- Email: [email protected]
Identification of materials hidden inside a container by using the 14 MeV tagged neutron beam
NIM B Beam Interactions with Materials & Atoms
Nuclear Instruments and Methods in Physics Research B 261 (2007) 321–325 www.elsevier.com/locate/nimb
Identification of materials hidden inside a container by using the 14 MeV tagged neutron beam Davorin Sudac
b
a,*
, Silvia Pesente b, Giancarlo Nebbia b, Giuseppe Viesti b, Vladivoj Valkovic a
a Rudjer Boskovic Institute, P.O. Box 180, 10002 Zagreb, Croatia Dipartimento di Fisica dell’Universita di Padova, Vai Marzolo 8, I-35131 Padova, Italy
Available online 5 April 2007
Abstract The results of the experiments aiming to confirm the presence of explosive inside the container by using the 14 MeV tagged neutron beam are presented. Measurements were performed with paper, sugar, flour, fertilizer, tobacco and explosive (Semtex1a) as target material placed in the center of an empty container. Additional measurements were done with paper and explosive placed in the center of the container filled with the iron matrix of 0.2 gcm3 density and with the paper target shielded by the 5.1 cm thick iron shield. The results of time of flight measurements and gamma ray spectra obtained by 14 MeV tagged neutron beam have showed that investigated materials could be well distinguished in the triangle plot with coordinates being the number of counts in the carbon peak, the number of counts in oxygen peak and the number of counts in transmitted neutron peak. By using such presentation we have been able to separate paper from Semtex1a, both hidden inside the 0.2 gcm3 iron matrix. We have also been able to confirm the presence of 64.4 kg of paper behind the 5.1 cm thick iron shield corresponding to the range of 300 keV X-rays. 2007 Elsevier B.V. All rights reserved. PACS: 25.40.Fq; 29.40. M; 82.80.Ej Keywords: Explosive detection; Fast neutrons; Container inspection
1. Introduction Method of inspection of cargo containers by using 14 MeV tagged neutron (TN) beam by detection of associated alpha particles has been recently discussed by several groups [1,2]. The aim of the proposed technique is to determine the chemical composition of the suspicious volume within the inspected container, which was previously identified by the X-ray imaging. Such a two sensors technique has been improved with the results promising it to become the powerful tool for container inspection ([3–5]). In the experiment presented, the 14 MeV tagged neutron tech-
*
Corresponding author. Tel.: +385 1 4561 161/1531; fax: +385 1 4680 239. E-mail address: [email protected] (D. Sudac). 0168-583X/$ - see front matter 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.nimb.2007.03.088
nique has been used to confirm the presence of the explosive Semtex1a within the inspected container. 2. Methods and results Fig. 1 shows the experimental set-up with the investigated target placed in the middle of the container. Target was the iron box dimensions 40 · 40 · 66 cm3 and mass 9.2 kg, filled with the different materials in different measurements: 86.2 kg of graphite, 64.4 kg of paper, 78 kg of sugar, 75 kg of flour, 75 kg of fertilizer, 23.2 kg of tobacco, 100 kg of sand and 100 kg of Semtex1a. Four 300 · 300 NaI(Tl) were put at the transmission position in the cone of the tagged neutrons beam. Each detector was shielded with the 5 cm thick lead. The alpha detector (YAP:Ce) was 8 cm away from the tritium target. A quadratic collimator, 1.8 cm long, was placed in front of YAP:Ce
322
D. Sudac et al. / Nucl. Instr. and Meth. in Phys. Res. B 261 (2007) 321–325 1000
2340
0.1 ns/ch 800
3'' x 3'' NaI (4 peac.)
400
1, 2
400
axis
Target
180
neutron source
1180
3, 4
beam
600 1130
Number of counts/ch
2350
1270
Tagged neutrons
600
400 Gammas from carbon 200
435
floor
0 20
40
60
80
Fig. 1. Experimental set-up inside the container (all units are in mm).
120
140
160
180
300 0.032 MeV/ch 250 1. escape
Number of counts/ch
scintilator. For data processing, a standard NIM electronics was used. The four fast outputs from the 4-segmented PMT connected to the YAP were fed through the octal constant fraction discriminator and fan-in fan-out to the STOP of the time to amplitude converters (TAC). Outputs from the NaI(Tl) detectors were fed through the constant fraction discriminators and ‘‘or’’ logic units to the START of the TACs. Slow signals from the gamma detectors were fed through the amplifiers, quad linear gates and delay amplifiers to the acquisition system. Neutron beam intensity was kept low, 2.5 · 106 n/s in 4p, in order to avoid the problems experienced in previous experiments [6]. Iron matrix was made from iron boxes filled with the iron wire. The axis of the tagged neutron cone was directed through the container approximately parallel with the container bottom (see Fig. 1). Vertical and horizontal profiles of the tagged neutron beam were measured with the 7.62 cm times 7.62 cm NE-213 neutron detector in coincidence with the alpha counter. The experimental diameter of the neutron spot in the middle of the container was (37 ± 2) cm and (35 ± 2) cm for the horizontal and the vertical profile, respectively. As a QC/QA measures the carbon target was used. The 86.2 kg of graphite with the volume of 30 · 30 · 40 cm3 was put in the middle of the container. Fig. 2 shows the corresponding time spectrum (a) and the c-ray spectrum (b) obtained with the appropriate time window and subtracted random background. The main peaks, corresponding to 4.4 MeV excited state as well as the first and the second escape peaks are clearly seen. Fig. 3 shows the time spectra from various materials taken with the same number of alphas. Integral number of the counts under the tagged neutron peaks gives the information on the densities of the various objects (and the surroundings matrices) while the time window set on the gamma peaks give the chemical compositions of the corresponding objects. All measurements with sand, sugar, tobacco, flour, paper, paper behind the 5 cm iron shield and fertilizer were done under the same conditions described in this article. The measurements with the explosive Semtex1a and paper in 0.2 g/cm3 iron matrix were done under the similar conditions, described in [3], and
100
Time (ns)
2. escape
200
4.4 MeV 150
100
50
0 0
1
2
3
4
5
6
7
8
9
10
11
12
Energy (MeV)
Fig. 2. The time spectrum (a) and the gamma ray spectrum (b) of the 86.2 kg of carbon. The gamma ray spectrum was obtained with the appropriate time window on gamma peak. (Target volume = 30 · 40 · 40 cm3, 2 · 108 tagged neutrons, Elapsed time = 17997 s, Average intensity = 0.3 · 107 n/s).
appropriately scaled (scaling factor was found with the help of paper which was measured in both cases). Fig. 4 shows the dependence of integral numbers of TN on density for various materials. Density of the materials was found by dividing the mass with the volume of the iron box. Experimentally obtained data for empty iron box, tobacco, paper, fertilizer, flour, sugar and sand were fitted according to the following formula: N ¼ AeBq
ð1Þ
where A and B are fitting parameters, N is integrated number of the tagged neutrons and q is density. Appropriately scaled Semtex1a nicely follows the exponential fit. Semtex1a in iron matrix, paper in iron matrix and paper behind the iron shield does not follow the exponential fit as was expected. Fig. 5(a) shows the gamma ray spectra obtained for paper and paper behind the iron shield, respectively. Fig. 5(b) shows the gamma ray spectra obtained for sugar and sand, respectively. The gamma ray spectra obtained for flour and tobacco are shown in Fig. 6(a) and gamma
D. Sudac et al. / Nucl. Instr. and Meth. in Phys. Res. B 261 (2007) 321–325 1000
323
300 0.033 MeV/ch
100
250
Carbon
3.3 3.9
Number of counts/ch
Tagged neutrons
Number of counts/ch
Oxygen
paper paper behind the iron shield flour
Gammas from target
Gammas from iron shield
200 4.4 Oxygen 150 5.6 5.1
100
6.1
50
10
0 200
400
600
800
1000
1200
1400
1600
1800
0
1
2
3
5
6
7
8
250
10000
0.033 MeV/ch
0.1 ns/ch
Oxygen
Tagged neutrons
3.8
3.9
Carbon
200
1000
Empty iron box tobacco sugar fertilzer sand
Gammas
100
Numbers of counts/ch
Number of counts/ch
4
Energy (MeV)
Channel
3.3
4.4 Oxygen
150 2.7
5.6 100
5.1
6.1
10
50
1
0 200
400
600
800
1000
1200
1400
1600
1800
0
1
Channel
2
3
4
5
6
7
8
Energy (MeV)
Fig. 3. The time spectra from various materials taken with the same number of alphas.
Fig. 5. The gamma ray spectra from paper (bold black, elapsed time = 39267 s, 4 · 108 tagged neutrons) and paper behind the iron shield (black, elapsed time = 50130 s, 5 · 108 tagged neutrons) (a) and sugar (bold black, elapsed time = 39456 s, 4 · 108 tagged neutrons) and sand (black, elapsed time = 20109 s, 2 · 108 tagged neutrons) (b), respectively.
Integral number of tagged neutrons
1e+6
ray spectrum obtained for fertilizer is shown in Fig. 6(b), respectively. Oxygen and carbon peaks are seen in each spectrum except in the case of fertilizer. In the case of fertilizer the nitrogen, sulphur and phosphorus peaks are visible. In the case of tobacco carbon peaks are visible together with the broad features around 6 MeV which is characteristic for oxygen.
1e+5
1e+4
3. Triangle diagram 1e+3 -0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
Density (g/cm3) Exponential fit empty iron box, tobacco, paper, fertilizer, flour, sugar, sand semtex1a paper behind 5 cm iron shield semtex1a in 0.2 g/cm 3 iron matrix paper in 0.2 g/cm3 iron matrix
Fig. 4. The integral number of the tagged neutrons versus density.
In the c-ray spectra of Semtex1a [3], it can be seen that the nitrogen peak at 5.1 MeV is overlapped with the second escape peak from the 6.1 MeV oxygen line while the peak at the 2.3 MeV is not recognizable at all. On the other side, some home-made explosives do not contain nitrogen, e.g. acetone peroxide (TATP). For that reason we used density as a parameter in triangle diagram instead the nitrogen.
324
D. Sudac et al. / Nucl. Instr. and Meth. in Phys. Res. B 261 (2007) 321–325 250 0.033 MeV/ch
Number of couts/ch
Carbon
3.9
200
4.4
C*
150
O*
Oxygen
5.1
100
5.6 6.1
50
Ro*
0 0
1
2
3
4
5
6
7
8
sand sugar tobacco flour paper fertilizer Semtex1a
Energy (MeV) 300 0.033 MeV/ch 2.2 ;2.3 Nitrogen, Sulfur, Phosphorus
Number of counts/ch
250
Semtex1a in iron matrix paper in iron matrix paper behind the iron shield
200 4.7; 4.6 ... Sulfur, Phosporus 150 1.escape
Fig. 7. Triangle diagram.
5.8; 6.0 Sulfur
100
50
0 0
1
2
3
4
5
6
7
8
Energy (MeV)
Fig. 6. The gamma ray spectra from flour (bold black, elapsed time = 41558 s, 4 · 108 tagged neutrons) and tobacco (black, elapsed time = 42230 s, 4 · 108 tagged neutrons) (a) and fertilizer (elapsed time = 41345 s, 4 · 108 tagged neutrons) (b), respectively.
Fig. 7 shows the triangle diagram. Axes are defined as: 2
Ro ¼ 2 N N
C ¼ 2 N N
2 2 þ CC þ OO 2 C C
þ
N N
2 C C
þ
2 O O
O ¼ 2
N N
þ
2 C C
þ
2
ð2Þ
O O
2 O O
where N is the average integral number of tagged neutrons obtained for sand, sugar, tobacco, flour, paper, Semtex1a and fertilizer, C is the sum of the 4.4 MeV main peak and his first escape. O is the sum of the 6.1 MeV peak, his first and the second escape. C and O are average values. The vertical axes in triangle diagram is Ro* and the horizontal axes is 0.57735 (1 + C* O*). Each peak used in
the calculations is found to be over the critical limit defined as [7]: rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi m ð3Þ Lc ¼ 1:645 B 1 þ 2 where B is the background under the peak and m is the number of channels covered by the peak. Most of the peaks used in the calculations are over the detection limit defined as: rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi m LD ¼ 2:71 þ 3:29 B 1 þ ð4Þ 2 4. Conclusions Semtex1a and Semtex1a in iron matrix are well separated in the triangle diagram from the other materials, even from the paper which has similar carbon to oxygen ratio. Semtex1a and Semtex1a in iron matrix are placed on the same positions in triangle diagram, as well as the paper (paper in iron matrix is a little bit off probably because of the scaling). This means that at the least for the presented cases the surrounding matrices do not influence the position of the point in the triangle diagram. Better statistic will improve the results and lower the error bar, which is important in order to identify various materials. Even with the limited statistics, the paper behind the 5 cm iron shield was successfully identified, which can not be done with the commercial 300 keV X-ray system. And finally, it was showed that with the presented set-up and the neutron beam intensity of 108 n/s, 100 kg of Semtex1a behind the 5 cm iron shield can be found in 20 min.
D. Sudac et al. / Nucl. Instr. and Meth. in Phys. Res. B 261 (2007) 321–325
Acknowledgement This work was supported by NATO Science Programme-Science for Peace Project Sfp – 980526, ‘‘Control of Illicit Trafficking in Threat Materials’’. References [1] S. Pesente et al., Nucl. Instr. and Meth. A 531 (2004) 657. [2] S. Blagus, D. Sudac, V. Valkovic, Nucl. Instr. and Meth.B 213 (2004) 434.
[3] [4] [5] [6]
325
D. Sudac, V. Valkovic, in: Proceedings of SPIE, Vol. 6204-02, 2006. B. Perot et al., in: Proceedings of SPIE, Vol. 6213-06, 2006. A.V. Kuznetsov et al., in: Proceedings of SPIE, Vol. 6213-21, 2006. D. Sudac, S. Blagus, V. Valkovic, Proceedings of the Sixth Topical Meeting on Industrial Radiation and Radioisotope Measurement Applications IRRMA6. [7] G. Gilmore, J. Hemingway, Practical Gamma Ray Spectrometry, John Wiley & Sons.
Nuclear Instruments and Methods in Physics Research B 261 (2007) 321–325 www.elsevier.com/locate/nimb
Identification of materials hidden inside a container by using the 14 MeV tagged neutron beam Davorin Sudac
b
a,*
, Silvia Pesente b, Giancarlo Nebbia b, Giuseppe Viesti b, Vladivoj Valkovic a
a Rudjer Boskovic Institute, P.O. Box 180, 10002 Zagreb, Croatia Dipartimento di Fisica dell’Universita di Padova, Vai Marzolo 8, I-35131 Padova, Italy
Available online 5 April 2007
Abstract The results of the experiments aiming to confirm the presence of explosive inside the container by using the 14 MeV tagged neutron beam are presented. Measurements were performed with paper, sugar, flour, fertilizer, tobacco and explosive (Semtex1a) as target material placed in the center of an empty container. Additional measurements were done with paper and explosive placed in the center of the container filled with the iron matrix of 0.2 gcm3 density and with the paper target shielded by the 5.1 cm thick iron shield. The results of time of flight measurements and gamma ray spectra obtained by 14 MeV tagged neutron beam have showed that investigated materials could be well distinguished in the triangle plot with coordinates being the number of counts in the carbon peak, the number of counts in oxygen peak and the number of counts in transmitted neutron peak. By using such presentation we have been able to separate paper from Semtex1a, both hidden inside the 0.2 gcm3 iron matrix. We have also been able to confirm the presence of 64.4 kg of paper behind the 5.1 cm thick iron shield corresponding to the range of 300 keV X-rays. 2007 Elsevier B.V. All rights reserved. PACS: 25.40.Fq; 29.40. M; 82.80.Ej Keywords: Explosive detection; Fast neutrons; Container inspection
1. Introduction Method of inspection of cargo containers by using 14 MeV tagged neutron (TN) beam by detection of associated alpha particles has been recently discussed by several groups [1,2]. The aim of the proposed technique is to determine the chemical composition of the suspicious volume within the inspected container, which was previously identified by the X-ray imaging. Such a two sensors technique has been improved with the results promising it to become the powerful tool for container inspection ([3–5]). In the experiment presented, the 14 MeV tagged neutron tech-
*
Corresponding author. Tel.: +385 1 4561 161/1531; fax: +385 1 4680 239. E-mail address: [email protected] (D. Sudac). 0168-583X/$ - see front matter 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.nimb.2007.03.088
nique has been used to confirm the presence of the explosive Semtex1a within the inspected container. 2. Methods and results Fig. 1 shows the experimental set-up with the investigated target placed in the middle of the container. Target was the iron box dimensions 40 · 40 · 66 cm3 and mass 9.2 kg, filled with the different materials in different measurements: 86.2 kg of graphite, 64.4 kg of paper, 78 kg of sugar, 75 kg of flour, 75 kg of fertilizer, 23.2 kg of tobacco, 100 kg of sand and 100 kg of Semtex1a. Four 300 · 300 NaI(Tl) were put at the transmission position in the cone of the tagged neutrons beam. Each detector was shielded with the 5 cm thick lead. The alpha detector (YAP:Ce) was 8 cm away from the tritium target. A quadratic collimator, 1.8 cm long, was placed in front of YAP:Ce
322
D. Sudac et al. / Nucl. Instr. and Meth. in Phys. Res. B 261 (2007) 321–325 1000
2340
0.1 ns/ch 800
3'' x 3'' NaI (4 peac.)
400
1, 2
400
axis
Target
180
neutron source
1180
3, 4
beam
600 1130
Number of counts/ch
2350
1270
Tagged neutrons
600
400 Gammas from carbon 200
435
floor
0 20
40
60
80
Fig. 1. Experimental set-up inside the container (all units are in mm).
120
140
160
180
300 0.032 MeV/ch 250 1. escape
Number of counts/ch
scintilator. For data processing, a standard NIM electronics was used. The four fast outputs from the 4-segmented PMT connected to the YAP were fed through the octal constant fraction discriminator and fan-in fan-out to the STOP of the time to amplitude converters (TAC). Outputs from the NaI(Tl) detectors were fed through the constant fraction discriminators and ‘‘or’’ logic units to the START of the TACs. Slow signals from the gamma detectors were fed through the amplifiers, quad linear gates and delay amplifiers to the acquisition system. Neutron beam intensity was kept low, 2.5 · 106 n/s in 4p, in order to avoid the problems experienced in previous experiments [6]. Iron matrix was made from iron boxes filled with the iron wire. The axis of the tagged neutron cone was directed through the container approximately parallel with the container bottom (see Fig. 1). Vertical and horizontal profiles of the tagged neutron beam were measured with the 7.62 cm times 7.62 cm NE-213 neutron detector in coincidence with the alpha counter. The experimental diameter of the neutron spot in the middle of the container was (37 ± 2) cm and (35 ± 2) cm for the horizontal and the vertical profile, respectively. As a QC/QA measures the carbon target was used. The 86.2 kg of graphite with the volume of 30 · 30 · 40 cm3 was put in the middle of the container. Fig. 2 shows the corresponding time spectrum (a) and the c-ray spectrum (b) obtained with the appropriate time window and subtracted random background. The main peaks, corresponding to 4.4 MeV excited state as well as the first and the second escape peaks are clearly seen. Fig. 3 shows the time spectra from various materials taken with the same number of alphas. Integral number of the counts under the tagged neutron peaks gives the information on the densities of the various objects (and the surroundings matrices) while the time window set on the gamma peaks give the chemical compositions of the corresponding objects. All measurements with sand, sugar, tobacco, flour, paper, paper behind the 5 cm iron shield and fertilizer were done under the same conditions described in this article. The measurements with the explosive Semtex1a and paper in 0.2 g/cm3 iron matrix were done under the similar conditions, described in [3], and
100
Time (ns)
2. escape
200
4.4 MeV 150
100
50
0 0
1
2
3
4
5
6
7
8
9
10
11
12
Energy (MeV)
Fig. 2. The time spectrum (a) and the gamma ray spectrum (b) of the 86.2 kg of carbon. The gamma ray spectrum was obtained with the appropriate time window on gamma peak. (Target volume = 30 · 40 · 40 cm3, 2 · 108 tagged neutrons, Elapsed time = 17997 s, Average intensity = 0.3 · 107 n/s).
appropriately scaled (scaling factor was found with the help of paper which was measured in both cases). Fig. 4 shows the dependence of integral numbers of TN on density for various materials. Density of the materials was found by dividing the mass with the volume of the iron box. Experimentally obtained data for empty iron box, tobacco, paper, fertilizer, flour, sugar and sand were fitted according to the following formula: N ¼ AeBq
ð1Þ
where A and B are fitting parameters, N is integrated number of the tagged neutrons and q is density. Appropriately scaled Semtex1a nicely follows the exponential fit. Semtex1a in iron matrix, paper in iron matrix and paper behind the iron shield does not follow the exponential fit as was expected. Fig. 5(a) shows the gamma ray spectra obtained for paper and paper behind the iron shield, respectively. Fig. 5(b) shows the gamma ray spectra obtained for sugar and sand, respectively. The gamma ray spectra obtained for flour and tobacco are shown in Fig. 6(a) and gamma
D. Sudac et al. / Nucl. Instr. and Meth. in Phys. Res. B 261 (2007) 321–325 1000
323
300 0.033 MeV/ch
100
250
Carbon
3.3 3.9
Number of counts/ch
Tagged neutrons
Number of counts/ch
Oxygen
paper paper behind the iron shield flour
Gammas from target
Gammas from iron shield
200 4.4 Oxygen 150 5.6 5.1
100
6.1
50
10
0 200
400
600
800
1000
1200
1400
1600
1800
0
1
2
3
5
6
7
8
250
10000
0.033 MeV/ch
0.1 ns/ch
Oxygen
Tagged neutrons
3.8
3.9
Carbon
200
1000
Empty iron box tobacco sugar fertilzer sand
Gammas
100
Numbers of counts/ch
Number of counts/ch
4
Energy (MeV)
Channel
3.3
4.4 Oxygen
150 2.7
5.6 100
5.1
6.1
10
50
1
0 200
400
600
800
1000
1200
1400
1600
1800
0
1
Channel
2
3
4
5
6
7
8
Energy (MeV)
Fig. 3. The time spectra from various materials taken with the same number of alphas.
Fig. 5. The gamma ray spectra from paper (bold black, elapsed time = 39267 s, 4 · 108 tagged neutrons) and paper behind the iron shield (black, elapsed time = 50130 s, 5 · 108 tagged neutrons) (a) and sugar (bold black, elapsed time = 39456 s, 4 · 108 tagged neutrons) and sand (black, elapsed time = 20109 s, 2 · 108 tagged neutrons) (b), respectively.
Integral number of tagged neutrons
1e+6
ray spectrum obtained for fertilizer is shown in Fig. 6(b), respectively. Oxygen and carbon peaks are seen in each spectrum except in the case of fertilizer. In the case of fertilizer the nitrogen, sulphur and phosphorus peaks are visible. In the case of tobacco carbon peaks are visible together with the broad features around 6 MeV which is characteristic for oxygen.
1e+5
1e+4
3. Triangle diagram 1e+3 -0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
Density (g/cm3) Exponential fit empty iron box, tobacco, paper, fertilizer, flour, sugar, sand semtex1a paper behind 5 cm iron shield semtex1a in 0.2 g/cm 3 iron matrix paper in 0.2 g/cm3 iron matrix
Fig. 4. The integral number of the tagged neutrons versus density.
In the c-ray spectra of Semtex1a [3], it can be seen that the nitrogen peak at 5.1 MeV is overlapped with the second escape peak from the 6.1 MeV oxygen line while the peak at the 2.3 MeV is not recognizable at all. On the other side, some home-made explosives do not contain nitrogen, e.g. acetone peroxide (TATP). For that reason we used density as a parameter in triangle diagram instead the nitrogen.
324
D. Sudac et al. / Nucl. Instr. and Meth. in Phys. Res. B 261 (2007) 321–325 250 0.033 MeV/ch
Number of couts/ch
Carbon
3.9
200
4.4
C*
150
O*
Oxygen
5.1
100
5.6 6.1
50
Ro*
0 0
1
2
3
4
5
6
7
8
sand sugar tobacco flour paper fertilizer Semtex1a
Energy (MeV) 300 0.033 MeV/ch 2.2 ;2.3 Nitrogen, Sulfur, Phosphorus
Number of counts/ch
250
Semtex1a in iron matrix paper in iron matrix paper behind the iron shield
200 4.7; 4.6 ... Sulfur, Phosporus 150 1.escape
Fig. 7. Triangle diagram.
5.8; 6.0 Sulfur
100
50
0 0
1
2
3
4
5
6
7
8
Energy (MeV)
Fig. 6. The gamma ray spectra from flour (bold black, elapsed time = 41558 s, 4 · 108 tagged neutrons) and tobacco (black, elapsed time = 42230 s, 4 · 108 tagged neutrons) (a) and fertilizer (elapsed time = 41345 s, 4 · 108 tagged neutrons) (b), respectively.
Fig. 7 shows the triangle diagram. Axes are defined as: 2
Ro ¼ 2 N N
C ¼ 2 N N
2 2 þ CC þ OO 2 C C
þ
N N
2 C C
þ
2 O O
O ¼ 2
N N
þ
2 C C
þ
2
ð2Þ
O O
2 O O
where N is the average integral number of tagged neutrons obtained for sand, sugar, tobacco, flour, paper, Semtex1a and fertilizer, C is the sum of the 4.4 MeV main peak and his first escape. O is the sum of the 6.1 MeV peak, his first and the second escape. C and O are average values. The vertical axes in triangle diagram is Ro* and the horizontal axes is 0.57735 (1 + C* O*). Each peak used in
the calculations is found to be over the critical limit defined as [7]: rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi m ð3Þ Lc ¼ 1:645 B 1 þ 2 where B is the background under the peak and m is the number of channels covered by the peak. Most of the peaks used in the calculations are over the detection limit defined as: rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi m LD ¼ 2:71 þ 3:29 B 1 þ ð4Þ 2 4. Conclusions Semtex1a and Semtex1a in iron matrix are well separated in the triangle diagram from the other materials, even from the paper which has similar carbon to oxygen ratio. Semtex1a and Semtex1a in iron matrix are placed on the same positions in triangle diagram, as well as the paper (paper in iron matrix is a little bit off probably because of the scaling). This means that at the least for the presented cases the surrounding matrices do not influence the position of the point in the triangle diagram. Better statistic will improve the results and lower the error bar, which is important in order to identify various materials. Even with the limited statistics, the paper behind the 5 cm iron shield was successfully identified, which can not be done with the commercial 300 keV X-ray system. And finally, it was showed that with the presented set-up and the neutron beam intensity of 108 n/s, 100 kg of Semtex1a behind the 5 cm iron shield can be found in 20 min.
D. Sudac et al. / Nucl. Instr. and Meth. in Phys. Res. B 261 (2007) 321–325
Acknowledgement This work was supported by NATO Science Programme-Science for Peace Project Sfp – 980526, ‘‘Control of Illicit Trafficking in Threat Materials’’. References [1] S. Pesente et al., Nucl. Instr. and Meth. A 531 (2004) 657. [2] S. Blagus, D. Sudac, V. Valkovic, Nucl. Instr. and Meth.B 213 (2004) 434.
[3] [4] [5] [6]
325
D. Sudac, V. Valkovic, in: Proceedings of SPIE, Vol. 6204-02, 2006. B. Perot et al., in: Proceedings of SPIE, Vol. 6213-06, 2006. A.V. Kuznetsov et al., in: Proceedings of SPIE, Vol. 6213-21, 2006. D. Sudac, S. Blagus, V. Valkovic, Proceedings of the Sixth Topical Meeting on Industrial Radiation and Radioisotope Measurement Applications IRRMA6. [7] G. Gilmore, J. Hemingway, Practical Gamma Ray Spectrometry, John Wiley & Sons.