24
Mg-24 (n,a) Ne-23

Note that Na-24 emits a 1369 keV gamma.

You'll have to pick this up and look through the CSISRS (EXFOR) database for supporting evidence, but it's certainly a possibility.

Thanks for the fine work!
-Carl

[Jon Rosenstiel]

Carl,

The magnesium blocks that I used in the inelastic scattering experiment were originally handlebar mounts from a motocross bike. I did an x-ray fluorescence analysis (XRF) on them some time back and found they also contain Zinc and Zirconium. Aluminum may also be there, but my crude XRF setup can’t detect it unless it’s in it pure form.

I just received some (supposedly) 99.99% pure magnesium from eBay seller "emovendo" and plan on giving that a test in the near future. I’ll post results here.

Thanks for the hints; hopefully I can get to the bottom of this…though it may be some months!

Jon Rosenstiel

[Richard Hull]

More fascinating reports that are consumed with glee and gusto.
Thanks, as always, for sharing the info. It is unfortunate that gammas abound and are spit out of a lot of bombarded stuff. figuring out the culprit or causitive agent is a detective story.

Richard Hull

[Jon Rosenstiel]

I received the 99.99% pure (as claimed by eBay seller “Emovendo”) magnesium bars and put them to the test in a fast neutron flux. Nice gamma peak at 1369 keV, just as before.

Then, the next day a Google search turned up an excellent paper on inelastic scattering in B-10, C, N, O, F, Mg, Al, S, Ca, Fe, Ni, Cu, Ta, Pb, and Bi. The paper was authored by Robert B. Day of Los Alamos Scientific Laboratory and dated May 1, 1956. The paper is titled: “Gamma Rays from Neutron Inelastic Scattering”. (Cost was $20.00)

For magnesium Day says: “The principal magnesium isotope, Mg-24, has a well known level at 1.370 meV. Figure 12 shows that this level is strongly excited by neutron inelastic scattering”. (Figure 12 shows a strong gamma peak labeled 1.365 meV).

So, it appears that the 1369 keV gamma I observed is in fact from Mg-24. Kind of odd that I had to dig up a 50 year old paper to find data on Mg-24’s inelastic scattering gammas.

If anyone is interested in inelastic scattering I highly recommend Day’s paper. Day gives a very thorough account of the setup used and the results obtained. I really liked what he had to say about his setup. (Remember, this paper is dated 1956) “Pulses from the scintillation counter were amplified in a Los Alamos Model 250 amplifier and preamplifier. This is a nonoverloading amplifier having very good linearity and stability. After amplification the pulses were analyzed by a 10-channel pulse-height analyzer designed by Johnstone. For some of the later work a much improved version of the Hutchinson-Scarrott 100-channel analyzer was used. Since this analyzer had an average dead-time of 500 µs, it was necessary to run a fast single-channel analyzer in parallel with it in order to make a correction for counting losses”.

Jon Rosenstiel

[Richard Hull]

It doesn't surprise me a bit that it took a paper that old to deal completely with the subject. The more I read the old stuff the more I find that the hard data is to be found there.

Often folks using lesser gear and materials post a better synopsis and more thorough treatment of something than a modern author. A lot of moderns would not necessarily revisit this subject thinking it had already been covered. In this case, it obviously was.

The older stuff still had the author's coupled to the real world giving their exposition to colleagues that were to be impressed by the simple results more than the fluff and bluster of the author telling why it was necessarily so. While there is no cutoff date ever on good stuff, you will find the most concise and useful material that is readily consumed and easily digested, came out prior to the 70's with the hard core, down to earth papers being those from the 50's or before. The older papers gave a lot more data on technique and experimental minutia.

I am glad that you were gratified by the rather complete treatment of inelastic neutron scattering in this older paper.

Richard Hull

[DaveC]

I am also inclined to agree with Richard that the better writing was in 70's or earlier. I think it is the publish or perish syndrome, and the journals which have to contend with far greater quantity, of not necessarily equal quality papers. Page restrictions, without extra publication charges have been around forever, but that number seems to slowly drop. This greatly restricts the authors' opportunity to explain, and... many are using equipment they hardly understand... which sends large warning flags up for me. This is especially true of the digital signal processing, whether, scopes (DSOs), or data acquisition units (DAQs), analog-digital converters A/D, etc.... the researcher needs to fully understand the signal flow, and prove to himself that the results are what should be seen. "Aliasing" and other sampling related phenomena are always around and can give fascinating, but utterly wrong and irrelevant results.

The earlier workers had a much harder go of it, and thus were more keenly aware (sometimes !) of the hardware limitations.

The descriptive experimental account is extremely useful to the experimenter as it smokes out procedural (and theoretical) errors, saving your reputation and buckets of time.

It's just too bad for those on limited budgets, that the price tag for these excellent papers is $20 - $30 each. But there are the University libraries, if you know what you want to find.

Dave Cooper

[Richard Hull]

I really believe the older experimenters were much closer to their work and much more personally and intimately involved. This probably helps explain a lot.

Thanks for your thoughts on this Dave.

Richard Hull