Experimental Investigations of Additives on Irradiation Performances of Oxide Fuel

Transcription

1 Experimental Investigations of Additives on Irradiation Performances of Oxide Fuel Boris Volkov* 1, Terje Tverberg 1, M. McGrath 1 1 Halden Reactor Project, Halden, P.O. Box 173, Norway Tel , Fax , borisv@hrp.no ABSTRACT: The fuel and material tests performed in the Halden reactor are aimed at supporting innovative fuel development which is aimed at improving the fuel cycle by: suppressing fission gas release; reducing pellet-cladding mechanical interaction; improving oxide fuel thermal conductivity; and as a consequence the in-pile fuel performance with burn-up accumulation. In general, all recent developments presented in the Halden experiments are based on the following additives: - Gd neutron absorbing isotopes, which suppress extra reactor reactivity at BOL for prolongation of the fuel cycles in LWRs - chromium oxides and aluminum silicates, which are able to increase fuel grain size during production and suppress fission gas release during irradiation - beryllium oxide, which increases composite oxide fuel thermal conductivity This paper reviews some of the experiments performed in the Halden reactor for understanding the effects of additives on oxide fuel performance during irradiation. The paper also presents future plans for the testing the Halden reactor other innovative fuels with increased tolerance to accident conditions. KEYWORDS: Halden Reactor, experiments, additives, large grain fuel, Gd fuel, chromium oxide, aluminum silicate, beryllium oxide fuel, thermal conductivity I. INTRODUCTION The safe and reliable operation of nuclear power plants benefits from R&D advances and related technical solutions. The OECD Halden Reactor Project (HRP) is a joint undertaking of more than 100 organizations from twenty countries across the world that together finance the experimental programs supporting not only safety and reliability but also innovative fuel development [1] Figure 1 Flags of the countries participarting in HRP Joint Progrme. The Joint Program is aimed at generating the following information for safety and licensing assessments: - data for fuel behavior model development, verification and validation - showing compliance of fuel performance with design, operational and safety criteria - assessing safety criteria with respect to available margins and reasonableness - a database for plant life extension and mitigation of core component ageing [2] The standard fuel pellets for Light Water Reactors (LWR), developed by many fuel vendors during many years, have to date provided excellent performance in most reactor operation conditions. However, in order to achieve even higher fuel reliability, further investigations are needed aimed at development of optimal design and fuel microstructure. Recent efforts in the area of nuclear fuel have been applied to investigate the behavior of fuel with Gd neutron absorbing isotopes implemented for suppressing excessive reactivity at BOL in NPPs with improved fuel cycles. Other modified fuels were also produced with some additives to increase fuel grain size, which is expected to reduce FGR at high burnup. Another type of composite fuel with BeO additive is developed to increase fuel thermal conductivity with consequent improvement of the fuel thermal performance. A review of some of the experiments performed in the Halden reactor for understanding the effects of additives on oxide fuel performance under irradiation is presented. II. EXPERIMENTAL INVESTIGATION OF ADDITIVES ON IRRADIATION PERFORMANCE OF OXIDE FUEL Fuel and material investigations at HRP are based on an integral approach which is provided by a reliable in-pile instrumentation enabling the following key in-pile parameters to be measured [3]: - fuel temperature to determine fuel thermal performance at power during burnup accumulation; - fuel elongation to detect fuel densification and swelling; - cladding elongation to measure deformation under irradiation and detect PCMI;

2 Relative power Relative power rate Temperature, O C Proceedings of WRFPM pressure in the fuel rods to detect FGR The analysis of the parameters measured and calculated gives some additional knowledge on the fuel behavior under irradiation. 1. Particularities of Gd-doped Fuel Behavior Several tests were performed in the Halden reactor with Gd-doped fuels containing of 2, 5 and 8 % absorbing isotopes. The isotopes are usually homogenously distributed in the oxide fuel pellets and calculations show that at the beginning of life, the radial power profile is strongly increased towards to fuel pellets surface unlike in ordinary uranium oxide fuel due to the neutron absorbing effect at BOL (Figure 2). 2,0 1,8 1,6 1,4 1,2 1,0 0,8 Radial power profiles: -Gd 2 O 3 0,6 0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0 Pellet radius, mm Figure 2 Calculated radial power distribution in fresh Gd-doped and fuels. The average power generated in Gd bearing rods is low at BOL, however, during irradiation the Gd absorbing isotopes burn out and the power steadily increases. At the same time the radial power profile is changing towards a normal radial power distribution with fuel burnup accumulation as shown in Figure 3 for the VVER 5% Gd-doped fuel tested at the Halden reactor. 3,0 2,5 2,0 1,5 1,0 Radius (mm): FTEMP3 code [5] at constant average power using the HELIOS data on power profile (Figure 3) is changing with burnup as shown in Figure 4. The same temperature measured in Gd-doped fuel follows to the theoretical calculations at very low power rating up to a certain burnup FTEMP3 calcualted at 4.5 kw/mtcent Measurements in 5% Gd-doped fuel Rod burnup, MWd/kg oxide Figure 4 Calculated and measured fuel centerline temperatures at constant power in Gd-doped fuel as a function of burnup. However, the measurements show that the centerline temperature in Gd-doped fuel is steadily reduced deviating from the theoretical trend at average power exceeding some value [6]. Obviously, this effect may be related to so-called late Gd-fuel relocation due to pellets cracking, which usually takes place in ordinary uranium oxide fuel at first rise to power. The fuel elongation measurements which are usually used to study fuel pellet dimensional stability like irradiation induced densification and swelling, were also affected by Gd-absorbing isotopes contained in the oxide fuel. No densification was found in the Gd doped fuel with content of absorbing isotopes higher than 5% but only swelling was seen from BOL as shown in Figure 5 0,5 0, Rod burnup, MWd/kg oxide Figure 3 Example of radial power variation vs. burnup calculated by the HELIOS code [4] for Gd-doped VVER fuel testing in the Halden reactor The strong change in radial power distribution and fuel temperature profile with burnup at BOL, plays a key role for Gd-doped fuel performance when the power generated in the rods with this fuel is lower compared to ordinary fuel. The fuel centerline temperature calculated by the Figure 5 Fuel elongation dimensional changes measured in the Gd-doped fuel and ordinary fuel tested in Halden.

3 In addition, the radial distribution of the fission power in Gd-doped fuel (see Fig. 2) affected the linear swelling rate against average fuel burnup. However, a correction against local burnup calculated for the Gd fuel periphery with dished PWR pellets allows a linear swelling rate for Gd doped fuel to be estimated. The data analysis shows (in the Figure 5 above) that the swelling rate for Gd-doped fuel is similar to ordinary fuel. A test with non-absorbing Gd isotopes was performed to determine Gd fuel thermal conductivity and its degradation with burnup in comparison with ordinary fuel. The fuel temperatures measured in the identical rods with Gd-doped and ordinary fuel during irradiation to a burnup of 60 MWd/kg oxide shown in Figure 6 indicated the lower Gd-doped fuel conductivity, however a similar degradation was determined in both fuel types. These data allowed the Halden fuel conductivity degradation correlation to be extended for Gd-doped fuel [7] Figure 6 Fuel centerline temperature measured in Gd-doped and ordinary fuels testing in Halden Magnitude and evolution of FGR as power gradually decreases due to fuel depletion; Recurrence of FGR after periods with low power. The main characteristics of the rods are described in Table 1. Fuel pellets and cladding materials were delivered by HRP organisations (Westinghouse Sweden, Areva ANP Germany and GNF USA). Two rods had fuel pellets doped with 200 ppm Al 2 O 3 in both rods and 900 ppm and 500 ppm Cr 2 O 3 in rod 1 and 5 respectively. The doped fuel had both higher density and larger grain size. Table 1 Fuel types tested in IFA-677 Rod Fuel type Ad. Ad. UO 2 Cr 2 O 3, ppm Al 2 O 3, ppm Grain size, μm Fill gas (He) pressure, bar Four of the rods (rods 1, 2, 5, and 6) had identical dimensions whereas the two remaining rods (rods 3 and 4) had a larger fuel diameter and larger gap. To study the influence of internal fill gas pressure on the FGR behaviour of the fuel, one of the large gap rods had a lower fill gas pressure than the other rods. The six rods were all fitted with pressure transducers (PF), fuel centreline thermocouples (TF) in both ends and fuel stack elongation detectors (EF). One rod was also furnished with a cladding extensometer (EC) to study PCMI. The rod instrumentation is shown in Figure 7. Experiments with Gd-doped fuel are continuing with the aim of determining the integral Gd-doped fuel performance with emphasis on FGR and PCMI studies at high fuel burnup and with different Gd contents. 2.I Irradiation performance of large grain fuel with additives Fuel vendors are developing fuels designed to suppress or delay FGR at high burnup. The main trend is to develop a technology for the production of large grained fuel with improved irradiation performance. It is expected that large grain fuel delays FGR due to altering the diffusion length and HBS structure formation. To achieve this, additives are used to stimulate grain growth during fuel production. Such fuels with additives like chromium oxide, aluminum oxide and aluminum silicate have been tested in the Halden reactor [8]. One of the tests in the Halden reactor containing six fresh, instrumented fuel rods, including fuels with Al 2 O 3 and Cr 2 O 3 additives, was performed to study the following phenomena: Onset of FGR for modern fuel with high initial power; Feedback effect from densification and fission gas release; Figure 7 Instrumentation of the test fuel rods in IFA-677. The irradiation started at a high power (~45 kw/m) and the power gradually decreased to ~25 kw/m by the end of the test when 30 MWd/kg, was achieved. The data showed that FGR had occurred in all the rods since the first cycle of operation due to the high initial rating and high fuel temperatures. In the beginning of the third cycle the FGR

4 fractions started to increase more significantly whereas the most significant increase in FGR occurred in the beginning of the 6 th cycle, when the rods achieved a higher linear power. The estimated FGR fractions during the four cycles of irradiation are shown in Figure 8 and the values at the end of irradiation are also summarized in Table 2. between 15 and 25 MWd/kg oxide. Data analysis showed that FGR from both fuels occurred at the measured fuel centerline temperatures exceeding the Halden (Vitanza) thermal FGR threshold [9]. The higher power in the rods with large grain fuel initiated larger FGR than in the fuel with reference VVER fuel as shown in Figure 9. Figure 8 Estimated fission gas release from on-line pressure measurements in IFA-677 Table 2 Estimate of FGR in IFA-677 Rod i 5 6 Fuel type Ad. Ad. UO UO 2 2 Burnup, MWd/kg FGR (%) It was noticed that FGR from the low pressurised rod (rod 3) was almost 30%. During the first two cycles, the rod pressure gradually increased, whereas in the beginning of the third cycle, significant FGR occurred, since the power was increased from its previous lower level. From the test results, no difference in FGRs between fuel with additives (large grain size fuel) and without additives (small grain size) was found. However, it should be noted that the power histories varied between the rods during irradiation as a result of varying radial flux distribution among the rods in the test assembly. It was suspected that the additives used for these types of fuel production enhanced the diffusion coefficients thereby compensating the effect from the increased diffusion length. Another test with fuel (supplied by JSC TVEL, Russia) produced with aluminum silicate additive (Al 2 SiO 5 ) for large grain fuel formation was also performed in the Halden reactor. Four rods from six (see Table 3) were included to the test matrix to study the effect of large grain fuel with additives on FGR at high burnup Table 3 Fuel types tested in IFA-676. Rod Fuel type W+ Wr W+ Wr Additives Al 2 SiO 5 - Al 2 SiO 5 - Grain size, μm In-pile PF,2TF, PF,2TF, ET;EC; ET;EC; detectors EF EF EF EF The first FGR was detected at a burnup in the range Figure 9 In-pile measurements of fuel temperatures and gas pressures used for studying FGR from large grain fuel (W+) and reference VVER fuel (Wr) testing in IFA-676. This fuel behavior was analyzed in detail using a fuel performance code with FGR modeling capability [10]. The FGR from the large grain fuel was explained by micro-cracking of the large grains during power uprating after the period with low power. The test rods continued irradiation also beyond the burnup where fission gas release was first seen, and at a burnup of 60MWd/kg oxide a special test to investigate the FGR from the same types of fuel was initiated again. The in-pile measurements of the fuel temperatures and rod pressures were used to evaluate the thermal FGR threshold and also to estimate the effect of large grain size on fission gas release at high burnup. FGR was detected at the measured fuel center temperature below the 1% FGR threshold for both fuels whereas the relative FGR from large grain fuel was lower than from reference VVER fuel, perhaps due to the lower fuel temperatures measured. Follow up investigations of large grain fuels in the Halden reactor are continuing in other tests with the aim to study FGR mechanism in fuels with or without additives. Another way to suppress FGR from fuel, which may limit the fuel discharge burnup, is a reduction of fuel temperature at the same power level. For this aim some additives with better conductivity than that of standard may be used. Some of these innovative fuel types have also been tested in the Halden reactor together with common fuel types. 3. Irradiation of composite uranium-beryllium oxide fuel A -BeO fuel is currently being tested in the Halden reactor together with other types of fuels for studying the FGR mechanism. The first in-pile data for this fuel type confirmed that the fuel temperatures measured in -BeO fuel are lower due to better thermal conductivity than in the

5 conventional fuel [8]. The in-pile measurements of the fuel centerline temperatures allowed evaluation of -BeO fuel conductivity against pure fuel. The fuel temperature measurements as a function of Linear Heat Rating (LHR) at the axial position of thermocouples are shown in Figure 10 in comparison with calculation using modified fuel conductivity correlation. -BeO fuel is proposed, with a microstructure schematically shown in Figure 12 where microspheres are embedded into the -BeO fuel matrix. Figure 12 Schematic view of microstructure of -BeO fuel produced with other technology [12]. Figure 10 In-pile fuel centerline measurements in the composite -BeO and pure fuels. The calculations indicated that the composite fuel with 3% BeO additives has a better conductivity by about 30% which is in good agreement with in-pile measurements. The following irradiation showed that FGR was not detected for this fuel type at the power ratings at which the FGR was substantial for both ordinary and large grain fuels. The conductivity improvement in this fuel type is achieved by implementation of the BeO powder (particles of μm with thermal conductivity of about 200 W/m K) into the matrix [11]. The basis of -BeO production is mechanical blending of and ВеО powders. The pellets are produced from the blended -ВеО powder by sintering at conventional conditions at С, with final microstructure shown in Figure 11. A comparison of the behavior of both of these fuels under irradiation will be interesting with a view to finding the -BeO fuel with the best in-reactor performance. III. CONTRIBUTION TO INNOVATIVEFUEL DEVELOPMENT AND TESTING After the Fukushima power plants accident in Japan in 2011, the discussion on the development of fuels with increased accident tolerance (ATF) for LWRs was initiated by the nuclear society [13]. Most attention is paid to new cladding types which are able to avoid hydrogen generation unlike Zr-based materials under overheating beyond DB LOCA. The main development is concentrated on some special materials like SiC and metals with for example high melting points like Mo alloys. However, some innovative fuel matrices which are able to trap excessive fission product release during both normal operation and accident situations have also been proposed for testing in the Halden reactor. One of these fuel matrix concepts is schematically shown in Figure 13 where the micro-cell fuel pellet concept is design to obstruct the diffusion of some fission products outside of the cell wall to the pellet periphery [14]. A test with such fuel amongst other proposed additive and high performance fuels is under consideration in the Halden Joint Programme. 10 m Figure 11 Microstructure of the composite UO2-BeO fuel produced in ULBA (Kazakhstan) and tested in the Halden reactor [11]. According to [12] an alternative technology for producing Figure 13 Proposed concept of the fuel matrix for accident tolerant fuel [14].

6 IV. CONCLUSIONS The main trend in the area of fuel developments for the current fleet of NPPs with PWR, BWR and VVER are based on conventional fuel production with some additives which are able to improve the irradiation performance of the fuel towards the increase of discharge burnup. The main concern is FGR and PCMI margins at high burnup which may limit the utilization of ordinary fuel. The OECD HRP has carried out extensive in-pile testing with fuels produced with different types of additives, studying in particular FGR mechanisms and PCMI. Innovative fuels and claddings being developed for increased accident tolerance are also under consideration by HRP member organizations for testing in the Halden Reactor in the near future. ACKNOWLEDGMENTS The authors would like to thank the HRP membership for their support of the Joint Program tests presented in the paper and the Halden staff for carry out and following up the long term irradiations of fuels with additives. Personal thanks to Radomir Jošek (Westinghouse Co., Sweden) for the detailed analysis of the data in the test IFA-677. NOMENCLATURE OECD- Organization for Economic Cooperation and Developments; HPR Halden Reactor Project BOL- Beginning of Life PF- pressure transducer for FGR detection EF fuel elongation sensor TF fuel thermocouple EC- cladding elongation detector; PCMI Pellet-cladding Mechanical interaction FGR- Fission Gas Release; HBS High Burnup Structure; BeO Beryllium Oxide DB LOCA Design Basis Loss of Coolant Accident ATF- Accident Tolerant Fuel. REFERENCES 1. C. Vitanza. Overview of the OECD Halden Reactor Project." Elsevier Science S.A." M.A. McGrath, B. Yu. Volkov Role of Halden Reactor Project for world-wide nuclear energy development, VI International Scientific Conference Contemporary issues of uranium industry, Almaty, Kazakhstan, September, 14-16, B. Volkov Integral Approach to Innovative Fuel and Material Investigations in the Halden Reactor, paper presented in 8-th International Conference on WWER Fuel Performance, Modelling and Experimental support, Bulgaria, R.J.J. Stamm ler et. al., "HELlOS Documentation; FMS the Scanpower Fuel Management System", T. J. Bjørlo, E. Kolstad, C. Vitanza, FUELTEMP-2, A compufer programme for analyzing the steady state thermal behaviour of oxide fuel rods, HPR A. S. Shcheglov, B. Yu. Volkov, V. N. Proselkov, Yu. Pimenov Specific Features of the WWER Uranium-Gadolinium Fuel Behavior at BOL, Conference WWER Fuel Modelling and Experimental Support, Sandanski, September, W. Wiesenack. Thermal Performance of High Burnup Fuel, In-Pile Temperature Data and Analysis. Presented at the ANS Meeting, Park City, Utah, USA, April R. Jošek The High Initial Rating Test IFA-677.1: Final Report On In-Pile Results, HWR-872, C. Vitanza, E. Kolstad, and U. Graziani; Fission Gas Release from Pellet Fuel at High Burnup; Proceedings of American Nuclear Society Topical Meeting on Light Water Reactor Fuel Performance, Portland, OR, D. A. Chulkin, V. I. Kuznetsov, A.V. Krupkin, V.V. Novikov «FGR Modeling of Large Grained Fuel During Transients by START-3A», ANS LWR Fuel Performance Meeting, TopFuel September 15-19, 2013, North Caroline, USA. p Y. Russin, Y. Shakhvorostov, M. McGrath Innovative fuel development in ULBA (Kazakhstan) and testing in the Halden reactor, 2011 WRFPM, TopFuel 2011, Chengdu, China, Sept , Sean M. McDeavitt, Chad Garcia, Jean C. Ragusa1, Joshua Smith, James Malone, Behavior Assessments For -BeO Enhanced Conductivity Fuel In A PWR. ANS LWR Fuel Performance Meeting, TopFuel September 15-19, 2013, North Caroline, USA. p Jon Carmack, Frank Goldner, Shannon M. Bragg-Sitton and Lance L. Snead, Overview Of The U.S. Doe Accident Tolerant Fuel Development Program, ANS LWR Fuel Performance Meeting, TopFuel September 15-19, 2013, North Caroline, USA. p Jae Ho YANG, Keon Sik KIM, Dong-Joo KIM, Jong Hun KIM, Jang Soo OH, Young Woo RHEE, Yang-Hyun KOO, Micro-Cell Pellets For Enhanced Accident Tolerant Fuel, ANS LWR Fuel Performance Meeting, TopFuel September 15-19, 2013, North Caroline, USA, p