Energy and Angular Responses of the Criticality Accident Alarm System Using Current-mode-operated Scintillation Detector

Transcription

1 Energy and Angular Responses of the Criticality Accident Alarm System Using Current-mode-operated Scintillation Detector N. Tsujimura 1, T. Yoshida 1 and S. Mikami 1 1 Japan Nuclear Cycle Development Institute, 4-33, Tokai-mura, Iraraki-ken, , Japan tujimura@tokai.jnc.go.jp Abstract. The Japan Nuclear Cycle Development Institute (JNC), Tokai Works, has operated the spent fuel reprocessing plant and the MOX (PuO 2 -UO 2 Mixed Oxide) fuel fabrication plants. In these facilities, the criticality accident alarm systems (CAAS) have been installed. The Toshiba RD120 CAAS consists of the current-mode-operated plastic scintillator and the 2 out of 3 voting system. The alarm triggering point was set to -2.0 mgy/h in photon dose to detect the minimum accident of concern. However, current-mode-operated plastic scintillator is principally sensitive not only to primary photons but also to secondary photons produced by neutron capture reactions in the detector itself and concrete wall behind the CAAS. The authors calculated energy and angular responses of the CAAS to primary and secondary photons. The response to primary photons was evaluated from MCNP-4B and EGS4 calculations and pure gamma and X-rays irradiation experiments. The response to secondary photons was computed with MCNP-4B. From the results of calculations and experiments, it was found that the response of the CAAS by neutron-photon interactions was 20% or more of the response against pure gamma rays. This result indicates that in neutron and photon mixed fields the criticality alarm signal would be actuated at the lower dose rate level than the alarm triggering point that was preset to detect pure primary photons. 1. Introduction Most facilities handling fissile materials have criticality accident alarm system (CAAS) to detect a criticality accident. The purpose of a criticality accident alarm system is, or should be, to reduce the risk associated with fissile material operations. In order to fulfil this purpose, the CAAS must be carefully designed to promptly and accurately respond to the accidents while minimizing false alarms. The International Standards [1,2] and ANSI Standard [3] prescribe the performance requirements for the CAAS, and the items to be considered in design of the CAAS are in terms of reliability, tolerance criterion, detection criterion, response time, etc. Compared with radiation protection instruments, the requirement for energy and angular response functions is not so important item. However, determining appropriate coverage area requires understanding the response of the detectors to the criticality accident radiation field. The evaluated response function of the CAAS makes it possible to optimize the number and placement of CAAS, and would serve the cost reduction of CAAS instrumentation. The Japan Nuclear Cycle Development Institute (JNC), Tokai Works, has operated the spent fuel reprocessing plant and the MOX (PuO 2 -UO 2 Mixed Oxide) fuel fabrication plants. In these facilities, the CAAS have been installed. In the early stage of the operation of plants, we introduced the CAAS manufactured by the SEIN Corporation of France. With an increase of the plants to be covered by the CAAS, in 1980 s, we initiated a domestic development and production of the CAAS under cooperation with Toshiba Corporation [4]. 105 detector assemblies have been presently installed in the spent fuel reprocessing plant, the Pu fuel fabrication facilities, the Pu conversion facility and the chemical process research facility. The CAAS, the Toshiba RD120, consists of the current-mode-operated plastic scintillator and the 2 out of 3 voting system. This detector was originally designed to detect gamma-rays and the alarm triggering point has been set to -2.0 mgy/h in photon dose to detect the minimum accident of concern. However, for a current-mode, a plastic scintillator is inevitably sensitive not only to primary photons but also to secondary photons produced by neutron capture reactions in the detector assembly itself and in concrete wall behind the CAAS. Hence, in neutron and photon mixed fields produced in criticality accidents, the actual triggering point of a criticality alarm has been fairly ambiguous. This paper discusses how the actual response function to photon and neutron for the CAAS was determined. First the CAAS detector system is described. Then, a model of the detector system is 1

2 developed to facilitate calculation of response function. Finally the results of the calculation are presented. 2. The Criticality Accident Alarm System The detector assembly, the Toshiba RD120, consists of a gamma-ray detector and related electronics circuits and power units. These are assembled in a steel box. A detector is a cylindrical plastic scintillator, the NE102 of 3.8cm diameter and 5.1cm long, coupled to the Hamamatsu R878 photomultiplier tube which is operated in a constant current mode. The scintillator and the photomultiplier tube are encased in a cylindrical Mu metal shell which acts as magnetic shield. The detector gives the alarm signal when its anode current exceeds a preset value due to gamma-rays. The alarm signal of the detector assembly is fed into the Toshiba RU726 two out of three voting unit installed in the process control room. This unit actuates the alarm sounds and warning lights when at least two detectors out of three ones give alarm signals within a coincidence time of 500 msec. FIG. 1 shows the external view of the RD120 CAAS detector assemblies. FIG. 1. External view of the Toshiba RD120 detector assemblies installed in the reprocessing plant. 3. Experiments and Calculations 3.1 Experiments The response of the CAAS to photons was investigated in the photon calibration fields produced with the 137 Cs gamma source and the X-ray generator. Photons with energies from 35 kev to 181 kev and 662 kev were irradiated in free-air condition. An anode signal of the photomultiplier tube was measured by the high precision digital multi-meter. 3.2 Calculations Calculation model The calculation of responses of CAAS was performed by Monte Carlo N-particle transport code, MCNP TM -4B [5],and the EGS4 [6]. In this study the MCNP code was mainly used for photon and neutron-photon transport calculations with the neutron cross section library of ENDF/B-VI and the photon transport library of MCPLIB02. The EGS4 was used for comparison and verification of the results computed by the energy deposition calculation tally (F6 tally) of MCNP. For modelling the detector assembly, complex three-dimensional models of physical objects were created. Material composition of plastic scintillator is a hydrogen-oxygen mixture in a ratio of : 1 with a density of 32 g.cm -3. The photomultiplier was approximated to a cylinder made of a mixture of borosilicate glass, koval TM, etc., with a uniform density of 5 g.cm -3 based on the elemental composition [7]. The steel box, detector housing, detector support steel, major electric 2

3 circuit board and wall-mounting-rack are included in the calculation model. The steel box was modelled with height, width, and depth of 17.5, 3 and 2cm, respectively. The thickness of front steel is 4mm. FIG.2 illustrates the calculation geometry model used in MCNP. To obtain a response, the energy deposited in the plastic scintillator was tallied with F6 tally. For simplicity of comparison with experimental results the calculated results in unit of MeV.g -1 were converted to 137 Cs gamma-ray equivalent output by putting a normalization factor obtained from 137 Cs pure gamma-ray irradiation experiments. NE102 PMT Bakelite Al housing Steel FIG.2. Schematic top view of the RD120 detector for MCNP4B calculation In-air response As for in-air responses, mono-energetic photons from 60 kev to 10 MeV and neutrons with energies from thermal Maxellian to 10 MeV were generated in aligned direction perpendicular to the surface of the detector assembly. For an evaluation of angular response, the geometry model was rotated by the coordinate transformation card in azimuthal and polar directions by 15 degree steps On-concrete-wall response Considering the actual situation of criticality accidents, the CAAS detector would measure not only primary radiations emitted from criticality assemblies but also secondary radiations reflected from the concrete wall on which the CAAS was installed. For calculation of on-concrete-wall responses, the detector assembly was set on the concrete wall having a 320cm diameter and a 30cm thickness. The determination of dimensions of the concrete wall as a backscattering material was based on the author s earlier work [8]. Elemental composition of concrete was taken from those of type 2A concrete with a density of 2.1g.cm -3. In the MCNP calculations the CAAS detector is mounted at the center of concrete wall, and mono-energetic photons from 60 kev to 10 MeV and neutrons with energies from thermal Maxellian to 10 MeV were uniformly irradiated on the wall. In addition, the SILENE reactor neutron spectra [9] were used as a source spectrum for estimation of response of the CAAS in realistic criticality accident radiation fields. 4. Results and Discussion 4.1 Photon response FIG. 3 shows the in-air response function of the CAAS detector per unit air kerma with normal photon incidence. In this figure, the response for 137 Cs gamma-rays is set to 1 by using a calibration factor obtained from the confirmatory measurements with 137 Cs gamma-irradiation experiments. The CAAS detector has an energy independent response over a several hundred kev but it sharply falls down 3

4 below 100 kev. This is because of an attenuation of photons due to the 4mm thickness steel of the casing box. The on-concrete-wall responses of the CAAS were about 10-20% higher than those in-air because of backscattered photons from the wall. Since the gamma-ray field in criticality accident is high energetic gamma-ray field mainly produced by prompt gamma-rays from fission and capture gamma radiation, the CAAS detector can respond those radiations regardless of the types of criticality system and surrounding materials. The azimuthal and polar angular dependencies of the CAAS for 60 Co gamma-rays were shown in FIG.4. Variation of responses with incident angle is small with an exception of an incidence from photo-multiplier tube position which acts as a photon shielding. Response per Air Kerma Experiment in-air MCNP4B in air MCNP4B on wall EGS4 in-air Photon energy [kev] FIG.3. Response of the RD120 detector assembly per air kerma in-air and on-concrete-wall conditions. Open and closed circles represent calculated response by MCNP-4B in-air and onconcrete-wall conditions, respectively. Squares are calculated by EGS4 in-air. Triangles are experimental results. 60 Co gamma-rays, in air 60 Co gamma-rays, on concrete wall Relative Response PMT Azimuthal angle θ (degrees) Polar angle θ (degrees) FIG.4. Calculated azimuthal and polar angular responses of RD120 detector assemblies to 60 Co gamma-rays. 4.2 Neutron response FIG.5 shows the in-air and on-concrete-wall response functions of the CAAS detector per unit neutron tissue kerma with normal neutron incidence. In this figure the response is expressed in unit of gamma- 4

5 equivalent response that gives the same photomultiplier output with gamma irradiation. The shape of the response function is primarily affected by two reactions; neutron capture reactions of H(n,γ)D in the plastic scintillator and 56 Fe(n,γ) in the steel box in lower neutron energy region and in-elastic scattering of Fe(n,n γ) in higher neutron energy region. Those reactions are enhanced by a concrete wall. The CAAS detector has a strongly energy-dependent response function and is also extremely dependent on the presence of the backscattering wall. The gamma-equivalent responses per neutron tissue kerma to thermal and epithermal neutrons were about 10 for in-air condition and about 50 for on-wall condition. The tendency of response function is very similar to that of albedo dosemeters used in personal monitoring. Variation of responses with incident angle is shown in FIG.6. These results greatly differ from angular responses to photon. The in-air response to thermal neutron becomes the maximum in the direction of 45 because of the variation of an area of steel exposed to thermal neutron. On the other hand, the neutron angular dependence of responses on concrete wall generally follows a bell-shape. The relative responses at 30 and 60 are about 0.9 and, respectively Gamma Equivalent Response per Tissue Kerma in air on concrete wall 137 Cs gamma response = Neutron energy [MeV] FIG.5. Response of the RD120 detector assembly per neutron tissue kerma in-air and on-concretewall conditions. Open and closed circles represent calculated response by MCNP-4B in-air and onconcrete-wall conditions, respectively. Data are expressed in unit of gamma-equivalent response that gives the same photomultiplier tube output with gamma-ray irradiations. Relative Response thermal neutrons, in air SILENE (Bare) spectrum on wall Azimuthal angle (degrees) PMT Polar angle (degrees) FIG.6. Calculated azimuthal and polar angular responses of RD120 detector assemblies to neutrons. 5

6 4.3 Response in neutron and photon mixed field The responses of the CAAS detector installed on the concrete wall in neutron and photon mixed field are summarized in Table I. When the response of the CAAS per air kerma in-air condition is set to 1, the responses per neutron tissue kerma were ranged from 1 to 5 in free-air condition and from 0.15 to 0.30 in on-concrete-wall condition. This result indicates that with the alarm signal level set to 1mGy/h in photon dose and assuming a neutron to photon dose ratio Dn/Dγ of 1, the CAAS detector which is faced to the criticality assembly gives the equivalent signal of total neutron-photon dose of 1.15 to 1.3 mgy/h in a realistic neutron and photon mixture field. Table I. Response of the CAAS per photon and neutron dose in-air and on-concrete-wall conditions. The response per air kerma to 137 Cs gamma-rays is set to 1. In-air On concrete wall Source Quality Response per air kerma Response per tissue kerma Response per tissue kerma 137 Cs Gamma Cf Neutron SILENE Bare Neutron SILENE Lead Neutron SILENE (CH 2 ) n Neutron SILENE Steel Neutron Conclusion The energy and angular responses of the Toshiba RD120 CAAS detector to photons and neutrons was obtained from experiments and calculations. As a result, the CAAS currently used in JNC has a very high response to thermal and epithermal neutrons than that to photons, and the CAAS detector would respond to about 20% of neutron dose by neutron-photon interactions in neutron fields. This result will indicate that in criticality accident radiation fields the criticality alarm signal would be actuated at the lower dose rate level than the alarm triggering point that was preset to intend to detect pure primary photons. In near future the authors will initiate the design of a new CAAS to respond to the same threshold triggering dose regardless of whether it is exposed to neutrons, gamma-rays and a mixture of the two radiations by using a neutron-photon converter and a hydrogenous moderator. References [1] ISO. Nuclear energy Performance and Testing Requirements for Criticality Detection and Alarm Systems. ISO International Standards 7753 (1987). [2] IEC. Warning Equipment for Criticality Accidents. IEC International Standards 860 (1987). [3] American National Standard for Criticality Accident Alarm Systems. ANSI/ANS , American Nuclear Society (1996). [4] Noda,K. Development of Criticality Accident Alarm System, PNC Technical Review, 81, pp (1992) [in Japanese]. [5] Briesmeister, J.F.(ed.) MCNP - A General Monte Carlo N-Particle Transport Code, Version 4B. LA M (Los Alamos National Laboratory, Los Alamos, NM) (1997). [6] Nelson,W.R., Hirayama,H. and Rogers,D.W.O. The EGS4 Code System, SLAC-Report-265 (Stanford Linear Accelerator Center) (1985). [7] Hamamatsu Photonics K.K.; private communication (2003). [8] Tsujimura,N., Yoshida,T. and Ishizuka,A. Energy and Angular Responses of the Two Types of Criticality Accident Alarm System - Plastic Scintillator and Fission Detector with Moderator. Proc.7 th NUCEF seminar, (2004). [in Japanese]. [9] IAEA. Compendium of Neutron Spectra and Detector Responses for Radiation Protection Purposes. Technical Report Series No.403, International Atomic Energy Agency (2001). 6