Neutron detection moderator embedded activation!

This area is for discussions involving any fusion related radiation metrology issues. Neutrons are the key signature of fusion, but other radiations are of interest to the amateur fusioneer as well.
User avatar
Richard Hull
Moderator
Posts: 12604
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Neutron detection moderator embedded activation!

Post by Richard Hull » Mon Aug 10, 2020 9:28 pm

I have scoped the tube during my assembly as I adjusted the proper GM tube voltage. An immaculate scope trace resulted no noise, a flat, zero volt base line save for the huge 5 volt detect pulse sent to the Arduino interrupt pin. Finally, the tube has zero tungsten in it. The shell is pure, very thin aluminum to let in the betas to which it is most sensitive along with very low energy gammas. High energy gammas arriving at the tube would zip right through undetected. An Am241 source makes it scream as those are only 60 keV gammas and are easily detected due to photo electrons from the Aluminum shell. So, there is zero noise in the electronics and zero tungsten present in the tube. This is why I refuse to even take a count with any fusor power applied or the supply even turned on at zero variac position. I have a total power kill switch to the HV control variac before I press the count begin button. The entire area is EMF free during the count.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Richard Hull
Moderator
Posts: 12604
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Neutron detection moderator embedded activation!

Post by Richard Hull » Thu Aug 13, 2020 7:09 am

Still battling the background. Lots of suggestion from Bill Kolb as well as Frank. I just can't kill background due to gamma P.E. and Compton radiation. I attach proof in images and graphs.

First I removed probe and ran a naked background out of moderator. This is really wild (Photo + graph)
Next I encased the naked probe in 5/8" thick lead plating. got it down low but still flyer values. (Photo + graph)
Next did a background with probe back in moderator with 5/8" external lead (Photo + graph)
Next an activation run with 5/8 lead (graph)
finally back to another activation run but with just the simple 1/16 inch external covering as normal (graph)

An interesting tid-bit of info. In all graphs I now quote the differential between the highest and lowest recorded value in each background measurement over the full 600 second graph and in the activation graphs the same, but near the tail end of the 600 second count time once the average settles due to end of useful activation readings. These "spans" are listed top graph down.
24 naked probe
14 naked probe fully encased in 5/8 lead
15 Probe in moderator with background taken with 5/8 external lead shielding.
16 Activation with 5/8 external lead shielding
16 Activation with only 1/16 external lead sheeting

What real good did the 5/8 lead do to the span extremes?? Sure, it lowered the averages but not the spans.

Enjoy, mumble, muse and cogitate. I think I will just go with the 1/16th inch sheeting and carry on. I was rather unimpressed with the external 5/8 lead.

Richard Hull
Attachments
Probe naked.JPG
naked removed probe for out-in-the-open background. Graph below.
NakedprobeBKgd8-12-20.JPG
Naked probe background out of moderator. Lots of counts and highly variable.
NakedprobeBKgd8-12-20.JPG (87.72 KiB) Viewed 1313 times
Probe58Pb.JPG
5/8-inch thick lead castle built around naked probe out of moderator. graph below
58heavyextnaked probe 8-12-20.JPG
The background with the naked probe out of moderator but castled with 5/8-inch lead all around. Very low but still highly variable.
58heavyextnaked probe 8-12-20.JPG (86.82 KiB) Viewed 1313 times
Probe ext58.JPG
Photo of external 5/8-inch upper and side shields in place to run a background.
58heavyextPb 8-12-20.JPG
Background with above external 5/8-inch lead shield probe in moderator. Low but not all that great.
58heavyextPb 8-12-20.JPG (104.72 KiB) Viewed 1313 times
Agactivation1 ext58Pb 8-12-20.JPG
the 5/8 heavy external shield - Again, a good activation run lower background but I think the lead did attenuate the fast neuts a tiny bit. Is it worth it??
Agactivation1 ext58Pb 8-12-20.JPG (92.48 KiB) Viewed 1313 times
Agactivation1-16Pbext 8-12-20.JPG
Just my normal 1/16-inch lead external shield - a good activation run
Agactivation1-16Pbext 8-12-20.JPG (93.07 KiB) Viewed 1313 times
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Richard Hull
Moderator
Posts: 12604
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Neutron detection moderator embedded activation!

Post by Richard Hull » Thu Aug 20, 2020 8:59 am

To beat the background back a bit in the resultant Ag activation effort, I will increase the surface area of activation by 800%! See image below.
The beauty of the variable and movable components of the "toy block" built up moderator is showing its value. I have already tested the tube in its assembly for background in the moderator. It is just a tiny bit lower on average than the old setup. The old setup will be kept in reserve of course.

I will cover the new rectangular form in the next day of two and test it out. I will keep all abreast of how it does.

Richard Hull
Attachments
Large GM Anno.jpg
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Richard Hull
Moderator
Posts: 12604
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Neutron detection moderator embedded activation!

Post by Richard Hull » Fri Aug 28, 2020 12:58 pm

Another great activation this evening. I was able to hit 3He numbers of 93,000 cpm for about 3 minutes after 6 or 7 minutes at rates over 40,000 cpm on the climb past 60,000 and 70,000 to 80,000. I find that skipping just one day of operation the system is difficult to push above 60,000 cpm. However after two days of rest, I have to run the system and struggle to get to 50,000 counts, however if I shut down and let the fusor chamber cool and start back up 3 hours later, I have no problem hitting 80,000. I dare not experiment and leave the system off for 2 weeks. I will prep the system prior to the big HEAS event in early October as folks will expect it to function well.

I attach the graph of the Excell file.
Attachments
NeutACT 8-28-20.JPG
Hit 110 counts in the first 10 sec count! I can start the count within 2 seconds of the shutdown of fusor power.
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

ChristofferBraestrup
Posts: 53
Joined: Sat Apr 11, 2020 1:47 pm
Real name: Christoffer Braestrup

Re: Neutron detection moderator embedded activation!

Post by ChristofferBraestrup » Tue Sep 01, 2020 10:47 am

Very intersting progress! There is one problem that have been nagging me ever since you made this thread: The STS-5/SBM-20 tubes are aluminium-cased, they will undoubtedly activate as well. For aluminium activation this might be of no consequence but for silver, shouldn't the corresponding decay be two superimposed decay functions? And given the aluminium casing, couldn't the tube by itself function as a complete assembly for Al-activation?

I think it would be interesting to try out a similar concept with a flat-plate plastic scintillator coupled to a PMT with a long-ish acryllic light pipe.

User avatar
Richard Hull
Moderator
Posts: 12604
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Neutron detection moderator embedded activation!

Post by Richard Hull » Wed Sep 02, 2020 5:09 am

A non-issue for sure. Aluminum requires a lot of neuts delivered over a much longer period to activate to any measurable degree compared to Silver. Fusors are operated over rather short periods. As such, the aluminum has no figure in this case. How could the tube of the GM boost the silver activation?? It will not emit neutrons as it is activated. The silver would absorb most of the thermal neutrons that try to reach the aluminum due to the cross sectional figures. In short, the silver would act as a neutron absorber, protecting the aluminum to a degree from activation.

Of note, there are indeed two simultaneous decays going on with Silver! There are two that can fully activate rather for any given flux within 600 seconds of exposure. This is why I chose to set the Arduino up to collect data for only 600 seconds. (see graph)

Ag110 will fully activate from activation time, t=0 and be in "activation equilibrium" in 130 seconds! Ag108 will fully activate to activation equilibrium from T=0 within ~600 seconds. The instant neutrons go away, (fusor turned off), both nuclides start to decay from decay time, T=0. Thus a complex curve results plotted by solving simultaneous, dual exponential decay equations. My friend Bill Kolb has joined me in this effort and I submit his efforts on my behalf in the image below.

Aluminum has no place in any of this unless one has access to a steady 10e5 neutron flux over the normal 600 second activation period as Al27 has an atrocious cross section of 0.2 compared to either of the silver nuclides, the worst of which is 150 times more likely to capture a neutron than the aluminum shell of the GM counter! Add to this, the fact that full activation of the aluminum would require an exposure time in any flux of 10,100 seconds 17 times longer than I expose the silver for....Figuring further.... at my peak first 10 second count, (see graph), I detect 10 bq of total silver activation radiation. The aluminum by comparison would contribute but .0006 bq to that count over the 600 second neutron activation period. I now consider Bill Kolb as a partner in this endeavor. As such, the graph is 100% attributable to my old pal,William Kolb.

Richard Hull
Attachments
Bill Kolbs best fit 8-16 and 8-30 activation.JPG
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Richard Hull
Moderator
Posts: 12604
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Neutron detection moderator embedded activation!

Post by Richard Hull » Wed Nov 04, 2020 9:46 pm

I have, since HEAS, been busy with my activation efforts following a degree of calibration for my system due to Bob Reite bringing his remball system to HEAS. (my remball system has a defective indicator electronics box). Many of the preceding images in this post have been made with my Arduino data logging box. This box took 60 successive reading of 10 second counts. This comma delimited data was placed in Xcel to generate all the graphs seen in all the preceding posts. Bill Kolb has assisted with his work on generating a computed best fit for the dual silver decay paths which agree very well with the data collected in the graphs.

As the silver 110 decays with a half life of only ~24 seconds, the first ten second count is a smear of counts taken over almost a quarter of the half life, I have changed two values in only two lines of code in the Arduino program to alter the count interval to only 5 seconds taken over 120 counts. (Still a 10 minute data collection period.). It is realized that statistically halving of the count, doubles the error. However, I feel that the long 10 second count will smear, by a large amount, the first half-life's peak of Ag108. It is unfortunate that the background is so high, yet, also halved along with the total interval's count. It is hoped that the peak of the AG110 will be more representative of the first 10 count intervals.

Note: If this shorter interval proves too noisy I will return to the 10 sec count interval. This may be the case once I get the Rhodium.

I submit my new background graph over the total 10 minute count interval. I also contribute the decay graph of a super run of fusor five with a peak TIER of 825,000 n/s run with a peak neutron count in my detector of 140,000 CPS.

Richard Hull
Attachments
NeutACT 11-4-20.JPG
NeutACT 11-4-20.JPG (54.14 KiB) Viewed 499 times
Background 11-1-20 5 sec.JPG
Background 11-1-20 5 sec.JPG (58.33 KiB) Viewed 514 times
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Richard Hull
Moderator
Posts: 12604
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Neutron detection moderator embedded activation!

Post by Richard Hull » Thu Nov 05, 2020 11:34 pm

To go with the above post, Bill Kolbs best fit for the Ag110 and Ag108. The run on the 4th was really great.
While the scatter is amplified due to the 5 second counts, the trend line clustering is obvious.

Richard Hull
Attachments
Bill Kolbs best fit 11-4-20 activation.JPG
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Richard Hull
Moderator
Posts: 12604
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Neutron detection moderator embedded activation!

Post by Richard Hull » Wed Nov 18, 2020 7:24 pm

There be Rhodium!

Yes, I bit the bullet, spent the long buck and the image is contained below. It cost exactly what I recently paid for the large 6" - 6 way conflat cross!!
I hope to set this up in my enclosed moderator activation scheme in some fashion soon. More later. Yes, the sting of $500 still hurts a bit, but the die is cast.

Edit 11/20: What a difference a day or two makes! Rhodium hit $13,800 per troy ounce!
Edit #2 11/23: Rh hit $15,300 today for a while.

Richard Hull
Attachments
Rhodium.anno.jpg
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

User avatar
Richard Hull
Moderator
Posts: 12604
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Neutron detection moderator embedded activation!

Post by Richard Hull » Thu Nov 26, 2020 8:42 pm

I finally ran the rhodium activation effort. I attach images of the probe modification and of a 20 minute comparison, activation run of silver versus rhodium.

A bit of background. I have not done any activation since a week of two after HEAS. As some may remember I did fantastic activation at HEAS with over 120,000 peak CPM neutrons in an activation run. Laying fallow for two months, the wall loading was zero. I spent about three days of running fusor V to just get the fusor back up into the 80k CPM range, at best! I have shortened the count to 5 seconds per data point in the Xcel graphs, to better capture the fast decays. Nonetheless, with fusor V producing less than optimum results and the lab getting colder, I decided to try the effort of comparison runs.

As some know, the Ag 110 has a half-life of 24 seconds and decays very rapidly. It has a neutron capture cross section of ~80. Rhodium has almost double the half-life with a cross section of ~145 These two characteristics make it the ideal fast activation metal with the highest cross section of its one isotope composing 100% of the metal. Silver 109, the fast activator to Ag 110 is only 50% of the metal. Rhodium is king, for sure.

I went into these runs worried that rhodium due to a number of factors would be crippled due to the factors listed below.

1. The old silver probe is seen to have 100% coverage of the activated metal as a nice cylinder of silver of 9999 purity completely covers the GM tube. The rhodium covers only two sides not even 50% coverage.
2. The old silver probe had 100% of it active detection length exposed to the activated metal. The rhodium strips do not cover the entire active detection length of the GM tube.
3. Worse still, was the fact that the silver run on the 24th had a fusor conditioned over its 3 day prior runs so that on the 24th, ~90,000 CPM neutron counts were obtained. I skipped the 25th not running the fusor at all and in just the 36 hours lost, the Rhodium run struggled to hit 74,000 CPM ( slight loss of wall loading.)

YET! The rhodium run beat the results of the silver run in spite of all the above handicaps! I love the longer single isotope decay time that spreads the decay out over time.

Net Results

Rhodium obtained 22% more activation with only 81% of the peak neutron activation that silver had and with only 45% of the tube counting versus 100% of the tube counting the silver....Wow!

Check out the photos and graphs. Rhodium is the the cat's pajamas!

Richard Hull
Attachments
Rhodium 1.jpg
Rhodium 2.jpg
Rhodium 3.jpg
NeutACT 11-24-20.JPG
Silver, silver, silver above for comparison as noted in the above test.
NeutACT 11-24-20.JPG (84.23 KiB) Viewed 150 times
NeutACT 11-26-20.JPG
Rhodium, Rhodium "Less is more with rhodium" speaking in the Orwellian
NeutACT 11-26-20.JPG (72.52 KiB) Viewed 150 times
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

Post Reply