Detection of Fissioning Materials Using a Neutron Source Based on a ...

3 downloads 0 Views 66KB Size Report
The possibility of using a neutron source based on a 10 MeV compact betatron and 1 kg heavy water or. 3.3 kg beryllium in the betratron bremsstrahlung beam ...
Atomic Energy, Vol. 96, No. 2, 2004

DETECTION OF FISSIONING MATERIALS USING A NEUTRON SOURCE BASED ON A COMPACT BETATRON

V. M. Golovkov,1 V. L. Chakhlov,2 and M. M. Shtein2

UDC 621.039.1

The possibility of using a neutron source based on a 10 MeV compact betatron and 1 kg heavy water or 3.3 kg beryllium in the betratron bremsstrahlung beam is investigated. The 235U detection limit in a 358 mm in diameter and 782 mm high container with probability 0.997 in an experimental apparatus with one epithermal-neutron detector is 40 mg with exposure 10 min. Increasing the mass of the neutron target to 10 kg heavy water and the number of neutron counters to 25 could decrease the 235U detection limit to 3 mg. A neutron detector based on a compact betatron can be expected to give 235U detection sensitivity 10–8 g/g.

Fissioning nuclear materials can be detected by irradiating an experimental object with a pulsed neutron flux from a D–T or D–D generator and detecting fast or epithermal neutrons which arise as a result of a fission reaction on thermal neutrons [1–3]. For certain reasons, a neutron source based on a compact betatron and a target for generating photoneutrons could be preferable. Such a source produces neutrons with lower energy and therefore a smaller amount of moderator is required to moderate the neutrons to thermal energy. In addition, a photoneutron source does not contain radioactive materials and can be easily switched off, its neutron flux has high long-term stability, and if necessary additional radiographic examination of an object can be performed using a bremsstrahlung beam from a betatron. The purpose of the present work is to investigate the possibility of detecting fissioning materials in a test object using a betatron-based photoneutron source. It is well known that the sensitivity of the method is characterized by the limit Ld, which is determined from the formula [4] Ld =

Kα 2 2

 8N b  1 + 1 + ,  α 2  

(1)

where K = ms/(Ns – Nb) is a calibration factor, equal to the ratio of the mass ms of the fissioning material and the difference of the number Ns of detector counts in the case where fissioning material is present in the object and the number of background counts of the detector Nb without fissioning material; α is a parameter which determines the probability of finding fissioning material P(α), for example, P(α = 1) = 0.67, P(α = 2) = 0.95, P(α = 3) = 0.997. Thus, if the quantities Ns and Nb are determined experimentally for different masses of fissioning material, then the detection limit for fissioning materials for a given experimental setup can be found. 1 2

State Unitary Enterprise, Scientific-Research Institute of Nuclear Physics at Tomsk Polytechnical University. State Unitary Enterprise, Scientific-Research Institute of Introscopy at Tomsk Polytechnical University. Translated from Atomnaya Énergiya, Vol. 96, No. 2, pp. 138–142, February, 2004. Original article submitted May 6,

2003. 1063-4258/04/9602-0127 ©2004 Plenum Publishing Corporation

127

D, rad/min 250 1 2 150

50

–20

0

ϕ, deg

20

Fig. 1. Angular distribution of the bremmstrahlung dose rate in the vertical (1) and horizontal (2) planes.

Yield, rad–1·g–1 105 1 2

104

3 103

4

102 2

4

6

8 10 Emax, MeV

Fig. 2. Yield of photoneutrons from lithium deuteride (1), beryllium (2), uranium (3), and lead (4) targets.

Experimental Procedure. A KRAB betatron was used to generate the bremsstrahlung beam [5]. This betatron can accelerate electrons up to 3–10 MeV with the possibility of smooth regulation and can operate continuously at 50 Hz. The instability of the betatron radiation after heating does not exceed 5% over 8 h of continuous operation. The betatron can operate at 100 Hz, being switched off periodically for cooling (1 h operation, 30 min cooling). The betatron consists of three basic parts – a radiation block, a power supply, and a control panel. The dimensions of the radiator block are 969 × 560 × 350 mm, its mass is 275 kg, and its power consumption is 4 kW. The operating time of the betatron before the sealed chamber needs to be replaced is several thousands of hours. The angular distribution of the bremsstrahlung intensity with maximum energy 10 MeV at the output window of the betatron is shown in Fig. 1. The dose rate 1 m from the target of the radiator is 0.16 Gy/min with pulse frequency 100 Hz. The bremsstrahlung beam with maximum energy 10 MeV irradiated a neutron target placed at the maximum of the dose rate 0.58 Gy/min 700 mm from the radiator target. The specific neutron yield for certain neutron targets is shown in Fig. 2 [6]. 128

ϕ, arb. units 0.3

0.2

0.1 1 0

2

2 4

6 E, MeV

Fig. 3. Energy spectrum of neutrons from lithium deuteride (1) and beryllium (2) targets under bremmstrahlung with Emax = 9.83 MeV.

The targets consisted of 1 kg D2O in the form of a 100 mm in diameter and 120 mm high cylinder and a 3340 g beryllium block in the form of a 139 × 139 × 99 mm parallelepiped with a 40 mm in diameter through opening. The neutron yield from the heavy-water and beryllium targets was estimated for the location of the target in these experiments to be (4.2 ± 0.9)·107 and (3.8 ± 0.8)·107 sec–1, respectively. The energy spectrum of neutrons generated in the lithium deuteride and beryllium targets under bremsstrahlung with Emax = 9.8 MeV is shown in Fig. 3; the spectrum was measured by the time-of-flight method using the procedure described in [7]. It is evident that the average energy of the neutrons produced is approximately 1 MeV. The SNM-18 3He-filled neutron counter 1, surrounded by a polyethylene moderator 2 and covered on all sides by a 1 mm thick sheet of cadmium 3, was used as a neutron detector (Fig. 4a). The polyethylene moderator in the form of a cylinder with an axial opening for the counter moderates fast neutrons and increases their detection efficiency. The cadmium layer decreases the sensitivity to thermal neutrons. The measured detection efficiency εd of fast neutrons from a 252 Cf source increases with moderator diameter in the range 58–128 mm (Fig. 5, curve 1). Aside from the high detection efficiency for fast neutrons and insensitivity of the detector to thermal neutrons, it is also necessary to attain a short damping time τd of the neutron flux in the detector as compared with the damping time τm of the neutron flux in the entire moderator surrounding the experimental volume. Depending on the moderator diameter, the damping times τd and τm were found from the temporal distributions of the signals recorded by the neutron counter with various moderators under the action of the neutron flux of the target, which was irradiated with pulsed bremmstrahlung (Fig. 5, curve 3). The quality of the neutron detector for detecting epithermal neutrons arising from the interaction of thermalized neutrons with a target containing a fissioning material can be characterized using the parameter F = εdexp(τd/τm) (Fig. 5, curve 2). For the present experimental conditions, the damping constant τm of the neutron flux in the entire moderator was 1.8·10–4 sec; the optimal diameter of the moderator in the detector is 58–78 mm. A detector with a moderator diameter 78 mm was used in subsequent experiments (see Fig. 4b). The electron beam 5 accelerated in the betatron 4 for 150 nsec was dumped onto the target 6 and generated pulsed bremsstrahlung 7. The neutron target 9 was placed in the moderator 10 consisting of polyethylene granules. The moderator filled the space between two coaxial cylinders, made of 2 mm thick steel (the outer cylinder was 565 mm in diameter and 895 mm high; the dimensions of the inner cylinder were 358 × 782 mm). The distance between the bottom of the first and second cylinders was 110 mm. The top edge of both cylinders was located at the same level and was covered with an 80 mm thick polyethylene plate. A neutron detector 11 was also present in the moderator behind the shield 8, which decreased the ionization overload of the sensitive volume of the counter as a result of the action of the bremsstrahlung pulse. Using a 250 mm thick lead shield made it possible to obtain complete restoration of the counter sensitivity 5.4·10–4 sec after the bremsstrahlung pulse. 129

Fig. 4. Neutron detector (a) and layout of the experimental setup (b).

10–1

Fd, εd, arb. units; τd, sec 3

10–3

2

1 10–5 50

70

90

110

130 d, cm

Fig. 5. The function F = εdexp(τd/τs) versus the diameter of the polyethylene moderator surrounding the SNM-18 counter.

After passing through a preamplifier, amplifier, and amplitude selector 12, the pulses from the neutron counter were fed into a multichannel temporal analyzer 14, which recorded the number of pulses in successive time intervals. The pulse from the photomultiplier 13, caused by the bremsstrahlung pulse, was used as a starting pulse for recording the temporal distribution of the pulses from the counter. The maximum signal counting rate did not exceed 5·105 sec–1. The recorded distribution of pulses was displayed as a histogram on a computer monitor. An ionization chamber 16 placed on the opposite side of the betatron was used to monitor the bremsstrahlung dose during exposure. The ionization current from the chamber was recorded using a clinical dosimeter 17. The samples of fissioning materials consisted of 100 × 100 mm aluminum foils on which a 235U layer with mass 542, 601, and 598 mg was deposited. In the experiments, the fissioning material 15 (1, 2, or 3 foils in the form of a stack) was placed at the center of the measuring chamber surrounded by the moderator. Results. The time distribution of the signals from the neutron source which were obtained in experiments with a 1 kg D2O neutron target and no fissioning material in the measuring cavity (background measurement) and three 235U samples is presented in Fig. 6. During the first 5.4·10–4 sec after the bremsstrahlung pulse, the neutron counter could not record 130

Counts 105

4 3

103

2 1 10

0

1000

2000

t, µsec

Fig. 6. Temporal distribution of neutron-detector counts for different amounts of 235U in the measurement cavity with a D2O neutron target: 1) background without 235U; 2, 3, 4) 235U content 542, 1143, 1741 mg, respectively.

TABLE 1. Experimental Results Measurement No.

Neutron target

235

U mass in the sample, mg

Exposure, min

Number of counts (N ± 2σN) in the interval 540–2300 µsec

K, mg

Ld (α = 3), mg

1

D2O, 1 kg

0, background

10

1434 ± 113

2

D2O, 1 kg

Be, 3.34 kg

10

1667 ± 120

3

D2O, 1 kg

542

10

5780 ± 130

0.125 ± 0.004

20.6 ± 0.7

4

D2O, 1 kg

1143

10

10990 ± 210

0.120 ± 0.003

19.8 ± 0.6

0.140 ± 0.003

23.1 ± 0.7

5

D2O, 1 kg

1741

10

13840 ± 235

6

Be, 3.34 kg

0, background

10

1839 ± 85

7

Be, 3.34 kg

542

10

6615 ± 160

0.113 ± 0.004

21.1 ± 0.7

8

Be, 3.34 kg

1143

10

11172 ± 210

0.122 ± 0.003

22.7 ± 0.6

9

Be, 3.34 kg

1741

10

15498 ± 250

0.127 ± 0.003

23.7 ± 0.6

undistorted signals of single neutrons because of ionization of the gas in the chamber volume. After this period of time, the sensitivity of the detector had the nominal value, and the number of recorded pulses depended on the amount of 235U placed inside the measuring cavity. The time distribution from 540 to 2500 µsec is exponential with a decay constant 180 µsec for the thermal neutron flux in the moderator surrounding the measurement cavity. The number of recorded pulses in the interval 540–2500 µsec for a series of experiments is presented in Table 1. The influence of beryllium on the detector indications was investigated in the second experiment. When 3.34 kg of beryllium was placed in the measurement cavity, the number of pulses increased by only 223 ± 164 over background. This increase corresponds to 31 ± 23 mg 235U. The calibration constant K and the detection limit Ld were estimated from the probability for detecting fissioning material 0.997 (α = 3) in accordance with relation (1) and are also presented in Table 1. The increase in K and Ld with the amount of 235U can be attributed to the screening of the neutron flux as a result of the increase in the 235U concentration in the samples from 5 to 15 mg/cm2. The number of signals from the neutron detector was also measured for the case where the fissioning material is placed at the center of a 300 mm in diameter and 400 mm high cylinder filled with sand. The influence of such an inert material reduces to decreasing the number of counts by approximately a factor of 2 compared with the case without such materi131

al. Thus, the influence of the inert material reduces to increasing the detection limit by approximately a factor of 2. The detection limit for fissioning materials for the present experimental setup, taking account of the influence of inert material and for an exposure of 10 min, was found to be about 40 mg 235U for both neutron targets. Conclusions. It has been shown for the first time that a small betatron can be used to detect fissioning materials. The sensitivity of this method can be increased. For example, the neutron yield from the target can be increased by at least a factor of 10 by increasing the mass of heavy water; the number of neutron detectors can be increased to 25. If graphite with the appropriate dimensions is used as the moderator, then neutron absorption and leakage in the moderator can be substantially decreased. Taking this into account, estimates show that the detection limit can be decreased by a factor of 15 and can reach 3 mg 235U with the indicated exposure. It can be expected that using such a source, based on a compact betatron, a sensitivity of detection of 235U at the level 10–8 g/g can be achieved. The objects for monitoring fissioning materials can now be containers with wastes from enterprises in the nuclear fuel cycle, baggage, and containers transported in automobiles and railroad cars. The advantage of this method of detection is that it can be combined with existing methods of radiographic monitoring of large objects using radiation from accelerated electrons.

REFERENCES 1. 2. 3. 4. 5.

6. 7.

132

T. Gozani, “Active nondestructive assay of nuclear materials: principles and applications,” NUREG/CR-0602, Washington, DC (1981). V. V. Frolov, Nuclear Physical Methods for Monitoring Fissioning Materials, Énergoatomizdat, Moscow (1988). W. Kunz, “A fissile waste or scrap assay system with 1 mg sensitivity,” Trans. ANS, 39, 341–346 (1981). V. T. Tustanovskii, Estimation of the Sensitivity and Accuracy of Activation Analysis, Atomizdat, Moscow (1976), pp. 160–164. V. L. Chakhlov, V. G. Volkov, V. S. Pushin, et al., “Application of small-size betatrons in NDT,” in: Proceedings of the 11th International Conference on Radiation Physics and Chemistry of Condensed Matter (Tomsk, September 24–29, 2000), TPU, Tomsk (2000), pp. 94–95. A. Yu. Basai, G. I. Burov, V. G. Volkov, et al., “Combined source of bremsstrahlung and neutrons,” Prib. Tekh. Éksp., No. 1, 43–45 (1998). V. L. Chakhlov, Z. W. Bell, V. M. Golovkov, and M. M. Shtein, “Photoneutron source based on a compact 10 MeV betatron,” Nucl. Instrum. Methods, A422, 5–9 (1999).