Hello there,
This past week my spectrometer has been busy with another piece of uranium glassware: a small cup that I bought more than a year ago, but for some reason I never tested it properly.
As usual the best way to look at a piece of uranium glassware is under UV light in the dark.
Under ordinary visible light the cup displays the typical pale yellow-green colour.
The cup is not very hot, the best I could get with the Geiger counter was less than 450 CPM (the background is around 50 CPM), so I expected to get very little gammas from it.
That was confirmed when I first tested it with the PDS: 54.80 CPS at contact with the body of the device, roughly 50 of which coming from the background, so that was in the region of 10% of the background in terms of counts, and roughly half of it in terms of exposure from gamma radiation since the kind of uranium used for these kind of objects is refined, it does not contain all the products of the decay chain like Radium, Lead and Bismuth which are responsible of the high energy part of natural uranium spectrum, so you are left with only the low energy part of it.
The PDS spectrum was of very little significance. The device doesn’t allow you to subtract the background therefore the signal from the sample is buried in the background curve. There’s maybe a hint of a counts excess in the 92 keV (Th234), but it’s too weak and messy to tell for sure. Linear first, then logarithmic.
Now the problem was that the sample was too big to be put into my shield’s chamber. I could make the chamber bigger taking away the copper and plastic layers and wrapping the lead and the pewter with a radius big enough to contain the cup inside, but in the end I decided against it. I wanted to see what I could do with such a weak sample with no shield at all so I went for a “no-shield” measurement.
I first recorded a 24 hour unshielded background, then I had a 96 hours with the sample and then an additional 48 hour background.
This is the resulting 72 background. It’s worth noting that in the middle of the week Italy got in the middle of a heat wave, probably the hottest days of the summer, and that inevitably affected the calibration. I saw the K40 peak moving south by more than 30 keV over the week. Luckily the interesting part of the spectrum of this sample is at lower energies which are less affected in absolute terms, simply because 2% of 185 is a lot less visible than 2% of 1460.
- 07 - Background - Stanza Fondo - 72 Hours - 0.03 Gauss - No Shield - 26-07 + 30-07_01-08-20.png (67.95 KiB) Viewed 7695 times
Here’s how the measurement was taken. I tried to optimize the geometry. As mentioned before, there was no shield.
And here’s the result, both in counts and energy per bin. The whole measurement lasted a week.
Once again it turned out to be Natural Uranium, you can tell by the fact that the U235 peaks are too strong to be depleted uranium, so this must be a pre-WW2 piece. You can see the thermal drift in the two Pa234m peaks which are considerably lower than they were supposed to be in terms of energy.
Again, there’s a 30-35 keV peak which I still cannot identify.
Also, since activity was so low and no shield was involved, the result is a bit noisy, which is particularly visible in the energy per bin view.
In the end the cup gave me 17.47 CPS, less than 7% of my background, so not an easy sample to test with no shield in place. The good thing with refined uranium is that counts are spread over a comparatively low number of peaks so you have a better chance to see something even with such a low activity.
Until next time.
Massimo
