Photomultiplier Scintillator Cosmic Ray level monitoring

[hardhacker]

I'm currently in the process of building a new muon detector using PMTs and Scintillators http://www.hardhack.org.au/scintillator_detector for testing the validity of my other cheaper detector projects.

But I was also thinking of building another detector for logging average Cosmic Ray levels over an extended period. As I'm also involved in an Amateur Radio Astronomy Group http://www.radio-assa.org.au/ and thought it would be interesting to see if there is any trend between cosmic ray showers and radio noise levels or other radio correlations.

My question - is it was absolutely necessary to use two detectors in a coincidence detection arrangement particularly monitored on a data logger over time.

A PMT can measure the energy of each strike, so I assume it would be possible to filter out terrestrial radiation, as they would a have lower energy, particularly with lead shielding around the scintillator. Where a muon would pass through easier and have a higher energy.

Thoughts welcome

Regards

Robert

 

[bcrowell]

Cool project!

A few thoughts:

Your scintillators are small compared to the ones I've seen used for this purpose. This will cut down a lot on your efficiency. On the other hand, it might make lead shielding more practical than it would be with a large scintillator.

Have you looked at sources of lead, and methods of disposal afterward? People are not as casual about this kind of thing as they used to be. Solid lead is not a hazard unless you ingest it or get it in an open cut, but I don't think you can just throw it in the trash and let it go to a landfill.

Scintillators have very poor energy resolution. Also, a muon is only going to deposit some of its energy as it passes through the plastic. So I'm not sure you'd be able to accurately discriminate between muons and natural gamma background just by looking at pulse height. Do you know what the energy spectrum of cosmic-ray muons is?

What is your reason for not wanting to do it in coincidence mode? Are you trying to increase your efficiency to make up for the small detectors? Is the cost of the coincidence electronics an issue?

 

[Bob S]

Lead is fairly good shielding; a 2" thick lead brick will attenuate cobalt-60 gammas (1.1 and 1.3 MeV) by about a factor of 10. But lead is often contaminated with radioactivity (except lead from medieval church roofs in Europe) (but fallout may have contaminated tis also). So using only one scintillator is not sufficient to remove background. If you use two scinitillators on a separated by about 20 cm of lead (or copper) on a pivot, you can do an azimuthal cosmic ray survey. A cosmic ray muon will lose about 2 MeV per gram/cm², or ~ 1 MeV in a 1/4" scintillator (similar to a cobalt-60 gamma). When you use two scintillators, you will also have opportunity to take coincidence curves to "time in" the two PMTs (as well as plateau them). You will need two adjustable 2 kV or 3 kV dc power supplies.
Bob S

 

[hardhacker]

Excellent feedback! I had similar thoughts, but not having the education background in this area, my knowledge is limited to what I can read or google.

We are very much the amateur tinkerer :-)

My main issue to-date has not been the photomultiplier tubes (thanks ebay), electronics, HV regulated power supplies or even lead or coper.

It is mostly the limited availability of affordable scintillator plastic in Australia and the shipping costs from anywhere else. The three pieces that I have got are ideal in terms of their sensitivity to muons.

They are much smaller than I would like, but tinkerers just work with what they can get.

I am in the process of building a coincidence detector with two panels, but I was hoping to get away with one for the Radio Astronomy Group.

Just need to find a cheaper source for scintillator plastic sheet.

 

[Bob S]

Just need to find a cheaper source for scintillator plastic sheet.

You should ideally have 8" square or 12" square by 1/4" plastic scintillators. If you have 2" PMTs (like the RCA 6342) then you will also need a Lucite (UVT) transition piece to match the scintillator to the PMT. Lots of details. I have seen two arrays (5 or 6 Geiger Mueller tubes side-by-side) of GM tubes used as a cosmic ray telescope.
Bob S

 

[hardhacker]

Here is a rough sketch of what the detector may look like, working with what I have available.

The http://www.hardhack.org.au/files/BC412.pdf" block measure 89mm x 89mm x 38mm and respond to an energy between 100 KEV to 5 MEV and emits light between 420nM and 450nM

The scintillator blocks will be coupled using Dowel Corning DC4 and the photomultiplier are a http://www.hardhack.org.au/files/s83020f.pdf" sensitive to light between 350nM to 500nM.

 

[Vanadium 50]

You are going to need some shielding at the top if you want to be monitoring cosmic ray muons. Otherwise you will be swamped with single hits. It doesn't have to be lead - it may be easier to get steel or brass.

I think you are going to be unhappy with your scintillator geometry. It's surface area is very small - you'll only get a fraction of a Hz of cosmic rays, and it's thicker than it needs to be. The more typical geometry is that of a large "paddle": say 100 cm x 10cm x 1 cm.

If you're willing to experiment, I might suggest making your own liquid scintillator. The most commonly available scintillator is quinine sulfate. You'll need to evaporate a lot of tonic water to get the concentration up, and may have to play a bit with the chemistry.

 

[hardhacker]

Thank you I really appreciate the feedback. The sizes of scintillator are not through choice, it is what I can scrounge from here an there, I would buy larger panels if I could find a supplier.

Nevertheless I'm still in an experimentation phase, so this will help me better understand what a verifiable muon strike really looks like. This will be enclosed in a light proof metal box. So I can move the shielding in and out, adjust spacings etc. Also helping to refine the electronics and data-logging systems.

 

[Vanadium 50]

Nevertheless I'm still in an experimentation phase, so this will help me better understand what a verifiable muon strike really looks like.

I'm not so sure. Your count rate depends on a lot of things - geometry, shielding, quantum efficiency, etc. Nonetheless, a good guess would be 0.05 Hz. Your singles rate from the tube is also hard to estimate, but you probably have several hundred Hz from thermionic emission alone: call it 1 kHz.

So you're starting off with a signal that is 20,000 smaller than your background.

Coincidence will help, of course - if you have a 50 ns window, that will reduce your background rate to 0.05 Hz: the same as your signal rate.

The actual numbers are there just to set the scale - your problem is that you expect accidental coincidence rates comparable to your cosmic ray rate. How can you tell you are seeing cosmic rays?

This is one reason people go to big counters - the tube noise rate is independent of size, but the signal grows as surface area.

This will be enclosed in a light proof metal box.

You've never done this before, have you?

Let me try and explain what you are up against. Sunlight has a few x 10¹⁷ photons/cm². Phototubes are sensitive to a single photon. So you need to have something like a rejection of outside photons of 10¹⁸.

If you look at most undergrad lab setups of this, you will find something like this:

• Scintillator wrapped in white paper.
• Next, scintillator wrapped in black paper.
• Scintillator wrapped in black paper again.
• Next, scintillator wrapped in black tape. Extra attention is paid the edges and corners.
• Then the setup is put in a dark box.
• Then a black blanket is placed over the dark box.

Well designed dark boxes are double boxes - one side has the external connections for the power and signals, then the cables are routed to a different side for the internal connections.

 

[hardhacker]

That is an excellent explanation why I need larger scintillators.

Does quinine sulphate diluted in distilled water work as a scintillator how many parts per ml?

Or do you need the sugar in tonic

You've never done this before, have you?

No not at all Its very interesting nonetheless. but I was going to wrap everything, shielded cables and enclose the electronics in separate die-cast boxes.

If you look at most undergrad lab setups of this, you will find something like

I have looked at a number of examples of undergrad lab setupsand none have lead shielding between the scintillators?

Thank you for the information

Robert

 

[Vanadium 50]

The reason you want shielding is so that a) soft particles don't cause a signal and b) so you don't have a Compton electron leave one scintillator and enter another, giving you a false coincidence. You don't need lead, and you don't need much thickness, but you should use something. I've had good success with thin (~1/8" maybe) sheets of brass.

It will take some experimenting to get the right light output from quinine. I know the solution needs to be slightly acidic, but I don't know what the right concentration is. If it were up to me, and I couldn't just buy quinine, I'd start with diet tonic water (no sugar), and experiment with a UV lamp to try and get the solution with the right brightness.

 

[hardhacker]

Thank for the tips I have some uv lamps I'll give it a try, make a good blog. Tonic and cosmic rays.

I guess I could custom make up some brass boxes to enclose the scintillators, I have some brass sheet in the workshop.

Regards

Robert

 

[Bob S]

In the not-too-recent past, I built a cosmic ray telescope as a senior in college, and later on, a senior in college under my supervision built one. Here is some useful information.
For shielding between scintillators, the photon attenuation coefficient in lead at 1 MeV is about 0.07 cm²/gram, or 0.791 per cm. so you really want ~ 3 cm minimum of lead (or equiv nonmagnetic material) between the two scintillators to minimize accidental coincidences from correlated events (Compton scattering, etc.).
Cosmic ray (muon) energy loss in scintillator or water is about ~ 2 MeV per gram/cm² (Bethe-Bloch ionization). Here is an estimate of signal yield per cm of radiator. Muons originate in proton hadronic cascades via pion decay in the upper atmosphere, and have a range of about βγcτ meters before decaying (β and γ are the usual relativistic parameters, c = 3 x 10⁸ meters/sec, τ = 2.2 microsec.) before decaying. Some reach the ground.
See pdf
http://beamdocs.fnal.gov/DocDB/0010/001068/001/A%20tutorial%20on%20beam%20loss%20monitoring.pdf
Muon energy loss in material is given by the Bethe-Bloch equation (1) on page 2 of pdf. The muon is generally considered a minimum ionizing particle (2 MeV per gram/cm²), with the minimum at ~ 200 MeV (See Fig. 1 on page 2 for protons). See Table on page 10 for scintillator photon yield, and Eq (11) on page 13 for Cerenkov light photon yield per cm. Cerenkov detectors are less sensitive to low energy radiation background radiation, but usually require 14-stage photomultipliers (e.g., RCA 6810).

Light output comparison; NE 102 plastic scintillator and water Cerenkov radiator.

Radiator---------------------------------- Ne 102-------------Cerenkov
Photons per cm in water--------------------20,000 ------------ 160
Light collection efficiency---------------------------- 10% ------------
Light collected (photons)--------------------2,000 --------------16
Photoelectron convers. eff. ---------------------------20% --------------
Photoelectrons-------------------------------400----------------4-------
Photomultiplier tube gain ----------------------------100,000----------
Anode electrons ----------------------------40 E6--------------400 E3
Output pulse width-------------------------------------50 ns------------
Peak current--------------------------------130 μA-------------1.3 μA
Cable & termination (RG-58)----------------------------50 ohms------------
Output volts------------------------------------6 mV-------------60 μV

You will need more signal before going into fast digital logic. Possible ways of increasing signal are: 1) Thicker radiator; 2) Better light collection efficiency (note: there is no internal reflection in liquid scintillators); 3) Increase photomultiplier gain by cranking up voltage; 4) Use higher impedance cable and terminator.
The output pulse should probably go into an ultrafast input discriminator/comparator with an adjustable threshold (e.g., see Linear Technology selection chart at http://parametric.linear.com/html/high_speed_comparators)
and maybe an ultrafast x10 preamp.
Bob S