NaI Scintillation Probe and Gamma Spectroscopy

Short Intro on Scintillation

Scintillation is the property of some materials to exhibit luminescence (light emission) when excited by ionizing radiation. Therefore it is possible to use Scintillation to detect ionizing radiation, by measuring the light emitted.
However the light resulting in scintillation is very low, and extreme amplification must be used to convert the few photons into a usable electronic signal. For this purpose we can use photomultipliers or semiconductors.

Progress summary:
2010-11-29, Introduction to scintillation
2011-07-07, A wrong design DIY counter attempt!
2011-07-08, Output pulses are too short!
2012-05-31, First good results
2012-10-06, More crystals

November 29, 2010

I recently acquired a Scintillation probe, it is almost 50cm long, it comes with a high voltage inverter, that requires -12V 0V +12V, it is regularized using a neon tube to -1200V, a signal amplifier using a few transistors, a photomultiplier (Philips XP2008) and a NaI scintillation crystal all enclosed in a nice aluminum container. This sensitive photomultiplier tube measures the light from the crystal. The PMT is attached to the electronic amplifier to quantify the amplitude of the signals produced by the photomultiplier.

The NaI Crystal (sodium iodide doped with thallium: NaI(Tl) ) is by far the most widely used scintillator material. It is available in single crystal form or the more rugged polycrystalline form (used in high vibration environments, e.g. wireline logging in the oil industry). NaI(Tl) is very hygroscopic and needs to be housed in an air-tight enclosure. The one I have puts out a nice blue glow if exposed under UV, see the photos below.

A few words on the functionality

When a charged particle strikes the scintillator, a flash of light is produced, which may or may not be in the visible region of the spectrum. Each charged particle produces a flash. If a flash is produced in a visible region, it can be observed through a microscope and counted – an impractical method. The association of a scintillator and photomultiplier with the counter circuits forms the basis of the scintillation counter apparatus. When a charged particle passes through the phosphor, some of the phosphor’s atoms get excited and emit photons. The intensity of the light flash depends on the energy of the charged particles. Sodium iodide (NaI) containing a small amount of thallium is used as a scintillator for the detection of gamma waves. (Cesium iodide (CsI) in crystalline form is used as the scintillator for the detection of protons and alpha particles).

The scintillation counter has a layer of phosphor cemented in one of the ends of the photomultiplier. Its inner surface is coated with a photo-emitter with less work potential. This photoelectric emitter is called as photo-cathode and is connected to the negative terminal of a high tension battery. A number of electrodes called dynodes are arranged in the tube at increasing positive potential. When a charged particle strikes the phosphor, a photon is emitted. This photon strikes the photo-cathode in the photomultiplier, releasing an electron. This electron accelerates towards the first dynode and hits it. Multiple secondary electrons are emitted, which accelerate towards the second dynode. More electrons are emitted and the chain continues, multiplying the effect of the first charged particle. By the time the electrons reach the last dynode, enough have been released to send a voltage pulse across the external resistors. This voltage pulse is amplified and recorded by the electronic counter. Source here.

So the NaI crystal can be used to count Gamma photons. This is also related to the tick aluminum recipient enclosing the NaI crystal, that excludes the possibility of using this probe for Alpha or Beta particles. However the crystal can be easily removed, and another scintillator can be placed on top of the photomultiplier. This probe is usable for a large spectrum of applications. Eg. using Intensifier Screens from x-ray cassettes turn this into a very sensitive x-ray probe.

While powered (energized), photomultipliers must be shielded from ambient light to prevent their destruction through overexcitation. Care must also be taken to avoid humidity damage, as NaI crystals are highly hygroscopic.
Here is a photo showing three NaI(Tl) scintillation crystals with quality varying from left to right. The yellowish nuances indicate humidity damage:

All three crystals in the above pictures, are functional.
This probe can be hooked to an oscilloscope to view the individual pulses. I will do that later and report back.

Update July 7, 2011 – A wrong design DIY counter attempt!

For some time now I’ve been looking for a good counter design, and I finally came up with a prototype based on a ATMega8 microcontroller. I’ve also connected a Bluetooth Radio Module, to have the probe data transmitted wirelessly and avoid exposure to dangerous levels of radiation.

The probe itself seems to be very sensitive to gamma below 100KeV . Unfortunately I get no response to isotopes like Cs137 (0.5MeV), so I won’t be able to use it for Gamma Spectrometry. Yet, I might get interesting results related to cathode metal emission characteristics.

Update July 8, 2011 – Output pulses are too short!

It appears my initial assumption that the probe is insensitive to radiation of higher energy levels was incorrect. The probe’s crystal is a cylinder of 3.2cm diameter and 2.5cm long. The Cesium 137 was not detected previously because of counter limitations. The radiation events caused by Cs137 are discreet in time, meaning that the counter must be capable of processing a high number of samples per second. The Atmega8 was reading data from the ADC port every 100ms which is totally inappropriate!

I’ve been running a few tests sending the probe data directly to the sound card of my PC (via line-in), since this card is capable of up to 48000 samples per second. The result is that the Cesium sample is now correctly detected:

This graph shows 30 seconds of background radiation followed by another 30 seconds of events caused by the Cs137 brought close to the scintillation probe.
Here is the emission spectrum, computed using FFT on the sound card data . Software used PRA and INTUNE (available for free online). This can be used to identify the radioisotope emitting the particles detected by the probe like in the expensive dosimeters:

The last graph is a energy/event frequency representation. There are two energy peaks. The isotope used was Cs137 (from an old spark gap tube). Here is the decay chain:

Update May 31, 2012 – First good results

A gamma spectrometer can be easily built with a photomultiplier tube coupled to a NaI scintillator crystal. The counting of the number of peaks corresponding to each energy to get the gamma spectrum can be done very quickly and easily, without additional hardware, using the freeware software PRA (and InTUNE as well) developed by Marek Dolleiser, both available online.

A short guide on how to use PRA is available on Youtube, here.

I have attached an audio recording representing the output of my NaI Scintillator. It lasts for 1 min, where the first half contains background radiation, while for the last 30 seconds a Cs137 source has been placed near the scintillator. The file is encoded as MP3 so it is only ment for hearing it in your headsets, if anyone needs the original uncompressed data, I can provide it as well:30sec background 30sec Cs137 mp3
Here is the signal representation as amplitude/time, where the Cs137 presence after the first half is clearly visible:

Using PRA, it is possible to generate the pulse Height histogram, that we can use to view the radiation energy spectrum, after the proper calibration.
Here is the spectrum for the 30sec Cs137 recording (attached here as mp3):

And here is another one for a longer recording, 10minutes with the Cs137 sample placed on the scintillator:

There is some signal contamination from the HV inverter, and some other sources. This is an industrial probe, I didn’t make it myself, so I’ll try to come with a few improvements. I already tested a different PMT preamplifier that comes with lower noise and better amplification. There’s a lot to do here.
The final goal is to have a portable unit, with a nice lcd showing the spectrum , to easily identify various nuclides.

Update October 06, 2012 – More crystals

I managed to get a few NaI(Tl) crystals that will help me with my future experiments on scintillation and gamma spectroscopy. I finally have a crystal with a big enough volume to be suitable for higher energy gamma detection:

This article has 14 Comments

  1. Hello Radu,

    This website is very nice!
    I saw the Gamma ND-302 PMT tube on this site. Do you have more information about it? What type of scintillator crystal do you have for this PMT?

    Where can you procure the NaI crystals?



  2. Thank you, Miklos.

    The ND-302 that I have is incomplete: it only contains the preamplifier / high voltage part, but not the

    The other two probes that I have, are complete and operational.

    I will check if I have additional info on the ND-302 and get back to you.

    It’s hard to find NaI crystals, and it took me a lot of time to gather those that you see in the pics.

  3. Hi Radu,

    Thanks for the answer! Do you have an aceptable price source for NaI crystal which is reachable for me too? From Ukraine or something similar?


  4. Do you know if I can cut an NaI(Tl) crystal with a band saw? I have a crystal bonded to a PMT and want to remove it to perform and experiment with another detector technology.

    Thanks … Chuck

  5. Chuck, if you do that you would have multiple problems to solve:
    – polishing the crystal to obtain the initial transparency
    – avoid contact with humidity, which would degrade it easily.

  6. Hello Radu, it’s the cat again 😀
    As i told you in another thread I am experimenting with gammaspectroscopy too. Over a year passed by while I built a suitable HV source, a dynode string for the PMT and a peak shaping circuit to supply the pulses to my PC soundcard. Just recently I got the first promising results with an old latern mantle. Now I’m looking for a Cs137 source for calibration purposes. You mention a spark gap tube – do you have any idea where to obtain such a thing? Maybe you can look up the component name or manufacturer? Any other ideas how to find an affordable Cs137 test source?

    @Chuck: – be careful and take precautions when you try to cut the crystal.

    Yours, Katze

  7. Hi katze,

    The cs137 source is a tg77 tube used as a spark gap in old radar technology. You might still find a few on ebay.

  8. We are specialized in making crystal pxiel and arrays and prism,windows for various detectors,

    Scintillator crystal:Ce:YAP,Ce:YAG,Ce:LuAG,CsI(Tl),NaI(Tl),CsI(Na),PbWO4,LYSO(Ce),BGO,YSO(Ce),Eu:CaF2 BaF2,CeF3,CdWO4,Plastics,CeBr,LaCl3,LaBr3,Plastics scintillator,

Leave a Reply