Improved Proton Recoil: Fast Neutron Detector?

Nuclear related topics

Improved Proton Recoil: Fast Neutron Detector?

Postby lutzhoffman » Mon Aug 02, 2010 8:29 pm

Proton recoil scintillation fast neutron detectors have been around now for quite some time, ever since the Hornyak button, it has had some improvements, and we know that they all pretty much work. The complaint however has always been the same: "Not sensitive enough" and the recommendation to follow is usually: "Go get an He3 tube, with a moderator". While this is true, and the advice is spot on, I am not completely satified that this proton recoil scintillation concept has been taken to anywhere near its limits.

First came the Hornyak, then Bicron came along with the target shaped rings of ZnS(Ag) scintillator, inside of a 5/8 x 2" clear disk. Finally came PRECILA, here ludlum took 4 of the 2" Bicron style scintillators, and attached them to a square Lucite block to combine the light. Ludlum also added a small Li doped plastic ZnS(Ag) thermal neutron section, to now cover the entire neutron energy range up to 20 MEV. This design is of course for commercial niche markets like weapons labs, and reactor facilities, etc, and not the home lab.

For the home lab, fast neutrons are what we are mainly after, like for fusors, potential drop accelerators etc. The new PRECILA detector head hits a sensitivity of .1mrem/hr, which is pretty darn good, but still nowhere near as good as an He3 tube with a moderator. This brings me to my point that the proton recoil scintillator, may still have a lot of room for optimization, and increased fast neutron sensitivity. The sensitivity of these is directly related to the scintillation light collecting area geometry, of the ZnS(Ag) doped scintillator, within the clear plastic matrix. All of the approaches that I have seen use only one dimension, maybe 2 if you include depth, of the clear plastic matrix, to fill with the ZnS scintillaton material. Yes they now have larger, up to 5" diameter units, for $750, all for $20 worth of plastic, and 10 bucks worth of scintillator. Below is an idea of how I propose to boost the sensitivity of a proton recoil scintillator, to a higher level:

Specifically I was thinking: What if you instead made a similar 5-6"D version of the target ring style button, and then you machined a Lucite cone shaped reducing adapter / light guide section. To collect all of the scintillation light, and to couple this to a common size 2" PMT, it could all even be a one piece design, with the target rings cut into the end of the Lucite cylinder, to be filled later with the optical epoxy 5% ZnS(Ag) mixture. This should help, but it is still not what is possible.

To take this a step further, what if your cut linear grooves into the side barrel surface of the Lucite cylindrical section, which attaches to the PMT, and then filled these same grooves with the scintillation mixture, that fills the end rings. These scintillator filled linear grooves on the side of the barrel, would be in addition to the target style ZnS filled ring grooves on the end away from the PMT, which normally contain the ZnS(Ag) epoxy scintillation mixture?

Various different end geometry ZnS(Ag) fill methods have been tried, but none have used the outside barrel surface, of a longer length clear cylinder. Figuring the ideal length and groove geometry would not be that hard, and this longer length collecting area should work, since Lucite is pretty clear to the blue scintillator wavelength.

The final idea is: Progressively deeper linear groove’s cut into the cylinder, instead of just the radial end ones, which deepen as you go away from the PMT end of the cylinder, going towards the end. Fabrication would be simple: A simple table saw, with an angle jig, could be used to cut the progressively deeper grooves into the sides of the Lucite cylinder. To boost the scintillation collecting area even more, alternating grooves could be cut to a different depth. The grooved cylinder could then be placed into an internally polished Al tube, with one end closed. It could then be back filled under vacuum, with the ZnS(Ag) optical epoxy mix. The only thing left would be to mount it to the PMT.

(Since posting this, I have learned that Doug already had an idea like this before: By packing plastic rods as a light guides, and then filling the space between them with the ZnS/Epoxy scintillation mix. On this variation do you frost the plastic rods with sandpaper first, for better filling and coupling to the epoxy scintillator mix? I would also think that there would be an optimum ratio between the rod diameter and the total length, do you remember the size used? It would seem that hexagon shaped plastic rods would be ideal in order to to better regulate the scintillation layer thickness? I am now thinking of using this method between my 2" Bicron copy, and the PMT for some added umpfff. Since Doug mentioned that the ZnS Has a high index of refraction, which drops the light transfer efficiency: Maybe a liquid filling with an high index gelled ZnS containing fluid would increase the light collection efficiency even more.)

(Sorry Doug, Now I get it "Meterology" Sorry I did not get the spelling, I thought this column was for weather related stuff, seriously. This is why I put the post where it is, I will pay more attention in the future.)

These ideas are just some thoughts on the matter, maybe someone else could think of a different approach to increase the available ZnS(Ag) epoxy scintillator filled area. Thanks for your thoughts.....Lutz

PS: Please let me know if you have 20 g to an ounce or so of ZnS(Ag) to sell, or if you want to combine an order, thanks. Attached are the detail specs for the Ludlum PRECILA device.
(1.76 MiB) Downloaded 220 times
Last edited by lutzhoffman on Tue Aug 03, 2010 1:32 pm, edited 1 time in total.
Posts: 31
Joined: Thu Jul 15, 2010 1:57 am

Re: Improved Proton Recoil: Fast Neutron Detector?

Postby Doug Coulter » Mon Aug 02, 2010 10:47 pm

Lutz -- I'm not trying to be harsh here, just save some time in discussion about all this since it's been done,
no need to speculate.
I would point out that we do have a thread for
Please try to put all posts about detectors, there, not everywhere else.

Well, been there and done that with hornyaks both bought, and made here. You'd do well to check this out:

This is also where we got the ZnS:Ag we used to make our own hornyak which worked about twice as well as the one they sell there,
in a direct comparison on the same fusor run with identical phototubes.

It costs several bucks a gram. 50 gram minimum, which is less than half a fluid ounce, it's dense.
That's not a ripoff -- the stuff needs to be really pure and very carefully made to be good, and their stuff is good.
It's what I wound up using to make my pinhole camera, about 20mg makes a nice screen for that.

But most of their PVT based scintillator plastics kick any hornyak around the block, with 10-100 times better sensitivity, as long as
you keep the X rays off them -- which is very easy since lead doesn't stop neutrons or even slow them down much.
They have 5ns response times. They have zero of the other disadvantages listed below for hornyaks. They don't eat
their own light, the phosphor is mixed on a molecular level, they are fast, they are sensitive, they are cheap.
(The DHS is letting their old portal monitors go surplus, and they have huge sheets of scint plastic in them, and at least Geo is selling it
as surplus, fairly cheap)

And you get zillions of times faster response along with the improved sensitivity.
This isn't theory or a guess -- I've tried it all, and the plain plastic scintillator blows them all away, no contest.
It's not even close.

There are severe basic, physic(s)al flaws in a hornyak:
1. ZnS:Ag has a high index of refraction (2+) which keeps the light inside it. There is no clear glue or plastic that matches the index, not even close.
2. ZnS:Ag absorbs its own light.
3. ZnS:Ag has a very long phosphorescent tail -- seconds. After seeing room light, it's useless for about 24 hours.
4. Recoil protons have very short range in plastic -- 0 to tiny fraction of an inch.
5. This means you have to mix the phosphor and plastic or glue real intimately.
6. See number 1,2 above -- this means you can't get the light out of it well at all, even for the few neutrons that do make light.
7. The phosphorescence builds up with fast count rates quickly to the point where you can't count the next pulse in the background of it.

You really cannot fix most of this geometrically as the neutrons aren't fast enough to make usable recoil protons at all after they go through even
a pretty thin layer of plastic (fraction of an inch). You need fairly fast protons to light off the phosphor if it's ZnS. Recoil protons have an energy distribution that is flat from
zero to the neutron energy as is, which means most of them don't make light in the phosphor even if created already touching it, much less after going
through a mm of plastic which stops them cold. You can't fix that. Other phosphors are better, but they are all X ray sensitive....

It works out that 10 bucks worth of scint plastic beats a $150+ dollar hornyak by at least factor 10 every which way, including sensitivity, speed,
easy to make, and ability to count fast. Unless you need a very numb detector that won't count more than once a minute on a Be/Po neutron source (which means you can't see
it in the cosmic ray background without trick statistics and hours of counting) they are not much use except for long term averaging in a very hot neutron flux, for
which say, silver is better yet, via activation in a moderated neutron oven. Or In. Hornyaks are really that numb. For that kind of thing, activation of silver or similar
has the advantage all the way -- and zero X ray/gamma ray sensitivity to boot, rather than simply "low" as with a hornyak.

Please do try this yourself. You'll be a believer in no time.

Even a thin layer of shellac over a small cheap BGO crystal works better than any hornyak. By quite a bit.

I'll try to put up a picture of the pretty useless one we got from Eljen, along with the slightly less useless one I made.
Both are in the junk pile someplace now. Almost everything else works so much better, they aren't worth the space they take up near the fusor gear.
Even a hydrogen or plain helium filled geiger-type tube works much better, and the output signal is faster and cleaner, easier to handle.

The main advantage of any of these is with no moderator, you can get closer to the source and see more neutrons in the first place.
None of them will tell you anything much about neutron energy, however, because the knock on proton spectrum is flat from neutron energy down
to zero. And the neutron energy is spread out anyway as not all neutron interactions with the hydrogenous material result in a knock on proton, but
slow the neutrons down anyway.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
Posts: 3038
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Return to Nuclear

Who is online

Users browsing this forum: No registered users and 1 guest