1 106

Fuel Cladding Chemical Interaction Tests of Irradiated Metallic Fuel

J.-S. Kim, J.-H. Kim, J.-S. Cheon, B.-O. Lee, J.-H. Kim, and C.-B. Lee

Korea Atomic Energy Research Institute, 989-111 Daedeokdaero, Yuseong-gu, Daejeon, 305-353, Republic of Korea

E-mail contact of main author: juskim@kaeri.re.kr

Abstract. To investigate the fuel cladding chemical interaction for the irradiated metallic fuel, high temperature heating tests were performed. The fuel rods consisting of U-10Zr and U-10Zr-5Ce fuel with T92 cladding were irradiated in HANARO reactor. After the irradiation, the fuels was cut into cylindrical specimens, the specimens were exposed at high temperature of 750 C and 800 C for one hour. Microstructural examinations were conducted by utilizing optical microscope, scanning electron microscope, and electron probe micro-analysis. In the case of U-10Zr with T92 specimen, migration phenomena of U, Zr, and Fe as well as Nd lanthanide fission product were observed at the eutectic melting region. Elements distribution at the eutectic melting region demonstrates that eutectic melting occurs during high temperature experiment. However, any eutectic reaction region was not observed in U-10Zr-5Ce with T92 specimen. To get an intended reaction, the rig for high temperature heating was newly designed and the rig showed good performance through a preliminary out-of-file examination. In the subsequent study, eutectic reaction of U-10Zr-5Ce with T92, HT9 and FC92 will be conducted using the rig.

Key Words: Sodium cooled fast reactor, Fuel cladding chemical interaction, U-10Zr, Migration of fuel constituent

1. Introduction

Sodium-cooled fast reactor (SFR) has been developed worldwide due to an enhanced safety, fuel cycle economy, a reduction of a high-level waste volume and toxicity by burning the minor actinides such as Np, Am and Cm [1, 2, 3, 4]. The development project of Prototype Gen-IV SFR (PGSFR) was initiated to achieve an ultimate goal of transmutation capability of transuranic (TRU) elements from the spent nuclear fuel in Korea [5, 6].

The U-10Zr and ferritic-martensitic stainless (FMS) steel cladding has been considered the most probable fuel for the initial core of PGSFR [7] since the thermal conductivity of uranium alloys is among the highest of nuclear fuels and addition of Zr improves material compatibility between fuel and cladding and increases the solidus temperature [2, 8]. However, as-fabricated distribution of fuel constituents changes with irradiation time due to the radial temperature gradient and fission leading phase change of material, and affecting the fuel performance [9, 10, 11]. Post-irradiation examinations have showed that the significant redistribution of U and Zr occurs even at a short radiation exposure or a low burn-up [3]. Initially, bond sodium is filled in a gap between fuel slug and cladding for increasing the thermal conductivity of the gap. As the burn-up increases, however, the volume of metallic fuel increases due to the fission product and gas swelling resulting in a contact of fuel with the cladding, and then, the fuel constituents U, Zr tend to interact metallurgically with iron-based cladding material at elevated temperatures. In other words, inter-diffusion between fuel slug and cladding occurs during normal operation conditions and off-normal reactor events, and eventually, can cause eutectic melting and reduction of the initial metal thickness [12, 13]. In

2 106

particular, rare earth elements such as Ce, La, Pr, Nd produced from fission migrates down the temperature gradient and reacts with the cladding material, which lowers the eutectic threshold temperature and increasing the thickness of reaction layer [14, 15]. To mitigate the eutectic melting, placing a thin barrier inside wall of cladding has been proposed, and its feasibility has been proved [16, 17, 18].

In this work, irradiated U-10Zr and U-10Zr-5Ce fuel slugs with T92 cladding at high-temperatures heating tests were performed at hot-cell in Korea Atomic Energy Research Insitute (KAERI). After heating test, the microstructures of fuel and cladding are observed through optical microscope (OM), scanning electron microscope (SEM), and element distribution was analysed by using electron probe micro analyzer (EPMA). After the 1st

experiment in hot-cell, however, it was recognized that the heating experiment in air condition can significantly affect the eutectic melting by formation of oxide layer. Therefore, the experimental method and environment was modified for minimizing oxidation.

2. Experimental

2.1. Materials

U-10Zr and U-10Zr-5Ce with T92 cladding fuels were irradiated at HANARO test reactor in KAERI. The fuel irradiation was performed during 182 effective full power days (EFPDs). Figure 1 shows the linear heat generation rate (LHGR) of the fuel rod (U-10Zr with T92 cladding) and corresponding burn-up during irradiation. The LHGR of fuel rod is between 120 W/cm and 245 W/cm and the average maximum burn-up of 2.44 at% can be achieved at the end of cycle operation

Max.: 245.1 W/cm Max.: 2.44 at.%

0

0.5

1

1.5

2

2.5

3

0

50

100

150

200

250

300

0 50 100 150 200

Bur

nup

(at.%

)

Lin

ear

pow

er (W

/cm

)

EFPD

FIG. 1. Linear power of fuel rod with time

3 106

FIG. 2. Element distribution (a) in the fuel slug, and (b) near fuel-cladding interface after irradiation

Based on the irradiation history presented in Fig.1, the temperature of fuel rods is calculated by using MACSIS (KAERI’s fuel performance code) [19]. The maximum centerline temperature is 549 C which is far below the phase transition temperature of U-Zr phase diagram [20], therefore, fuel slug would be in the + phase during Hanaro irradiation. Fig. 2 shows the elements distribution in fuel slug and near the fuel and cladding interface of the irradiated U-10Zr specimen. As shown in the figure, uniformly distributed Zr elements migrate into the fuel center. It is general tendency that Zr tends to diffuse hotter area if the

4 106

fuel was operated in the + phase [3, 21]. It should be noted that fission product of Nd is observed at the perirphery of the fuel slug and significant Nd migration occurs to the colder area.

2.2. Experimental

The irradiated fuel rods were cut into short segments of 10 mm and were inserted into the chamber of an electric furnace in the hot-cell of KAERI. To evaluate the fuel cladding chemical interaction (FCCI) under the transient condition of SFR, the U-10Zr and U-10Zr-5Ce fuel rods were heated to 750 C, and 800 C, respectively. The heating rate from room temperature to the reaction temperature is around 0.2 C/sec. Dwell time at test temperature is 1 hour, and then the specimens were furnace-cooled to room temperature. During the test, inert He gas was injected into the chamber at flow rate of 150 mL/min to minimize oxidation of test specimens. Moreover, Zircaloy-4 sheet was inserted in the chamber as an oxygen getter. Microstructures of test specimens were observed by using the OM, SEM and EPMA after completing the heating test.

3. Results and discussion

3.1 Post-irradiation examinations

Figure 3 shows a cross section of U-10Zr/T92 specimen after heating test. As shown in Fig. 3, eutectic melting region was locally observed at the interface of fuel and cladding. It seems that the mechanical contact is not uniform in entire fuel region. Figure 4 shows the fuel constituent distributions and eutectic penetration depth of U-10Zr/T92 specimen. As shown in the figure, inter-diffusion of fuel constituent occurs, i.e., U and Zr penetrates into the cladding region while Fe component of cladding slightly penetrated into fuel slug region at the eutectic melting region. Interestingly, the lanthanide fission product, the Nd at periphery of the fuel region is distributed at the eutectic melting region.

Figure 5 shows the experimental results with the literature value [2]. The measured maximum eutectic melting depth and corresponding penetration rate are 115 m and 0.032 m/sec, respectively. The penetration rate is almost similar but slightly higher than the value of 0.026 m/sec predicted by existing eutectic penetration rate currently being used for design and modeling purposes, and other irradiated ones. It is thought that the difference comes from the additional reaction during the slow furnace cooling. On the other hand, the eutectic penetration rates of this study are much greater than those of the un-irradiated U-10Zr specimens with FMS [16, 17, 18]. Though the dependency of penetration depth on lanthanide fission product has not been quantitatively assessed, this result indicates that the lanthanide fission products play a significant role in accelerating the eutectic melting reaction.

On the other hand, any reaction layer was not observed in U-10Zr-5Ce/T92 specimen. Figure 6 shows the cross-section of the U-10Zr-5Ce/T92 after the heating test. As shown in the fig-ure, a significant gap was found between the fuel slug and cladding. A couple of tests were re-peated to evaluate the eutectic reaction between U-10Zr-5Ce and T92 cladding under air con-dition; however, such a gap was observed in all of the specimens. It seems that the inert He environment and Zircaloy-4 oxygen getter do not prevent the oxidation reaction.

5 106

FIG. 3. Microscopic photograph of U-10Zr/T92 specimen after heating test (750 C, 1 hr).

FIG. 4. Constituent distributions and eutectic penetration depth of U-10Zr/T92 specimen after heating test (750 C, 1 hr).

6 106

0.00001

0.0001

0.001

0.01

0.1

1

650700750800850900

Max

. cla

ddin

g pe

netra

tion

rate

(m

/s)

Temperature (oC)

5.6 at.% (U-19Pu-10Zr) [2]

11.1 at.% (U-19Pu-10Zr) [2]

2.44 at.% (U-10Zr): in this work

Unirradiated U-10Zr [16-18]

/ exp 11.646 15665 / ( )m s T K

FIG. 5. Maximum eutectic penetration rate along cladding direction.

FIG. 6. Microscopic photograph of U-10Zr-5Ce/T92 specimen after heating test (800 C, 1 hr).

7 106

3.2 Modification of experimental methods

As discussed in the previous section, it is found that the current system (He inert gas and oxygen getter) does not prevent oxidation of the metallic fuel. Therefore, the experimental method and procedure were re-established. By focusing on preventing oxidation, vacuum tube was introduced; moreover, the special rig for transient test was designed and manufactured. Figure 7 shows the schematic diagram of the rig. As shown in the figure, the top and the bottom plates of the specimens come into contacts with FC92 and HT9 plates, which enables to induce eutectic reaction of fuel slug with T92, FC, and HT9 simultaneously.

FIG. 7. Schematic diagram of the rig

In order to test the performance of the rig, the rig was heat up to 800 C and stay for one hour under air and vacuum environment (10-5 torr). In the test, instead of the irradiated fuel rod, mischmetal (70Ce-30La) with T92 cladding was inserted into the rig. After one hour exposure at 800 C, the specimen was extracted from the furnce, and then the specimen was air-cooled for minimizing any additional reaction during furnace cooling. Figure 8 shows the cross sectional view of the reaction between mischmetal and FMSs. The reaction layers were clearly seen in Fig. 8(a); however, in the case of heating test under air, not reaction layers but large gaps and cavity were observed as shown in Fig. 8(b).

FIG. 8. OM image of the cross section of reaction between mischmetal and FMSs after heating test(800C, 1hour) under (a) vacuum, and (b) air

8 106

4. Summary

The chemical interaction of irradiated metallic fuel and T92 cladding was conducted at 750 C. After heating metallurgical microstructure was observed by using OM, SEM. In the case of U-10Zr slug with T92 specimen, eutectic reaction layer was observed and element distribution indicates that significant migration of elements occurs and Nd plays a significant role in increasing penetration depth. The measured penetration rate is almost similar but slightly higher than the reference value. It is thought that the difference comes from the furnace cooling. On the other hand, no eutectic melting region was found in the case of irradiated U-10Zr-5Ce with T92 specimens by fuel slug oxidation. Therefore, to minimize the oxidation of specimen at high temperatures, special rig for the heating test was made. Preliminary examination using the rig showed that rig is effective in preventing the oxidation. In the subsequent study, irradiated U-10Zr-5Ce with FMS at high temperature heating tests will be conducted with the developed rig under the vacuum environment.

Acknowledgement

This project has been carried out under the Nuclear R&D program by Ministry of Science, ICT & Future Planning

Reference

[1] Sofu T. A review of inherent safety characteristics of metal alloy sodium-cooled fast reactor fuel against postulated accidents. Nuclear Engineering and Technology. 2015;47:227-39.[2] Cohen AB, Tsai H, Neimark LA. Fuel/cladding compatibility in U-19Pu-10Zr/HT9-clad fuel at elevated temperatures. Journal of Nuclear Materials. 1993;204:244-51.[3] Hofman GL, Hayes SL, Petri MC. Temperature gradient driven constituent redistribution in UZr alloys. Journal of Nuclear Materials. 1996;227:277-86.[4] Brémier S, Inagaki K, Capriotti L, Poeml P, Ogata T, Ohta H, Rondinella VV. Electron probe microanalysis of a METAPHIX UPuZr metallic alloy fuel irradiated to 7.0 at.% burn-up. Journal of Nuclear Materials. 2016;480:109-19.[5] Yoo J, Chang J, Lim J-Y, Cheon J-S, Lee T-H, Kim SK, Lee KL, Joo H-K. Overall System Description and Safety Characteristics of Prototype Gen IV Sodium Cooled Fast Reactor in Korea. Nuclear Engineering and Technology. 2016;48:1059-70.[6] Kim J-B, Jeong J-Y, Lee T-H, Kim S, Euh D-J, Joo H-K. On the Safety and Performance Demonstration Tests of Prototype Gen-IV Sodium-Cooled Fast Reactor and Validation and Verification of Computational Codes. Nuclear Engineering and Technology. 2016;48:1083-95.[7] Lee CB, Cheon JS, Kim SH, Park J-Y, Joo H-K. Metal Fuel Development and Verification for Prototype Generation IV Sodium-Cooled Fast Reactor. Nuclear Engineering and Technology. 2016;48:1096-108.[8] Hofman GL, Walters LC, Bauer TH. The Technology of the Integral Fast Reactor and its Associated Fuel CycleMetallic fast reactor fuels. Progress in Nuclear Energy. 1997;31:83-110.[9] Pahl RG, Lahm CE, Hayes SL. Performance of HT9 clad metallic fuel at high temperature. Journal of Nuclear Materials. 1993;204:141-7.[10] Pahl RG, Porter DL, Lahm CE, Hofman GL. Experimental studies of U-Pu-Zr fast reactor fuel pins in the experimental breeder reactor-ll. MTA. 1990;21:1863-70.

9 106

[11] Porter DL, Lahm CE, Pahl RG. Fuel constituent redistribution during the early stages of U-Pu-Zr irradiation. MTA. 1990;21:1871-6.[12] Ogata T, Nakamura K, Itoh A, Akabori M. Reactions between U–Zr alloys and Fe at 1003 K. Journal of Nuclear Materials. 2013;441:579-82.[13] Ogata T, Nakamura K, Kurata M, Yokoo T, Mignanelli MA. Reactions between U-Pu-Zr Alloys and Fe at 923 K. Journal of Nuclear Science and Technology. 2000;37:244-52.[14] Kim JH, Cheon JS, Lee BO, Kim JH. Interaction behavior between binary xCe-yNd alloy and HT9. Journal of Nuclear Materials. 2016;479:394-401.[15] Inagaki K, Ogata T. Reaction of lanthanide elements with Fe–Cr alloy. Journal of Nuclear Materials. 2013;441:574-8.[16] Ryu HJ, Lee BO, Oh SJ, Kim JH, Lee CB. Performance of FCCI barrier foils for U–Zr–X metallic fuel. Journal of Nuclear Materials. 2009;392:206-12.[17] Kim JH, Ryu HJ, Baek JH, Oh SJ, Lee BO, Lee CB, Yoon YS. Performance of a diffusion barrier under a fuel–clad chemical interaction (FCCI). Journal of Nuclear Materials. 2009;394:144-50.[18] Yang SW, Ryu HJ, Kim JH, Lee BO, Lee CB. FCCI barrier performance of electroplated Cr for metallic fuel. Journal of Nuclear Materials. 2010;401:98-103.[19] Lee BO, Cheon JS, Lee CB. Performance limit analysis of a metallic fuel for Kalimer: American Nuclear Society - ANS, La Grange Park (United States); American Nuclear Society, 555 North Kensington Avenue, La Grange Park, IL 60526 (United States); 2007.[20] Leibowitz L, Blomquist RA, Pelton AD. Thermodynamics of the uranium-zirconium system. Journal of Nuclear Materials. 1989;167:76-81.[21] Ogawa T, Iwai T, Kurata M. Demixing of U-Zr alloys under a thermal gradient. Journal of the Less Common Metals. 1991;175:59-69.