2x2" CsI(Tl) Detector

[GigaBecquerel]

Hello!

Luuk had a wonderful 2x2" CsI(Tl) Crystal to sell, and I am always looking to build more detectors, so I just had to get it ;-)
He did some work on it (I'd love to hear more details about that!), wrapped it in reflector and measured it, then sent it my way with some test sheets.

photo_2021-03-15_12-01-25.jpg

The results were: 13.6% @ 59 keV, 10.8% @ 122 keV, 6.2% @ 662 keV and 4.4% @ 1332 keV respectively.

First thing I did was pair it up with my R6231, adding a few layers of black tape as well as a layer of aluminum foil, since I was warned that the reflector isn't perfectly light tight.
With this setup I did my first tests to confirm my PMT is good and the crystal didn't get damaged in transport etc. This is where I measured my first energy resolution of 6.4% @ 662 keV, so it was safe to assume all was good.

photo_2021-03-15_19-51-27.jpg

photo_2021-03-15_19-51-31.jpg

This was my first time working with CsI, and I was not aware just how long its afterglow lasts. Within a day or so the resolution improved to the measured 6.2%. I was also amazed how linear CsI(Tl) actually is, as it needed no quadratic factor to fit the complete ²³²Th spectrum!
The whole detector was then hidden inside a mumetal shield.

After those tests I started building the detector housing, I went with my tried and tested method of a stainless steel tube, and aluminium endcaps.
I am especially proud of the endcap this time, as it is only 1 mm thick on every wall, perfectly encapsulating the crystal while not adding much shielding to low energy gammas.

photo_2021-03-15_12-27-17.jpg

photo_2021-03-15_12-27-24.jpg

photo_2021-03-15_12-27-27.jpg

Details of the detector endcap

photo_2021-03-15_12-27-30.jpg

All the components side by side

photo_2021-03-15_12-27-32.jpg

Fit of the assembly inside the case. Between the mu-metal and SS tube is foam padding to prevent rattling and damage.

The complete detector is very heavy, about 1.4 kg in total!
500g if it is the crystal itself, around 500g is the housing, and the rest are the PMT, socket, divider and all the small bits inside.

Now for my resolution measurements:
15.5% for the 59 keV line of ²⁴¹Am
I don't have any ⁵⁷Co, closest thing is ¹⁷⁶Lu with 9.1% @ 201 keV & 8.0% @ 306 keV
5.85% (!!!) at the 662 keV line of ¹³⁷Cs
And I also don't have any ⁶⁰Co.
I am very happy with these measurements!

But those are just numbers, in my opinion you can compare a detector best with a spectrum everyone knows and recognizes. Here's the gammaspec of ²³²Th:

Screenshot from 2021-03-14 20-43-15.png

You can see how well the peaks separate, and how linear the spectrum fits!

To be honest, I never really considered CsI(Tl) for "real" gammaspec, as I always associated it with cheap and small SiPM at 8% or worse.
I have fallen in love with the material due to it's high stopping power, rigidity and most importantly very good resolution!
This detector will very likely replace the 6.1% NaI(Tl) in my lead castle, as it is just better in every way.
I am very happy with how this detector turned out and want too thank Luuk again for the awesome crystal!

Finally, here are some pics of the finished detector, side by side with his smaller brother, my 20x10 mm LaBr.

photo_2021-03-15_19-37-43.jpg

photo_2021-03-15_19-37-44.jpg

photo_2021-03-15_19-37-44 (2).jpg

I hope you enjoyed this article, and as always feel free to ask any question!
Lukas

[luuk]

Hi Lukas,
That nice detector you made with this crystal, looks great and works of course great to.
The endcap is nicely done, good job!
Thanks for this great report do hope you have lots of fun with your detector, keep the crystal dry and it will last for ever.
Luuk

[GigaBecquerel]

Hi Luuk,

Thank you for the kind words!
I added a pack of silica gel inside the detector, I guess that it will be enough to keep the crystal dry.
I was a bit surprised about the loss of resolution at 59 keV, while all the others got better... There are still some things I will have to figure out with this one!

Lukas