Gamma Spectrometer Build
In 2023, I set out to put together a gamma spectrometer with a high resolution and very low background. Commercial spectrometers with similar goals are extremely expensive and can weigh a couple of tons with the lead castle. I chose to sacrifice the overall shielding capability for the weight and cost reduction.
Detector System
At the heart of the system is a Bicron 1.5M2.25/1.5L NaI(Tl) scintillation detector. This was purchased second-hand, but it was in great condition and extremely sensitive. This type of detector is designed for gamma spectroscopy and is normally used by nuclear labs for sample analysis. This is powered by a commercial Multi Channel Analyzer (MCA) that records the spectrum.
Lead Castle Build
One of the most important parts of a gamma spectrometer system is the shielding that encapsulates the detector. This helps to block out natural radiation from the environment, such as granite or cosmic rays. The thicker the shield is, the less noise you will have in the system.
Commercial systems sometimes use interlocking lead blocks that have a characteristic shape to help prevent something called “shine-through” where the cracks between normal blocks let radiation through. These blocks are expensive to buy, so I decided to try casting my own interlocking lead bricks.
It is important to remember that molten lead is very dangerous not only because of its temperature but also because of the vapors it gives off. Because of this, melting lead needs to be done outside with appropriate PPE. I had a leather fire resistant apron and gloves as well as a vapor-rated respirator.
The blocks were designed in CAD based around the detector and sample chamber. These were 3D printed and used as negatives for the sand mold. The mold was ordinary oil-based casting sand that was packed around the 3D printed parts in order to create the chamber for the lead to flow. The lead was poured and allowed to cool, but I wasn’t able to get a very consistent top surface, so I had to re-cast a couple of the blocks. This whole process came out significantly better than I was expecting. I had created a whole Lead Castle from lead ingots and they fit well together.
One of the design considerations for a lead Castle is that lead has a naturally occurring isotope of Pb-210. So even if you made a lead chamber with walls a mile thick, the detector would still pick up the lead isotope as noise. The emission of Pb-210 is a low-energy X-ray peak so it is possible to block this with the right materials. Copper is the most common option as it is well suited for absorbing the X-rays.
The best design is one of several layers with the outer layer being thick lead blocks, then inside of that a Copper layer, and lastly a layer of Tin to absorb the last of the X-rays. Sometimes a final interior layer of thick plastic is used to decrease the effect of beta particles generating X-rays inside of the chamber, but I do not plan on measuring samples with a high beta emission. The chamber included a removable door and plastic lining to reduce the risk of lead contamination.
This build ended up being very successful. Outside of the chamber the scintillation probe read about 900 counts per minute (CPM), but when it was placed inside and closed this dropped to just about 30 CPM, or about one count every two seconds. This is an incredible reduction for a 1.5×2.25″ NaI(Tl) crystal.
After calibrating the detector and running some measurements, I calculated the resolution of the detector to be just about 6.8% which is great for the NaI(Tl) system. I took measurements of several materials including calibration sources of the standard Cs-137 and Lu-176. These spectra are shown below:
