2017 Water Reactor Fuel Performance Meeting September 10 (Sun) ~ 14 (Thu), 2017 Ramada Plaza Jeju Jeju Island, Korea

Transcription

1 FUEL ROD WITH E110 CLADDING OVERPRESSURE/LIFT-OFF EXPERIMENT AT HALDEN. IN-PILE DATA AND MODELING WITH START-3A CODE D. A. Chulkin 1, P.G. Demianov 1, V. I. Kuznetsov 1, V. V. Novikov 1, Yu. V. Pimenov 2, B. Yu. Volkov 3, J. E. Hansen 3, Ø. Brennvall 3 1 JSC VNIINM (Russia) 2 JSC TVEL (Russia) 3 IFE (Norway) ABSTRACT: The bilateral LIFT-OFF experiment was carried out in the Halden reactor (Norway) with the fuel rod which was also pre-irradiated to a burnup of 58 MWd/kgU in the PWR loop in Halden. The main objective of the experiment was to study the behaviour of the E-110 fuel rod cladding under the irradiation when internal pressure in the fuel rod was increased with stepwise levels exceeding coolant pressure. The main goal of this overpressure experiment was to find the pressure threshold at which the cladding departures from the fuel and as a consequence the fuel temperature rate are substantially increased (LIFT-OFF effect) due to the fuel-cladding gap re-opening. The cladding of the fuel rod was produced in TVEL from E110 alloy (from sponge zirconium) with diameter of 9.5 mm. At the end of the base irradiation, the average oxide film thickness was measured by Eddy current about 25 microns and the HSB effective fuel-cladding diametral gap was measured around 30 microns before the LIFT-OFF test. One of the main distinctive features of this experiment unlike to the other standard LIFT-OFF tests performed in the Halden reactor, was the measurements of the cladding diameter during irradiation, which was provided due to a new design test rig instrumented with movable cladding diameter gauge. The linear heat rate of the fuel rod was in the range kw/m, the internal pressure in the fuel rod was stepwise increased from 150 bar to 300 bar (see Fig. 1) by means of an ultra-high pressure system using argon gas for the rod pressurisation during irradiation. The fuel performance code START-3A [2] was verified against the data recorded during this LIFT-OFF experiment in the Halden reactor using the base irradiation data derived from other Halden test where the test rod was preirradiated. The calculation and experimental results have been compared and it was found the reasonable agreement. Analysis of the experimental data and calculations using the START-3A code indicate the absence of the LIFT-OFF effect during the experiment when the final overpressure of 14 MPa is reached. The calculated prognosis indicates the start of LIFT-OFF at the overpressure of 16 MPa. KEYWORDS: in-pile test, LIFT-OFF effect, fuel rod overpressure threshold, Halden reactor. I. INTRODUCTION The Institute of Energy Technology (Norway) on the request of Halden Reactor Project has carried out series of experiments to investigate a high burnup fuel response on the rod pressure development which may be exceeded the coolant system pressure at high burnup during irradiation [1]. This phenomenon, so called LIFT-OFF effect is related to the issue of the excessive fission gases release from highly irradiated fuel at reduced rod free volume due to both fuel swelling and cladding creep down during the pre-irradiation period. According to the Russian safety regulation the gas pressure in VVER fuel rods must not exceeded the system pressure whereas for PWRs, the regulation requires that the fuel-clad gap must not be re-opened due to cladding reversal creep-out under overpressure conditions with consequence of the fuel temperature increase and possible feedback effect on fission gas release. The main goal of the experiments is to establish the overpressure threshold to onset of increasing fuel temperature. Standard in-pile measurements for these series tests IFA-610 such as cladding elongation and fuel centerline temperature with periodically measured hydraulic diameter which is associated with fuel-cladding gap were supplemented in the present bilateral test with in-pile cladding diameter gauge measurements in order to complete a picture of the fuel performance under the rod overpressure conditions. The paper gives an overview of the results obtained from the experiment for fuel segment produced from E110 alloy and PWR fuel design, and pre-irradiated to a burnup of about 58 MWd/kg U in 1

2 the Halden reactor. The data were used for verification of the E110 cladding behavior and validation of the full scale fuel performance code START-3A under rod overpressure for PWR fuel design. II. EXPERIMENTAL A. Fuel rod characterisation The fuel rod nominal design parameters before basic irradiation in the PWR test loop of the Halden Reactor are given in Table 1. TABLE I. Design data on the rod for LIFT-OFF test - nominal outer diameter (mm) 9.50 CLADDING E110 (sponge) - measured inner diameter (mm) nominal wall thickness (mm) enrichment (wt%) at BOL 14 FUEL UO 2 Burn-up, 58 MWd/kgU - density (% of T.D./ g/cm 3 ) 96 / initial pellet outer diameter (mm) 8.15 initial fuel cladding diametric gap (mm) ~ dishing / chamfer Yes / No initial fill gas pressure of test rod, bar 15 This rod was irradiated more than 900 operational days up to a burnup of about 58 MW. day/kgu at power rating of about 35 kw/m. The irradiation history including fast flux and power rating are shown in Fig. 1. The water chemistry in the loop were performed with enhanced level of Lithium which is more aggressive than for the standard PWR conditions, however, the corrosion level of the E110 cladding was detected by Eddy current measurements below 25 micron at the end of irradiation. Fig. 1. Fast flux and Average Linear Heat Rating pre-irradiation history of the rod used for LIFT-OFF test. 2

3 B. Advanced test rig for LIFT-OFF test The Instrumented Fuel Assembly (IFA-788) for TVEL Lift-off test is designed for Fuel Flask Assembly (FFA) to be connected to the loop in the Halden Boiling Water Reactor (HBWR) and contains one replaceable fuel rod with the following rig instrumentation: Downcomer, inlet and outlet coolant thermocouples, self-powered neutron detectors, LVDT for cladding elongation measurements (EC), diameter Gauge (DG), Linear Resolver to detect DG position during measurements, The schematic view of the rig is shown in Fig. 2. The rig was surrounded by booster fuel rods in order to amplify the neutron fast flux. Fig. 2. Schematic view of the LIFT-OFF test rig with diameter gauge. The test rod was instrumented by EC detector in the bottom and fuel centreline thermocouple (TF) at the top. The centre hole for TF was drilled in the solid pellets of the pre-irradiated rod. The rod was produced with gas pipe lines in the both ends, which used for the test rod pressurisation and depressurisation during the experiment. The fuel thermocouple is connected to cables using specially designed in-core connector. The pipes from the fuel rod are connected to gas ultra-high pressure system with special Swagelok valves, which can be used several times for the rod re-loading. III. IN-PILE MEASUREMENTS AND DATA ANALYSIS The in-pile instrumentation allowed the rod dimensional changes (cladding elongation (EC) and rod diameter (DG)) for different overpressure levels at full power to be measured together with fuel centreline temperature which response on possible clad-fuel gap re-opening to be determined during irradiation. The test is not finished and data analysis presented here is preliminary. A. In-pile measurements at BOL The test was started with DG measurements at hot stand by conditions (HSB), which determined rod diameter profile at BOL. The hydraulic diameter measurements were performed by He gas flushing through the test rod, which provided an information on the fuel-cladding gap both at HSB and at first rise to power due to change gas flow resistance. The diametric fuel-clad gap as a function Average Linear Heat Rating (ALHR) is shown in Fig. 3. Upon completion of reactor operation cycle, the hydraulic diameter measurements were also carried out at hot standby and cold conditions for comparison with the results prior to irradiation test. B. In-pile measurements during irradiation The data recorded are based on the direct measurements of coolant and fuel temperatures, gas and coolant pressure, cladding elongation and NDs signals as well as parameters calculated like power, LHR at fuel thermocouple position and fast fluxes. The plots of some important for the test parameters are shown in Fig. 4 as a function of operational time. 3

4 The Hydraulic Diameter (HD) and periodically cladding diameter (DG) measurements together with on-line cladding elongation recording during irradiation have provided useful information on the fuel-cladding gap and cladding two dimensional changes under overpressure conditions. The compiled cladding diameter (DG) data showed that the cladding diameter was steadily increased under overpressure conditions. The rod diameter was changed to total increment of about 30 micron. The compiled DG measurements showed that cladding creep rate increased with the rod overpressure increased. The test provided the experimental data on LIFT-OFF E110 (sponge) cladding in the Halden reactor under representative PWR conditions. Fig. 3 Hydraulic diameter measurements in tests rod vs. ALHR Fig 4. Power and overpressure history, fuel temperature measurements during LIFT-OFF test. 4

5 IV. START-3A CALCULATIONS Below is a comparison of experimental data and calculations by the code START-3A and numerical analysis of the LIFT- OFF experiment. A. Base irradiation The assembly of IFA-728 underwent six cycles of irradiation in the Halden PWR loop during the period from 2010 to 2015 (see point I A). By the end of the sixth cycle, it was irradiated with 906 effective days, the history of the change in linear power is presented in Fig. 1. For numerical calculations, a special version of the START-3A code was prepared with a fuel re-fabrication module. The standard models of oxidation and creep of a cladding made of E110 material on a spongy basis were used in the simulation. The volumetric swelling of the fuel was taken equal to 0.56% per 10 MW day/kgu. Since the irradiation took place at a fairly high average power throughout the entire time (~ 35 kw/m), the temperatures at the center of the fuel and the yield of the fission gas release (FGR) after irradiation turned out to be high. The results of calculating the yield of the FGR and the temperature change in the center of the fuel are shown in Figure 5. The calculated gas release at the end of the base irradiation was 29%, the cold pressure was 7.7 MPa, the hot pressure was 11.7 MPa. Fig. 5. The history of the center temperature change for the experimental fuel rod by the code START-3A and the gas release during the basic irradiation period. Figure 6 shows the distribution of the thickness of the oxide film of the fuel rod #4 from the assembly IFA-728 and the calculated values according to the code START-3A. The linear heat rate was about 20 kw/m for the rod pre-irradiated to a burnup of 58 MW. day/kgu. The diameter at BOL provided by DG measurements is shown in Fig. 6 alongside with START-3A calculation results. The DG measurements performed during start up showed small diameter changes, due to the thermal expansion, while further DG measurements indicated rod diameter change at different overpressure levels due to cladding creep out during irradiation time. 5

6 Fig. 6. The thickness of the oxide film of the fuel rod #4 after the base irradiation in the assembly IFA-728 and the change in the outer diameter of the cladding of the experimental fuel element by the START-3A code in comparison with the experimental data according to the movable diameter measurement gauge before the LIFT-OFF test. B. LIFT-OFF TEST In Fig. 7 presents the calculation results of the temperature change at the center of the experimental fuel rod by the START-3A code in comparison with the experimental data according to the thermocouple at the center of the fuel column during the LIFT-OFF experiment. After the fuel rod reached a power rate of ~ 20 kw/m, the temperature at the center of the fuel was ~ 800 C when filled with helium and ~ 870 C when filled with argon. To eliminate the uncertainty introduced by the additional heat transfer of the gaseous medium, the boundary conditions of the first kind are applied in the calculations. 6

7 Fig. 7. Fuel Center Temperature of the experimental fuel rod change by the START-3A code in comparison with the experimental data according to the thermocouple at the center of the fuel column during the LIFT-OFF experiment. Figure 8 shows the results of calculating the change in the average outer diameter of the cladding of the experimental fuel rod by the START-3A code in comparison with the experimental data according to the mobile diameter measurement gauge. The calculated change in diameter during the LIFT-OFF experiment correlates well with the experimental data. 7

8 Fig. 8. Change in the average outer diameter of the cladding of the experimental fuel element calculated by the START-3A code in comparison with the experimental data according to the movable diameter measurement gauge. LIFT-OFF Fig.9. Calculation results of the gap between the fuel and the cladding of the experimental fuel element using the START-3A code. 8

9 C. Numerical Data analysis Experimental measurements of the temperature of the fuel center and the linear heat rate were numerically processed. The Halden group investigated the normalized temperature in Figure 4. The VNIINM group studied the time series x(t) = T (t) / ql(t), the ratio of the Temperature at the center of the fuel to the linear heat rate, shown in Figure 10 on the left. The aim of the analysis was to reveal the increasing tendency of the x(t). Areas with a stable positive derivative of x(t) would indicate a higher temperature growth rate relative to the growth rate of LHR and it would serve as an indirect confirmation of the presence of the LIFT-OFF effect. Fig.10. T/LHR signal and temperature increase rate during LIFT-OFF experiment. To exclude the effect of noise, the moving average method was used. The data covered by the sliding window was approximated, a linear trend was calculated. From the processed data a temperature increment per 1000 hours is obtained with a linear power of 20 kw / m, shown in Figure 10 on the right. Temperature change associated with the change in thermal conductivity during the experiment due to the gain of burnup was revealed and accounted. For this, additional calculations were made using the START-3A code with the "frozen" burnup at the beginning of the LIFT-OFF test at the burnup interval from 58 63MW. day/kgu. The calculation showed that the change in the temperature of the fuel center associated with this effect lies in the range of 3-12 K and does not affect the trend of temperature change as a whole, the result is shown in Fig. 10. Independent calculations of the Halden and VNIINM groups showed a tendency to the beginning of LIFT-OFF effect at an overpressure of 104 and 113 bar, the result is shown in Figure 11. Fig.11. Temperature increase rates and LIFT-OFF thresholds by Halden and VNIINM calculations. 9

10 V. RESULTS DISCUSSION The LIFT-OFF experiment was modeled using the START-3A code. The calculation is in satisfactory agreement with the available experimental data, namely, the change in the diameter of the shell, the growth of the oxide film, the temperature of the fuel center, the hydraulic diameter. From the results of experimental measurements and calculations using the START-3A code, a step-like growth of the fuel rod diameter during the LIFT-OFF test is seen in Fig. 8, in addition, calculation by the START-3A code indicates the growth of the gap between the fuel and the cladding, beginning with an excess pressure of 140 bar, fig. 9. Numerical analysis of the experimental data, carried out by the Halden and VNIINM groups, indicates a tendency to increase the fuel temperature at a linear heat rate, but there is no significant temperature increase at the moment. It was also revealed and taken into account the temperature change effect associated with the change in thermal conductivity during the experiment due to the gain of burnup. For this, additional calculations were made using the START-3A code with the "frozen" burnup at the burnup interval from MW. day/kgu. The calculation showed that the change in the temperature of the fuel center associated with this effect lies in the range of 3-12 K and does not affect the trend of temperature change as a whole, the result is shown in Fig. 10. The calculated prognosis using the START-3A code of further carrying out of LIFT-OFF test indicates a significant increase in the gap and gladding diameter. VI. CONCLUSIONS The LIFT-OFF experiment was conducted on a re-fabricated fuel element with a burn-up of 58 MW.day/kgU, which underwent basic irradiation in the Halden reactor in the assembly of IFA-728 in a loop with the conditions of irradiation of the PWR reactor. At the moment, the gas overpressure inside the fuel rod is 14 MPa. Analysis of the experimental data and calculations using the START-3A code indicates an increase in the cladding diameter and the gap between the fuel and the cladding at an excess pressure of 140 bar, in addition, a numerical analysis of temperatures indicates a steady tendency to increase in temperature, starting at 120 bar. The processing of the experimental data and the numerical calculation using the START-3A code indicate the beginning of the LIFT-OFF effect. It is expected that an increase in excess pressure should reveal a more significant growth in the gap and cladding diameter, starting at a pressure of 160 MPa and an increase in temperature at a pressure of bar, which is in agreement with the earlier experiment [1]. At the moment, the experiment is continuing, further increase in overpressure is planned and the PIE will be conducted. VII. ACKNOWLEDGMENTS The authors are grateful to Dr. Mikhail Peregud (JSC VNIINM) for fruitful discussions and consultations. VIII. REFERENCES 1. W. Wiesenack, T. Tverberg, E. Kolstad, S. Beguin Rod overpressure / LIFT-OFF testing at Halden- - in-pile data and analysis, Journal of NUCLEAR SCIENCE and TECHNOLOGY, Vol. 43, No. 9, p (2006). 2. Shestopalov A.A., Chulkin D.A., Demyanov P.G., Kuznetsov V.I., Novikov V.V., Zhitilev V.A., Ovchinnikov V.A. Modelling of the VVER High-Burnup Fuel Behavior under Ramp Conditions, EHPG, 7-11 September 2014, Roros, Norway 10