Page 1 of 1

Archived - Lutetium gamma spectrum

Posted: Tue Apr 13, 2004 9:22 pm
by Carl Willis
Archived due to the superb presentation and info on Gamma spec work. RH

FYI...just an example of a gamma spectrum that illustrates some tricky peaks.

This is a gamma energy spectrum taken from a 23 g lump of 0.9999 Lu metal, using a Bicron 1.5x3" NaI(Tl) detector. Lu-176 (natural abundance = 2.6%) has a half-life of 3.8 x 10^10 years and emits beta / gamma radiations. The activity is easily picked up with a Geiger counter.

The ENSDF database of nuclear data (ie.lbl.gov) lists four gamma energies for Lu-176:

88.36 keV (emission intensity of 15.6)
201.83 keV (83.3)
306.84 keV (100)
401.1 keV (0.4)

The first three of these can be discerned in the spectrum above (within a small bit of uncertainty of course). The peak at 510.4 keV is a "sum peak," recorded whenever 202 and a 307 keV gammas are simultaneously absorbed in the detector. The peak at 389.6 keV is ANOTHER sum peak, superimposed on the weak 401.1 keV line. This sum peak occurs when a 307 keV gamma and an 88 keV gamma enter the detector at the same time. Why so many sum peaks? The point here is that one has to know not just the decay energies but the decay scheme in order to know when sum peaks should be expected. It so happens that in Lu-176, the transitions in which gamma rays are emitted at 202, 307, and 88 keV are all sequential following a single beta decay, and hence these radiations are emitted almost simultaneously. The level diagrams for just about any radionuclide are accessible from ENSDF (link above).

Finally, the peak at 47.2 keV is from Hf and Lu K x-rays. The broad peaking from ~130 - 170 keV results from backscattered gammas, and the Compton edges for both the two big peaks. Where to look for the "backscatter peak" and the Compton edge are simple mathematical relations to the full energy peaks; I use a spreadsheet to automatically predict where these will be.

How do you know if you have a sum peak or backscatter peak without knowing the decay scheme or math formula, respectively? The answer is that the relative intensity of sum peaks and backscatter peaks is very strongly dependent on the physical configuration of the sample and detector. If you move the sample closer and note a peak that doubles in height relative to another, suspect that it might be a sum peak. If you put a lead sheet behind your sample and note the sudden appearance of a new peak, consider whether it makes sense for it to be a backscatter peak.

Anyway...gamma spectroscopy can be misleading sometimes and it helps to know the physical principles well.

-Carl

Re: Lutetium gamma spectrum

Posted: Tue Apr 13, 2004 10:08 pm
by Jon Rosenstiel
Nice spectrum, Carl.

Couple of questions.... How long was your count? Background subtracted?

This is good stuff, keep it coming.

Jon Rosenstiel

Ps: I could sure use some help on getting around the ENSDF database. Maybe you could post an tutorial? (Hint, hint)!

Re: Lutetium gamma spectrum

Posted: Thu Apr 15, 2004 1:05 am
by Carl Willis
Hi Jon,

This was gathered over about 5 hours, but I'd set the MCA to a max counts cutoff of 10,000 counts (so that Excel would plot it nicely on a log scale) and didn't record the live time. No background was subtracted. However, background is very low because I use an old 2" thick steel-lined lead counting chamber for shielding.

I'm happy to post some links to all the useful databases I have used, all in one place, and a little blurb on how to use each. Probably belongs in the "interesting links" section. May take me a little while to get around to it!

-Carl

Re: Lutetium gamma spectrum

Posted: Fri Apr 16, 2004 12:30 am
by Jon Rosenstiel
Carl,

I managed to duplicate your results fairly well using a 5g piece of Lu. Lots of background crept into the spectrum, but all of the peaks were there.

I think I’ll visit the local metal yard this weekend, see if I can find some thick steel suitable for shielding.

Jon Rosenstiel

Re: Lutetium gamma spectrum

Posted: Fri Apr 16, 2004 11:21 am
by Richard Hull
Odd, Both Bill Kolb and myself ran recent spectrums of Lutetium chunks we have and e-mailed back and forth on it. We got the same results as you guys. Nice stuff Huh?

Richard Hull

Re: Lutetium gamma spectrum

Posted: Sun Apr 18, 2004 11:05 am
by Carl Willis
Hi Richard,

I saw Bill Kolb's Lu spectrum taken with his well detector, and ultimately it prompted me to get a piece. Hence the apparent coincidence! As this metal is rare and has no real commercial importance (that I know of), it is indeed a bit of an oddity to have around. Lanthanum is much cheaper and supposedly has some gamma emissions as well, so I'm going to try that next. And samarium is actually surprisingly radioactive, although it only emits alphas. I'm going to try and get samples of some of these lesser-known natural radionuclides.

-Carl

Re: Lutetium gamma spectrum

Posted: Sun Apr 18, 2004 12:40 pm
by Jon Rosenstiel
Hi Carl:

That's odd.... I've got a 100g chunk of samarium, did absolutely nothing on my pancake detector. (Though it did respond well to PGNA).

Jon Rosenstiel

Re: Lutetium gamma spectrum

Posted: Tue Jun 01, 2004 9:33 pm
by polcott
> As this metal is rare and has no real commercial importance

Lu as a radiative source does not have a commertial importance, but lutetium does. It is a very important element in LSO, or Lu2SiO5, lutetium orthosilicate.

LSO is an excellent scintillation crystal for 511keV gamma detection for PET (Positron Emission Tomography) because of the large production of photons (~15000/MEV) and fast decay constant (~40ns)

I happened to be trying to calculate the number of incorrect events for a 511keV interaction with LSO with a 10% energy window would record from the background of Lu. Thanks for the timely energy spectrum!

Did this gamma ray spectrum not include the beta spectrum also associated with Lu-176? (Since you used a large 23g sample, did the beta's escape to produce interactions with the crystal?)

Re: Lutetium gamma spectrum

Posted: Wed Jun 02, 2004 9:55 am
by Richard Hull
Betas don't interact in a significant way with NaI:Tl crystals and most are in aluminum or stainless cups keeping such radiation out, anyway.

Richard Hull

Re: Lutetium gamma spectrum

Posted: Wed Jun 09, 2004 12:27 pm
by Carl Willis
Hi Polcott,

Glad that spectrum helped out. There's a great book online in several places by Heath, with many sample NaI(Tl) spectra of natural and artificial gamma emitters. Should be easy to locate with Google if you want to see more of this kind of thing.

Richard mentioned that the betas don't influence this measurement because they are stopped prior to getting to the crystal. Beta radiation can be a complication if it is energetic and the scintillator has only a thin aluminum can around it.

LSO has got to be the first commercial application of lutetium that I have heard of, and I'm sure it's pricey. You're saying it's both fast and bright. Do you know if it's doped with something?

-Carl

Re: Lutetium gamma spectrum- Richard's run

Posted: Tue Jun 15, 2004 1:51 am
by Richard Hull
My spectrum is included here. I used a giant piece of Lutetium (~53 gram) and this reduced the collection time and made a bold spectrum.

I used a 2"X2" Bicron detector head and my model 40 Canberra.
Standard log-lin presentation with no background subtraction, as shown.

Thanks to Richard Hester for the XL to gif conversion.


Richard Hull