Silicon Carbide and Silicon Carbide Composites for Fusion Reactor Application

Tatsuya Hinoki1,+, Yutai Katoh2, Lance L. Snead2, Hun-Chae Jung1, Kazumi Ozawa2,Hirokazu Katsui3, Zhi-Hong Zhong1, Sosuke Kondo2, Yi-Hyun Park1, Chunghao Shih2,Chad M. Parish2, Roberta A. Meisner2 and Akira Hasegawa4

1Institute of Advanced Energy, Kyoto University, Uji 611-0011, Japan2Materials Science & Technology Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6138, USA3Institute for Material Research, Tohoku University, Sendai 980-8577, Japan4Department of Quantum Science and Energy Engineering, Tohoku University, Sendai 980-8579, Japan

This paper reviews recent achievements as to “nuclear-grade” SiC composites in particular for materials-system integration. SiC compositecomponent development are reviewed including VHTR control rod scale model and compact intermediate heat exchanger scale mode by currentjoining and assembly techniques. Joining methods for SiC to metal and results of characterization of joint shear strength by the torsion tests usingsmall specimens were also reviewed. The recent results of neutron irradiation experiments were also reviewed including detailed analysis ofmechanical properties, irradiation creep and preliminary results on tritium behavior in SiC. [doi:10.2320/matertrans.MG201206]

(Received September 10, 2012; Accepted December 11, 2012; Published February 16, 2013)

Keywords: SiC composites, fusion reactor, assembly technique, joining, irradiation effect, irradiation creep, tritum

1. Introduction

A silicon carbide (SiC) composite is considered as apromising material for the fusion reactor system due to itsenhanced toughness by fiber reinforcement and intrinsicproperties of SiC such as low activation, chemical andenvironmental inertness, irradiation stability, and very hightemperature performance. High-purity silicon carbide com-posites consisting of the high purity and highly crystallinefibers and matrices have proven the excellent mechanicalperformance following neutron irradiation and recognized as“nuclear-grade” SiC composites.

The primary application for SiC composites has been theintegrated first-wall and blanket structure,1) and materialdevelopment has focused on stable mechanical performanceand high thermal conductivity under high temperatureirradiation environment. A nearer-term fusion applicationfor SiC composites is the flow channel insert of the dualcoolant system,2) and lower thermal conductivity and lowerelectrical conductivity are contrarily required. The otherpotential application is a divertor structure in form ofcombination with tungsten.

This material system has undergone a feasibility checkwithin the international fusion materials program over thepast two decades, progressing from a laboratory curiosityto a material on the verge of engineering application.Manufacturing and assembly techniques including joininghave been developed using nuclear-grade SiC compositesin a decade. Historic and substantial development effortshave spanned fundamental modeling to materials fabricationto irradiation performance and properties databasing.3)

This paper reviews the current status of international anddomestic activities in particular for materials-system inte-gration for fusion DEMO blanket. This paper focuses onthe issues as a system using SiC materials includingassembly and joining techniques, compatibility and irradi-ation effect.

2. Material Development

Nuclear-grade SiC composites can be fabricated bychemical vapor infiltration (CVI) process4) or Nano-powderInfiltration and Transient Eutectic-phase (NITE) process5)

using high purity SiC fibers such as Hi-Nicalon-S or TyrannoSA. Both processes can form highly crystalline SiC structurewith limited impurity, which is a key requirement for materialapplication in neutron irradiation environment.6) Currentinterests in material development are development ofmanufacturing and assembly techniques including joiningusing nuclear-grade SiC composites rather than optimizationof the materials with small laboratory scale for materials-system integration.

2.1 Joining and assembly techniques for SiC compositesJoining and assembly techniques have been demonstrated

using nuclear-grade SiC composites in US and Japanrecently. A SiC composite control rod demonstrator withmechanical joints was successfully produced for the VeryHigh-Temperature Reactor (VHTR) application in US.1) Twocontrol rod sheath half-segments, one coupling tube/sheathconnector, two shear pins, one reactivity control retention/support plate, and four retention clips were produced usingHi-Nicalon-S braided preforms by the CVI technique. Allcomponents following machining were assembled andmechanically joined.

A hermetic joining technique utilizing the SiC slurry sheet,which consists of SiC powder and sintering additives, wasalso developed to assemble SiC composite compact inter-mediate heat exchanger (IHX).7) The slurry sheets were cutto fit to joint region. The SiC composites fabricated by theNITE process were machined to plates, frames and pins.These parts were assembled for IHX using the slurry sheetsas joint inserts. The plates were grooved slightly to keepalignment of the pin position and to prevent slurry from beingforced out to the channel. The assembled parts includingthe slurry sheets were then hot-pressed. Uniform joint layerthickness is important to prevent stress concentration in+Corresponding author, E-mail: hinoki@iae.kyoto-u.ac.jp

Materials Transactions, Vol. 54, No. 4 (2013) pp. 472 to 476Special Issue on Materials-System Integration for Fusion DEMO Blanket©2013 The Japan Institute of Metals OVERVIEW

particular for ceramic joint. The IHX scale model with 10 cmsquare was fabricated successfully utilizing the slurry sheetwith uniform thickness as shown in Fig. 1.

2.2 Joining of SiC and metalTungsten is a candidate for plasma facing materials. It is a

key technique to develop robust joining technique for SiCand tungsten to utilize SiC materials for first-wall, blanketstructure blanket and divertor. The coefficient of thermalexpansion (CTE) for SiC and tungsten are 4.7 and4.5 © 10¹6/K, respectively, at ambient temperature. It istherefore believed that a high temperature joining process canbe applied without significant residual stress induced at theW/SiC interface during cooling. Direct diffusion bondingwas successfully applied for joining SiC with tungsten.8)

Figure 2 shows typical SEM and EDS images for the jointinterface. According to the EDS analysis, WC and W5Si3

were formed at the interface. The joining strength for thediffusion bonding was around 100MPa by a shear test usingdouble-notched specimen.

Joining techniques for SiC and ferritic steel were alsodeveloped to extend design flexibility,9) since the ferritic steelis a primary candidate fusion structural material. The CTEof F82H, which is the ferritic steel developed for fusionapplication by Japan Atomic Energy Agency, Japan, is14 © 10¹6/K and much different from that of SiC. Interlayermetals including Ni and Cu were used to mitigate CTEmismatch. Tungsten was also used between SiC and theinterlayer metals to prevent extra reaction. Figure 3 is anexample of joining of SiC and F82H using tungsten andcopper as interlayer metals.

2.3 Development of torsion shear test methodLarge stress concentration at edge or loading point was an

important issue for conventional characterization techniquesfor joint shear strength. Specimen size must be sufficientlysmall for neutron irradiation experiments. Of many shearevaluation methods, the torsion shear test method wasdeveloped using small specimens as one of the most viablemethods.10) The torsion test has advantages in terms ofapplied pure shear stress and small specimen size. The shearstrength obtained by the torsion test is the most reliable shearstrength considering the large stress concentration for theother conventional shear tests.11) The specimen size and shapewere determined by the FEM analysis and experiments interms of pure shear stress and applicable to large shearstrength over 200MPa. The determined specimen geometry isshown in Fig. 4. The characterized shear strength of SiC jointby hot pressing using SiC slurry was approximately 250MPa.

3. Irradiation Effect

The material system including joining is used underneutron irradiation environment. Irradiation effect on the

Fig. 1 SiC composite IHX scale model and schematic joining image ofeach part.

Fig. 2 SEM (left figure) and EDS analysis elementary mapping (right figures) for W, C, Si and O images for SiC and tungsten interface.

Silicon Carbide and Silicon Carbide Composites for Fusion Reactor Application 473

material system isn’t determined by only irradiation effect oneach material but also the interaction between each material.The interaction includes applied stress at interface attributedfrom CTE mismatch and different irradiation-inducedswelling and chemical interaction under neutron irradiation.This paper reviews the recent results of the irradiation effectson the SiC composites, the irradiation-induced creep of SiCmaterials and the compatibility of SiC with the solid breedingmaterials.

3.1 Irradiation effect on mechanical properties ofnuclear-grade SiC composites

It has been reported that nuclear grade SiC composites arevery stable to neutron irradiation.12) It is attributed fromlimited change of mechanical properties and dimension ofSiC constituents by neutron irradiation. Figure 5 shows theirradiation effect on the tensile stress-strain behavior ofNITE-SiC composites irradiated at 800°C up to 5.9 dpa and1300°C up to 5.8 dpa in High Flux Isotope Reactor (HFIR).No significant change in ultimate tensile strength andproportional limit stress was identified. However it was

found that some parameters including loop width and damageparameter changed and the axial residual stress decreased byirradiation according to hysteresis loop analysis.13) It wasassumed that the residual stress of the composites wasaffected by differential irradiation induced-swelling ofconstituents14) and irradiation induced creep, as well as theCTE mismatch as clearly shown for the non-irradiated case.

3.2 Irradiation creep of monolithic SiC ceramicsThe bend stress relaxation technique was applied for an

irradiation creep measurement of various type of SiC.15) Aconstant bend strain was applied to thin strip samples duringneutron irradiation to various fluences at various temper-atures. It was reported that irradiation creep strain at up toaround 0.1 dpa exhibited only a limited dependence onirradiation temperature as shown in Fig. 6, in which stressretention ratio, m, is described as ratio of applied bend stressfollowing irradiation and that before irradiation required for

Fig. 5 Neutron irradiation effects on tensile stress­strain curve of NITE-SiC composites irradiated at 800°C up to 5.9 dpa and 1300°C up to5.8 dpa.

Fig. 3 Joining of SiC and F82H using tungsten and copper as interlayers.

Fig. 4 Torsion shear test geometry.Fig. 6 BSR Irradiation Creep of polycrystalline CVD SiC provided by

Rohm and Haas and Coorstek and monocrystalline SiC provided by Creein which m is described as ratio of applied bend stress followingirradiation and that before irradiation required for same strain.

T. Hinoki et al.474

same strain. However the creep strain depended on materialproperties at around 1 dpa or beyond significantly. The SiCfabricated by liquid phase sintering (LPS), which is referencematerial for NITE-SiC composite matrix, had remainedsintering additive at grain boundaries and triple junctions16)

and showed significant large amount of creep compared tothe high purity CVD SiC. The reduced sintering additiveswere used for LPS 2 SiC, which was half of the sinteringadditives used for LPS 1. Effect of the amount of sinteringadditive on creep wasn’t observed.

3.3 Compatibility of SiC with the solid breedingmaterials under neutron irradiation

It is important to understand tritium behavior in SiC underirradiation of neutrons and energetic particles includingtritons created by nuclear reactions of Li(n, ¡)T in breedingmaterials to establish the material system. Compatibility ofSiC with ternary lithium ceramics under neutron irradiationwas characterized and reported.17) The ternary lithiumceramics used were LiAlO2, Li1.9TiO3, Li2ZrO3 and Li4SiO4.SiC disk samples were irradiated in contact with the ternarylithium ceramics in HFIR at 800°C up to 5.9 dpa. Theirradiated and non-irradiated ternary lithium ceramics andSiC surfaces at the interface were shown in Fig. 7. Althoughsignificant chemical reactions were observed for SiC/Li2ZrO3 and SiC/Li4SiO4 interfaces, chemical reactions werelimited for LiAlO2 and Li1.9TiO3 interfaces by irradiation interms of the surface roughness and the chemical reactionproducts. The effect of lithium burn up due to the (n, ¡)nuclear reaction was also examined. More reaction productswere observed on the SiC surface in contact with LiAlO2 withthe lower lithium to metal ratio (Li/Al). It was consideredthat the formation of LiAl5O8 phase due to lithium loss coulddeteriorate the compatibility of SiC and LiAlO2.

4. Summary

SiC material system has undergone a feasibility checkwithin the international fusion materials program over the

past two decades and “nuclear-grade” SiC composites wereestablished. Significant progresses for joining and assemblytechniques have been confirmed. Precise characterization ofmechanical properties including irradiation effects andirradiation creep were carried out. A part of mechanisms ofstable mechanical properties following neutron irradiationwas discussed. Preliminary results on tritium behavior in SiCunder neutron irradiation were reported.

The current interests in material evaluation are practicalenvironment related issues including lifetime, anisotropy,physical properties and irradiation effects including trans-mutation effects rather than fundamental properties, withadvancements yielding proven performance near prototypicalfusion power reactor conditions.

Acknowledgments

Part of this work was performed as a part of US/JapanTITAN collaboration on fusion blanket and materialsresearch and development.

REFERENCES

1) L. L. Snead, T. Nozawa, M. Ferraris, Y. Katoh, R. Shinavski and M.Sawan: J. Nucl. Mater. 417 (2011) 330­339.

2) F. Najmabadi and The ARIES Team: Fusion Eng. Des. 65 (2003) 143­164.

3) T. Nozawa, Y. Choi, T. Hinoki, H. Kishimoto, A. Kohyama and H.Tanigawa: IOP Conf. Series-Mater. Sci. Eng. 18 (2011) 162011.

4) T. Hinoki, L. L. Snead, T. Taguchi, N. Igawa, W. Yang, T. Nozawa, Y.Katoh and A. Kohyama: Ceram. Trans. 144 (2002) 55­67.

5) Y. Katoh, A. Kohyama, T. Nozawa and M. Sato: J. Nucl. Mater. 329­333 (2004) 587­591.

6) T. Hinoki, L. L. Snead, Y. Katoh, A. Hasegawa, T. Nozawa and A.Kohyama: J. Nucl. Mater. 307­311 (2002) 1157­1162.

7) H. C. Jung, Y. H. Park, J. S. Park, T. Hinoki and A. Kohyama: J. Nucl.Mater. 386 (2009) 847­851.

8) T. Hinoki and A. Kohyama: Ann. Chim. 30 (2005) 659­671.9) Z. H. Zhong, T. Hinoki and A. Kohyama: J. Nucl. Mater. 417 (2011)

395­399.10) H. C. Jung, T. Hinoki, Y. Katoh and A. Kohyama: J. Nucl. Mater. 417

(2011) 383­386.

Fig. 7 Irradiated (800°C, 5.9 dpa) and non-irradiated ternary lithium ceramics and SiC surfaces at the interface.

Silicon Carbide and Silicon Carbide Composites for Fusion Reactor Application 475

11) M. Ferraris, M. Salvo, S. Rizzo, V. Casalegno, S. H. Han, A. Ventrella,T. Hinoki and Y. Katoh: Int. J. App. Ceram. Tech. 9 (2012) 786­794.

12) T. Hinoki, L. L. Snead, Y. Katoh, A. Hasegawa, T. Nozawa and A.Kohyama: J. Nucl. Mater. 307­311 (2002) 1157­1162.

13) E. Vagaggini, J. M. Domergue and A. G. Evans: J. Am. Ceram. Soc. 78(1995) 2709­2720.

14) T. Koyanagi, S. Kondo and T. Hinoki: IOP Conf. Series-Mater. Sci.

Eng. 18 (2011) 162008.15) Y. Katoh, L. L. Snead, T. Hinoki, S. Kondo and A. Kohyama: J. Nucl.

Mater. 367 (2007) 758­763.16) T. Koyanagi, S. Kondo and T. Hinoki: J. Nucl. Mater. 417 (2011) 435­

439.17) H. Katsui, A. Hasegawa, Y. Katoh, Y. Hatano, T. Tanaka, S. Nogami, T.

Hinoki and T. Shikama: Fusion Sci. Tech. 60 (2011) 288­291.

T. Hinoki et al.476