Thermal Conductivity Change in High Burnup MOX Fuel Pellet

Transcription

1 Journal of Nuclear Science and Technology ISSN: (Print) (Online) Journal homepage: Thermal Conductivity Change in High Burnup MOX Fuel Pellet Jinichi NAKAMURA, Masaki AMAYA, Fumihisa NAGASE & Toyoshi FUKETA To cite this article: Jinichi NAKAMURA, Masaki AMAYA, Fumihisa NAGASE & Toyoshi FUKETA (9) Thermal Conductivity Change in High Burnup MOX Fuel Pellet, Journal of Nuclear Science and Technology, 46:9, To link to this article: Published online: 16 Mar 212. Submit your article to this journal Article views: 357 Citing articles: 4 View citing articles Full Terms & Conditions of access and use can be found at

2 Journal of NUCLEAR SCIENCE and TECHNOLOGY, Vol. 46, No. 9, p (9) ARTICLE Thermal Conductivity Change in High Burnup MOX Fuel Pellet Jinichi NAKAMURA, Masaki AMAYA, Fumihisa NAGASE and Toyoshi FUKETA Fuel Safety Research Group, Nuclear Safety Research Center, Japan Atomic Energy Agency, Tokai-mura, Naka-gun, Ibaraki , Japan (Received December 5, 8 and accepted in revised form May 28, 9) High burnup MOX and UO 2 test rods were prepared from the fuel rods irradiated in commercial BWRs. Each test rod was equipped with a fuel center thermocouple and reirradiated in the Halden boiling water reactor (HBWR) in Norway. The burnups of MOX and UO 2 test rods reached about 84 GWd/tHM and 72 GWd/t, respectively. Fuel temperature was measured continuously during the re-irradiation tests. Thermal conductivity change in high burnup fuel was evaluated from the results of comparison between the measured fuel temperature and the data calculated by using the fuel analysis code FEMAXI-6. The comparison results suggested that the thermal conductivity of MOX fuel pellets is comparable to that of UO 2 fuel pellets in the high burnup region around 8 GWd/t. It is probable that the impurity effect of Pu atoms gradually diminishes with increasing burnup because other factors that affect pellet thermal conductivity, such as the accumulation effect of soluble fission products and irradiation-induced defects in crystal lattice, become dominant in a high burnup region. KEYWORDS: thermal conductivity, nuclear fuel, high burnup, MOX, UO 2, BWR, FEMAXI-6, fuel temperature, fission products, irradiation-induced defects I. Introduction Mixed oxide fuel (MOX) is being used worldwide in light water reactors (LWRs) from the viewpoint of fuel cycle cost reduction and efficient use of resources. It is considered that MOX fuel use in LWRs has the following advantages: as for the fissile contents, there is an enrichment limitation at 5% for UO 2 fuel, while no limitation for MOX fuel as a Pu fissile content, and the power reduction in MOX fuel in a high burnup region is quite less than that in UO 2 fuel. In order to expand the burnup range of MOX fuel in LWRs, it is important to investigate the behavior of MOX fuel at high burnup. However, few data concerning the MOX fuel behavior at high burnup have been obtained. Fuel temperature during irradiation is one of the most important factors that control fuel behaviors, such as fission gas release and pellet-cladding mechanical interaction (PCMI). Since fuel temperature is affected by pellet thermal conductivity, it is necessary to evaluate the effects of burnup and Pu content on pellet thermal conductivity precisely. Although the effects of burnup and Pu addition on the thermal conductivity of UO 2 pellet have been reported, 1,2) the thermal conductivity data are quite limited for the high burnup MOX fuel pellet irradiated up to high burnup region in LWRs. 3,4) Corresponding author, nakamura.jinichi@jaea.go.jp ÓAtomic Energy Society of Japan The purpose of this study is to investigate the thermal conductivity change in high burnup MOX fuel pellet. The thermal conductivity change was evaluated by comparing the temperature changes measured continuously during irradiation tests with the values calculated by using a fuel analysis code. II. Reirradiation Tests of Fuel Rods 1. Test Fuel Rods The test fuels consist of two high burnup UO 2 fuel rods (1 1 type) irradiated in a boiling water reactor (BWR) at Leibstadt, Switzerland and four high burnup MOX fuel rods (MIMAS, 9 9 type) irradiated in a BWR at Gundremmingen, Germany. The test rods for the reirradiation tests were refabricated from the full-length rods. Both ends of the test rod were cut, and the pellets near these ends were removed in order to weld the upper and lower end plugs. A central hole was drilled in fuel pellets in a test rod, and each test rod was instrumented with a fuel centerline thermocouple and also equipped with an inner pressure gauge or a cladding elongation detector. The main specifications of the test fuel rods for reirradiation tests are summarized in Table 1. The fuel stack length is 3 mm, and instrumented end plugs were welded at both ends. The two UO 2 fuel rods were equipped with fuel centerline thermocouples: one was fitted with an inner pressure gauge and the other with a cladding elongation detector. On 944

3 Thermal Conductivity Change in High Burnup MOX Fuel Pellet 945 Table 1 Main specification of fuel rods for reirradiation test Rig No. IFA-687 IFA-688 Rod No Fuel type UO 2 MOX Enrichment (wt%) 4.46 Fissile Pu content (wt%) 5.5 Stack length (mm) 3 3 Fuel weight (g) 166 Burnup (GWd/t) Filler gas composition and Ar(64%)-He(36%), Ar(95%)-He(5%), pressure at 1.44 MPa.5 MPa refabrication Cladding outer diameter (mm) Upper instrumentation Lower instrumentation Fuel thermocouple (TF) Cladding elongation (EC) Rod inner pressure (PF) Cladding elongation (EC) Fuel thermocouple (TF) Rod inner pressure (PF) the other hand, each MOX fuel rod was equipped with a fuel center thermocouple. In addition, one MOX fuel rod was fitted with a cladding elongation detector and three MOX fuel rods with rod inner pressure gauges. Post-irradiation examinations (PIEs) were carried out after the reirradiation tests. The PIE results showed that the gap between the cladding and the pellet was closed and both materials were fully bonded along the circumference direction for all UO 2 and MOX fuel rods. On the basis of the short reirradiation period, it is considered that the full bonding between the cladding and the pellet occurred at the start point of reirradiation tests. 2. Outline of the Irradiation Rigs The reirradiation tests of the high burnup UO 2 and MOX fuels were conducted by using two irradiation rigs and pressure flasks depicted in Figs. 1 and 2. Two UO 2 fuel rods were installed into the irradiation rig called Instrumented Fuel Assembly-687 (IFA-687), and the irradiation rig was installed into the corresponding pressure flask called Fuelled Flask Assembly-28 (FFA-28). Moreover, four MOX fuel rods were installed into the irradiation rig called IFA-688, and the corresponding pressure flask FFA-29 was used. In order to simulate the neutron flux spectrum in a typical BWR, twelve booster fuel rods with fuel pellets of 5% enrichment can be installed between the outside of the thermal insulation tube surrounding the pressure flask and the inside of the thin shroud wall, which is the outer boundary of the irradiation channel. The active fuel stack of the booster fuel rod is mm. The pressure flasks and booster fuel Table 2 Coolant condition for the irradiation test at the Halden reactor Coolant chemistry Coolant temperature ( C) 28 Coolant pressure (MPa) 7.2 Dissolved oxygen concentration (ppm).2 Dissolved hydrogen concentration (ppm).5 Electric conductivity (ms/cm) <:3 rods were designed to satisfy the demands for mechanical strength, thermal hydraulic condition of a coolant (28 C, 7.2 MPa), and rod linear heat rate (12 22 kw/m). The irradiation rigs that included the test fuel rods were irradiated in the Halden boiling water reactor (HBWR) in Norway under the coolant condition shown in Table 2. Coolant temperatures and pressures during the reirradiation tests were monitored continuously and confirmed that the coolant condition during the irradiation test simulated a typical BWR operation condition well. Water chemistry conditions in both FFAs were nearly the same during the reirradiation test. 3. Irradiation Conditions during Base Irradiation The power histories of the UO 2 fuel rods in a commercial BWR at Leibstadt are shown in Fig. 3. The power histories of the MOX fuel rods in a commercial BWR at Gundremmingen are shown in Fig. 4. The UO 2 rods were irradiated at a power level of 2 23 kw/m for the first three VOL. 46, NO. 9, SEPTEMBER 9

4 946 J. NAKAMURA et al. Fig. 1 Layout of the test rig for UO 2 test rods Fig. 2 Layout of the test rig for MOX test rods Linear heat rate (kw/m) Linear heat rate (kw/m) Time (day) Time (day) (a) Rod 1 (b) Rod 2 Fig. 3 Irradiation histories of UO 2 test rods in a BWR at Leibstadt cycles and then at low power levels of less than 1 kw/m for the four cycles. On the other hand, MOX rods were irradiated for 6 cycles with a similar power history in each cycle. The rods were operated at relatively high power levels at startup and then power was decreased during the last part of the cycle. JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY

5 Thermal Conductivity Change in High Burnup MOX Fuel Pellet ROD 1 3 ROD 2 Linear power (W/cm) Linear power (W/cm) Time (days) Time (days) (a) Rod 1 (b) Rod 2 3 ROD 3 3 ROD 4 Linear power (W/cm) Linear power (W/cm) Time (days) Time (days) (c) Rod 3 (d) Rod 4 Fig. 4 Irradiation histories of MOX test rods in a BWR at Gundremmingen III. Results The reirradiation tests were conducted for two irradiation cycles in the HBWR. At the beginning of an irradiation test, rig power calibration was carried out for each irradiation rig in order to determine the relationship between the thermal power of the rig and the outputs from the neutron detectors in the rig. The HBWR was operated under steady state condition at about 18 MW during the irradiation cycles. The maximum burnups of UO 2 and MOX fuel rods reached about 72 GWd/t and 84 GWd/tHM, respectively, at the end of the second irradiation cycle. 1. Rod Average Linear Heat Rate and Burnup Histories during the Irradiation Tests The histories of the average linear heat rate and burnup of UO 2 rods during the reirradiation test are shown in Fig. 5. The rod average linear heat rate was kept at kw/m during the steady state operation in the reirradiation test. The rod average burnups reached 71.5 and 7.9 GWd/t for rods 1 and 2, respectively, at the end of the reirradiation test. The histories of the average linear heat rate and burnup of MOX rods during the reirradiation cycles are shown in Fig. 6. The rod linear heat rate was kept at kw/m during the steady state operation in the reirradiation test. The rod average burnups reached 8.3, 81.9, 81.6, and 83.7 GWd/tHM for rods 1, 2, 3, and 4, respectively, at the end of the reirradiation test. 2. Measured Temperature Histories of Fuel Rods (1) UO 2 Test Rods The measured fuel temperatures peaked at the startup of the first irradiation cycle in connection with the highest rod power and gradually decreased with irradiation time. The measured fuel temperatures were 68 7 C at the startup of the first irradiation cycle and 58 6 C during the second irradiation cycle. The temperature of rod 2 was 2 3 C higher than that of rod 1 due to the rod power difference. In the second half of the second cycle, the temperature difference between the rods decreased, while the temperature of rod 1 hardly decreased. This suggests that the temperature of rod 2 decreased during this period. Since the rod inner pressure gauge showed an anomalous increase at the beginning of this period, it is possible that the fuel failure of rod 2 occurred at that time. It is also possible that the heat transfer condition was affected by the hydrogen generated from the steam that entered the fuel rod after fuel rod failure, and the temperature decrease of rod 2 may be due to the heat transfer increase between pellet and cladding. (2) MOX Test Rods The measured fuel temperatures peaked at the startup of the first cycle due to the highest rod power in the whole irradiation period and gradually decreased with time. The measured fuel temperatures were 8 8 C at the startup of the first irradiation cycle and 7 8 C during the second irradiation cycle. The fuel temperature of rod 1 was the highest among those of the rods in IFA-688, and the temperatures of rods 2, 3, and 4 were similar. By considering the VOL. 46, NO. 9, SEPTEMBER 9

6 948 J. NAKAMURA et al. (a) first cycle (b) second cycle Fig. 5 Histories of average linear heat rates and burnups of the UO2 test rods and assembly (a) first cycle (b) second cycle Fig. 6 Histories of average linear heat rates and burnups of the MOX rods and assembly full bonding between the cladding and the fuel pellet that was observed in the PIE results of MOX fuel rods, it was estimated that the effects of fission gas release on the fuel center temperature change are small because the thermal conductance between the cladding and the pellet is controlled not by the gap gas but by the bonding layer, that is, a solid. It is likely that this rod temperature difference was mainly due to the difference in rod power. JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY

7 Thermal Conductivity Change in High Burnup MOX Fuel Pellet (a) start of cycle (b) end of cycle (c) start of cycle (d) end of cycle 2 (TF:Measured fuel temperature, LHRTF: LHR at the position of thermocouple tip) Fig. 7 Linear heat rate dependence of the measured fuel temperature of UO 2 test rods at the position of thermocouple tip 3. Rod Linear Heat Rate Dependence of the Measured Fuel Temperature (1) UO 2 Test Rods Rod linear heat rate (LHR) dependences of the measured fuel temperatures at the startup and end of the first and second cycles are depicted in Fig. 7. In this figure, the horizontal axis shows the local linear heat rate at the thermocouple tip position. The measured fuel temperature increased monotonically with increasing fuel rod power. The difference in the linear heat rate dependence of the measured fuel temperature is not very clear between rods 1 and 2. The measured fuel temperature at 1 kw/m was about 5 C through the whole irradiation period. (2) MOX Test Rods The rod linear heat rate dependence of fuel temperature at the start and end of the first and second cycles is depicted in Fig. 8. In this figure, the horizontal axis shows the local linear heat rate at the thermocouple tip position. The fuel temperature increased monotonically with increasing rod power. The difference in LHR dependence was not clearly observed between rods 2, 3, and 4, and the fuel temperatures of these fuel rods were about 58 C at 1 kw/m through the whole irradiation period. The reason why the linear heat rate dependence of the fuel temperature of rod 1 is different from those of the other rods is not clear at present. It has been reported that pellet thermal conductivity decreases with burnup 1) and Pu content. 2) In this reirradiation test, since the MOX fuel, which has a higher burnup and a higher Pu content than the UO 2 fuel, has a lower thermal conductivity than the UO 2 fuel, the fuel center temperature of the MOX fuel tends to be higher than that of the UO 2 fuel. The present experimental results agree well with this tendency. The effect of Pu addition on the thermal conductivity of high burnup UO 2 fuel pellet will be discussed in detail below. 4. Analysis by Using the Fuel Analysis Code FEMAXI-6 The obtained irradiation data were analyzed by using the fuel analysis code FEMAXI-6. 5) The number of data points of the power history in the commercial reactors was reduced VOL. 46, NO. 9, SEPTEMBER 9

8 9 J. NAKAMURA et al TF4 TF TF4 TF (a) start of cycle 1 (b) end of cycle TF4 TF TF4 TF (c) start of cycle 2 (d) end of cycle 2 (TF: Measured fuel temperature, LHRTF: LHR at thermocouple tip position) Fig. 8 Linear heat rate dependence of the measured fuel temperature of MOX test rods at the position of thermocouple tip to 3 6. Since the data were recorded every 15 min in the HBWR, the number of data points is quite large. Therefore, the number of recorded data points was reduced to about 3. These data were used as the input data for FEMAXI-6 calculation. The radial power distribution affects the fuel temperature calculation directly. In high burnup fuels, the Pu contents at the pellet periphery and in Pu agglomerates in MOX affect the radial power distribution. The radial power distribution was evaluated by using the nuclear calculation code PLUTON. 6) At PIE, the gap between cladding and pellets in all UO 2 and MOX rods was revealed to be closed. Therefore, in the calculation, the gap was postulated to be closed and the thermal conductivity of solid ZrO 2 was used for thermal resistance calculation at the gap location. The temperature increment at the bonding position was calculated to be about 7 C at 2 kw/m. IV. Discussion 1. Rod Linear Heat Rate Dependence of Fuel Temperature (1) UO 2 Fuel Figure 9 shows the fuel temperature calculated by using the linear heat rate at the thermocouple tip position of rod 2. Since the linear heat rate dependence of fuel temperature in rod 1 was similar to that in rod 2, only the dependence of rod 2 is shown by a solid line in the figure for simplicity. Here, the thermal conductivity model for UO 2 pellets, which was developed by the Halden Project, 7) was used. In addition, by considering PIE results, it is assumed that the gap conductance between the cladding and the pellet can be expressed only by using the thermal conductance through the bonding layer formed in the gap region, indicating that the effect of gas conductance on the gap can be negligible. JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY

9 Thermal Conductivity Change in High Burnup MOX Fuel Pellet 951 Fuel temperature ( C) Calculated temperature by FEMAXI-6 code with Halden thermal conductivity model for UO 2 Fuel temperature ( C) Halden thermal conductivity model for UO 2 Halden thermal conductivity model for MOX Baron (EdF) thermal conductivity model for MOX Linear heat rate at thermocouple tip position (kw/m) Linear heat rate at thermocouple tip position (kw/m) Fig. 9 Comparison between measured fuel temperatures of the UO 2 rods and values calculated by using the code FEMAXI-6 Fig. 1 Comparison between measured fuel temperatures of the MOX rods and values calculated by using the code FEMAXI-6 Figure 9 also compares the linear heat rate dependences of the measured and calculated fuel temperatures at the thermocouple tip position during the first irradiation cycle. The linear heat rate at the thermocouple tip position was estimated by interpolating the signals from two neutron detectors located at different axial positions. As shown in the figure, the fuel temperatures of rods 1 and 2 agree well within about 1 C with the calculated values. Consequently, it is considered that the calculation conditions including the pellet thermal conductivity model of the Halden Project and the gap thermal conductance model, in which the effect of the bonding layer was considered, were adequate. (2) MOX Fuel The fuel temperatures of rod 3 at the thermocouple tip position were calculated by using the Halden MOX, 8) Baron MOX, 9) and Halden UO 2 7) models as pellet thermal conductivities. Here, on the basis of the bonding condition between the cladding and the pellet in the MOX fuel rods used in this study, the gap thermal conductance model, in which the effect of the bonding layer was considered, was also applied to the MOX fuel rods. The measured fuel center temperatures of all MOX rods were compared with the values calculated by using the above-mentioned three models, as shown in Fig. 1. While only the measured temperatures of rod 1 show a different trend, those of the other three rods agree well with each other within about 1 K. As seen in Fig. 8, the relationship between the measured fuel temperature and rod linear heat rate was almost the same during the first and second irradiation cycles. The reason why rod 1 showed a lower temperature is not clear at present. It is possible that the actual rod power deviated from the evaluated rod power due to the tilting of neutron flux in the rig, which cannot be evaluated by the arrangement of neutron detectors in the rig. It is also possible that the thermocouple tip was located at the pellet-pellet interface. 2. Effect of Plutonium on the Thermal Conductivity of High Burnup Fuel Pellets As seen in Fig. 1, the fuel temperatures calculated by using the Halden MOX and Baron models are 7 and 13 C higher than the measured values at 15 kw/m, respectively. Here, the thermal conductivity of the MOX fuel pellet used in the Halden MOX model is 8% lower than that of the UO 2 fuel pellet. On the other hand, the temperatures calculated by using the Halden UO 2 model agree well with the measured values. This suggests that the pellet thermal conductivity in these high burnup MOX fuel rods is nearly the same as that in high burnup UO 2 fuel rods. In other words, although it has been reported that the thermal conductivity of unirradiated MOX pellet is lower than that of UO 2 pellet due to the impurity effect of Pu atoms, 2) it is possible that the impurity effect of Pu atoms gradually diminishes with increasing burnup because other factors, such as the effect of the accumulation of soluble fission products and irradiation defects in crystal lattice, become dominant over the pellet thermal conductivity change in a high burnup region. V. Conclusion High burnup MOX and UO 2 fuels equipped with fuel centerline thermocouples were reirradiated in the Halden boiling water reactor, and fuel temperatures were measured continuously during the reirradiation test. The thermal conductivity change in high burnup fuels was also evaluated by using the fuel analysis code FEMAXI-6. Based on the different thermal conductivity models, the rod linear heat rate dependence of fuel temperature was calculated by using FEMAXI-6, and the dependence was compared with the measured values. From the obtained results, the thermal conductivity of the MOX fuel was determined to be comparable to that of the UO 2 fuel in the high burnup region around 8 GWd/t. This trend implies that the effect of Pu VOL. 46, NO. 9, SEPTEMBER 9

10 952 J. NAKAMURA et al. addition on the thermal conductivity of UO 2 fuel pellet is negligible in the high burnup region around 8 GWd/t. It is possible that the impurity effect of Pu atoms gradually diminishes with increasing burnup because other factors affecting the pellet thermal conductivity, such as the accumulation effect of soluble fission products and irradiationinduced defects in crystal lattice, become dominant in a high burnup region. Acknowledgements The present study was conducted as a part of the program sponsored and organized by the Nuclear and Industrial Safety Agency, Ministry of Economy, Trade and Industry. The authors would like to acknowledge and express their appreciation for the efforts devoted by members of the OECD Halden Reactor Project and also for the efforts devoted by Messrs M. Suzuki and Y. Udagawa of the Japan Atomic Energy Agency for FEMAXI-6 calculation. References 1) M. Amaya, M. Hirai, H. Sakurai et al., Thermal conductivities of irradiated UO 2 and (U, Gd)O 2 pellets, J. Nucl. Mater., 3, (2). 2) J. R. Topliss, I. D. Palmer, S. Abeta et al., Measurement and analysis of MOX physical properties, Technical Committee Mtg. on Recycling of Plutonium and Uranium in Water Reactor Fuel, Windermere, UK, Jul (1995). 3) H. Fujii, H. Teshima, K. Kanasugi et al., Final assessment of MOX fuel performance experiment with Japanese PWR specification fuel in the HBWR, Proc. 7 Int. LWR Fuel Performance Mtg., San Francisco, California, Sep. 3 Oct. 3, 7, (7). 4) D. Boulanger, M. Lippens, L. Mertens et al., High burnup PWR and BWR MOX fuel performance: A review of BERGONUCLEAIRE recent experimental programs, Proc. 4 Int. LWR Fuel Performance Mtg., Orlando, Florida, Sep , 4, (4). 5) M. Suzuki, Light Water Reactor Fuel Analysis Code FEMAXI-6 (Ver. 1) Detailed Structure and User Manual, JAEA-Data/ Code 5-3, Japan Atomic Energy Agency (5). 6) S. E. Lemehov, M. Suzuki, PLUTON-Three Group Neutronic Code for Burnup Analysis of Isotope Generation and Depletion in Highly Irradiated LWR Fuel Rod, JAERI-Data/Code 1-25, Japan Atomic Energy Agency (1). 7) W. Wiesenack, M. Vankeerbergehn, R. Thankaappan, Assessment of UO 2 Conductivity Degradation Based on In-Pile Temperature Data, HWR-469, Halden Reactor Project (1996). 8) A. Gates, K. Takano, R. J. White, Thermal Performance of MOX Fuel, HWR-589, Halden Reactor Project (1999). 9) D. Baron, About the modeling of fuel thermal conductivity degradation at high burnup accounting for recovering process with temperature, Proc. Seminar on Thermal Performance of High Burnup LWR Fuel, Cadarache, France, Mar. 3 6, 1998, (1998). JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY