Trinitite – The Quest for Europium and More: A 17 Day Gamma Spectrum

[cicastol]

If you look at pg. 9 table 2 of your's linked paper, the estimated 133Ba activity concentrations recalculated to the date of explosion was 9.1 Bq/g (LMS 2010) now in 2020 it should be less than 0.1 Bq/g , i think it is well under MDA of your's setup , 137Cs to Ba133 ratio now should be near 100/1 and from the spectra given this is not the case.

[Go-Figure]

If you look at pg. 9 table 2 of your's linked paper, the estimated 133Ba activity concentrations recalculated to the date of explosion was 9.1 Bq/g (LMS 2010) now in 2020 it should be less than 0.1 Bq/g , i think it is well under MDA of your's setup , 137Cs to Ba133 ratio now should be near 100/1 and from the spectra given this is not the case.

Do you mean the table at page 6?
Yes, I've seen that, but Trinitite samples are not all the same. The paper itself mentions an uncertainty of 1σ.
I tested enough Trinitite samples to know that differences between them can be pretty big depending on where the sample comes from in the blast area. For example, I have several Trinitite samples where the Eu152's peak at 121 keV is not even visible with my setup, you can barely spot a gaussian correlation, while in this "type 5" (to use the Minresco's terminology) its strenght is comparable with that of Cs137 at 661.67 keV.
I would add that in the spectrum showed in the paper Ba133's peak is strong enough to be comparable with others that you can usually spot. That spectra should date back to 2010. Ba133's half life is 10.51 years, while Eu152 has an half life of 13.52 years, so their relative activity didn't change dramatically since then.

Again, that peak has likely several other things in it, some Pb212 and Pb214, and some XRF too, but according to my experience there's something else, because I never had something like that in any of the Trinitite sample I tested before with the same setup (I have seven of them), and around 80 keV there's nothing else I can think about.

Am I absolutely sure? Of course not, but it seems the best candidate to me.

[cicastol]

Yes pg 6 , it is listed as median taken from 11 trinitite samples , one thing for sure, your's sample look really interesting !!
Now i'll take again the spectrum from my sample but i remember only a hint of Eu peaks and not much more ,not quite interesting as yours....

[Go-Figure]

Yes pg 6 , it is listed as median taken from 11 trinitite samples , one thing for sure, your's sample look really interesting !!
Now i'll take again the spectrum from my sample but i remember only a hint of Eu peaks and not much more ,not quite interesting as yours....

Actually I have my fair share of "uninteresting" samples. Purchasing that "type 5" was an act of desperation after a number of attempts with disappointing results (some of them I posted here). I was skeptical that it could really make any difference, but to my surprise it actually did.

[Go-Figure]

Hi again,
A new spectrum from the lockdown.
Not really a new one, it's my second attempt with the type 5 trinitite sample.
Why did I do it again? Well, first of all the spectrum I got from this sample was so much richer than anything I got before from my mostly miserable attempts with other trinitite specimens that I wanted to make sure it wasn't a hallucination.
Ok, that doesn't sound very scientific, but I also had another reason, I wanted to repeat the measurement doubling the background acquisition time to see how the final result would improve in terms of "smoothness".

So this time I did it this way:
3 days of background acquisition- 1-4 March.
14 days of sample acquisition - 4-18 March.
3 further days of background - 18-21 March.
I divided the background acquisition in two because in 20 days you can have not irrelevant changes in temperature and boundary conditions in general and recording it all at the beginning or all the end I would have been more prone to calibration drifts.
If I will ever do it again I will try 14 days of background + 14 days of sample.

I tried to recreate the exact same conditions but of course it's never really the same.
The results:
CPS from the sample was 8.49 versus 8.38 last time, nearly identical.
Peaks were mostly identical too, with only two differences worth mentoning:
- The controversial Ba133 peak around 80 keV is weaker this time, which will possibly make it even more controversial. As it was forming it looked again as two distinct peaks with something around 80 keV and something else around 85 keV, just like last time. But just like last time I couldn't find any realistic candidate for the latter (see first post of the thread for more details).

- The Eu152 peak at 1408 keV. This has been the biggest question mark in all my Trinitite spectra. It's pretty weak and it's in a region where there's K40 too, so I never know what I am looking at there. I am glad to inform you that this further attempt did nothing to clarify my doubts as far as this peak is concerned. So in the end, since I know they are both there, I put them both in the same label, and I guess it will stay like this until someone gives me an HPGe detector.

One last thing, the apparent peak visible around 180-200 keV is most likely the back scatter peak from Cs137. I should have put a label there too.

Hope you are all fine and healthy, wherever you are.

Until next time.

Massimo

[Peter-1]

hello Massimo,
I have now also tried to find the peak at 1408 keV. I'm not sure if it's the Eu 152 line. Unfortunately only a measuring time of 11 hours. What do you think about the measurement ?

Peter

[cicastol]

- The controversial Ba133 peak around 80 keV is weaker this time, which will possibly make it even more controversial. As it was forming it looked again as two distinct peaks with something around 80 keV and something else around 85 keV, just like last time.

Hi Massimo, IMHO this reinforce me the idea that the peaks are the Pb classic 75-85 keV lines, if you look my spectra this double peaks are visible.

[Go-Figure]

hello Massimo,
I have now also tried to find the peak at 1408 keV. I'm not sure if it's the Eu 152 line. Unfortunately only a measuring time of 11 hours. What do you think about the measurement ?
2020_03_30-Trinitit-C.jpg
Peter

Hi Peter,
After testing no less than seven different Trinitite samples I still can't make up my mind about the Eu152 peak at 1408.
Anyway, we know there's K40 in Trinitite, because it was naturally present in the desert's sand, and we also know there's Eu152, although not always there's enough to be clearly visibile, but that's not your case apparently since the peak at 121 keV is clear enough.
So in the end my conclusion, as far as my measurements are concerned, was that both are there but I am currently unable to tell which one of the two is contributing the most to what I see in the 1410-1460 keV region. Until that's the case I will put both labels.

Hi Massimo, IMHO this reinforce me the idea that the peaks are the Pb classic 75-85 keV lines, if you look my spectra this double peaks are visible.

I don't if anyone is surprised about by that but.....I just started another measurement on the same sample, this time I will have 14 days of background and 14 days of accumulation with the sample. That's mostly because in that region I have a visible difference between my two previous attempts, despite the setup being basically identical. This is going to be interesting.

[Go-Figure]

A promise is a promise....this is the result of my latest Trinitite measurement.
The sample is the same “type 5” about which I already posted two different spectra, so this is the third one.

This time it’s a 28 days measurement:
7 days of shielded background.
14 days with the sample inside the shield.
7 further days of shielded background.

An accident during the test allowed me to discover an interesting thing:
I was more or less in the middle of my 14 day acquisition with the sample when there was a blackout. There’s nothing like a blackout during a lockdown. Besides, it was in the middle of the night, at about 2 AM when I was about to go to sleep.
Luckily the battery of the laptop prevented it from shutting down. I didn’t know when the light would be back so I turned everything I could into saving mode in order to extend the life of the battery as long as I could, in the hope that it would be enough.
At around 4 AM my battery only had minutes left and I was resigned it wasn't going to be enough. I saved the result so I wouldn’t have ended up totally empty handed. But I knew my plan for a 28 days measurement was gone…I was back to square one….or so I thought.

The next day I tried something desperate to try to resume measurement from where I left instead of starting from zero again. I wondered…what if I load the spectrum I saved last night and resume measuring again from there? The scintillator was still inside the shielding, so in terms of boundary condition everything was the same, but what about calibration?
I always calibrate my system before starting a new measurement and it never crossed my mind to try to use a previous calibration after restarting the computer.
But this time I had little to lose so I tried. I loaded the spectrum (and the calibration with it) and then I started to accumulate a background inside the shield in order to see how much the peaks would be off.
After almost a day of accumulation, to my surprise, I found the peaks weren’t off by any clearly noticeable amount, so I decided to resume my measurement with the sample and finish the job.
In the end about half of the measurement was carried out before the blackout and the other half after that. I wouldn’t advise shutting down the system in the middle of a test, but in this particular case it doesn’t look like it had a significant impact on the final result, which is the following:

Background, 61.54 CPS vs 61.86 of my previous measurement, almost identical. The setup was the same.

Background - 336 Hours - Counts x Bin - Shield V2-1 - 0.036 Gauss - 31-03_07-04+01-05_8-05-20.png (20.49 KiB) Viewed 8783 times

The main reason for testing this sample again was to try to have a clearer picture of the 75-85 keV zone, which, I have to say, wasn’t the case. This third attempt gave me the third different result in that very zone. This time the Am241 peak at 59 keV doesn’t come down almost vertically all the way through and there is a “slope” where you can see a peak in the 80 keV region, but the slope makes it less clear than in my previous attempts.

So unfortunately this wasn’t enough to make the situation clearer. I am trying to acquire more pewter in order to repeat the test further improving my XRF attenuation.
The potential Ba133 peak is therefore still under scrutiny

As for the rest, as expected a longer background was key to make the spectrum smoother and this allowed me to identify a further Eu152 peak at 867 keV which was always there, but not as clear as it was this time around.

That’s it for now but….that’s not the end of the story.

Massimo

[Svilen]

You are the king of the long measurements (and patience), Massimo!
Maybe you can this way determine the optimal test period for these samples in your conditions, a good balance between time and result quality ;)