HEAS Activation Experiment

[Ryan Ginter]

Richard has been kind enough to provide me with some space next to his fusor at HEAS next week, so I'll be taking advantage of that to run an activation experiment within the fusor's neutron flux. This will allow me to get some experience working with a verified neutron source, which I later hope to apply to my fusor once it's completed.

The experiment will be conducted using a Z-Graded lead castle, a block of paraffin moderator, a one-ounce ingot of pure Indium metal, and my Radiacode-102 scintillation detector. The whole assembly will be fairly compact, roughly a cube with six inches to each side. The goal of the experiment is as follows:

• Capture gamma spectrum data of prompt emissions by Indium under bombardment of fusion neutrons.

• Capture the gamma spectrum of Indium-116m1 decay after the fusion run has ended.

• Record the gamma activity of the sample over time to experimentally confirm the half-life of In-116m1.

There are various advantages and disadvantages associated with using the Radiacode for this experiment. The obvious negative is the 1 cm^3 scintillation crystal, as this will severely limit the sensitivity of detection as well as the resolution of the gaussian photopeaks. The advantages however are the ability to save collected data and reset the count accumulations wirelessly without opening the lead castle, the constant rolling data collection that will allow for simultaneous measurement of the gamma spectrum and activity decay rate, and the compact size allowing the device to be fully contained within the lead castle. Having it fully encased will greatly reduce background counts, improving the ability to resolve activation peaks from the noise.

Now I shall overview the materials used in the assembly's construction, starting with the Z-Graded lead castle. I built the castle about a year ago, stemming from a desire to record a clean spectrum of the trinitite sample I had purchased from Richard at HEAS #34. The Castle was built around a 3D-printed PLA box. The inner volume for placing samples was roughly 4.5" X 4.5" X 5.0".

The box was printed with no infill, such that it would be solid all the way through and the walls were made 1 cm thick. This both provided a central structure for the lead castle and helped to attenuate the radioactive emissions of samples placed within before they could reach the denser elements in the castle walls and potentially cause X-Ray fluorescence to shine back at the detector.

Aluminum Sheets of 0.02" thickness were epoxied to the outer walls of the PLA box. The aluminum sheets were sized such that they stuck out from two sides of the box by 0.02", allowing the sheets to meet flush at each edge. Multiple layers of aluminum tape were then placed over each edge, providing some additional attenuation for any X-Rays that might slip between the seems.

The copper layer was made in the same fashion, epoxying 0.02" thick copper plates over the aluminum and then sealing the edges with copper tape. Bonded to the outside of the copper faces were lead-free pewter sheets of 0.1" thickness. The composition of the pewter was 92% Tin, 7.5% Antimony, and 0.5% Copper. These sheets were again cut such that they would meet flush around the edges, but the edges were hammered together to reduce the gap as opposed to using metallic tape.

Finally, The outer lead walls were epoxied to the pewter, each measuring 0.25" in thickness. Just as with the pewter, the edges were hammered together to reduce the size of gaps between the walls. The walls were then wrapped in Kapton tape to allow the castle to be handled without risk of lead contamination.

The thickness of lead was chosen as a compromise between cost, mobility, and attenuation. The thickness of all subsequent layers was calculated to provide at least 95% attenuation of the neighboring layer's fluorescence, most having been made a bit thicker than necessary.

Pictured here is the lead castle. (Note: I'm currently working on finishing the lid today. Up to this point I've been laying the various layers freely on top the castle.)

Background measurements were taken outside and within the castle after its completion to evaluate its performance. The original background photon energy outside the castle peaked between 80 and 90 KeV, but after attenuation, this peak was reduced by over 30X. The attenuation factor was weaker at slightly higher energies, allowing for a new peak to form around 150 KeV. The intensity of this 150 KeV peak was still reduced by over 13X compared to the unshielded background. The original total activity of the background at the location of measurement was 3.25 CPS. Placing the detector within the castle dropped that reading to 0.42 CPS, a reduction in total activity of 7.7X.

The following graphs compare the original background in orange to the Z-Graded castle background in green. The measurements have been adjusted so that their total counts per channel are based on the same duration of time.

Returning now to a few days ago, I started working on the moderator. An insert was designed for my castle to act as a carrier, enabling it to be switched between general gamma measurement and neutron activation with no mess or hassle. The carrier was printed out using SLA photo-polymer resin.

The part was designed with 2 mm thick walls and a number of support structures so the box could be handled roughly without risk of damaging it. Channels passed through the bottoms of these support structures so the wax could flow evenly throughout them. A central tunnel passed completely through the carrier, providing space for the Radiacode and activation material to be housed. SLA resin was chosen for its thermal resistance, allowing the wax to be cast directly inside.

Once the part had cured I began melting the paraffin wax and filled the carrier to the top. The volume of wax shrunk considerably as it began to solidify, so I would periodically stab a hole through the top surface and pour new wax overtop of it.

Here is an image of the moderator right after I had finished casting.

The carrier was then cleaned up and the wax was shaved flush to the top surface. After completion, the carrier was inserted into the castle as pictured here.

The moderating materials of this assembly consist of PLA, Photo-resin, and Paraffin wax. The thickness of materials between the first wall and the activation sample are as follows: 0.4 cm photo-resin, 1 cm PLA, and 4.1 cm of paraffin wax. Unfortunately, the effectiveness of the PLA and photo-resin are not known to me. They will not perform to the same degree as the paraffin wax, but they do contain significant amounts of hydrogen.

Adding the three materials together I'm probably still a bit on the low end for neutron thermalization, but that's all the space I have available within the castle. I believe it should be sufficient for these experiments.

Indium was chosen to be the activation material for the large number of photo-peaks from In-116m1 decay at a wide range of energies. A one-ounce ingot of the metal was hammered over an anvil to produce a round about 1" in diameter. A mock-up of my Radiacode was then 3D-printed, and the round was hammered over the end of it and shaped into a cap. The sample weighed in at 24.63 grams once completed. The indium cap was inserted over the portion of the scintillation detector that housed the crystal, which provided a perfect fit.

Image of the Indium cap over the Radiacode-102

The Radiacode and Indium cap placed within the moderator

Having finished constructing the assembly, I inserted the Radiacode and Indium target into the moderator and placed the lid back on the castle. The counter was left collecting background data for the next day, just to ensure there weren't any unexpected sources of noise in the measurement.

To my surprise, a photopeak was in fact waiting for me when I returned. Comparing it to the normal background for my castle, a prominent peak had formed between 20 and 30 KeV. I was confused by the nature of the source for a moment, but soon realized it might have been fluorescence from the Indium. Sure enough, I looked up Indium's characteristics X-Rays and it fluoresces at 24 and 27 KeV.

1.5 day spectrum of the Indium sample shown in orange and the typical lead castle background in green.

This was fairly unexpected, I knew the whole point of Z-grading the walls was to remove outside sources of X-Ray fluorescence from inside the chamber, but I never thought it would reduce them to such an extent that characteristic X-rays could be measured using nothing but background radiation as the X-ray energy source.

While I find it interesting that the characteristic X-rays are visible, the poor definition and sensitivity of the Radiacode for signals in the low KeV range render it unsuitable for reliably identifying elements in this manner.

That all aside, it would appear as though I have everything ready for next week's experiment. I'll be performing control measurements Friday night and will activate Indium during the Saturday demonstration run. Data collected during these runs will be added to this post in the days following the event.

[Richard Hull]

Fabulous and very complete exposition of your effort to activate indium and record decay gamma curves. I look forward to supplying neutrons for your experiment.

Richard Hull

[Cinar Kagan]

I am in the process of designing and building a neutron producing fusor (not enough progress yet to make a post about my progress).

However, neutron detection is truly a problem for me since I live in Turkey and there is no ebay to buy Russian He-3 tubes, rem-balls etc. or even a simple Ludlum Model 3 without a probe without paying 10k+ euro to just buy some of the parts that are necessary. I cannot order from ebay USA or anything due to the custom limit being 30 euro. Thankfully, I managed to get a Radiacode 102 when I went to other parts of Europe.

I had a similar idea on vaction like putting a moderator + Pb shielding around the crystal of a Radiacode 102 to create a ludlum hammer probe-like neutron detector but it detects the activation gammas of indium instead of recoil protons from neutrons. I hope you get positive results so I can do the same in my fusor for neutron detection.

Also I'd like to point out that the manufacturers of the Radiacode device claim that the radiacode 102 device can measure neutrons via recoil but I'm not sure how efficiently. If you were to test that out and look for a peak in the high MeV region where it is close to the proton energy, I would appreciate it.

Best of lucks in your experiment

Cinar Kagan

[Dennis P Brown]

Great idea and so very kind of Richard to help you. But that is Richard. He is always trying to improve access and knowledge in science as well as encourage people! HEAS is certainly a huge proof of his major dedication to those worthy goals.

Good luck with your work. Sorry I will miss it.

[Cinar Kagan]

I wish I was there too. Sadly I have school as well as visa issues (I have no visa nor applied to it yet, maybe next year).

Hopefully I will come to HEAS 2 years after this HEAS (HEAS #35?).

Cinar Kagan

[Ryan Ginter]

I started recording a gamma spectrum of Thorium two days ago in preparation for the experiment this weekend. A half-gram sample of elemental Thorium was placed roughly 1 cm from the Radiacode and left to accumulate while I was off at work.

This is the spectrum that had collected when I returned home

I haven't needed to make any precise measurements in some time, so the device hasn't been calibrated since March of this year. The low range of the spectrum had remained fairly accurate, but the old coefficients became worse as the measured energy increased.

The Radiacode was left to collect the Thorium sample for an additional day to further resolve the 2614 keV peak, after which the data was used to recalibrate the device. The software has a built-in calibration function using three data points to create a polynomial fit. The channels residing on the photopeaks for 238, 583, and 2614 keV were used to recalibrate the device across the full spectrum.

Here is an image of the Thorium spectrum after recalibrating.

Short-duration measurements were then taken from Cesium-137 and Americium-241. These isotopes were used to verify the calibration at the low and middle ranges of the spectrum.

The Cesium-137 check source spectrum

The Americium-241 spectrum, taken from the outer housing of a smoke detector.

Results indicated the isotopes were identified within +/- 1 keV of the target energy, which is typical for the Radiacode-102. The 26 and 32 keV peaks for Americium and Cesium respectively both read high by around 5 keV. This is also typical for the 102 model, as it has poor resolution below 50 keV. The accuracy can be improved by placing the three calibration points lower on the spectrum, but Indium-116m1 won't have any peaks below 138 keV, so the accuracy at and above this value should be prioritized.

The reason I am performing the calibration now and not on the day of the experiment is due to the small size of the scintillation crystal. The 1 cm^3 CsI crystal requires long accumulation times to resolve high energy peaks, so I won't have time to prepare on-site. Thankfully the Radiacode's measurements are temperature-compensated, so the reading is unlikely to drift significantly over the coming days.

I will be taking along the Cesium-137 check source to quickly verify the low and middle ranges of the spectrum before the experiment is started. If required, the low and middle calibration points can be edited beforehand, and the original coefficient for the third (high-energy) point can be left unchanged.

[Alexey Khrushchev]

Were you able to do an activation experiment? It seems to me that the thickness of lead will be insufficient to block X-rays, but I may be wrong.

[Richard Hull]

I will allow Ryan to comment on the results of the work done at HEAS.

Alas, radicode is a very, (really extremely), torpid detector unless near deadly levels of gamma radiation when being investigated. ( my belief from seeing it in action for the first time and from past reports on it made here.

If you have a survivable gamma source of far less then 1 mrem/h, like most humans might come across, and many, many hours or days to set the radicode near the weak source to record gamma spectrograms, it is great.

Fusors really kicking out neutrons run for a few minute length runs. The x-rays never exceed 50-60kv. The higher energy gammas from fusor activated materials are difficult to record on a radicode in so short a period of activation. This is especially the case with short lived isotopes once activated.

In short, the radicode, in my opinion, is intrinsically worthless of activation gamma recording of amateur level fusor operated activation experimental work.

Only a large volume, sensitive gamma scintillation crystal on a PMT, backed by a first rate electronic gamma ray spectrograph is capable of gamma activated isotopes,, in amateur fusor hands, is capable of doing the work desired.

Richard Hull