Shielding Design to Obtain Compact

Transcription

1 Journal of NUCLEAR SCIENCE and TECHNOLOGY, 31[6], pp. 510,-520 (June 1994). Shielding Design to Obtain Compact Marine Reactor Akio YAMAJI and Kiyoshi SAKO Tokai Research Establishment, Japan Atomic Energy Research Institute* (Received July 7, 1993), (Revised November 9, 1993) The marine reactors equipped in previously constructed nuclear ships are in need of the secondary shield which is installed outside the containment vessel. Most of the weight and volume of the reactor plants are occupied by this secondary shield. An advanced marine reactor called MRX (Marine Reactor X) has been designed to obtain a more compact and lightweight marine reactor with enhanced safety. The MRX is a new type of marine reactor which is an integral PWR (The steam generator is installed in the pressure vessel.) with adopting a waterfilled containment vessel and a new shielding design method of no installation of the secondary shield. As a result, MRX is considerably lighter in weight and more compact in size as compared with the reactors equipped in previously constructed nuclear ships. For instance, the plant weight and volume of the containment vessel of MRX are about 50% and 70% of those of the Nuclear Ship MUTSU, in spite of the power of MRX is 2.8 times as large as the MUTSU's reactor. The shielding design calculation was made using the ANISN, DOT3,5, QAD-CGGP2 and ORIGEN codes. The computational accuracy was confirmed by experimental analyses. KEYWORDS: ship propulsion reactors, shields, shielding, design, computer calculations, computer codes, weight, volume, nuclear ships I. INTRODUCTION A marine reactor should be compact and lightweight since it has to be installed in narrow and limited space in a ship, and also for economical viewpoint of a ship. The reactor type of previously constructed nuclear ships SAVANNAH and MUTSU is a conventional pressurized water reactor (PWR), and that of the nuclear ship OTTO HAHN an integral PWR**. These reactors are in need of the primary shield installed around the pressure vessel and the secondary shield outside the containment vessel. On the PWR of nuclear ships SAVANNAH and MUTSU, the secondary shield is necessary during normal reactor operation and in the time of reactor shutdown, since the primary coolant loop is installed outside the primary shield, and also necessary in the event of an accident with a consequential release of radioactive fission products from the core into the containment vessel. On the integral PWR of nuclear ship OTTO HAHN, the main purpose of the secondary shield is to prevent the crew and others from radiation during the time of an accident. The secondary shield of nuclear ships SAVANNAH, OTTO HAHN and MUTSU occupies the major part of the whole shieldo)-(8). Most of the weight of the marine reactors is due to the secondary shield. For instance, in the case of the Nuclear Ship MUTSU (N. S. MUTSU), the weight of the secondary shield is 88% of the whole shield, and the total shield weight of the N. S. MUTSU exceeds 70% of that of the reactor plant. To obtain a more compact and lightweight * Tokai-mura, Ibaraki-ken ** The steam generator of integral PWR is in - stalled in the pressure vessel. There is no large primary coolant pipe outside the pressure vessel

2 Vol. 31, No. 6 (June 1994) 511 marine reactor with enhanced safety, a design study is being performed on an advanced marine reactor called MRX (Marine Reactor X). A view of MRX is shown in Fig. 1. The following features of MRX design can be identified as specific(4)-(6) : Integral Fig. 1 Conceptual view of MRX PWR -El iminates the possibility of a large LOCA, simplifies the engineered safety system and makes a reactor plant compact. In-pressure vessel type control rod drive mechanisms -Eli minate the possibility of "control rod ejection" accident and simplify the engineered safety system. Water-filled containment vessel assively maintains core -P flooding, and is of great advantage to make a compact reactor plant. Passive decay heat removal system implifies the engineered safety -S system. The water in the containment vessel is useful for, (1) attenuation of radiation transmitted through the pressure vessel and/or generated in the containment vessel, (2) prevention of excessive radiation streaming through shield penetration such as an air gap around the pressure vessel, and (3) shielding in the case of an accident with a consequential release of radioactive fission products from the core into the containment vessel. Due to the adoption of the integral PWR and water-filled containment vessel, it becomes possible to keep the exposure dose to the crew or others on board or in the vicinity of the ship within a permissible level without any bulk shield outside the containment vessel during normal reactor operation, in the time of reactor shutdown and in the event of a hypothetical accident, which means the realization of a reactor which is not only small in size but also lightweight. A new concept, to design without secondary shield, was introduced into the shielding design of MRX. Namely, MRX has been structured so as to satisfy the design criteria without any bulk shield outside the containment vessel. As a consequence, the size of the containment vessel should also be designed so as to satisfy the shielding requirement. Figure 2 shows an example of a two-mrx Fig. 2 Two-MRX equipped icebreaker for use as a scientific observation ship 15

3 512 J. Nucl. Sci. Technol., equipped icebreaker for scientific observations. An icebreaker is used as it is the most conceivable nuclear ship to be equipped with advanced marine reactors in the near future. So for the time being, MRX is designed for testing on an icebreaker. This paper is organized as follows : Chap. II shows the design criteria with the setting conditions. In Chap. III the reactor and shield are described. Chap. IV shows the shielding analyses during normal reactor operation, in the time of reactor shutdown and in the event of a hypothetical accident. The computational accuracy is also described in Chap. IV. The conclusion is presented in Chap. V. II. DESIGN CRITERIA The specifications for the radiation levels on board are as follows : (a) The reactor room outside the containment vessel is classified as the controlled area. The maximum permissible design dose rate equivalent is specified as 10 msv/h for 48 working hours per week. Working in long hours is possible in this room with a satisfactory low dose equivalent. (b) In the double bottom under the containment vessel, which is classified as the controlled area, the maximum permissible dose rate equivalent is specified as 50 msv/h for 10 working hours per week. (c) In the engine room, which is classified as the surveillance area, the maximum permissible design dose rate equivalent is specified as 6 msv/h for 50 working hours per week. The criteria (a), (b) and (c) are set based on the Japanese regulations. (d) In the accommodation space, which is classified as the uncontrolled area, the dose rate equivalent has to be lower than ms/h*, resulting in 50 msv/yr for an unlimited stay in this area ; namely, staying for 365 x 24 h/yr. This criterion is based on the guidelines for the safety examination of the Nuclear Safety Commission of Japan. (e) Activation of the bottom shell due to neutrons from the core is restricted to a value under 3.7x 10-2 Bq/g(7) which is equal to the maximum radioactivity of steel used in nonnuclear facilities. (f) Long work hours have to be considered in the containment vessel at a reasonable time after the reactor has been shut down. (g) Repair work has to be possible for the ship's bottom with an unlimited stay after the reactor has been shut down. (h) It has to protect the reactor control room and wheel house from radiation during the time of an accident with a consequential release of radioactive fission products from the core into the containment vessel, and limited operation should be possible in the control room and the wheel house during the accident. (i) On the basis of an expected 20- year plant life, a maximum fast-neutron irradiation of 1019 nvt (E>=1 MeV) is used as the fast-neutron exposure limit of the pressure vessel. III DESCRIPTION OF REACTOR AND SHIELD The reactor generates 100 MWt with UO2 fuel enriched 4.3 wt% on the average. The active fuel length and equivalent lateral core diameter are 140 and 149 cm, respectively. The outer diameter of the pressure vessel is 400 cm. Table 1 shows the major specifications of MRX. The shielding system is designed to satisfy the design criteria during normal operation, in the time of reactor shutdown and in the event of a hypothetical accident, with consideration given to minimum weight, size and cost. For use of an integral PWR, it must be considered that the 16N and 17N-activation of secondary water due to the reactions 16O(n, p)16n and 17O(n, p)17n will occur in the steam generator, because of the very small distance existing between the core and steam generator as shown in Fig. 1, and may raise the dose rate equivalent around the secondary loop in the engine and reactor rooms. This activation must be kept sufficiently low to maintain the dose rate equivalents within the design criteria of these rooms. To satisfy this condition, an iron shield is installed between the core and steam generator, since * The value of msv/h is 1/50 of that of the N.S. MUTSU. 16

4 Vol. 31, No. 6 (June 1994) 513 Table 1 Major specifications of MRX only neutrons with an energy above 10 MeV can produce the above mentioned reactions, and iron has an excellent shielding ability against these high energy neutrons. This iron shield also has a role, (1) to decrease the g-ray flux of the diagonally upward direction, and to keep the dose rate equivalents at points D and E (shown in Fig. 1) on the outer-surface of the containment vessel within a permissible level and to maintain a low exposure dose during the work at point G (shown in Fig. 1) in the containment vessel for a reasonable time after the reactor has been shut down, and (2) to reduce the generation of radioactive corrosion products in the secondary water in the steam generator and to maintain the dose rate equivalent around the secondary loop in the engine and reactor rooms within a permissible level. The containment vessel is filled with water in which the cast steel shield is arranged. The outer diameter and height of the containment vessel are 7 and 13.3 m, respectively. The pressure vessel is immersed in water in the containment vessel. Figure 3 shows a schematic section in the horizontal plane at the core's center position. In the time of reactor shutdown, the water level in the containment vessel can be lowered for maintenance work. The thickness and arrangement of the cast steel shield, and the size of the containment vessel were determined based on shielding calculations by varying the arrangements systematically. The containment vessel can stand under high pressure at LOCA which was confirmed by analysis(6). Maintenance/repair work is capable under this structure(6). Fig. 3 Schematic section in horizontal plane of MRX's core center position (A case of a rectangular fuel assembly. Another case shows a hexagonal fuel assembly which is presented in Ref. (6).) There are an air gap and thermal insulator between the pressure vessel and water in the containment vessel. As described later, radiation streaming through this air gap does not reach any significant level, because of the water filled containment vessel and relatively large radius of the pressure vessel, and there are no other shield irregularities through which radiation streaming rises significantly. Around the pressure vessel, a steel shield 5 cm in thickness is placed to allow for space for the air gap and the thermal insulator, and to lower the exposure dose during repair work in the containment vessel after the reactor has been shut down. 17

5 514 J. Nucl. Sci. Technol., The position of the cast steel installed in water of the containment vessel was determined so as to effectively reduce secondary r-rays, to minimize shield weight and keep sufficient working space in the containment vessel. The water filled containment vessel serves also to reduce the exposure dose to the crew and others on board or in the vicinity of the ship in the event of an accident with a consequential release of radioactive fission products from the core into the containment vessel. The problems for the downwards are, (a) scattered radiation from the bottom shell and seawater getting into the rooms outside the containment vessel, (b) generation of induced activity in the structures outside the containment vessel, and (c) exposure during repair work on the ship's bottom after the reactor has been shut down. The space below the reactor, at the bottom of the ship, does not need to be completely shielded, since usually work in this space is not necessary during reactor operation. From this reason, it is possible to design for the downward direction with a relatively small distance between the core and containment vessel. MRX does not provide any bulk shield outside the containment vessel. As a result, MRX is lighter in weight and more compact in size as compared with the reactors equipped in previously constructed nuclear ships. group constant made by ANISN calculations with this library. The energy group structures used in ANISN and DOT calculations are shown in Table 2. Table 2 Energy group structures used in ANISN and DOT calculations IV. SHIELDING ANALYSIS The shielding design calculations were performed for normal reactor operation, reactor shutdown and a hypothetical accident, using the discrete ordinates codes ANISN(8) and DOT3.5(9), the point kernel code QAD- CGGP2(10) and the fission product generation code ORIGEN(11). The calculations were made with one-dimensional spherical and cylindrical geometries for ANISN, the twodimensional RZ geometry for DOT, and threedimensional combinatorial geometry for QAD. A data library, DLC-23E, was used for the calculations of ANISN. The DOT calculations were performed using the collapsed 18

6 Vol. 31, No. 6 (June 1994) 515 These discrete ordinates codes were used with P3 Legendre polynomial expansion and S16(ANISN calculations) and S48 (DOT calculations) angular quadrature sets. Evaluation of the computational accuracy is provided in this chapter. 1. Shielding Analysis during Normal Operation The conceptual shield structure and size of containment vessel were determined based on the calculations using the ANISN code with one-dimensional spherical geometries. Many ANISN calculations were made concerning the shielding arrangements of the radial, downward, and diagonally upward and downward directions because of importance of these directions as regards shielding design. Figure 1 shows the determined structure with the ANISN computational directions. The ANISN geometrical arrangements are shown in Table 3 along the radial and downward directions for the determined structure with the numbers of computational mesh points in each region, and Fig. 4 shows the calculated results along these directions. The ratios between dose rate equivalents of primary and secondary g-rays are 1/50 and 1/300 at the outer-surface of the pressure vessel and containment vessel in the radial direction. The calculated dose rate equi- Table 3 Shield sequence in radial and downward directions used in AN1SN calculation 19 --

7 516 J. Nucl. Sci. Technol., valents are presented in Table 4 at points A, B, C, D and E outside the containment vessel shown in Fig. 1. Fig. 4 ANISN calculation along radial and downward directions of MRX Table 4 ANISN calculated dose rate equivalent on outer-surface of containment vessel and in double bottom (msv/h), and ratio between DOT and ANISN calculated total dose rate equivalents The ANISN calculation in the downward direction shows that the maximum value of thermal neutron flux at the bottom shell is 2 x10-1 cm2/s during full power, from which the activation of the bottom shell is estimated to be 3 x 10-4 Bq/g two weeks after the reactor shutdown after continuous 20 year full power operation. This value is sufficiently small compared with the design criteria of 3.7 x 10-2 Bq/g described in Chap. II. The ANISN calculations in the radial, and diagonally upward and downward directions also show that the activation values in structures outside the containment vessel are under 1/100 of 3.7x10-2 Bq/g two weeks after the reactor shutdown after continuous 20 year full power operation. The DOT calculation was performed with fission neutrons and secondary g-rays to confirm the overall radiation level with this final structure. Figure 5 shows the computational geometry and the contours of the dose rate equivalent due to fission neutrons and secondary g-rays. This figure shows that no significant radiation streaming occurs in the containment vessel and the radiation levels are adequate at all positions around the containment vessel. Table 4 also shows the ratios between the DOT and ANISN values outside the containment vessel. The ANISN and DOT calculated values of the irradiation of fast neutrons are below 8 X 1015 nvt (E>= 1.11 MeV) at the innersurface of the pressure vessel for full power reactor operation for 20 continuous years. This value is sufficiently low compared with the design 20

8 Vol. 31, No. 6 (June 1994) 517 Fig. 5 Contour of dose rate equivalent due to fission neutrons and secondary -rays obtained by DOT RZ calculation (Sv/h at full power) g criteria of 1019 nvt (.E>=1 MeV) for radiation damage of the pressure vessel. This small value is due to the relatively large radius of the pressure vessel. For the calculations in diagonally upward direction, it is important to give consideration to fast neutron flux above 10 MeV, since these neutrons contribute to the generations of 16N g-rays and 17N neutrons in the secondary water in the steam generator. For this reason, another DOT calculation was made for the diagonally upward direction with a neutron energy structure of 14 groups above 10 MeV to obtain P(E) with a fine energy structure above 10 MeV in the steam generator. The calculations showed that a fast neutron flux above 10 MeV is below 106 n/cm2-5 at the undersurface of the steam generator. Using fast neutron fluxes for each energy group in the steam generator and the cross sections of 16N(n, p)16o and 17O(n, p)17n, the source densities due to 16N and 17N-decays were calculated in the high pressure steam, where the densities were assumed to be constant up to the turbine because of the high velocity of the steam. For the cross section of the 16O(n, p)16n reaction, the experimental data obtained by Martin("' was used, and for (n, p)17n reaction, the cross section data 17O in the Reactor Shielding Design Manual.(13). These activities will result in a dose rate equivalent of 2 x 10-5 msv/h at a distance of 30 cm from the steam pipeline, and even a small value at the turbine which are based on the QAD-CGGP2 calculations with the 16N -ray source density and ANISN one-dimen- g sional cylindrical calculations with the 17N neutron source density. At the hotwell, the dose rate equivalent will increase to 2 x 10-4 Sv/h at a distance of 30 cm due to mthe higher source density. The dose rate equivalent due to the corrosion products in the secondary loop is estimated to be sufficiently small in comparison with the design criteria of 10 msv/h at the surface of all components of the secondary system, where the estimation was made based on the calculated neutron flux in the steam generator and dose measurements in actual plants(14). 2. Shielding Analysis in Time of Reactor Shutdown To satisfy the design criteria, repair or maintenance work must be possible outside 21

9 518 J. Nucl. Sci. Technol., the pressure vessel, namely, in the containment vessel, in the double bottom, under the bottom shell, etc., 24 h after the reactor has been shut down. The fission products in the core were calculated using the ORIGEN code at 24 h after continuous reactor operation at full power to the maximum degree of burnup (25,000 MWD/t), and dose rate equivalents were calculated using the ANISN code with same geometrical condition in the calculations of normal reactor operation. The values are sufficiently low, namely (1) 5 x 10-6 msv/h and 1 x 10-6 msv/h at points F and G (shown in Fig. 1) in the containment vessel, (2) 3 x 10-9 Sv/h at point A on the outer-surface of m the containment vessel in the radial direction, (3) 7 x 10-8 msv/h at point C in the double bottom and (4) 3 x 10-9 msv/h at the under-surface of the bottom shell. The corrosion products in the primary coolant and deposited on the equipment of the primary system, and activation of the reactor components near the core are not evaluated precisely at the present designing stage, it seems however from the N. S. MUTSU shielding modification design calculation(14) that the dose rate equivalent due to these source terms does not become large enough to interfere with the work in containment vessel, since there are thick iron between these source positions and the working area in the containment vessel. From the above-mentioned radiation level due to the fission products in the core, and the discussion on the other source terms, it can be said that the maintenance and/or repair work in the containment vessel which are performed with a low level of water in the containment vessel are possible at a low exposure dose, and it is also possible to classify the area of seawater under the bottom shell into the uncontrolled area during the reactor shutdown. 3. Shielding Analysis for Accident For shielding design, it is necessary to evaluate the exposure dose equivalent of the crew and others due to direct g-rays and skyshine r-rays during a hypothetical accident with fission products being released into the containment vessel. The computational conditions are as follows : (1) The accident is assumed to occur immediately after continuous reactor operation at full power to the maximum degree of burn-up (25,000 MWD/t). (2) The ratios between the amount of fission products released into the containment vessel and those existing in the core before the release are assumed to be 100% for noble gas, 5% for iodine, 1% for halogen except for iodine and 0.01% for the other nuclei in the gaseous region in the containment vessel, on the other hand 50% for halogen and 1% for the other nuclei in the liquid region, these values were chosen based on the guidelines for the safety examination of the Nuclear Safety Commission of Japan. The isotope generation and depletion were calculated using the ORIGEN code. Thereafter, the dose rate equivalents on board and in the vicinity were calculated with the QAD- CGGP2 code using the above-mentioned source conditions in the containment vessel. The calculation shows the following which satisfies the Japanese regulations : (1) Installed a shield for the control room if necessary ; the exposure dose of crew can be restricted below 100 msv which is the limit of dose equivalent for an accident in which the crew are subjected to exposure. (2) For instance, if an unhabitation area is set within a distance of 200 m from the reactor, the dose equivalent at the boundary becomes 0.01 Sv which is sufficiently small compared with the dose limit of 0.25 Sv for public during the time of an accident. 4. Computational Accuracy Special attention was paid to evaluating of the accuracy of the calculation. The accuracy of calculations from the core to outside of the containment vessel was estimated by analyzing the primary shield tank experiment of the Nuclear Ship OTTO HAHN(15) by means of the ANISN code with the same conditions used in the design calculations, for order of Sn and Pi, mesh width, energy group structure and geometrical modeling method. A comparison between the experiment and calculation was made for the reac- 22

10 Vol. 31, No. 6 (June 1994) 519 tion rates of fast neutron detectors, epithermal and thermal neutron flux, and 7-ray dose rate equivalent. Figure 6 shows an example of the comparison between the measured and calculated values of 115In(n, n')115min reaction rate. Based on the experimental analysis, three times as large as the ANISN calculated dose rate equivalents were used as evaluated values at points A and C outer-surface of the containment vessel of MRX. The shielding design of MRX was made with considering this accuracy. Namely, the ANISN calculated dose rate equivalents on the outer-surface of the containment vessel are under 1/3 of the design criteria as shown in Table 4. V. CONCLUSION MRX has been designed to obtain a compact and lightweight marine reactor with enhanced safety. This was accomplished by adopting an integral PWR and a water-filled containment vessel. These adoptions enabled the authors to design without any bulk shield outside the containment vessel. A new concept, to design without secondary shield, was introduced into the shielding design of MRX. Namely, MRX has been structured so as to satisfy the design criteria without any bulk shield outside the containment vessel. As a result, MRX is considerably lighter in weight and more compact in size as compared with the reactors equipped in previously constructed nuclear ships SAVANNAH, OTTO HAHN and MUTSU. For instance, the plant weight and volume of the containment vessel of MRX are about 50% and 70% of those of the N.S. MUTSU, in spite of the power of MRX is 2.8 times as large as the MUTSU's reactor and the design criteria of dose rate equivalent in the accommodation space of MRX equipped nuclear ship is small as 1/50 of that of the N.S. MUTSU. The computational accuracy of the shielding design calculations was confirmed by experimental analyses. ACKNOWLEDGMENT Fig. 6 Comparison of 115In(n, n')115min reaction rate between ANISN calculation and experiment in shield tank of N.S. OTTO HAHN The analyses using the QAD code were already made by the authors for the following experiments, (1) shield mock-up tests for the first nuclear ship of Japan(16), (2) T- ray experiment using the LIDO nitrogen 16 facility at Harwell(17)(18) and (3) g-ray experiment for shield effect of ship structure(19). These analyses show an overestimation of the QAD values(20). In the MRX design, QAD calculated values were used without any corrections. The authors would like to thank Mr. H. Kawasaki and Mr. K. Ohuchi of CRC Research Inst. Inc. for assisting with the calculations. --REFERENCES- (1) SMITH, W. R., TURNER, M. A.: BA W , (1959). (2) JAEGER, R. G., et al. (ed.) : "Engineering Compendium on Reactor Shielding", Vol. III, (1970), Springer-Verlag. (3) YAMAJI, A., et el.: Proc. 6th Int. Conf. Radiation Shielding, Tokyo, 617 (1983). (4) SAKO, K., et al.: SMiRT 11 Trans., Vol. SD2, 357 (1991). (5) ISHIZAKA, Y., et al.: ibid., 363. (6) SAKO, K., et al.: Proc. Int. Conf. on Design and Safety of Advanced Nuclear Power Plants. Tokyo, (1992). 23

11 520 J. Nucl. Sci Technol., (7) Radioactivity : Recommendations of the International Commission on Radiological Units and Measurements, ICRU Rep. 10c, NBS Handbook 86, (1963). (8) ENGLE, Jr., W. W.: K-1693, (1967). (9) RHODES, W.A. : ORNL-TM-4280, (1973) ; NEA- CPL (1977). (10) SAKAMOTO, Y., TANAKA, S.: JAERI-M 90110, (1990). (11) BELL, M. J.: ORNL-4628, (1973), and KOYAMA, K., et al.: JAERI-M 6954, (1977). 2) MARTIN, B.C.: Phys. Rev., (1 93(3), 498 (1954). (13) ROCKWELL III, T. (Ed.) : "Reactor Shielding Design Manual", (1956), McGraw-Hill Book. (14) YAMAJI, A., et al.: To be published in JAERI- M. (15) FIEBIG, R., et al.: Atomkernenergie, 18(1), 71 (1971). (16) Japan Nucl. Ship Dev. Agency : JNS-4-1~6 (in Japanese), (1967). (17) BISHOP, G. B., SMITTON, C., PACKWOOD, A. : Ann. Nucl. Energy, 3, 65 (1976). ) SMITTON, C., BISHOP, G.B. (18 : J. Br. Nucl. Energy Soc., 14(1), 89 (1975). (19) YAMAKOSHI, H., UEKI, K., NAKATA, M.: J. Nucl. Sci. Technol., 20(2) 127 (1983). (20) YAMAJI, A., et al.: J. At. Energy Soc. Jpn., (in Japanese), 26(2), 139 (1984). 24