Search in the document preview
The characteristics of BF3 detector were studied in this experiment by making slow neutrons detection.
The operating voltage for the detector was measured by plotting the characteristics curve of the BF3
detector. The integral and differential pulse height spectra were obtained using SCA. The differential
pulse height spectrum was also obtained by using MCA. Finally, the branching ratio of Boron-neutron
reaction was determined form the differential spectrum.
Neutrons are neutral particles. They do not produce ionization directly. For their detection, they may be
allowed to interact with a material so that the resultant is charged particles capable of producing
ionization. Therefore, all types of neutron detector involve a converter material designed to carry out this
conversion together with one of the conventional radiation detectors.
The detection of neutrons may be divided into two groups depending upon their energies.
1. Fast neutron detection
2. Slow neutron detection
For fast neutron detection, one may use a material in which the threshold energy for the reaction is
slightly less than the neutron energies. For this U 238
, Th 232
, Np 237
etc. may be used depending upon
neutron energies. The light nuclei such as hydrogen may be used in such a way that its nucleus is knocked
out as a result of neutron interaction. Plastic scintillators used for fast neutron detection are based on this
proton recoil method.
For slow neutrons, chambers filled with a gas such as BF3 or He
3 or chambers coated with U
235 , U
, B 10
, etc. may be used. The different types of detectors for slow neutron detection are
1. BF3 detector
2. He3 detector
3. Fission chamber
4. Activation foils etc.
Among these BF3 detector is widely used for detection of slow neutrons. In this experiment our aim as to
study the characteristics of a BF3 detector.
BF3 detector characteristics
A widely used detector for neutron measurements is the BF3 proportional counter. The characteristics
study of BF3 counter gives a functional knowledge of working of the detector and various electronic
modules (NIM) which follow the detector for signal processing. The characteristics studied are effect of
applied voltage on count rate (characteristic curve), the effect of discriminator level (integral bias curve)
and the differential spectrum.
In this detector BF3 services both as the target for the slow neutrons conversion into secondary particles
as well as proportional gas. Other gases containing boron can be used but BF3 is near-universal choice
because of its superior properties as a proportional gas and very large cross section for thermal neutrons.
BF3 detector is also preferred because it has good gamma ray discrimination properties and is less costly
then 3 He or fission chamber. In most commercial detectors BF3 is 90% enriched in
The chamber is filled with boron tri-fluoride gas with a fine anode wire mounted in a cylindrical cathode
shell, and is operated at a sufficiently high voltage to cause internal gas amplification, producing an
output pulse directly proportional to the energy of the charged particle created . The best way to
determine the operating voltage of a BF3 detector is to generate a characteristic curve (count rate
versus high voltage). The curve increases with increasing voltage for some time but then becomes
nearly flat. The operating voltage of the BF3 detector should be selected on center of the plateau. If
the operating voltage is set too high, electronic noise and pulses due to background gamma rays can
exceed the threshold setting and generate spurious counts.
The neutron interacts with the 10B in the gas in the following ways:
10 B + n
7 Li + α Q = 2.792 MeV
10 B + n
* + α Q = 2.310 MeV
The resulting ions in each case share the available momentum, resulting in a continuum of energies for
each charged particle. The 7 Li is preferentially left in the excited state (94% dominant), with the reaction
to the ground state occurring 6% of the time. The 7 Li
* ion subsequently de-excites via the emission of a
447keV γ-ray. A good characteristic of this kind of detector is therefore its excellent discrimination
against γ-radiation. Another characteristic is the BF3 counters efficiency as a function of neutron energy.
At thermal energies, the 10
B (n,α) reaction has a cross section of 3840 barns.
The pulse height spectrum collected for any set of neutron interactions will have a number of
characteristic features. In the case of the 10
B (n,α) reaction there will be two full energy peaks due to the
deposition of energy from the two reactions in equations. The relative intensities of these two peaks
should provide an indication of the branching ratio of the interactions. In most practical cases, the size of
the detector tube is comparable to the range of the α-particles and recoil lithium nucleus produced in the
reaction with boron. This means that some of the events will not deposit their full energy in the gas due
to collisions with the chamber wall. This is known as the wall effect and results in a continuum of partial
energy depositions in the detector. Also present are two energy steps due to the loss of the α-particle or
Li 7 ion in the counter wall. Since the incoming neutrons have energies in the thermal range, they
carry no appreciable momentum, therefore the two reaction products must travel in opposite directions.
If one of the ions travels towards the chamber wall, resulting in a partial deposition of energy in the
gas, the other must travel away from it, and is therefore likely to deposit its full energy in the gas. It
is expected that wall losses are seen for only one reaction product at a time, where the combined
energy deposition distribution off all events where one reaction product strikes a wall is simply the
sum of the energy deposition for each event.
Figure 1.Differential Pulse Height Spectrum from A BF3 Detector
Experimental setup and results
Experimental setup was arranged as shown in figure 2. The applied voltage was increased in small steps
starting from zero volts and counter started giving counts at 1800 volts. Increasing voltage was kept
continued in small steps up to about 2500 volts and counts per 20 seconds were noted at each level of the
voltage. Count rate vs. applied voltage was plotted which is shown below in figure 3.
Figure 3.BF3 Voltage Characteristics Curve
1700 1800 1900 2000 2100 2200 2300 2400 2500 2600
Figure 2.Experimental Setup
The operating voltage was found to be 2200 volts and was fixed at this level from the high voltage
supply. The discriminator level was increased from 0 volts to 2.5 volts in small steps (0.1volts).
Counts/10 seconds were taken thrice at each level of the discriminator up to a level (2.5 volts) where no
count rate was observed. The average count rate was plotted as a function of discriminator level. The
observations, calculations and results are shown below:
Table 1.Integral Pulse Height Spectrum
0 1635 1632 1651 1639
0.1 1624 1652 1660 1645
0.2 1587 1539 1587 1571
0.3 1513 1550 1501 1521
0.4 1487 1449 1533 1490
0.5 1511 1503 1497 1504
0.6 1393 1454 1414 1420
0.7 1443 1379 1345 1389
0.8 1380 1386 1323 1363
0.9 1242 1292 1274 1269
1 1198 1297 1233 1243
1.1 1138 1196 1179 1171
1.2 1098 1075 1124 1099
1.3 970 1027 1012 1003
1.4 891 853 921 888
1.5 808 862 804 825
1.6 707 663 705 692
1.7 529 507 536 524
1.8 316 332 366 338
1.9 199 185 202 195
2 93 69 79 80
2.1 52 56 39 49
2.2 35 27 38 33
2.3 23 31 20 25
2.4 11 20 21 17
2.5 7 6 4 6
Then SCA window level was set at 0.1 volts and the lower discriminator level was increased from 0 volts
to 2.5 volts in small steps (0.1volts). Counts/minute were taken at each level of the discriminator up to a
level (2.5 volts) where no count rate was observed. The count rate was plotted as a function of
discriminator level. The observations, calculations and results are shown below:
Table 2.Differential Pulse Height Spectrum
0 369 0.7 304 1.4 598 2.1 72
0.1 255 0.8 343 1.5 659 2.2 59
0.2 218 0.9 368 1.6 919 2.3 66
0.3 267 1 435 1.7 1051 2.4 67
0.4 244 1.1 415 1.8 901 2.5 45
0.5 235 1.2 444 1.9 769
0.6 262 1.3 491 2 354
0 0.5 1 1.5 2 2.5 3
Figure 4.Integral Bias Curve
Figure 5.Differential Pulse Height Spectrum
Integrated counts under peak A = 4725
Integrated counts under peak B = 237
Ratio of peak B to peak A = 4725/237 = 19.94
Branching ratio of B 10
(n,α)Li 7 reactions = 94/6 = 15.67
The differential pulse height spectrum was also collected on MCA and the above data was taken from
there given below:
Integrated counts under peak A = 73285
Integrated counts under peak B = 2897
Ratio of peak B to peak A = 73285/2897 = 25.3
The energies corresponding to the wall effect edges C and D were calculated as follows:
Energy of edge C = (2.792-2.310)*1/(2.4-1.7) = 0.688 MeV
Energy of edge D = (2.792-2.310)*0.3/(2.4-1.7) = 0.206 MeV
0 0.5 1 1.5 2 2.5 3
The operating voltage for the BF3 detector was found to be 2200 volts by plotting the voltage
characteristics curve in which 2200 volts corresponded to about the center of the plateau region. This
central point selection as the operating voltage has the reason that the count rate is nearly independent of
the applied voltage at this point and so the electronic fluctuations do not affect our observations
significantly. It can be seen from the curve that the count rate decreases after about 2500 volts. This is due
the fact that SCA window when fully opened has the value of 10 volts and when the height of the pulses
of the maximum amplitude exceeds 10 volts, they are not counted and so by increasing the applied
voltage more and more, more pulses go beyond the SCA range and count rate continues to decrease.
The differential pulse height spectrum for BF3 detector was obtained using SCA window set at 0.1 volts
level and increasing the lower level in steps of 0.1 volts. The spectrum is the plot of count rate vs.
discriminator level. We start analyzing the spectrum from left side. At the most left the peak touching the
count rate axis is due the electronic noise pulses and the pulses formed due to the interaction of some
fraction of gammas emitted in the Li 7 excited state reactions. Next, on the right, an edge is formed due to
the wall effect when alpha particle strikes the detector’s wall and Li 7 is fully absorbed in the gas volume.
So this edge should give the energy acquired by the Li 7 ion out of the Q value of the reaction 2.310 MeV.
(Not considering reactions of Q value 2.792 MeV has the reason that those reactions occur with very
small percentage 6% so their contribution can be neglected). Since, location of the reaction in the detector
volume can be such that its separation from the detectors wall can extend from zero up to the maximum
range of the alpha particle so any energy in the range from ELi to ELi+Ealpha=2.310MeV can be deposited
in the gas volume. Therefore a continuum appears up to the next edge which corresponds to the wall
effect in which Li 7 ion strikes the wall and alpha particle is fully absorbed in the gas volume. Due to the
similar theory, a continuum next this second edge extends up to the full energy peak B. it is worth
mentioning that the second edge and corresponding continuum ride over the continuum corresponding to
the first edge previously discussed. The peak B corresponds to the full energy 2.310 MeV deposition
within the gas volume. On its right, the small peak A corresponds to the energy 2.792 MeV which is the
Q value of the Li 7 ground state reactions. The small height of this peak is due to the fact that the Li
ground state reactions occur with very small percentage.
The B 10
(n,α)Li 7 reaction branching ratio obtained from the differential spectrum is 19.94 which is ratio of
the integrated counts under the peak B and those under the peak A. Actual value of this ratio is 15.67.
The error may be introduced by not correctly locating the peaks end points while finding the integrated
counts under the peaks. Also there may be the contribution of background counts.
The operating voltage of the BF3 counting system was found to be 2200 volts. It was concluded that to
plot the voltage characteristics curve correctly the applied voltage must be increased in small steps of 20
or 30 volts. The integral pulse height spectrum was obtained using SCA and differential pulse height
spectrum was obtained using SCA as well as MCA. Form the spectra it can be concluded that BF3
counting system gives no information about the energy of the neutrons. It can only be used to detect
neutrons and to measure their flux.
1. Knoll, G.F. ; Radiation Detection and Measurement, John Wiley & Sons (1999)
2. Nasir Ahmad ; Experimental Radiation Detection, CNS-20, (1987)