Then the source is
taped to the end of a wooden block, this becomes the jig which holds everything
stable while we change
reflectors.
The test jig is positioned
a short distance in front of the beta probe (pancake probe, thin end window, or
thin sidewall GM) in such a way that none of the radiation reaches the probe directly. Once the jig is assembled, take
a baseline background count with no reflector inserted. This count will be
subtracted from all further counts, to yield a net contribution of only the reflected/backscattered
betas.
Set the scaler to run for
60 seconds, totaling the results when finished. A longer run time will be
statistically more significant.
Start by taking a baseline
background reading of the jig and source but no target. There is always a little
reflection from the jig itself,
and even your own body, so take an
accurate count and subtract this number from the individual reflector target
counts to yield a net CPM.
Gross CPM - Background CPM = Net CPM due to target
backscatter.
expressed
as: G^CPM - B^CPM = N^CPM
Next, one at a time, different reflector materials are
placed in the same carefully measured
position in front of the beta source.
Space it away about an inch.
Notice the space mark on the jig base.
What we are looking for is
the effect of different atomic number (Z) materials on a beta stream from the
source, only those
betas that are scattered
180 degrees or close to it are detected.
Our test results:
13 Al- 759 net
CPM
http://www.qsl.net/k/k0ff//Beta%20Backscatter/Aluminum%20Al.jpg
26 Fe-1038 net
CPM
27 Co- 1056 net
CPM
28 Ni- 1055 net
CPM
29 Cu- 1114 net
CPM
47 Ag- 1320 net
CPM
http://www.qsl.net/k/k0ff//Beta%20Backscatter/Silver%20Ag.jpg
50 Sn- 1320 net CPM
82 Pb- 1821 net
CPM
Conclusions:
Betas will bounce off
(backscatter) more as the Z of the reflector is increased. Elements that are
very close in Z number
show little difference form
one another, due to the randomness of the radioactive decay process and
measuring uncertainties.
Elements far apart in Z
number have markedly different reflective properties to the beta
particles.
Chart of
Data
The correlation coefficient (Pearson's r) =
0.99
Note: Above chart custom prepared
for this paper by Dr R J Prettyman and used with permission.
Note, a timed run was
performed with Pb in the jig and a beta shield consisting of a 1.8 mm aluminum
sheet covering the probe. This is done to verify it is betas we are
seeing reflected and not Bremsstrahlung or PIXE generated characteristic X-Rays.
These do of course exist with our setup but their
contribution is minuscule with this detector, and totally swamped put by the
betas.
Suggestions for further
study:
Once you jig is set up and
tested, try using other household materials as a target. Plastic, wood, glass
are just a few.
*(note 1): Gamma rays tend to be
emitted in discreet energies, as do alphas. Betas on the other hand are emitted
in a BAND of energies, from zero, to the MAX for that particular isotope
(source). Sr-90 has a 100% yield of a beta of 546 keV MAX, 195.8 Average. Even
though backscattered betas do lose some energy in the scattering process, the
probe used here is very sensitive to betas, even of relatively low
energies. when betas are bent via a magnetic filed, they do not lose
energy, but the degree of bending has a direct relationship to the energy of
each beta. Refer to article below "Bending Betas" for projects and
details.
Associated articles in this series by George Dowell:
PIXE- Element identification
using Particle induced X-Ray Emission.
Measuring Beta Efficiency of a
Pancake Probe - Using beta sources to test beta efficiency of
probes.
Bending Betas- Bending betas
using magnets. demonstrates the principles of Magnetic Beta
Spectroscopy.
Nuclear Radiation Lab
Experiments With Absorbers- Using calibrated absorbers to determine beta
energy, alpha, beta and gamma penetration.
Have fun
George
Dowell
