GAMMA RAY SPECTROSCOPY - PowerPoint PPT Presentation

1 / 1
About This Presentation
Title:

GAMMA RAY SPECTROSCOPY

Description:

... 00 12951.00 14153.00 15221.00 16180.00 17167.00 18356.00 19406.00 20355.00 21435.00 22273.00 22610.00 23286.00 23591.00 24145.00 24226.00 24448.00 24165.00 ... – PowerPoint PPT presentation

Number of Views:44
Avg rating:3.0/5.0
Slides: 2
Provided by: RayT6
Category:

less

Transcript and Presenter's Notes

Title: GAMMA RAY SPECTROSCOPY


1
GAMMA RAY SPECTROSCOPY Zhengqing Qi
DISCUSSION
METHOD
ABSTRACT
We expect the backscattering difference to show
only a rise at the backscatter peak because the
Pb body absorbs more ? rays than Al causing a
difference in the two backscatter peaks. The
oscillation at 650 KeV is due to a possible
difference in source placement giving a shift in
peaks.
A study on the spectra produced from gamma
rays of three different sources, namely Cs-137,
Co-60 and Na-22, is considered though the use of
a scintillator and photomultiplier. An aluminum
and lead backscatter body was used to observe the
gamma rays interaction with a different
backscatter material. A background run was also
done to subtract noise contributions from data
samples. Distinct peaks signaling the
photopeak, Compton edge, and backscatter peak are
noticed for all three sources. Oscillations from
the backscatter difference can be attributed to
the misplacement of the source. A noticeable drop
in the backscatter peak for the lead body is
observed for all three sources due to leads
higher atomic number. Thoughts are also made for
unidentifiable peaks.
Excited nucleus gives off one or more gamma rays
in its decay.

Radioactive source
NaI(Tl) crystal scintillator absorbs the gamma
ray causing the ejection of an electron. The
electron loses its energy in the crystal
producing light. The light emitted is
proportional to the energy deposited by the
original gamma ray.
Photopeak
Backscatter Peak
Compton Edge
Scintillator
Photocathode
Light from the scintillator undergoes the
photoelectric effect after hitting the
photocathode. Photoelectrons are produced and are
accelerated to the dynode.
Possible detector geometry contribution
Compton scattering is responsible for the
backscatter peak and Compton edge. The
backscatter peak is formed from energy of a ? ray
that has been scattered 180o while the Compton
edge is formed from the energy of an electron
scattered 180o by a ? ray.
INTRODUCTION
PMT
Experimentation in nuclear physics relies
not only on counting techniques but also energy
measurements. A gamma spectrometer will measure
the count rate and energy of gamma rays emitted
from a radioactive source allowing a study on
gamma-ray interaction with matter and also
properties of the radioactive source.
Excited nuclei have quantized energy states that
can emit one or more gamma rays when it
de-excites to the ground state. An investigation
of the gamma ray spectra from three radioactive
sources will be considered though the use of a
scintillation detector. A study on the spectra
will include the consideration of photopeaks,
Compton edge, backscatter peaks, and backscatter
difference. Possible contributions to unknown
peaks will also be discussed.
The setup for the experiment contains a lead
oven to host the radioactive source. The oven is
situated below the scintillator and
photomultiplier tube (PMT). The PMT outputs a
signal which is ran though the amplifier and
multi-channel analyzer (MCA). The analyzer is fed
to the computer which collects the data for
analysis. The methods for data collection
will be discussed along with data analysis and
suggestions for future study. The radioactive
sources used for this study include Cs-137, Co-60
and Na-22.
4 secondary electrons are emitted for every
electron that hits the dynode. There are 10
dynodes so the overall gain will be 410 resulting
in a large pulse of e- hitting the anode. The
pulse is proportional to energy of the original
gamma ray.
Backscatter Peak
Dynodes
Photopeak
Compton Edge
Anode
Amplifier
After amplification, the signal is fed to the
multi-channel analyzer. The analyzer sorts the
pulse according to size, recording bigger pulses
in higher numbered channels. There are 1024
channels.
Multi-channel analyzer
Conclusively, higher channels means higher gamma
ray energy. A file of counts and channel number
will be stored by the computer for plotting and
analysis.
e/e- annihilation (rest mass of an e/e- is 511
KeV, the other e/e- is scattered 180o so peak is
not at 1.22 GeV)
Backscatter Peak
Compton Edge
Photopeak
PROCEDURE
  • Three different sources were used Cs-137, Co-60
    and Na-22
  • The individual sources were placed in the oven
    for 30 minutes for data collection with Aluminum
    and then Lead as backscattering bodies. A total
    of seven 30 minute runs were taken. Two for each
    source with the different backscattering
    material, and one more for the background with no
    source.
  • The Pascal program nucspec.exe was used for
    data acquisition.
  • Coarse and fine grain settings on the amplifier
    are set to the lowest values for maximum channel
    coverage. Cuts on channels were made during
    plotting, that is, channels with only zero hits
    after a certain channel number will be discarded.
  • We identify and calibrate the peaks according to
    the decay schemes of each radioactive source
  • Cs-137?exticed Ba-137? ? (662Kev)?Ba-137
  • Co-60?? (1.17 MeV) ? (1.33 MeV) ? N-60
  • Na-22?e emission? excited Ne-22??
    (1.277MeV)?Ne-22

CONCLUSION
  • The spectra for three different radioactive
    sources were plotted and analyzed. The decrease
    in the backscatter peak for Lead is observed due
    to its higher atomic number.
  • Possible errors include the off placement of the
    source (resulting in oscillations for backscatter
    difference).
  • Consistent peaks at 130 KeV hints at possible
    geometry contributions or noise that requires
    further investigation.
  • The identification of an unknown radioactive
    material can be done though gamma spectroscopy
    and comparing the spectra with literature values.
Write a Comment
User Comments (0)
About PowerShow.com