- Email: [email protected]
Chemical compatibility of Eurofer steel with sodium-potassium NaK-78 eutectic alloy
Fusion Engineering and Design 103 (2016) 31–37
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Chemical compatibility of Eurofer steel with sodium-potassium NaK-78 eutectic alloy A. Abou-Sena a,∗ , F. Arbeiter a , S. Baumgaertner b , T. Boettcher a , A. Heinzel c , H. Piecha d , K. Zinn a a
Institute for Neutron Physics and Reactor Technology (INR), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany Institute for Applied Materials – Applied Materials Physics (IAM-AWP), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany c Institute for Pulsed Power and Microwave Technology (IHM), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany d Institute for Nuclear and Energy Technologies (IKET), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany b
h i g h l i g h t s • • • •
Eurofer SSTT specimens were immersed in static NaK-78 for 6 months at 480–500 ◦ C. The Eurofer steel did not suffer any degradation in its mechanical properties. The chemical interaction between Eurofer specimens and NaK-78 was minimal. Using NaK-78, with low oxygen content, is feasible for the IFMIF HFTM capsules.
a r t i c l e
i n f o
Article history: Received 29 June 2015 Received in revised form 4 December 2015 Accepted 5 December 2015 Available online 24 December 2015
a b s t r a c t In the high flux area of the International Fusion Materials Irradiation Facility (IFMIF) neutron source, the capsules of the High Flux Test Module (HFTM) contain SSTT Eurofer specimens for fusion relevant irradiation at temperature up to 550 ◦ C. Using the sodium potassium eutectic alloy NaK-78 to fill the gaps among the Eurofer specimens stacked inside the HFTM capsules was introduced in order to improve the thermal conduction among all specimens and have uniform and predictable temperature distribution. Therefore the objective of this study is to investigate the chemical compatibility between Eurofer steel and NaK-78 to evaluate the applicability of this concept. In the present experiment, the SSTT Eurofer specimens were immersed in static NaK-78 inside a capsule made of Eurofer and kept under IFMIF HFTMrelevant conditions including high temperature (cycling between 480 ◦ C and 500 ◦ C) and duration of six months. Following the experiment, mechanical tests (tensile and Charpy impact) of the Eurofer specimens were performed in addition to surface and microstructure analyses to detect any relevant corrosion or degradation. The mechanical tests revealed that the Eurofer specimens did not show any degradation in their mechanical properties. Also, the surface and microstructure analyses showed that the chemical interaction between the Eurofer steel and NaK-78 was minimal after six month of exposure at cyclic temperature between 480 and 500 ◦ C. © 2015 Elsevier B.V. All rights reserved.
1. Introduction The structural materials of the fusion reactor in-vessel components including the first wall and blanket should cope with the severe working conditions without any degradation in their mechanical properties or dimensional stability beyond the allowable design limits. Selecting these structural materials requires a
∗ Corresponding author. E-mail address: [email protected] (A. Abou-Sena). http://dx.doi.org/10.1016/j.fusengdes.2015.12.006 0920-3796/© 2015 Elsevier B.V. All rights reserved.
dedicated testing at fusion-relevant irradiation conditions. Therefore building the International Fusion Materials Irradiation Facility (IFMIF) has become an unavoidable step in the way to design and construct a fusion reactor. IFMIF is an accelerator-based neutron source that uses lithium-deuterium Li(d,xn) nuclear reactions to produce a neutron flux similar to that expected at the fusion reactor blanket. The task of IFMIF is to irradiate the fusion materials in irradiation conditions similar to those of the future fusion reactor DEMO in order to: (i) provide data for the engineering design of DEMO, (ii) contribute to the selection of candidate fusion materials, and (iii) validate the understanding of the irradiation effects on the
32
A. Abou-Sena et al. / Fusion Engineering and Design 103 (2016) 31–37
fusion functional materials. Next to the IFMIF neutron source in the high flux region, the irradiation of the structural material (e.g. Eurofer steel) specimens will be performed in the High Flux Test Module (HFTM) [1]. Eurofer steel belongs to the group of the Reduced Activation Ferritic Martensitic (RAFM) steels that have swelling resistance and low activation properties. Therefore Eurofer steel has been developed and optimized to serve as a structural material for the fusion reactor blanket. The Eurofer specimens are stacked inside the HFTM capsules and should have a specific uniform temperature therefore the heat transfer, mainly conduction, should be well understood and predictable. The idea of using the sodiumpotassium eutectic alloy NaK-78 (with 78 wt% K and 22 wt% Na) to fill the small gaps among the Eurofer specimens has been considered in order to: (i) improve the thermal conduction among all capsule parts and the heat transfer predictability, (ii) establish a uniform temperature distribution, and (iii) ensure that the thermocouples are in contact with NaK-78 or the specimens and not a gap to obtain accurate representative temperatures. However, the chemical compatibility of Eurofer with NaK-78 needs to be experimentally investigated to evaluate the applicability of this concept. For instance, the corrosion level and its influence on the mechanical properties and microstructure of the Eurofer specimens should be evaluated. 2. Literature review When the steel is exposed to a liquid metal (e.g. NaK or Na), the corrosion possibly occurs due to: (i) a mass transfer of non-metallic elements such as oxygen, carbon, nitrogen, and hydrogen, and (ii) a mass transfer of metallic elements as some of the steel constituents may dissolve into the liquid metal and react with its impurities; therefore the solubility of the steel constituents in the liquid metal plays a significant role in the corrosion process. The mass transfer of non-metallic elements can change the microstructure and hence the mechanical properties of the steel on the long run. Hence, it is preferred to have low concentrations of the non-metallic elements in the liquid metal. The chemical reactions between the steel and the surrounding liquid metal with its impurities, in most cases oxidation of the steel, cause the corrosion process. The oxygen content in the liquid metal is an important factor that can accelerate the corrosion of the structural materials; therefore it should be minimized. A previous study [2] reported that the corrosion of metals by a liquid metal is affected by some parameters including system temperature, oscillation of cyclic temperature, ratio of exposed solid surface area to volume of liquid metal, impurities in the liquid metal, flow velocity of the liquid metal, conditions of the container, and composition and microstructure of the materials involved. For instance, increasing the system temperature leads to higher solubility of the steel constituents into the liquid metal, higher diffusion rates and smaller viscosities and consequently results in a higher corrosion rate. Also, the cyclic temperature oscillation is a catalytic factor; for example the corrosion rate of the Cu-Bi system at 500 ± 5 ◦ C is several times higher than that at 500 ± 0.5 ◦ C as reported in the literature [2]. Another previous work [3] studied the compatibility between the RAFM steel and sodium, where the data of sodium except for oxygen solubility were used to evaluate the NaK compatibility. A mass transfer of a metallic element between the RAFM steel specimens and the Na (or NaK) was observed. It was concluded that the corrosion depth of RAFM specimens, due to the dissolution of metallic elements into the liquid metal and the oxidation by oxygen present in the liquid metal, is less than 1 m if the liquid metal is purified using the cold trap, and hence it is unlikely to affect the RAFM steel performance. It was stated in the literature [4] that the corrosive properties of sodium and NaK are similar and no distinction between the two liquid metals can be observed.
A fusion relevant study [5] presented an assessment of using alkali metals as a coolant for the ITER blanket. It was reported that the austenitic stainless steels did not produce iron or chromium oxides in the environments of liquid Na and NaK particularly at oxygen levels that correspond to a cold-trapped liquid. Hence, the corrosion rates of the steels are fairly low under the aforementioned conditions. In another study [6], the experimental results indicated that all stainless steels that were tested showed a good performance in static NaK capsules up to a temperature of 760 ◦ C. For the stainless steel SS310 in a very pure NaK, there was very small initial weight loss, which decreased quickly with the increase of exposure time at 760 ◦ C. The authors concluded that stainless steels are extremely resistant to a chemical attack by NaK at temperatures up to 760 ◦ C. In addition, experimental examinations were performed [7] to evaluate the performance of a stainless steel 316 loop for circulating the NaK-78 at temperature of 760 ◦ C for a long-term exposure of 32,600 h. It was concluded that the SS316 performed satisfactorily as a structural material for the loop and some changes in its mechanical properties and metallurgical effects occurred only when circulating non-isothermal NaK-78. However, these changes did not significantly affect the material properties in a way that can result in a material failure. At Oak Ridge National Laboratory, eleven experimental loops were built for the SNAP-8 corrosion program [8] to study the compatibility of NaK-78 with some structural materials including the nickel-based alloy Hastelloy N as well as the stainless steels 316 and 347. The loops were operated with a minimum NaK-78 temperature of 1100 ◦ F and for a running time of about 2000 h. The metallurgical analyses showed that the deposits consisted mainly of nickel, chromium and manganese in the low oxygen loops while the deposits were basically iron and chromium in the high oxygen loops. It was concluded that the oxygen level in NaK-78 is the most important factor that affects the corrosion of steels. Also, it was noted that corrosion of stainless steels 316 and 347 was very low compared with that of Hastelloy N.
3. Experimental setup The experimental setup, shown in Fig. 1, mainly consists of: (i) the capsule that contains the Eurofer specimens and NaK78, (ii) two heaters, seven thermocouples, an insulation block, structural supports and a metallic base, and (iii) a gas-tight cylinder to contain most of the setup’s components. The capsule has a shape of rectangular prism with inner and outer dimensions of 24 mm × 28 mm × 66 mm and 36 mm × 40 mm × 78 mm respectively. The capsule walls, cover, and base were welded together by electron beam welding because it is performed in vacuum; and hence it minimizes any contamination and produces consistent welds. The capsule has three tubes for filling, bleeding and pressurizing; and they penetrate from its cover and are used for the NaK filling process. Fig. 2 shows the capsule packed with the Eurofer specimens as well as its tubes and cover. The capsule, its tubes, and the Eurofer specimens were manufactured from the same Eurofer plate (Eurofer 97-2, batch 993391, thickness of 25 mm) and its properties are given elsewhere [9]. The capsule encloses 20 Eurofer specimens, similar to the IFMIF HFTM specimens, for post experiment mechanical tests. The Eurofer specimens were manufactured according to the Small Specimen Test Technology (SSTT) which was developed to investigate mechanical properties of nuclear materials taking into account the limited irradiation volumes in the test nuclear reactors and accelerator-based neutron sources. The tensile test specimen has a diameter of 2 mm and a gauge length of 7.6 mm. The Charpy impact specimen has dimensions of 27 mm × 4 mm × 3 mm with a v-notch depth of 1 mm, 60◦ notch angle, and 0.1 mm notch tip radius. In addition, the capsule contains two specimens, with a shape of rectangular prism
A. Abou-Sena et al. / Fusion Engineering and Design 103 (2016) 31–37
33
Fig. 1. Setup assembly: (a) picture during assembly, (b) sectional view of the setup.
heat loss by conduction. The capsule is enclosed by a gas-tight cylinder to control the atmosphere around the capsule and to serve as a secondary containment for the NaK-78. The cylinder consists of two CF blank flanges DN200, CF straight connector DN200, and two CF copper gaskets. This cylinder is evacuated and then filled with nitrogen (99.999 vol.%) at a pressure of around 1.2 bar absolute to have an inert shielding atmosphere around the capsule containing NaK78. Also, nitrogen has low thermal conductivity of 0.056 W/m K at 500 ◦ C [10] which helps to reduce the heat loss from the hot capsule to the surroundings. In addition, a pressure gauge is connected to the cylinder to observe its pressure. Temperatures of the capsule core and outer surfaces are measured and recorded continuously (24 h/day and 7 days/week) during the experimental campaign of 6 month. Four thermocouples are attached to the capsule outer surfaces in addition to three thermocouples immersed in NaK-78 inside the capsule. All thermocouples are type K with 1 mm diameter and can be used in temperature up to 900 ◦ C with deviations of ±2.5 ◦ C up to 333 ◦ C and ±0.75% above 333 ◦ C as reported by the manufacturer. The sheath material of the thermocouples is stainless steel AISI 321.
4. Experimental work Fig. 2. (a) Packing the capsule with specimens, (b) filling the capsule with NaK-78.
and dimensions of 12 mm × 10 mm × 4 mm, dedicated to the surface and microstructure analyses. The capsule is heated up to the required temperature range of 480–500 ◦ C by two electrical band heaters that have the following specifications: (i) internal diameter of 63.5 mm and width of 63.5 mm, (ii) voltage of 240 V and power of 1000 W, and (iii) stainless steel cover and maximum operating temperature of 760 ◦ C. A heat flux smoother, manufactured from a copper rod, is tightly placed between the capsule and the heaters in order to improve the heat transfer because any area of low heat conduction, between the capsule and the heaters, and thus chance of overheating should be prevented. An insulation block is placed underneath the capsule at the bottom of the setup to minimize the
4.1. Thermal management A thermal management of the experiment was needed to: (i) cycle the capsule temperature between 480 and 500 ◦ C, and (ii) protect the experimental setup from overheating during the experiment running time of six months. Beside the capsule’s heaters and thermocouples, a variable transformer and a temperature limiter were implemented in the setup. The temperature limiter receives a temperature reading from one of the capsule’s core thermocouples as input and compares it to the adjusted safety temperature limit (590 ◦ C) and then provides an output to the variable transformer which controls the input power to the capsule’s heaters. If the temperature passes the aforementioned safety temperature, the temperature limiter cuts the input power to the heaters to prevent
A. Abou-Sena et al. / Fusion Engineering and Design 103 (2016) 31–37
The difficulty of using NaK-78 lies in the volatile reactions it has with air and water. NaK-78 reacts with oxygen rapidly forming potassium oxide and superoxide which make a crust on its surface. In addition, NaK-78 reacts with water to release hydrogen and form potassium hydroxide and sodium hydroxide while the reaction heat may cause a fire. Therefore NaK-78 should be handled with safety measures and it is usually stored and studied in a controlled space such as a glove box filled with dry argon or nitrogen. NaK-78 has a melting temperature of −12.6 ◦ C hence it is liquid at room temperature. Its density is 867 kg/m3 at 20 ◦ C and decreases to 749 kg/m3 at 550 ◦ C [11]. In this study, NaK-78 does not flow but rather fills the capsule and occupies all spaces around the specimens. Filling the capsule with NaK-78 was performed inside a glove box filled with a mixture of argon and helium. The filling process was performed as follows. First, the capsule was moved into the glove box and then purged with a gas stream of argon and helium. Second, a filling mechanism (consists of metal syringe, piston, linear bearing and guides, see Fig. 2b) was used to fill the capsule through the filling tube. Third, the capsule was wrapped with a band heater and an insulation blanket to heat it up to 350 ◦ C. In addition, two thermocouples were attached to the capsule to monitor its temperature. Finally, the capsule was filled with 20 cm3 of NaK-78 sufficient for the specimens to be immersed completely. 4.3. Specimens retrieval After completing the experiment (i.e. keeping the capsule at temperature of 480–500 ◦ C for 6 month), the subsequent step was to retrieve the Eurofer specimens from the capsule. First, all tubes and cables attached to the gas-tight cylinder were disconnected, and then the cylinder was carefully transported to a glove box, where the capsule was taken out from the cylinder and the NaK-78 inside the capsule was drained through its tubes. Second, the capsule was opened by cutting off its cover and the specimens were
Yield strength (MPa)
4.2. NaK filling
600 500 400 300 NaK exposed -1st set NaK exposed -2nd set Base material -1st set Base material -2nd set Base material [9]
200 100 0 0
100
200
300
400
500
600
700
Temperature (°C) Fig. 4. The yield strength of the Eurofer specimens.
700
Ultimate strength (MPa)
an overheating. In addition, data acquisition software (LabView 2012) was used to: (i) record the capsule temperatures versus time and (ii) turn the heaters on/off to achieve the cyclic temperature using the capsule temperature readings as inputs for the software. The input power to the capsule’s heaters is adjusted manually using the variable transformer.
600 500 400 NaK exposed -1st set NaK exposed -2nd set Base material -1st set Base material -2nd set Base material [9]
300 200 100 0 0
100
200
300
400
500
600
700
Temperature (°C) Fig. 5. The ultimate tensile strength of the Eurofer specimens.
12 10
Impact energy (J)
34
8 6
NaK exposed -1st set NaK exposed -2nd set
4
Base material -1st set Base material -2nd set
2
Base material [13] 0 -120 -100 -80
-60
-40
-20
0
20
40
Temperature (°C) Fig. 6. The impact energy for the Eurofer specimens.
taken out to be cleaned from the NaK-78 adherent to their surfaces by immersing in ethanol. Fig. 3 shows the procedures performed to retrieve the specimens inside the glove box. After the specimens retrieval, it was clear that the specimens were totally wetted by NaK-78, see Fig. 3b. Finally, the specimens were moved from the glove box for further cleaning using ethanol and an ultrasonic cleaner. 5. Results and discussion
Fig. 3. (a) Opening the capsule, (b) NaK-wetted specimens, and (c) immersing in ethanol.
The post experiment tests and analyses of the Eurofer specimens are necessary to investigate any relevant chemical interaction between Eurofer and NaK-78 such as corrosion and its impact on
A. Abou-Sena et al. / Fusion Engineering and Design 103 (2016) 31–37
35
Fig. 7. LOM microstructure image (left) and SEM surface image (right) of base material.
Fig. 8. SEM micrograph of the surface of the NaK-78 exposed specimen.
the Eurofer’s mechanical properties. In the following sections, the results of the mechanical tests (tensile and Charpy impact) as well as the surface and microstructure analyses of the Eurofer specimens are presented. 5.1. Mechanical tests The tensile and Charpy impact tests of Eurofer specimens were performed to investigate any degradation in its mechanical properties due to NaK-78 exposure under a cyclic temperature of
480–500 ◦ C for 6 month. The tensile tests were implemented to measure the ultimate tensile strength and yield strength of the Eurofer specimens at temperatures (20, 300, 400, 500, 600 ◦ C) relevant to the fusion blanket. The Eurofer specimens were divided into two groups: (i) base material specimens which were not exposed to NaK-78 and (ii) NaK-78 exposed specimens. Each group has two specimens for each testing temperature leading to two sets of results. The values of the yield strength (YS) and the ultimate tensile strength (UTS) are presented in Figs. 4 and 5 respectively for these sets and also for Eurofer base material of a previous study [9]. The values of the UTS and YS for the NaK-78 exposed specimens are comparable with those of the base material specimens; however, they are higher by 1–7%. These deviations in the UTS and YS values are within the range one would expect from the variation of the tensile tests [12]. From the obtained results, one can conclude that the implemented NaK-78 exposure has no effect on the UTS and YS of the Eurofer specimens. Two groups (namely base material and NaK exposed) of the Charpy impact specimens were tested at temperatures of −100, −80, −60, −40, 20 ◦ C using two specimens per temperature. As shown in Fig. 6, the values of the impact energy for the NaK-78 exposed specimens are comparable with those of the present and previous [13] base material specimens indicating that the NaK78 exposed specimens did not suffer any reduction in the upper shelf energy (USE) or shift in the ductile-to-brittle transition temperature (DBTT) within the given temperature range. The DBTT is important for the fusion blanket structural components because of the potential for embrittlement due to the neutron irradiation. Based on the results of the tensile and Charpy impact tests, one
Fig. 9. SEM surface image of the NaK-78 exposed specimen and EDX measurements for the three marked points in the accompanying table.
36
A. Abou-Sena et al. / Fusion Engineering and Design 103 (2016) 31–37
Fig. 10. Optical micrograph of cross section of the NaK-78 exposed specimen (etched).
may conclude that the existence of NaK-78 in contact with the Eurofer specimens does not affect its mechanical properties. This conclusion is in agreement with the results of a previous study [3] about the compatibility of reduced activation ferritic martensitic steel specimens with liquid Na and NaK. 5.2. Surface and microstructure analyses Surfaces of the NaK-78 exposed specimens (12 mm × 10 mm × 4 mm), dedicated to surface and microstructure analyses, were examined using a scanning electron microscope (SEM) that has an integrated energy dispersive X-ray (EDX) analysis system. Afterwards, the specimens were cut with a diamond disk saw perpendicular to the exposed surface, then embedded, ground and polished for metallographic examinations. The cross sections were analyzed using a SEM equipped with an EDX system and a light optical microscope (LOM). These metallographic analyses were performed in order to detect any surface corrosion or diffusion of NaK-78 elements into the Eurofer specimens. In addition, a specimen of the Eurofer base material (i.e. without NaK-78 exposure) was examined and Fig. 7 shows images of its microstructure and surface. Fig. 8 shows a SEM surface image of the NaK-78 exposed specimen at a magnification of 15,000×. The formation of plate-shaped particles is obvious at the Eurofer specimen surface after NaK-78 exposure in comparison with the base material surface. In addition, Fig. 9 shows another surface image with a higher magnification (30,000×) and three marked points for the EDX measurements which are given in the accompanying table. Point 1 was selected directly on a plate-shaped particle while points 2 and 3 were located on areas with a smoother surface. Due to the high O content, the increased Cr content compared to that of the base metal, and the observed Na content it can be assumed that these plate-shaped particles are most likely sodium chromite. The Fe signal is most likely a result from the X-ray generation volume. The formation of ternary oxides when steels are exposed to Na or NaK is expected, for instance, the formation of sodium chromite (NaCrO2 ) on a steel surface exposed to Na or NaK, with dissolved oxygen, are reported previously in the literature [3,14,16]. Ref. [16] gives an important equation to calculate the threshold oxygen concentration required for the formation of ternary oxides in NaK-78 for the two stainless steels 304 and 316. For example, to form sodium chromite on a stainless steel 304 exposed to NaK-78, an oxygen content of about 0.152 ppm in NaK-78 at 480 ◦ C (and 0.189 ppm at 500 ◦ C) is required. Examination of the cross section of the NaK-78 exposed specimen with the light optical microscopy showed no visible corrosion
Fig. 11. EDX line scans of the NaK-78 exposed specimen.
A. Abou-Sena et al. / Fusion Engineering and Design 103 (2016) 31–37
37
Table 1 Alloying elements of the Eurofer steel [15]. Element
Cr
C
Mn
P
V
W
Ta
Fe
wt%
8.82–8.96
0.11–0.12
0.38–0.49
0.004
0.18–0.20
1.07–1.15
0.13–0.15
Balance
attack, see Fig. 10. The EDX semi-quantitative analyses (line scans) for the relevant elements (O, Na, K, Cr, Mn, Fe and W) were performed on the cross section of the NaK-78 exposed specimen and are presented in Fig. 11. With the begin of the bulk material, the oxygen and sodium contents are going to zero while the alloying elements of the steel are showing the initial composition of the Eurofer steel. The main alloying elements of the Eurofer steel, as given in [15], are presented in Table 1. There is no oxygen enrichment into the steel or a depletion of an alloying element detectable in the cross section of the NaK-78 exposed specimen. The fact that sodium chromite was found at a 2 m-thick-surface layer and the corrosion attack did not go further into the bulk Eurofer hint on that only at the beginning of the NaK-78 exposure sufficient oxygen was available. Due to the fact that several alloying elements form ternary oxides at much lower oxygen levels in NaK than those required for the formation of the respective binary oxides, it is recommended to reduce the oxygen content in NaK-78 as low as possible when using it to fill the gaps in the IFMIF HFTM capsules. 6. Conclusions The present experiment successfully completed six months of exposing many Eurofer steel specimens to the sodium potassium eutectic alloy (NaK-78) at cyclic temperature between 480 ◦ C and 500 ◦ C in order to investigate the chemical compatibility of Eurofer with NaK-78. The mechanical tests (Charpy impact and tensile) of the Eurofer specimens revealed that Eurofer did not suffer any degradation in its mechanical properties due to the NaK-78 exposure. In addition, the results of the surface and microstructure analyses showed that the chemical interaction between the Eurofer steel and NaK-78 was minimal under the mentioned exposure conditions. With the results of this study, one can conclude that implementing NaK-78 with low oxygen content, under the continuous service conditions of high temperature (between 480 ◦ C and 500 ◦ C) for six months, is feasible for the IFMIF HFTM capsules.
Acknowledgments The framework project of this paper was funded by the German Ministry for Education and Research (known as Bundesministerium für Bildung und Forschung BMBF) under the grant number 03FUS0008, as a contribution to the “Broader Approach” activities. The responsibility for the contents of this publication is with the authors. References [1] F. Arbeiter, et al., Development and validation status of the IFMIF High Flux Test Module, Fusion Eng. Des. 86 (2011) 607–610. [2] W.D. Manly, Fundamentals of Liquid-Metal Corrosion, Report ORNL-2055, 1956. [3] T. Yutani, et al., Compatibility of Reduced Activation Ferritic Martensitic Steel Specimens With Liquid Na and NaK in Irradiation Rig of IFMIF. JAERI-Tech 2005-036, 2005. [4] R.C. Asher, Corrosion in Nuclear Reactors and How it is Combated. Corrosion Technology, 1964, October. [5] K. Natesan, et al., Assessment of alkali metal coolants for the ITER blanket, Fusion Eng. Des. 27 (1995) 457–466. [6] L.R. Kelman, et al., Resistance of Materials to Attack by Liquid Metals. ANL-4417, Argonne National Laboratory, 1950. [7] T.L. Hoffman, Performance of type 316 stainless steel in sodium-potassium eutectic alloy during 32,600 hours service, in: National Association of Corrosion Engineers High Temperature Symposium, Chicago, Illinois, March, 1971. [8] H.W. Savage, et al., SNAP-8 Corrosion Program Summary Report. Oak Ridge National Laboratory Report ORNL-3898, 1965. [9] E. Materna-Morris, et al., Structural Material Eurofer 97-2, Characterization of Rod and Plate Material: Tensile, Charpy, and Creep Properties, FZK Fusion Report 276, 2007. [10] J. Millat, W.A. Wakeham, The thermal conductivity of nitrogen and carbon monoxide in the limit of zero density, J. Phys. Chem. Ref. Data 18 (2) (1989). [11] O.J. Foust, Sodium-NaK Engineering Handbook Sodium Chemistry and Physical Properties, vol. I, Gordon and Breach Science Publishers, 1972. [12] N.E. Dowling, Mechanical Behavior of Materials, Prentice Hall, Upper Saddle River, 1999, Appendix B. [13] A. von der Weth, et al., Optimization of the EUROFER uniaxial diffusion weld, J. Nucl. Mater. 367–370 (2007) 1203–1207. [14] P.L.F. Rademakers, B.H. Kolster, Corrosion of various ferritic steels in an isothermal sodium loop system, J. Nucl. Mater. 97 (1981) 309–318. [15] R. Lindau, et al., Present development status of Eurofer and ODS-Eurofer for application in blanket concepts, Fusion Eng. Des. 75–79 (2005) 989–996. [16] J. Zhang, R. Kapernick, Oxygen chemistry in liquid sodium-potassium systems, Prog. Nucl. Energy 51 (2009) 614–623.
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Chemical compatibility of Eurofer steel with sodium-potassium NaK-78 eutectic alloy A. Abou-Sena a,∗ , F. Arbeiter a , S. Baumgaertner b , T. Boettcher a , A. Heinzel c , H. Piecha d , K. Zinn a a
Institute for Neutron Physics and Reactor Technology (INR), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany Institute for Applied Materials – Applied Materials Physics (IAM-AWP), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany c Institute for Pulsed Power and Microwave Technology (IHM), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany d Institute for Nuclear and Energy Technologies (IKET), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany b
h i g h l i g h t s • • • •
Eurofer SSTT specimens were immersed in static NaK-78 for 6 months at 480–500 ◦ C. The Eurofer steel did not suffer any degradation in its mechanical properties. The chemical interaction between Eurofer specimens and NaK-78 was minimal. Using NaK-78, with low oxygen content, is feasible for the IFMIF HFTM capsules.
a r t i c l e
i n f o
Article history: Received 29 June 2015 Received in revised form 4 December 2015 Accepted 5 December 2015 Available online 24 December 2015
a b s t r a c t In the high flux area of the International Fusion Materials Irradiation Facility (IFMIF) neutron source, the capsules of the High Flux Test Module (HFTM) contain SSTT Eurofer specimens for fusion relevant irradiation at temperature up to 550 ◦ C. Using the sodium potassium eutectic alloy NaK-78 to fill the gaps among the Eurofer specimens stacked inside the HFTM capsules was introduced in order to improve the thermal conduction among all specimens and have uniform and predictable temperature distribution. Therefore the objective of this study is to investigate the chemical compatibility between Eurofer steel and NaK-78 to evaluate the applicability of this concept. In the present experiment, the SSTT Eurofer specimens were immersed in static NaK-78 inside a capsule made of Eurofer and kept under IFMIF HFTMrelevant conditions including high temperature (cycling between 480 ◦ C and 500 ◦ C) and duration of six months. Following the experiment, mechanical tests (tensile and Charpy impact) of the Eurofer specimens were performed in addition to surface and microstructure analyses to detect any relevant corrosion or degradation. The mechanical tests revealed that the Eurofer specimens did not show any degradation in their mechanical properties. Also, the surface and microstructure analyses showed that the chemical interaction between the Eurofer steel and NaK-78 was minimal after six month of exposure at cyclic temperature between 480 and 500 ◦ C. © 2015 Elsevier B.V. All rights reserved.
1. Introduction The structural materials of the fusion reactor in-vessel components including the first wall and blanket should cope with the severe working conditions without any degradation in their mechanical properties or dimensional stability beyond the allowable design limits. Selecting these structural materials requires a
∗ Corresponding author. E-mail address: [email protected] (A. Abou-Sena). http://dx.doi.org/10.1016/j.fusengdes.2015.12.006 0920-3796/© 2015 Elsevier B.V. All rights reserved.
dedicated testing at fusion-relevant irradiation conditions. Therefore building the International Fusion Materials Irradiation Facility (IFMIF) has become an unavoidable step in the way to design and construct a fusion reactor. IFMIF is an accelerator-based neutron source that uses lithium-deuterium Li(d,xn) nuclear reactions to produce a neutron flux similar to that expected at the fusion reactor blanket. The task of IFMIF is to irradiate the fusion materials in irradiation conditions similar to those of the future fusion reactor DEMO in order to: (i) provide data for the engineering design of DEMO, (ii) contribute to the selection of candidate fusion materials, and (iii) validate the understanding of the irradiation effects on the
32
A. Abou-Sena et al. / Fusion Engineering and Design 103 (2016) 31–37
fusion functional materials. Next to the IFMIF neutron source in the high flux region, the irradiation of the structural material (e.g. Eurofer steel) specimens will be performed in the High Flux Test Module (HFTM) [1]. Eurofer steel belongs to the group of the Reduced Activation Ferritic Martensitic (RAFM) steels that have swelling resistance and low activation properties. Therefore Eurofer steel has been developed and optimized to serve as a structural material for the fusion reactor blanket. The Eurofer specimens are stacked inside the HFTM capsules and should have a specific uniform temperature therefore the heat transfer, mainly conduction, should be well understood and predictable. The idea of using the sodiumpotassium eutectic alloy NaK-78 (with 78 wt% K and 22 wt% Na) to fill the small gaps among the Eurofer specimens has been considered in order to: (i) improve the thermal conduction among all capsule parts and the heat transfer predictability, (ii) establish a uniform temperature distribution, and (iii) ensure that the thermocouples are in contact with NaK-78 or the specimens and not a gap to obtain accurate representative temperatures. However, the chemical compatibility of Eurofer with NaK-78 needs to be experimentally investigated to evaluate the applicability of this concept. For instance, the corrosion level and its influence on the mechanical properties and microstructure of the Eurofer specimens should be evaluated. 2. Literature review When the steel is exposed to a liquid metal (e.g. NaK or Na), the corrosion possibly occurs due to: (i) a mass transfer of non-metallic elements such as oxygen, carbon, nitrogen, and hydrogen, and (ii) a mass transfer of metallic elements as some of the steel constituents may dissolve into the liquid metal and react with its impurities; therefore the solubility of the steel constituents in the liquid metal plays a significant role in the corrosion process. The mass transfer of non-metallic elements can change the microstructure and hence the mechanical properties of the steel on the long run. Hence, it is preferred to have low concentrations of the non-metallic elements in the liquid metal. The chemical reactions between the steel and the surrounding liquid metal with its impurities, in most cases oxidation of the steel, cause the corrosion process. The oxygen content in the liquid metal is an important factor that can accelerate the corrosion of the structural materials; therefore it should be minimized. A previous study [2] reported that the corrosion of metals by a liquid metal is affected by some parameters including system temperature, oscillation of cyclic temperature, ratio of exposed solid surface area to volume of liquid metal, impurities in the liquid metal, flow velocity of the liquid metal, conditions of the container, and composition and microstructure of the materials involved. For instance, increasing the system temperature leads to higher solubility of the steel constituents into the liquid metal, higher diffusion rates and smaller viscosities and consequently results in a higher corrosion rate. Also, the cyclic temperature oscillation is a catalytic factor; for example the corrosion rate of the Cu-Bi system at 500 ± 5 ◦ C is several times higher than that at 500 ± 0.5 ◦ C as reported in the literature [2]. Another previous work [3] studied the compatibility between the RAFM steel and sodium, where the data of sodium except for oxygen solubility were used to evaluate the NaK compatibility. A mass transfer of a metallic element between the RAFM steel specimens and the Na (or NaK) was observed. It was concluded that the corrosion depth of RAFM specimens, due to the dissolution of metallic elements into the liquid metal and the oxidation by oxygen present in the liquid metal, is less than 1 m if the liquid metal is purified using the cold trap, and hence it is unlikely to affect the RAFM steel performance. It was stated in the literature [4] that the corrosive properties of sodium and NaK are similar and no distinction between the two liquid metals can be observed.
A fusion relevant study [5] presented an assessment of using alkali metals as a coolant for the ITER blanket. It was reported that the austenitic stainless steels did not produce iron or chromium oxides in the environments of liquid Na and NaK particularly at oxygen levels that correspond to a cold-trapped liquid. Hence, the corrosion rates of the steels are fairly low under the aforementioned conditions. In another study [6], the experimental results indicated that all stainless steels that were tested showed a good performance in static NaK capsules up to a temperature of 760 ◦ C. For the stainless steel SS310 in a very pure NaK, there was very small initial weight loss, which decreased quickly with the increase of exposure time at 760 ◦ C. The authors concluded that stainless steels are extremely resistant to a chemical attack by NaK at temperatures up to 760 ◦ C. In addition, experimental examinations were performed [7] to evaluate the performance of a stainless steel 316 loop for circulating the NaK-78 at temperature of 760 ◦ C for a long-term exposure of 32,600 h. It was concluded that the SS316 performed satisfactorily as a structural material for the loop and some changes in its mechanical properties and metallurgical effects occurred only when circulating non-isothermal NaK-78. However, these changes did not significantly affect the material properties in a way that can result in a material failure. At Oak Ridge National Laboratory, eleven experimental loops were built for the SNAP-8 corrosion program [8] to study the compatibility of NaK-78 with some structural materials including the nickel-based alloy Hastelloy N as well as the stainless steels 316 and 347. The loops were operated with a minimum NaK-78 temperature of 1100 ◦ F and for a running time of about 2000 h. The metallurgical analyses showed that the deposits consisted mainly of nickel, chromium and manganese in the low oxygen loops while the deposits were basically iron and chromium in the high oxygen loops. It was concluded that the oxygen level in NaK-78 is the most important factor that affects the corrosion of steels. Also, it was noted that corrosion of stainless steels 316 and 347 was very low compared with that of Hastelloy N.
3. Experimental setup The experimental setup, shown in Fig. 1, mainly consists of: (i) the capsule that contains the Eurofer specimens and NaK78, (ii) two heaters, seven thermocouples, an insulation block, structural supports and a metallic base, and (iii) a gas-tight cylinder to contain most of the setup’s components. The capsule has a shape of rectangular prism with inner and outer dimensions of 24 mm × 28 mm × 66 mm and 36 mm × 40 mm × 78 mm respectively. The capsule walls, cover, and base were welded together by electron beam welding because it is performed in vacuum; and hence it minimizes any contamination and produces consistent welds. The capsule has three tubes for filling, bleeding and pressurizing; and they penetrate from its cover and are used for the NaK filling process. Fig. 2 shows the capsule packed with the Eurofer specimens as well as its tubes and cover. The capsule, its tubes, and the Eurofer specimens were manufactured from the same Eurofer plate (Eurofer 97-2, batch 993391, thickness of 25 mm) and its properties are given elsewhere [9]. The capsule encloses 20 Eurofer specimens, similar to the IFMIF HFTM specimens, for post experiment mechanical tests. The Eurofer specimens were manufactured according to the Small Specimen Test Technology (SSTT) which was developed to investigate mechanical properties of nuclear materials taking into account the limited irradiation volumes in the test nuclear reactors and accelerator-based neutron sources. The tensile test specimen has a diameter of 2 mm and a gauge length of 7.6 mm. The Charpy impact specimen has dimensions of 27 mm × 4 mm × 3 mm with a v-notch depth of 1 mm, 60◦ notch angle, and 0.1 mm notch tip radius. In addition, the capsule contains two specimens, with a shape of rectangular prism
A. Abou-Sena et al. / Fusion Engineering and Design 103 (2016) 31–37
33
Fig. 1. Setup assembly: (a) picture during assembly, (b) sectional view of the setup.
heat loss by conduction. The capsule is enclosed by a gas-tight cylinder to control the atmosphere around the capsule and to serve as a secondary containment for the NaK-78. The cylinder consists of two CF blank flanges DN200, CF straight connector DN200, and two CF copper gaskets. This cylinder is evacuated and then filled with nitrogen (99.999 vol.%) at a pressure of around 1.2 bar absolute to have an inert shielding atmosphere around the capsule containing NaK78. Also, nitrogen has low thermal conductivity of 0.056 W/m K at 500 ◦ C [10] which helps to reduce the heat loss from the hot capsule to the surroundings. In addition, a pressure gauge is connected to the cylinder to observe its pressure. Temperatures of the capsule core and outer surfaces are measured and recorded continuously (24 h/day and 7 days/week) during the experimental campaign of 6 month. Four thermocouples are attached to the capsule outer surfaces in addition to three thermocouples immersed in NaK-78 inside the capsule. All thermocouples are type K with 1 mm diameter and can be used in temperature up to 900 ◦ C with deviations of ±2.5 ◦ C up to 333 ◦ C and ±0.75% above 333 ◦ C as reported by the manufacturer. The sheath material of the thermocouples is stainless steel AISI 321.
4. Experimental work Fig. 2. (a) Packing the capsule with specimens, (b) filling the capsule with NaK-78.
and dimensions of 12 mm × 10 mm × 4 mm, dedicated to the surface and microstructure analyses. The capsule is heated up to the required temperature range of 480–500 ◦ C by two electrical band heaters that have the following specifications: (i) internal diameter of 63.5 mm and width of 63.5 mm, (ii) voltage of 240 V and power of 1000 W, and (iii) stainless steel cover and maximum operating temperature of 760 ◦ C. A heat flux smoother, manufactured from a copper rod, is tightly placed between the capsule and the heaters in order to improve the heat transfer because any area of low heat conduction, between the capsule and the heaters, and thus chance of overheating should be prevented. An insulation block is placed underneath the capsule at the bottom of the setup to minimize the
4.1. Thermal management A thermal management of the experiment was needed to: (i) cycle the capsule temperature between 480 and 500 ◦ C, and (ii) protect the experimental setup from overheating during the experiment running time of six months. Beside the capsule’s heaters and thermocouples, a variable transformer and a temperature limiter were implemented in the setup. The temperature limiter receives a temperature reading from one of the capsule’s core thermocouples as input and compares it to the adjusted safety temperature limit (590 ◦ C) and then provides an output to the variable transformer which controls the input power to the capsule’s heaters. If the temperature passes the aforementioned safety temperature, the temperature limiter cuts the input power to the heaters to prevent
A. Abou-Sena et al. / Fusion Engineering and Design 103 (2016) 31–37
The difficulty of using NaK-78 lies in the volatile reactions it has with air and water. NaK-78 reacts with oxygen rapidly forming potassium oxide and superoxide which make a crust on its surface. In addition, NaK-78 reacts with water to release hydrogen and form potassium hydroxide and sodium hydroxide while the reaction heat may cause a fire. Therefore NaK-78 should be handled with safety measures and it is usually stored and studied in a controlled space such as a glove box filled with dry argon or nitrogen. NaK-78 has a melting temperature of −12.6 ◦ C hence it is liquid at room temperature. Its density is 867 kg/m3 at 20 ◦ C and decreases to 749 kg/m3 at 550 ◦ C [11]. In this study, NaK-78 does not flow but rather fills the capsule and occupies all spaces around the specimens. Filling the capsule with NaK-78 was performed inside a glove box filled with a mixture of argon and helium. The filling process was performed as follows. First, the capsule was moved into the glove box and then purged with a gas stream of argon and helium. Second, a filling mechanism (consists of metal syringe, piston, linear bearing and guides, see Fig. 2b) was used to fill the capsule through the filling tube. Third, the capsule was wrapped with a band heater and an insulation blanket to heat it up to 350 ◦ C. In addition, two thermocouples were attached to the capsule to monitor its temperature. Finally, the capsule was filled with 20 cm3 of NaK-78 sufficient for the specimens to be immersed completely. 4.3. Specimens retrieval After completing the experiment (i.e. keeping the capsule at temperature of 480–500 ◦ C for 6 month), the subsequent step was to retrieve the Eurofer specimens from the capsule. First, all tubes and cables attached to the gas-tight cylinder were disconnected, and then the cylinder was carefully transported to a glove box, where the capsule was taken out from the cylinder and the NaK-78 inside the capsule was drained through its tubes. Second, the capsule was opened by cutting off its cover and the specimens were
Yield strength (MPa)
4.2. NaK filling
600 500 400 300 NaK exposed -1st set NaK exposed -2nd set Base material -1st set Base material -2nd set Base material [9]
200 100 0 0
100
200
300
400
500
600
700
Temperature (°C) Fig. 4. The yield strength of the Eurofer specimens.
700
Ultimate strength (MPa)
an overheating. In addition, data acquisition software (LabView 2012) was used to: (i) record the capsule temperatures versus time and (ii) turn the heaters on/off to achieve the cyclic temperature using the capsule temperature readings as inputs for the software. The input power to the capsule’s heaters is adjusted manually using the variable transformer.
600 500 400 NaK exposed -1st set NaK exposed -2nd set Base material -1st set Base material -2nd set Base material [9]
300 200 100 0 0
100
200
300
400
500
600
700
Temperature (°C) Fig. 5. The ultimate tensile strength of the Eurofer specimens.
12 10
Impact energy (J)
34
8 6
NaK exposed -1st set NaK exposed -2nd set
4
Base material -1st set Base material -2nd set
2
Base material [13] 0 -120 -100 -80
-60
-40
-20
0
20
40
Temperature (°C) Fig. 6. The impact energy for the Eurofer specimens.
taken out to be cleaned from the NaK-78 adherent to their surfaces by immersing in ethanol. Fig. 3 shows the procedures performed to retrieve the specimens inside the glove box. After the specimens retrieval, it was clear that the specimens were totally wetted by NaK-78, see Fig. 3b. Finally, the specimens were moved from the glove box for further cleaning using ethanol and an ultrasonic cleaner. 5. Results and discussion
Fig. 3. (a) Opening the capsule, (b) NaK-wetted specimens, and (c) immersing in ethanol.
The post experiment tests and analyses of the Eurofer specimens are necessary to investigate any relevant chemical interaction between Eurofer and NaK-78 such as corrosion and its impact on
A. Abou-Sena et al. / Fusion Engineering and Design 103 (2016) 31–37
35
Fig. 7. LOM microstructure image (left) and SEM surface image (right) of base material.
Fig. 8. SEM micrograph of the surface of the NaK-78 exposed specimen.
the Eurofer’s mechanical properties. In the following sections, the results of the mechanical tests (tensile and Charpy impact) as well as the surface and microstructure analyses of the Eurofer specimens are presented. 5.1. Mechanical tests The tensile and Charpy impact tests of Eurofer specimens were performed to investigate any degradation in its mechanical properties due to NaK-78 exposure under a cyclic temperature of
480–500 ◦ C for 6 month. The tensile tests were implemented to measure the ultimate tensile strength and yield strength of the Eurofer specimens at temperatures (20, 300, 400, 500, 600 ◦ C) relevant to the fusion blanket. The Eurofer specimens were divided into two groups: (i) base material specimens which were not exposed to NaK-78 and (ii) NaK-78 exposed specimens. Each group has two specimens for each testing temperature leading to two sets of results. The values of the yield strength (YS) and the ultimate tensile strength (UTS) are presented in Figs. 4 and 5 respectively for these sets and also for Eurofer base material of a previous study [9]. The values of the UTS and YS for the NaK-78 exposed specimens are comparable with those of the base material specimens; however, they are higher by 1–7%. These deviations in the UTS and YS values are within the range one would expect from the variation of the tensile tests [12]. From the obtained results, one can conclude that the implemented NaK-78 exposure has no effect on the UTS and YS of the Eurofer specimens. Two groups (namely base material and NaK exposed) of the Charpy impact specimens were tested at temperatures of −100, −80, −60, −40, 20 ◦ C using two specimens per temperature. As shown in Fig. 6, the values of the impact energy for the NaK-78 exposed specimens are comparable with those of the present and previous [13] base material specimens indicating that the NaK78 exposed specimens did not suffer any reduction in the upper shelf energy (USE) or shift in the ductile-to-brittle transition temperature (DBTT) within the given temperature range. The DBTT is important for the fusion blanket structural components because of the potential for embrittlement due to the neutron irradiation. Based on the results of the tensile and Charpy impact tests, one
Fig. 9. SEM surface image of the NaK-78 exposed specimen and EDX measurements for the three marked points in the accompanying table.
36
A. Abou-Sena et al. / Fusion Engineering and Design 103 (2016) 31–37
Fig. 10. Optical micrograph of cross section of the NaK-78 exposed specimen (etched).
may conclude that the existence of NaK-78 in contact with the Eurofer specimens does not affect its mechanical properties. This conclusion is in agreement with the results of a previous study [3] about the compatibility of reduced activation ferritic martensitic steel specimens with liquid Na and NaK. 5.2. Surface and microstructure analyses Surfaces of the NaK-78 exposed specimens (12 mm × 10 mm × 4 mm), dedicated to surface and microstructure analyses, were examined using a scanning electron microscope (SEM) that has an integrated energy dispersive X-ray (EDX) analysis system. Afterwards, the specimens were cut with a diamond disk saw perpendicular to the exposed surface, then embedded, ground and polished for metallographic examinations. The cross sections were analyzed using a SEM equipped with an EDX system and a light optical microscope (LOM). These metallographic analyses were performed in order to detect any surface corrosion or diffusion of NaK-78 elements into the Eurofer specimens. In addition, a specimen of the Eurofer base material (i.e. without NaK-78 exposure) was examined and Fig. 7 shows images of its microstructure and surface. Fig. 8 shows a SEM surface image of the NaK-78 exposed specimen at a magnification of 15,000×. The formation of plate-shaped particles is obvious at the Eurofer specimen surface after NaK-78 exposure in comparison with the base material surface. In addition, Fig. 9 shows another surface image with a higher magnification (30,000×) and three marked points for the EDX measurements which are given in the accompanying table. Point 1 was selected directly on a plate-shaped particle while points 2 and 3 were located on areas with a smoother surface. Due to the high O content, the increased Cr content compared to that of the base metal, and the observed Na content it can be assumed that these plate-shaped particles are most likely sodium chromite. The Fe signal is most likely a result from the X-ray generation volume. The formation of ternary oxides when steels are exposed to Na or NaK is expected, for instance, the formation of sodium chromite (NaCrO2 ) on a steel surface exposed to Na or NaK, with dissolved oxygen, are reported previously in the literature [3,14,16]. Ref. [16] gives an important equation to calculate the threshold oxygen concentration required for the formation of ternary oxides in NaK-78 for the two stainless steels 304 and 316. For example, to form sodium chromite on a stainless steel 304 exposed to NaK-78, an oxygen content of about 0.152 ppm in NaK-78 at 480 ◦ C (and 0.189 ppm at 500 ◦ C) is required. Examination of the cross section of the NaK-78 exposed specimen with the light optical microscopy showed no visible corrosion
Fig. 11. EDX line scans of the NaK-78 exposed specimen.
A. Abou-Sena et al. / Fusion Engineering and Design 103 (2016) 31–37
37
Table 1 Alloying elements of the Eurofer steel [15]. Element
Cr
C
Mn
P
V
W
Ta
Fe
wt%
8.82–8.96
0.11–0.12
0.38–0.49
0.004
0.18–0.20
1.07–1.15
0.13–0.15
Balance
attack, see Fig. 10. The EDX semi-quantitative analyses (line scans) for the relevant elements (O, Na, K, Cr, Mn, Fe and W) were performed on the cross section of the NaK-78 exposed specimen and are presented in Fig. 11. With the begin of the bulk material, the oxygen and sodium contents are going to zero while the alloying elements of the steel are showing the initial composition of the Eurofer steel. The main alloying elements of the Eurofer steel, as given in [15], are presented in Table 1. There is no oxygen enrichment into the steel or a depletion of an alloying element detectable in the cross section of the NaK-78 exposed specimen. The fact that sodium chromite was found at a 2 m-thick-surface layer and the corrosion attack did not go further into the bulk Eurofer hint on that only at the beginning of the NaK-78 exposure sufficient oxygen was available. Due to the fact that several alloying elements form ternary oxides at much lower oxygen levels in NaK than those required for the formation of the respective binary oxides, it is recommended to reduce the oxygen content in NaK-78 as low as possible when using it to fill the gaps in the IFMIF HFTM capsules. 6. Conclusions The present experiment successfully completed six months of exposing many Eurofer steel specimens to the sodium potassium eutectic alloy (NaK-78) at cyclic temperature between 480 ◦ C and 500 ◦ C in order to investigate the chemical compatibility of Eurofer with NaK-78. The mechanical tests (Charpy impact and tensile) of the Eurofer specimens revealed that Eurofer did not suffer any degradation in its mechanical properties due to the NaK-78 exposure. In addition, the results of the surface and microstructure analyses showed that the chemical interaction between the Eurofer steel and NaK-78 was minimal under the mentioned exposure conditions. With the results of this study, one can conclude that implementing NaK-78 with low oxygen content, under the continuous service conditions of high temperature (between 480 ◦ C and 500 ◦ C) for six months, is feasible for the IFMIF HFTM capsules.
Acknowledgments The framework project of this paper was funded by the German Ministry for Education and Research (known as Bundesministerium für Bildung und Forschung BMBF) under the grant number 03FUS0008, as a contribution to the “Broader Approach” activities. The responsibility for the contents of this publication is with the authors. References [1] F. Arbeiter, et al., Development and validation status of the IFMIF High Flux Test Module, Fusion Eng. Des. 86 (2011) 607–610. [2] W.D. Manly, Fundamentals of Liquid-Metal Corrosion, Report ORNL-2055, 1956. [3] T. Yutani, et al., Compatibility of Reduced Activation Ferritic Martensitic Steel Specimens With Liquid Na and NaK in Irradiation Rig of IFMIF. JAERI-Tech 2005-036, 2005. [4] R.C. Asher, Corrosion in Nuclear Reactors and How it is Combated. Corrosion Technology, 1964, October. [5] K. Natesan, et al., Assessment of alkali metal coolants for the ITER blanket, Fusion Eng. Des. 27 (1995) 457–466. [6] L.R. Kelman, et al., Resistance of Materials to Attack by Liquid Metals. ANL-4417, Argonne National Laboratory, 1950. [7] T.L. Hoffman, Performance of type 316 stainless steel in sodium-potassium eutectic alloy during 32,600 hours service, in: National Association of Corrosion Engineers High Temperature Symposium, Chicago, Illinois, March, 1971. [8] H.W. Savage, et al., SNAP-8 Corrosion Program Summary Report. Oak Ridge National Laboratory Report ORNL-3898, 1965. [9] E. Materna-Morris, et al., Structural Material Eurofer 97-2, Characterization of Rod and Plate Material: Tensile, Charpy, and Creep Properties, FZK Fusion Report 276, 2007. [10] J. Millat, W.A. Wakeham, The thermal conductivity of nitrogen and carbon monoxide in the limit of zero density, J. Phys. Chem. Ref. Data 18 (2) (1989). [11] O.J. Foust, Sodium-NaK Engineering Handbook Sodium Chemistry and Physical Properties, vol. I, Gordon and Breach Science Publishers, 1972. [12] N.E. Dowling, Mechanical Behavior of Materials, Prentice Hall, Upper Saddle River, 1999, Appendix B. [13] A. von der Weth, et al., Optimization of the EUROFER uniaxial diffusion weld, J. Nucl. Mater. 367–370 (2007) 1203–1207. [14] P.L.F. Rademakers, B.H. Kolster, Corrosion of various ferritic steels in an isothermal sodium loop system, J. Nucl. Mater. 97 (1981) 309–318. [15] R. Lindau, et al., Present development status of Eurofer and ODS-Eurofer for application in blanket concepts, Fusion Eng. Des. 75–79 (2005) 989–996. [16] J. Zhang, R. Kapernick, Oxygen chemistry in liquid sodium-potassium systems, Prog. Nucl. Energy 51 (2009) 614–623.