A week or two ago I set out to detect the infamous Polonium 210 gamma ray at 803.1 keV using a NaI(Tl) scintillation detector and a mere 3.7 kBq of Po210. This isn’t too hard, or so I have been told, to do with a HPGe or using my equipment and a significantly larger activity of Polonium. The problem is that only a few gammas will be emitted, only some will actually reach the detector, and of those only a tiny fraction will be detected! But… Who am I to listen? Lol I think that I found the gamma and it stands out in the spectrum, but I thought I would go a little further. This is not a formal paper (note my informal tone). I just thought that I would post a little more than a simple message.
I have a temperature stabilized environment with a average temperature fluctuation of 0.8 c. from the mean over a period of six hours. The detector was allowed a full day to warm up and become stable. Calibration was performed several times and with several sources, including Co60 and Cs137. Redundant calibrations were performed and tested against each other to detect any changes.
A new (less than seven days old) small circular plastic disk containing approximately (+/- 20%) 3.7 kBq of Po210 was placed directly in front of the detection crystal at a distance of 1 cm. Between the source and the crystal, a thin Pyrex glass layer was placed. The test was allowed to run for six hours and then repeated without the Po 210 source to account for background. The background was removed from the sample spectrum to produce the results.
A scientific result which can credibly called “true”, insomuch as any result is true”, requires at least five standard deviations from the mean of a set of data to rule out likely error. Given the very low amount of data logged and the generally entropic nature of the testing setup, such an outcome is unlikely. As a result, a positive declaration of the detection of Po210 gammas using the experiment as performed is unlikely.
A set of 60 data points was taken before and after background removal. These data sets were treated as a population set from which a simple population standard deviation was calculated, for both before and after background removal. Based upon calibration of the unit, the channel numberd 832 was the most likely channel to detect the gamma in. For both the origional data and the data with background removed, the channel, 832, displayed a clearly greater than other channels near it and for the set. For the raw data, channel 832 was 3.3636861676 deviations from the mean and with the background removed, the same channel was 3.2797495046 deviations from the mean. The variance between the data with and without background was 2.56%.