Proceedings of the International Symposium on Nuclear Physics (2009)
722
Neutron Backscattering Technique for Landmine Detection
*Saroj
Bishnoi,Tushar Roy, Ashish Agrawal,Tarun Patel and Amar Sinha
Laser & Neutron Physics Section,
Bhabha Atomic Research Centre, Trombay, India
.
* email: saroj@ barc.gov.in
Introduction:
To develop & design a portable neutron
source based detection system for
hydrogenous rich materials such as
landmines, explosive, licit drugs etc, we have
carried out feasibility experiments using PuBe neutron source for detection of landmines.
Fast neutrons from the Pu-Be source are back
scattered more by the buried landmines than
the surrounding soil. Landmines include antitank mines (ATM) and anti-personnel mines
(APM). ATM are typically about 5 kg or
more in mass while APM are much smaller,
often less than 300 g, thus making them
difficult to detect. Although conventional
methods like metal detectors are available,
but they are not very successful for nonmetallic landmines and/or landmines buried
at relatively greater depth. Therefore a more
confirmation method is needed to
unambiguously detect landmines. To
supplement metal detector, we are conducting
feasibility study using isotopic neutron
source which primarily detected back
scattered neutrons. This way one can
supplement the information of metal detector
through neutron sensors. Additionally we are
also exploring the possibility of the using
prompt gamma sensor.
Experimental results with explosive simulant
melamine and High Density Polyethylene
(HDPE) are presented, along with effect on
count rates due to the variations in the distance
between source and soil (standoff distance).
studied melamine (C3H6N6) sample (which is
very good explosive simulant) of 300 g placed
in a cylindrical holder of 5cm diameter and 10
cm height embedded into the surface of the
sand layer. Other hydrogen rich substances like
HDPE (9cm X 9cm X 5cm) and coal
(bitumen), were also studied
He3 detectors
Standoff distance
n source
Sand
Fig (1) Experimental Set up
Experimental results & discussion:
We have investigated effects on back scattered
neutrons due to variation in standoff distance,
sample depth, spatial distribution of count
rates. Our measurement (fig2.) shows that
small variations in the standoff distance
significantly influence the count rate. It was
therefore decided to use two identical sets of
detector, with a certain distance (14cm)
between them and a source positioned exactly
in the middle. In this way, there is always a
reference value available if one detector is
above the hydrogenous sample (fig3.) and the
other is not. Also the distribution of the count
rates becomes symmetrical in both side of
sample and the measured points can be well
fitted by a Gaussian curve.
( variation with standoff distance)
9000
Methodology:
sample=hdpe in dry sand
Buried depth= 5cm
Acquisition Time=10sec
8000
Pu-Be (1x10 n/s)) neutron source with two
identical sets of He-3 detectors (each set
containing three detectors) placed above the
sand (See fig1). A wooden box (1m X 0.5m X
0.5m) filled with the dry sand is designed in
such a way that it acts as a soil bed. The scans
were made over sand with a step size of 5cm
and a measuring time of counts is 10s per step.
When detectors are moved over the sand
horizontally, the hydrogenous samples buried
in sand were detected by observing the
anomalies in the back scattered neutrons. We
Back Scattered counts
6
7000
6000
5000
4000
3000
2000
0
2
4
6
8
10
12
14
16
18
Standoff distance(cm)
Fig (2) Count Rates with standoff distance
Proceedings of the International Symposium on Nuclear Physics (2009)
4500
detector set 1
detector set 2
set1+set 2
3000
2500
Melamine we can see that the signals are
much higher in HDPE compared to that of the
melamine
since
melamine
is
less
hydrogenous. Further experiments are being
planned to repeat the experiments with D-D
neutron source.
18000
2000
16000
14000
0
-30
-20
-10
0
10
20
30
sample posion from centre(cm)
Fig. (3) Spatial distribution of count rates for
melamine by individual detectors set.
An important feature of the scans shown in
Fig.3 is that the distance between the two
peaks is approximately equal the distance
between the detectors. This is due to the fact
that the highest back scattered counts are
measured when the sample is positioned
approximately in front of each detector set.
Also when the sample is placed at the centre
position then we get peak at sample position
by taking the sum of both detectors. From
result of explosive stimulant melamine (fig4)
and the HDPE (fig5), it is evident that the
peaks values get reduce as well as get
broaden, with increase of the sample depth in
sand and
Back scattered counts
1000
500
1cm
5cm
10cm
15cm
Acquisition time =10sec
Hdpe buried in sand
at various depths
6Detector He-3
1500
12000
10000
8000
6000
4000
2000
0
-30
-20
-10
0
10
20
30
sample position from center(cm)
Fig (5). Spatial distributions of count rates
for the HDPE sample
18000
hdpe(1cm))
hdpe(5cm))
melamine(1cm)
melamine(5cm)
16000
14000
Back Scattered Counts
Back Scattered counts
Sample(melamine)depth
4000
in dry sand=1cm
Aquisition time=10s
3500 Standoff distance=5cm
723
12000
10000
8000
6000
4000
2000
0
-30
Back Scattered Counts
4000
1cm
5cm
8cm
10cm
(Melamine(300 gm))
Aquisition time=10s
-20
-10
0
10
20
30
sample position from centre(cm)
Fig (6) Comparison of the spatial
distributions for HDPE &melamine
3000
Conclusion:
2000
1000
0
-30
-20
-10
0
10
20
30
Sample Position(cm)from center
Fig(4). Spatial distributions of count rates for
the melamine sample.
After a certain depth (10cm in case of
melamine and 15cm in case of HDPE) it gets
flatten as equal to background. Also on
comparing the results (fig 6) of the hdpe and
The feasibility experiments with Pu-Be has
successfully been used for the detection of
landmines. Additionally, it may be possible
using prompt gamma detector techniques to
try to identify the chemical composition of
various elements at a range. This could allow
the detection of materials within landmines
such as various plastics or explosive.
References:
[1] Cor P.Datema et al, IEEE Transactions on
Nuclear Science, Vol. 48, No. 4, August2001
[2]B. Kirlaly et al, Radiation Physics and
Chemistry 61 (2001) 781–784