Cs137 Contamination in Italian Alps's Soil

[Go-Figure]

Hello there,
This weekend I tested a sample of Italian north-west Alps’ soil sent me by Ciro (Cicastol).
First of all thanks Ciro!

The sample is about 300 grams of soil, it has contamination from Cs137 dating back to Chernobyl accident and comes from one of the areas of Piemonte most affected by the fallout.

I honestly didn’t know if my current shield was enough to see anything at all, right now I can only get rid of 72-75% of background. So initially I thought to leave it there waiting for the upgrade (namely: more lead) which I will get next month. But then I decided this was still going to be interesting in order to compare what I can do with my current shield vs what I’ll be able to do with the upgraded one.
I decided I was going to post something only if I could at least have a glimpse of Cs137 (which wasn’t for granted).
If I am here posting then it means Cs137 showed up, although the spectrum isn’t very good looking.

While I was recording my 12 hours background (which I later subtracted) I made two 30 minutes counts with my Ranger, with and without the sample. That allowed me to identify an extra activity due to the sample of 16 CPM (76 vs 60).

I would have liked to test the whole sample but I couldn’t fit it inside the little room at the end of the shield so I had to take a small portion of it (about 60 grams) and put it into a smaller plastic bags.
In hindsight I could have put some more.

The shield is the one already used for the Tyuyamunite test. 5 mm of plastic, 1 mm of copper , 4 mm of pewter and 13-15 mm of lead. The lead needs to be doubled, plus I need some of it to effectively close the “door” of the room. Right now I am using a piece of Radmax which was sent to me as a free sample. I folded it in two so it’s just 2 mm, besides it can’t even cover the whole diameter of the inner plastic tube. Plenty to improve there, and there’s nothing better than a weak sample to expose a weak shielding.

In my 12 hours background I got pretty much the same counts as last time, between 69 and 70 CPS. The surprising thing is that this time the lead fluorescence seemed to be reduced to almost zero (while last time according to my estimate it was reduced by only 70-75%). My shield is made of separate parts which are assembled for each experiment, therefore it’s never exactly the same from one test to the next. It was a nice surprise but honestly I have no idea of what I did differently. The only difference I can identify is that there’s probably a bit more room between the pewter and the lead.

Here’s the shielded background I got last time

Background - CPS - 12 Hours - Counts x Bin - Shield V2 - 0.042 Gauss+CPS - 29-09-19.png (9.72 KiB) Viewed 4331 times

And here’s the shielded background from this test, the shield is made of the same components.

Background - CPS - 12 Hours - Counts x Bin - Shield V2 - 0.045 Gauss - 19-10-19.png (17.03 KiB) Viewed 4331 times

I also tested the sample with the PDS, but as expected I couldn’t really see anything there. Yes, in the LOG view you can see the slope around 660 keV is maybe a bit more flat than usual, but there’s no real “bump”. Acquisition time was 12000 seconds.

Alps' Soil - 12000 Secs - 191019 - LOG.png (92.82 KiB) Viewed 4331 times

And now for the 2’’x2’’ NaI(Tl): the Cs137 peak is the only one clearly formed and also have the most clear Gaussian correlation. I can identify several NORM peaks which appear to be more from Th232 rather than U238, although I have Gaussian correlations rather than actual visible peaks.

Piemonte Soil - 24 Hours - BG Subtraction - Counts x Bin - Graded Shield V2 - 0.045 Gauss - 19-20 -10-19.png (76.92 KiB) Viewed 4331 times

The highest point of the spectrum corresponds to just 0.064 CPS, while the central (and highest) point of Cs137 peak is 0.009 CPS.
The whole spectrum is the result of an excess of counts, net of the shielded background, of less than 7 CPS (roughly 10% of the reduced background), so even small background fluctuations can easily interfere with the result. And as you can see, particularly at low energies, it looks more like a background spectrum than a sample spectrum.

The acquisition time was 24 hours (86400 seconds), but even like that the whole spectrum is only a little more than 600,000 counts, the equivalent of what I get from my ordinary background in 40 minutes.
Exact pick up location has been provided by Ciro. The map comes from the website of ARPA (Regional Environmental Protection Agency) http://www.arpa.piemonte.it/news/le-tra ... n-piemonte

Since I was going after Cs137 and I was only able to start the experiment with the background recording pretty late on Friday night, my calibration was pretty bad at high energy, because I only calibrated with Cs137 without correcting with Th232 as I usually do. As a result Cs137 peak is where it supposed to be, but background peak from K40 is at about 1415 keV and the small excess from the sample is even lower than that. I didn’t label it out of decency.
I have a pretty well formed guassian correlation around 850 keV, given the calibration drift already present in that region of the spectrum that is consistent with Tl208 peak at 860.56 keV which is expected from Th232 but I think it’s the first time I identify it in one of my spectra.

Quantitative Analysis with some of the numbers aready mentioned.

There’s not much to add.
I will have to repeat the experiment with a better shielding, maybe introducing a bit more sample material inside the “room” (there’s some room left) to increase the total activity.

So in summary, Cs137 identified, but plenty of room from improvement, more lead is needed.

Massimo

[cicastol]

Nice work Massimo,
this weak sample is not an easy task ,yes more lead is needed but you did it!
Try to use the whole sample "wrapped" around the detector in a 4Pi geometry, doing this improve a lot the detection.

[Go-Figure]

Nice work Massimo,
this weak sample is not an easy task ,yes more lead is needed but you did it!
Try to use the whole sample "wrapped" around the detector in a 4Pi geometry, doing this improve a lot the detection.

Yeah, I have a couple of ideas for a better geometry next time, also I am going to try to position the scintillator vertically,
Once I have more lead I will have a few more options and I expect some improvements.

I will keep you posted, but I leave for Japan in a couple of weeks (a day taking spectra and dosimetric measurement in Fukushima is part of the plan) so I am not going to be able to post much in the next few weeks.
Maybe I will have a new uranium ore next week!

Thanks again for the sample!
I owe you a favour.

Massimo