Department of Physics, Creighton University

PHY 562 : Nuclear Instruments and Methods

GAMMA RAY SPECTROSCOPY

Author: Shani Perera

Partners: John Higgins
Date: 04th Oct 2019
 1. Abstract
This experiment mainly focuses upon features of gamma-ray spectroscopy in relation to
nuclear processes. In this report, spectrometer calibration using Co-57, Cs-137 and Co-60
and effects due to dead time and due to surrounding metals (Al and Pb) has been discussed.
Furthermore, the energy of x-ray peaks, relative efficiencies, absolute activity and energy
resolution has been calculated. Finally, two unknown sources were identified as Ba-133
and Cd-109. Theoretical concepts are discussed in detail with appropriate derivations.

2. Introduction
Most radioactive sources produce gamma rays, which are of various energies and
intensities. When these emissions are detected and analyzed with a spectroscopy system, a
gamma-ray energy spectrum can be produced. NaI Detector has come to be accepted as the
standard scintillation material for routine gamma-ray spectroscopy. From the various ways
gamma rays can interact in matter, only three interaction mechanisms have any real
significance in gamma-ray spectroscopy: photoelectric absorption, Compton scattering,
and pair production.

Photoelectric absorption predominates for low energy gamma rays (up to several hundred
keV), pair production predominates for high-energy gamma rays (above 5-10 MeV), and
Compton scattering is the most probable process over the range of energies between these
extremes. The atomic number of the interaction medium has a strong influence on the
relative probabilities of these three interactions. [1]

Most of the radiation measurement systems use pulse-height analysis to sort out the
different radiation energies striking the detector. Thereby, to discriminate against
background radiation, and scattered radiation, and to identify the emission energies of
unknown radionuclides. [2]

3. Theory
The energy resolution (R) is the closest energy separation between two detected particles
such that the amplitude distributions of each is separately noticeable in the composite
spectrum. [3]
∆𝐸 (1)
R=
𝐸

PAGE 1
where ∆E is the full width at half maximum height (FWHM) of the full energy peak and E
is the full energy value. The peak is often fitted to a Gaussian distribution and the FWHM
for a Gaussian distribution is approximately equal to 2.35𝜎. [4] From the Poisson
distribution,

∆𝐸 =2.35 √𝑁 (2)

Since N is proportional to energy,

2.35 (3)
R=
√𝑁

The probability of full energy conversion in the detector material may be close to unity for
low 𝐸𝛾 , but it becomes progressively lower at higher energies as the importance of
photoelectric absorption decreases relative to Compton scattering.

➢ Compton scattering: The interaction process takes place when a photon interacts by
scattering from a loosely bound atomic electron. [3]
➢ Photoelectric absorption: In this process, a photon undergoes an interaction with a
tightly bound inner-shell atomic electron in which the photon completely
disappears after transferring all its energy to the electron.
➢ Pair production: In this interaction the gamma-ray photon disappears and is
replaced by an electron-positron pair.

Figure 1: γ-ray pulse height spectrum for a shielded setup of source and detector

PAGE 2
In producing the spectra, the detector output signal (pulse height) is converted
electronically into an integer, which is indicated as channel number in the abscissae of the
graphs.

X-ray peak: When γ-ray photons interact with the high-Z material of the shield, they can
produce x-ray photons. The energy of these photons is substantially lower than that of the
original γ-ray photons. Therefore, as shown in Figure 1, the energy deposited by them in
the detector produces a peak at the lower end of the energy spectrum. [5]

Backscatter peak: This is caused by detection of γ rays that have been scattered toward the
detector after undergoing a 180° scattering outside the detector.

Compton continuum: The curve below the Compton edge is called the Compton
continuum. This corresponds to the distribution of energy between incident photons and
the scattered electrons during Compton scattering at different angles. Since Compton
scattering is possible at all angles, this continuum can extend up to the beginning of the
spectrum.

Full energy peak: The peak at the upper end of each graph, corresponds to events in which
the γ ray transfers all its energy to electrons in the crystal.

The energy of the backscatter peak, Eb, is the energy of the scattered γ ray after 180°
scattering, whereas the energy of the Compton edge, Ece, is the energy given to the recoil
electron in a 180° scattering event.

𝐸𝛾 = 𝐸𝑏 +𝐸𝑐𝑒 (4)
The fraction of detected gamma-rays that contribute to the full energy peak is very
dependent on the size and shape of the detector and on atomic number Z and γ ray energy.

Peak efficiency is given by,

𝜀𝑇 (5)
𝜀𝑝 =
𝜀𝑃𝑇
where 𝜀𝑇 is total efficiency and 𝜀𝑃𝑇 is peak/total ratio.

Source activity is given by,

𝑁0 (6)
A=
𝑡

where t is the counting time for the measurement and 𝑁0 is given by,

𝑁(𝐸1 ) 𝑁(𝐸2 ) (7)

𝑁0 = T +
𝑁(𝑠)

PAGE 3
where T is the sum of all three peaks and given by,

T = 𝑁(𝐸1 ) + 𝑁(𝐸2 ) + 𝑁(𝑠) (8)

Absolute efficiency at 𝐸1 is given by,

𝑁(𝐸1 )+ 𝑁(𝑠) (9)

𝜀(𝐸1 ) =
𝑓(𝐸1 )𝑁0

where 𝑓(𝐸1 )is the number of gammas of energy 𝐸1 produced per decay of the source.

The detection efficiency is given by,

𝑁 (10)
𝜀=
𝑓𝐴𝑡

where N is the net count in a peak measured for a time t, f is the branching ratio.

The dead time is the time after each event during which the system is not able to record
another event. [6]

It is calculated by,

Dead time = Real time – Live time (11)

4. Procedure

Figure 2: Block diagram of the set up

Part A

• Arrange the set up as shown in the Figure 2 and set the MCA mode to PHA.
• Initially HV was set to zero then switched the HV settings to +800 V. But the
peak was right shifted in the screen. Therefore, HV was reduced to +700 V.
• Then placed the Cs-137 source and changed the gain until the main photopeak lie
about a channel that is 1/3 of the maximum channel.
• Course gain was set to 2 and Fine gain was set to 2.5.

PAGE 4
 • After that, using the Cs-137 and Co-57 calibrated the MCA to automatically
convert channels into energy.
Cs – 137 → 661.64 MeV
Co – 57 → 122.06 MeV & 136.48 Mev
• Measured the energies of Co-60.
• Finally, calibrated the feature of MCA using all three sources.
Co – 60 → 1170 MeV &1330 MeV

Part B

• Experiment was conducted according to the lab manual.

• After using Pb sheets it was important to wash hands.

Part C

• Recalibrated the MCA after changing the gain to observe the low energy portion
of the Cs-137 spectrum.

Part D, E, F and G

• Experiment was conducted according to the lab manual.

5. Data Analysis
Part A

1400

1200 y = 2.3618x - 5.058

R² = 1
1000
Peak energy

Actual data
800

600 Experimental data

400
Linear (Experimental
data)
200

0
0 200 400 600
Channel

Figure 3: Calibration Curve

PAGE 5
R2 is a measure of goodness of the fit which should be equal to 1 for a straight line. In this
graph slope = 2.361765248 ± 0.015737672 (2.36 ±0.02), intercept = -5.05799 ±5.2312.
R2 = 0.999955599.

Part B

Table 1: Real time, Live time comparison for Co- 60 source

Co- 60 Source Real time Live time Dead time
Top Shelf 101.57 100 1.57
Bottom Shelf 100.63 100 0.63
According to equation 10, dead time was calculated for Co-60 source placing it on top shelf
and bottom shelf. When it was placed at top shelf dead time was higher than when it was
placed at bottom shelf.

Spectras obtained by keeping the Al and Pb sheets beneath the source (Figure 4) and on
top of the source (Figure 5) were as below.

900
with Pb sheets beneath
source only
800 with Al sheets beneath

700

600

500
Counts

400

300

200

100

0
0 500 1000 1500 2000 2500
energy (keV)

Figure 4: Count vs energy graph when Al and Pb sheets were kept beneath the source

According to the figure 4, it can be seen that, a low energy peak has been created after
placing metal sheets beneath the source.

PAGE 6
 800
source only
with Pb sheets above
700 with Al sheets above

600

500
Counts

400

300

200

100

0
0 500 1000 1500 2000 2500
energy (keV)

Figure 5: Count vs energy graph when Al and Pb sheets were kept on top of the source

When metal sheets were placed above the source, number of counts corresponding to the
energy has reduced. A backscatter peak also can be seen. When comparing Al sheets and
Pb sheets, Al sheets have blocked more gamma-rays than Pb sheets.

Low Energy Peak

30000

25000

20000
Count

15000

10000

5000

0
0 200 400 600 800
Energy (keV)

Figure 6: Characteristic low energy peak of Co-57

PAGE 7
Part C

Low energy x-ray peak was observed at 27.702 keV. (Refer appendix C)

Part D

Counts
6000

5000

4000
Counts

3000

2000

1000

0
0 500 1000 1500 2000 2500
Energy (keV)

Figure 7: Na-22 Spectrum

After deducting the background counts from the total counts the above spectrum was
obtained. (Figure 7)

Table 2: Details about Na-22 two peaks.

Peak 1 Peak 2
Centroid 506.44 keV 1280.89
Net counts 82081 16227
Gross counts 92641 17768
FWHM 38.58 62.70
𝐸
From the equation, I = where A is source activity and t is time, it can be seen that
𝐴𝑡
intensity, I ∝ E.

Table 3: Comparison of ratios

Experimental Expected
Ratio of intensities 2.529 1.78
Ratio of efficiencies 2.835 3.4

PAGE 8
Experimental ratio of efficiencies was calculated using equation 10.
0.044
The accepted ratio of efficiencies = 0.013 = 3.4 (2″×2″ detector with the source separation
of 2.5 cm) was calculated using equation 5.

Part E

Table 4: Details related to sum peaks

Na-22 Co - 60
Live time 167 s 214 s
Net count 1411 100
Gross count 1675 147
FWHM 62.70 12.307
Centroid 1808.787 2523.823
Energy range 1720.671 – 1899.126 keV 2450.00 – 2591.538 keV

Table 5:Calculation of source activity

Na-22 Co-60
Sum of all 3 peaks (T) 99719 43722
N0 1043680 4799498
Source Activity (A) 6249.578 22427.56
Using the equations 6,7 and 8 source activity of Na-22 and Co- 60 was calculated.

Table 6: Calculation of absolute efficiency

Na-22 Co-60
Peak 1 Peak 2 Peak 1 Peak 2
Centroid (keV) 506.444 1280.892 1167.852 1328.759
Net count 82081 16227 22188 21434
Gross count 92641 17768 28899 23463
FWHM 38.584 62.7 61.538 64.615
Absolute
efficiency 0.079998 0.009473 0.004644 0.002515
Using the equation 9 absolute efficiency of the detector for two gamma-ray energies of
Na-22 and Co-60 was calculated.

PAGE 9
Part F

Table 7: Calculation of Energy resolution (R)

Source Energy FWHM Resolution (R) Percentage R

Cs – 137 652.63 45.985 0.070461 7%
Na – 22 506.44 38.584 0.076187 8%
1280.89 62.700 0.04895 5%
Co – 60 1167.85 61.538 0.052693 5%
1328.76 64.615 0.048628 5%
From the resolution values, an idea about how close in energy the two gamma rays can be
and still be identified as separate peaks in the overall gamma spectrum could be taken. [3]

4500
4000 y = 3.1796x - 39.905
R² = 0.9938
3500
3000
FWHM^2

2500
2000
1500
1000
500
0
0 200 400 600 800 1000 1200 1400
Energy (keV)

Figure 8: Graph of FWHM2 vs Energy

Gradient = 3.18 and Intercept = -39.91.

According to the equation, FWHM2 = CE + B, where B is the electronic noise

contribution to FWHM.

C= 3.18 and B= -39.91.

PAGE 10
Part G

1800

1600

1400

1200

1000
counts

800

600

400

200

0
0 50 100 150 200 250 300 350 400 450 500
energy (keV)

Figure 9: Graph of count vs energy for the first unknown

Energies
Observed 45.504 96.89 171.06 307 360
(keV)
Channel 20 57 110 206 243
First unknown source was predicted as Ba-133 after comparing the energies. (Appendix
G)

Paired 53.16 ( 2) 80.99 ( 35) 160.60 ( 1) 302.85 ( 18) 383.85 ( 9)

radiations

PAGE 11
 900

800

700

600

500
counts

400

300

200

100

0
0 100 200 300 400 500 600 700
energy (keV)

Figure 10: Graph of count vs energy for the second unknown

Energies Observed (keV) 22 93

Second unknown source was predicted as Cd-109 after comparing the energies.
(Appendix G)

Paired radiations 25.00 ( 18) 87.70 ( 4)

6. Results and Discussion

At first the spectrometer was calibrated using two sources of Cs-137 and Co-57, which was
called to point calibration. From the next step onwards, three-point calibration was used to
calibrate the spectrometer. Three-point calibration was better than two-point calibration.
The system could have been calibrated using one point, if the null points of the MCA bin
and the energy scale coincide. In general, this is not true due to the presence of an offset
on the MCA bin axis. That is, the zero energy does not correspond to the zero pulse height.
This offset is due to electronics and can be determined by using an external pulsar.
However, if one uses several known peaks in the spectrum and fits a straight line to the

PAGE 12
data, the offset can be determined by extrapolating the line to the x-axis. [5] When
calibrating it was assumed that the energy increases linearly with the channel number.

In the part B, dead time and external effects were observed. The dead time is the time that
the ADC is busy processing a signal and is not able to accept another pulse. Live time is
the time that the ADC is not busy processing a signal. When the source is very near to the
detector, activity is very high. Therefore, it takes more time to process. But when the source
is on the bottom shelf furthest from the detector, activity is less. So that, it can process one
signal after the other. Therefore, the dead time is also less.

Figure 11: Variation of scattered gamma ray energy with scattering angle [1]
According to figure 4 and figure 5, when metal sheets were placed beneath or on top of the
source, back scattered peak could be observed. From the shape of the curves in figure 11,
it can be seen that, any scattering angle greater than about 110-120° results in scattered

PAGE 13
photons of nearly identical energy. Therefore, a monoenergetic source will give rise to
many scattered gamma-rays whose energy is near this minimum value, and a peak will
appear in the recorded spectrum. [1] (Appendix B) Since these photons lose most of their
energy during the collision, these peaks appear at the lower end of the spectrum.

Figure 12: Actual pulse-height spectrum recorded with a NaI(Tl) detector and 137Cs

Cs-137 emits a single gamma line of 662 keV. The only decay channel of Cs-137 (half-life
of 30.07 years) is by β− decay (that is the conversion of a neutron into a proton with the
release of an electron and electron anti-neutrino) which makes the daughter Ba-137 in an
excited state called Ba-137m (half-life of 2.6 minutes). (Appendix C) The decay of that
excited state is accompanied by the production of gamma rays in the same way that the
decay of an excited atomic state produces visible or infra-red photons. This is actually the
origin of the 662 keV photon line of the Cesium, but there are other, lower energy photons
from the settling of that excited state as well such as, low energy x radiation due to internal
conversion of the gamma ray backscatter at the low energy end of the Compton
distribution.
The photo peaks observed were not sharp lines, as expected, but a somewhat broadened
peaks, and the Compton edge is rounded. This was caused by the imperfect energy
resolution of the NaI(Tl) detector. Since some radioactivity is present everywhere (i.e.,

PAGE 14
background radiation), the spectrum should be analyzed when no source is present. The
background radiation must then be subtracted from the actual measurements.

In the Terrestrial Background Radiation, the major radionuclides of concern are potassium,
uranium and thorium and their decay products, some of which, like radium and radon are
intensely radioactive but occur in low concentrations. [7]

For typical gamma-ray energies, the photoelectron is most likely to emerge from the K
shell, for which typical binding energies range from a few keV for low-Z materials to tens
of keV for materials with higher atomic number. Conservation of momentum requires that
the atom recoils in this process, but its recoil energy is very small and usually can be
neglected. [1]

Finally, the two unknown sources were identified as Ba-133 and Cd-109. The energy
values obtained were compared with the energy values of the sources to identify the
unknowns.

7. Conclusion
The characteristics of gamma spectroscopy were studied using the Co-57, Co-60, CS-137
and Na-22 sources. These features include photo peaks corresponding to gamma-rays
emitted by the nuclear transitions, backscatter peaks formed due to metal sheets and low
energy X-ray peaks. Many sources of possible errors exist, including distance calibration
errors and background radiation errors. A very good knowledge about the gamma-ray
spectroscopy was gained from this experiment.

8. References

[1] G. Knoll, Raation Detection and Measurement, John Wiley & Sons , 2010.

[2] "Pulse-Height Spectrometry," Radiology Key, 26 02 2016. [Online]. Available:

https://radiologykey.com/pulse-height-spectrometry/. [Accessed 02 10 2019].

[3] S. J. Cipolla, "A Laboratory course in Nuclear Physics," vol. 2.4, no. 5, 2019.

[4] J. Lilley, Nuclear Physics: Principles and Applications, Wiley, 2001.

PAGE 15
[5] .. N. A. S, Physics and Engineering of Radiation Detection, 2015.

[6] "Dead time of detectors," Nuclear power, [Online]. Available: https://www.nuclear-

power.net/nuclear-engineering/radiation-detection/detectors-of-ionization-
radiation/dead-time-of-detectors/. [Accessed 04 10 2019].

[7] Theremino System , "Gamma Spectrometry," 03 09 2015. [Online]. [Accessed 10

2019].

[8] W. M. v. W. B R S Simpson, "Absolute activity of (133)Ba by liquid scintillation

coincidence counting using the 4pi(e,X)-gamma extrapolation technique," Applied
radiation and isotopes : including data, instrumentation and methods for use in
agriculture, industry and medicine, 2008.

Appendix A

Cs-137 Spectrum
2500

2000

1500
Count

1000

500

0
0 200 400 600 800 1000
Channel

Figure 13: Cs-137 Spectrum obtained before calibrating the MCA

PAGE 16
 2. Calibration curve
1400

1200

1000
Counts

800

600

400

200

0
0 500 1000 1500 2000 2500
Energy (MeV)

Figure 14: 2 point Calibration curve

80000

70000

60000

50000
Counts

40000

30000

20000

10000

0
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Energy (MeV)

Figure 15: 3-point calibration curve

PAGE 17
 Appendix B

900
with Pb sheets beneath
w/o Pb sheets
800

700

600

500
Counts

400

300

200

100

0
0 500 1000 1500 2000 2500
energy (MeV)

Figure 16: Comparison of gamma spectroscopy by keeping Pb sheets beneath the source

PAGE 18
 800
source only
with Pb sheets above
700

600

500
Counts

400

300

200

100

0
0 500 1000 1500 2000 2500
energy (MeV)

Figure 17: Comparison of gamma-spectroscopy by keeping the Pb sheets above the

source

PAGE 19
 Appendix C

Cs-137
1200

1000

800
Counts

600

400

200

0
0 100 200 300 400 500 600 700 800
Energy (keV)

Figure 18: Cs-137 x-ray peak

Figure 19: Cs-137 decay scheme

PAGE 20
 Appendix D

Figure 20: calculation of FWHM of two Na-22 peaks and the sum peak

Appendix E

Na-22
6000

5000

4000
Counts

3000

2000

1000

0
0 500 1000 1500 2000 2500
Energy (keV)

Figure 21: Peaks of Na-22

PAGE 21
 Co - 60
2000
1800
1600
1400
1200
Counts

1000
800
600
400
200
0
0 500 1000 1500 2000 2500 3000 3500
Energy (keV)

Figure 22: Peaks of Co-60

Appendix F

Figure 23: Graph of FWHM^2 vs energy for Cs-137

PAGE 22
 Appendix G

Figure 24: Simplified Ba-133 decay scheme [8]

Figure 25: Simplified Cd-109 decay scheme

PAGE 23