Recent progress in R&D on tungsten alloys for divertor structural and plasma facing materials

TL;DR: In this paper, the fracture behavior is improved by using tungsten laminated materials and wire reinforced materials, which can achieve self-passivation, which is essential in case of loss-of-coolant accidents for plasma facing materials.

About: This article is published in Journal of Nuclear Materials.The article was published on 2013-11-01 and is currently open access. It has received 267 citations till now. The article focuses on the topics: Tungsten & Structural material.

Summary

1. Introduction

• W-based materials feature several advantageous properties for fusion application, e.g., high melting point, good thermal conductivity, high creep resistance, good high-temperature strength and low vapor pressure.
• Several approaches to increase ductility and fracture toughness and to decrease the DBTT will be discussed in Section 2.
• Two approaches presented in Sections 2.4 and 2.5. are very promising for an improved fracture behavior of W-based components.
• Section 4 outlines the formation and compares the properties of two different joining processes between ODS steels and W. Recent campaigns on neutron irradiation effects on several W-based materials are described in Section 5.

2.1. Tungsten alloying

• Even for small additions of Re, the solid solution softening process [10] is active and leads to ductilization and toughening [11].
• This is not the case for W-Ta alloys.
• Since Nb and Mo transmute to very longlived radioactive isotopes, they cannot be used for fusion applications; this leaves Ta, V, and Ti as possible candidates.
• Methods to stabilize advantageous microstructures will be discussed in the next subsections.

2.2. Oxide dispersion strengthened / stabilized materials

• W-based materials for structural applications are developed based on the idea that alloys and especially nano-grained materials should be more ductile than pure W and standard grain-sized materials [22].
• Furthermore, nano-grained materials and dispersion strengthened materials should be more radiation resistant than pure W and standard grain-sized materials, as the numerous grain boundaries and interfaces between matrix and particles are expected to act as sinks for the irradiation-induced defects [23].
• Three-point bending (3PB) tests showed that the material is brittle at 298 K and ductile at 673 K and above.

2.3. Recent progress in nanostructured W-TiC materials

• Radiation and recrystallization embrittlement are the major concerns about W-materials to be used as plasma facing materials.
• A promising solution to these problems is to develop nanostructures that can offer much-improved mechanical properties in the recrystallized state.
• TiC was selected because of its high melting point (~3400 K) and a self-adjustment capability of the lattice constant by forming a solid solution with W and non-stoichiometric TiCx.
• The method is based on significant activation of GB sliding by superplastic deformation that is attributed to the UFG structures and thus defined as “superplasticity-based microstructural modification (SPMM)” [4,16,35].
• Thermal shock loadings under ITER/ELM conditions do not cause any cracking or surface roughening in TFGR W-1.1TiC with a low O content [34], although the thermal conductivity of TFGR W-1.1TiC was almost the same as that of UFG W-0.5TiC [34].

2.4. W laminates made of W foils

• It is well known that body-centered cubic (bcc) metals can be ductilized by prior cold work [39].
• Furthermore, it was shown for ODS W that recrystallization and fracture toughness increase with increasing degree of deformation [40].
• Each layer had a thickness of 0.1 mm and the W foil was oriented in rolling direction (0°).
• This W laminate was characterized by Charpy impact tests; more details on this testing procedure can be found elsewhere [42–44].
• Because the microstructure of as-received foils is completely different from that of the annealed state, there might be different reasons for the ductility of W foils.

2.5. Tungsten-wire-reinforced tungsten composites with pseudo-toughness

• You et al. proposed a novel toughening concept based on W-wire reinforcement and demonstrated the potential benefit of corresponding W composites [45].
• Since this mechanism is a purely mechanics-based process not needing plasticity, it is supposed to function even under neutron irradiation conditions.
• Upon incremental extension of the primary crack, the strong wires collectively bridge the main crack flanks, suppressing dynamic fracture.
• Here, the fracture mechanical properties of the interface are the key factor.
• The wire-reinforced composites offer a positive prospect for making a breakthrough towards a tough material [45].

3. Self-passivating tungsten alloys

• The use of W represents an important safety concern in case of a loss-of-coolant accident with simultaneous air ingress into the reactor vessel.
• Temperatures up to 1473 K can be achieved in the in-vessel components within a few weeks due to the decay heat [51], which would lead to fast oxidation with the release of volatile, activated W oxides [52].
• During normal operation, the surface will consist of W, owing to preferential sputtering of alloying elements.
• The microstructure consists mainly of (W,Cr)5Si3 mixed crystal, pure W grains and isolated SiO2 grains.
• Based on the results of [56], introducing W-Cr-Ti thin-film alloys with even lower oxidation rates, WCr12Ti2.5 bulk alloys have been manufactured by MA + HIP, obtaining again 100% density [57].

4. Joining W plate with ODS steel

• ODS steels show excellent elevated-temperature strength, corrosion resistance and radiation resistance [59–61].
• Diffusion bonding processes were chosen for this work on pure W and ODS steel, because it is an attractive technique to obtain W-layered ODS-steel components with excellent interfacial strength.
• The melting point of the insert material was 1423 K; B and Si act as melting point depressants.
• The microstructural examination revealed that both LPDB and SSDB were successful; no voids were visible in the bonding interfaces.
• This suggests that the LPDB joint needed higher fracture energy under torsional shear stress than the SSDB joint, and this fact corresponds well to the evaluated shear strength.

5. Neutron irradiation of tungsten

• Neutron irradiation data on microstructural development in W and W-Re alloys were obtained from the JOYO reactor, the Japan Materials Testing Reactor (JMTR) and the High Flux Isotope Reactor (HFIR) irradiation experiments at 773 K to 1073 K.
• For low damage levels, void and dislocation loop formation occurs.
• The irradiation hardening increased with the formation of the precipitates [64–67].
• Furthermore, neutron irradiation behavior of La-doped W, K-doped W and UFG W-TiC were examined in JOYO and JMTR using disk-type specimens (804-1073 K, 0.15–0.47 dpa) for transmission electron microscopy.
• The effects of the La and K doping on the electrical resistivity were small compared with the UFG-W alloys.

6.1. Modeling of heat load and helium bombardment damage

• The presence of defects inside the material (e.g.
• He bubbles [70], voids, self-interstitial atoms) is responsible for the discrepancy between the actual and theoretical strengths of the material and the degradation of these properties.
• A multiphysics computational model has been developed within a finite-element framework to study the synergistic effects of transient plasma events and the existence of He bubbles on the thermomechanical damage of W.
• The effects of different heat fluxes were simulated, and the results showed surface damage by the formation of primary and secondary intergranular microcracks.
• The developed model provided satisfactory results that are in general agreement with the experimental observations.

6.2. Ab initio study of transition metals on tungsten grain boundary cohesion

• It was understood early that brittle fracture in W mainly occurs along GBs.
• In subsection 2.1. the outstanding properties of W-Re alloys were already discussed.
• The ground state structure from MD simulations was subsequently optimized, using an ab initio method, to get the final structure (Fig. 9a).
• The plotted energy represents the Griffith ideal cleavage energy with the fracture plane denoted by a dashed line in Fig. 9a.
• To assess the net effect for each element, the formation energies at different GB sites were compared.

7. Summary

• Smart W materials have to be developed that can withstand the harsh environment in the divertor region.
• For improving mechanical properties, alloying with Re has proven by experiments and calculations to be advantageous; however, the application is inhibited due to the element’s low availability.
• Other solid solution elements that dramatically improve the properties are not in sight.
• The next step is to use these improved materials, featuring a self-passivation layer on top for withstanding accidental air ingress, in a sophisticated configuration.
• This was shown for Wreinforced materials with W wires in crack arrester alignment and W laminate material that extends the excellent ductile properties of thin foils to bulk materials.

Figures

Fig. 5: Scanning electron micrograph made during an in situ 3PB fracture experiment. The detail view shows the propagating crack bypassing the wire.

Fig. 3: Temperature-dependent tensile stress-strain curves of sintered and hot-forged W2Y2O3.

Fig. 4: Results of Charpy impact tests for W laminates compared to W plate material in asreceived and annealed condition. The inset shows a Charpy test sample made of 20 layers of W foil (0.1 mm, 0°) and 19 layers of AgCu brazing filler (0.1 mm).

Fig. 7: Hardness distributions on the cross-sectional area of W and ODS-steel joints: (a) SSDB, (b) LPDB [63].

Fig. 6: Scanning electron micrograph using backscattered electrons of a WCr12Ti2.5 sample after HIP [55,57]. Cr is in solid solution (SS) in W and vice versa.

Fig 1: Transition of dislocation-core symmetry introduced by Re alloying. Empty circles are atoms projected on the (111) surface; displacements along the [111] direction are represented by arrows connecting neighboring atoms.

Fig. 9: a) Model structure of the W 27<110>{525} general twin GB as viewed from the <110> direction. The dashed line denotes the position of the GB cleavage plane. Different colors represent different coordinates along the <110> direction. b) Cleavage energy as a function of substitutional elements at different positions.

Fig. 8: Temperature distribution in K and damage due to differential thermal expansion/contraction and crack formation. The surface was subjected to a heat flux of 1 x 104 MW/m2 for 0.1 ms are indicated. Axes in mm.

Fig. 2: Charpy test results of pure W, W-Ta, and W-V alloys. The specimens were fabricated out of round blanks that were forged to a height reduction of 80%. The chosen orientations are sketched.

Frequently asked questions

Q1. How much heat can be achieved in the in-vessel components?

Temperatures up to 1473 K can be achieved in the in-vessel components within a few weeks due to the decay heat [51], which would lead to fast oxidation with the release of volatile, activated W oxides [52]. 

Q2. What contributions have the authors mentioned in the paper "Recent progress in r&d on tungsten alloys for divertor structural and plasma facing materials" ?

In this paper, Schmid et al. presented two approaches to increase ductility and fracture toughness and to decrease the DBTT of W-based components. 

Q3. What is the method for avoiding this important safety issue?

A possible method for avoiding this important safety issue is the addition of oxide-forming alloying elements leading to the growth of a self-passivating layer at high temperature in the presence of oxygen. 

Q4. What is the reason for the restriction of Re additions?

Re additions have to be restricted to fulfill low activation requirements [16] and to avoid the formation of brittle phases due to significant transmutation of W into Re. 

Q5. What is the main idea behind the W-based materials for structural applications?

W-based materials for structural applications are developed based on the idea that alloys and especially nano-grained materials should be more ductile than pure W and standard grain-sized materials [22]. 

Q6. What is the effect of irradiation on the surface of the UFG W-?

(2) Irradiation with 1 keV H3 containing ~0.8% C did not causesignificant blistering but produced small holes on the surface, probably by ejection of grains [32]. 

Q7. What is the effect of the residual insert material on the hardness of the LPDB joint?

The hardness in the area of residual insert material decreased to ~3.3 GPa because of the large grains of the residual insert material. 

Q8. Why is the application of Re inhibited?

For improving mechanical properties, alloying with Re has proven by experiments and calculations to be advantageous; however, the application is inhibited due to the element’s low availability. 

Q9. What is the effect of irradiation on the grain boundaries of nano-grained?

In addition, the particles are expected to stabilize the grain boundaries in nano-grained materials upon thermal annealing and/or irradiation. 

Q10. What is the main reason for the brittleness of the main phase?

The brittleness of the main phase (W,Cr)5Si3 has a detrimental effect on its workability, which has motivated the development of Si-free alloys. 

Q11. What is the effect of diffusion on the hardness of the LPDB joint?

It appears that a diffusion-affected zone (DAZ) exists in the ODS steel close to the interface, in which W diffused along grain boundaries of the ODS steel. 

Q12. What is the thermal conductivity of WCr12Ti2.5?

The thermal conductivity of this alloy is close to 50 W/mK at 873 K, which is enough for the intended application, and is significantly higher than that of WCr10Si10 [57]. 

Q13. What is the reason for the ductility of W laminates?

One reason might be the ‘foil effect’ which is the dislocation annihilation on the free surface respectively on the interface between the hard W and the soft brazing filler. 

Q14. What is the DBTT of TFGR W-1.1TiC?

The DBTT, defined as the no-ductility temperature, of TFGR W-1.1TiC decreases with decreasing O-impurity content and lies below RT when the O content is below 400 wppm. 

Q15. What are the properties of the UFG W-TiC compacts?

densified W-(0.25-1.5)%TiC compacts with recrystallized, equiaxed grain sizes of 50-200 nm and fine TiC dispersoids at grain boundaries (GBs) were developed by PM methods utilizing MA and HIPping [26,27]. 

Q16. What is the optimum density of WCr12Ti2.5 bulk alloys?

Based on the results of [56], introducing W-Cr-Ti thin-film alloys with even lower oxidation rates, WCr12Ti2.5 bulk alloys have been manufactured by MA + HIP, obtaining again 100% density [57]. 

Q17. Why was the LPDB joint different from JOYO?

In the case of HFIR irradiation, the microstructure and electrical resistivity of irradiated pure W were dramatically different from JOYO-irradiated W, because larger amounts of Re were generated by nuclear transmutation of W to Re, which is due to HFIR’s higher flux of thermal neutrons, leading to a transmutation rate 10 times larger than in JOYO [68]. 

Q18. What is the method for forming multifilament bulk composites?

For fabrication of multifilament bulk composites, a new processing route was developed based on a gas-phase infiltration technique.