Elsevier

Fusion Engineering and Design

Volume 122, November 2017, Pages 124-130
Fusion Engineering and Design

Fracture mechanical analysis of a tungsten monoblock-type plasma-facing component without macroscopic interlayer for high-heat-flux divertor target

https://doi.org/10.1016/j.fusengdes.2017.09.002Get rights and content

Highlights

  • Tungsten monoblock target without macroscopic interlayer has been modelled.

  • Fracture mechanical studies on fatigue and brittle fracture were performed.

  • Numerical results of targets w/o macroscopic interlayer were compared.

Abstract

In the framework of the European DEMO divertor project, several novel design concepts of the plasma-facing components of vertical targets are being developed. One of those concepts is the tungsten monoblock design (similar to the ITER divertor) but with a very thin interlayer (roughly 25 mm thick only) at the armor/tube bond interface instead of a thick (1 mm) copper interlayer as has been the case in the conventional tungsten monoblock design developed for ITER divertor. The thin interlayer serves as bonding agent, but not as structural constituent. The reasoning for this novel design concept to omit the thick soft copper interlayer, which has been used as stress-relaxing buffer between the stiff armor block and tube, is to prevent plastic fatigue damage under cyclic high heat flux loads and irradiation embrittlement which the soft copper interlayer is predicted to undergo. On the other hand, the desirable stress relaxation effect on a global scale is abandoned. In this study, such trade-off effects are computationally investigated in a comparative assessment of structural impact, which the presence (or absence) of the thick copper interlayer is expected to bring forth, in terms of the fracture and fatigue behaviour of the armour block and cooling tube in two representative cases of tungsten monoblock plasma facing component design, namely, with and without a thick copper interlayer. Quantitative results of cyclic plastic strain history and crack tip fracture energy are presented for the armour surface, bond interface and tube of the respective plasma facing component models. The positive and negative implications of these impacts on the structural integrity are discussed.

Introduction

Divertor is one of the major in-vessel components in a fusion power reactor. Being responsible for power exhaust and plasma particle removal, divertor shall be subjected to significant thermal loads by surface heat flux as well as volumetric nuclear heating. Particularly, the plasma-facing components (PFCs) of the vertical targets will be exposed to severe high heat flux (HHF) loads reaching about 20 MW/m2 or higher locally at the strike point in the next generation fusion reactors such as ITER and DEMO [1], [2].

For both ITER and DEMO, the divertor PFCs will be fully armoured with tungsten (W) [1], [2]. The standard design of a ITER divertor target PFC is based on a joint structure consisting of W monoblocks as armour, copper alloy (CuCrZr) tube as water-cooled heat sink and a rather thick (typically 1 mm) copper (Cu) interlayer between the armour and the tube (see Fig. 1) [1]. Such an ‘ITER-like’ PFC design concept was also adopted for the European DEMO divertor target as a baseline design model while several other novel design variants are being developed in parallel in the framework of the EUROfusion Work Package “Divertor” [2], [3].

One of those novel PFC design concepts is the W monoblock design with a very thin interlayer (roughly 20 μm thick) at the bond interface as illustrated in Fig. 2. The characteristic feature of this design concept proposed by a group at French Alternative Energies and Atomic Energy Commission (CEA) is that there is no macroscopic constituent between the armour and tube, which is in contrast to the ITER divertor PFC where a thick Cu interlayer is integrated. The thin interlayer serves actually as metallurgical bonding agent at the bond interface, but no more as structural entity. In order to improve adhesion and reduce thermal strain mismatch, the interlayer was compositionally graded on microscopic scale starting from full W layer on the armour side and ending up in full Cu on the tube side. Recently, Richou et al. demonstrated that real PFC mock-ups of this design concept could be successfully manufactured by means of Physical Vapour Deposition (PVD) coating of graded W/Cu interlayer and Hot Isostatic Pressing (HIP) joining process [3]. The mock-ups exhibited a sound joining quality and quite robust HHF fatigue performance at least up to 300 load cycles at 20 MW/m2 (screening test up to 25 MW/m2 without visible damage) [4]. Actually, this fairly positive HHF test result is quite remarkable and even surprising when considering the fact that it has been a widely accepted practice in the PFC engineering to employ a thick Cu interlayer in order to mitigate thermal stresses with the help of the softness and ductility of annealed Cu. Now, the recent HHF test result indicates that to incorporate a thick Cu interlayer may not necessarily be an indispensable requirement for achieving reliable HHF performance.

Moreover, there are two material issues supporting the idea to omit thick Cu interlayer as in the present design concept:

  • 1)

    Under cyclic HHF loads (15–20 MW/m2), a thick Cu interlayer tends to undergo pronounced plastic fatigue leading to a premature low cycle fatigue failure of a PFC [5]. This failure feature was also experimentally observed where ductile strain damage was revealed in form of growth and coalescence of voids [6]. With increasing dpa dose, plastic fatigue behaviour of Cu will be affected by the equilibrium between irradiation hardening and thermal recovery.

  • 2)

    A previous neutron irradiation test indicated that the uniform elongation of irradiated pure Cu diminished drastically even at elevated temperatures (350–400 °C) due to helium embrittlement resulting from transmutation [7]. The originally expected effect of stress relaxation via ductile yield of soft Cu will gradually disappear.

These two empirical findings put the effectiveness of the thick Cu interlayer into question. As mentioned above, the promising HHF test result of the PFC mock-ups having no thick Cu layer puts the indispensability of the thick Cu interlayer into question. From this background arose the motivation to interpret the HHF performance and to understand the structure mechanical benefits of the underlying design concept to omit a thick Cu interlayer. In this paper, a rigorous numerical study is reported on the structural effect caused by the absence of a thick Cu interlayer in terms of armour cracking and plastic fatigue of tube. To this end, fracture mechanical and cyclic-plastic simulations based on finite element method (FEM) were carried out.

Section snippets

FEM model, materials & boundary conditions

In this work, the FEM model (see Fig. 1) was built according to the novel divertor target with a very thin graded interlayer proposed by a group at CEA. The ITER-like baseline design with an interlayer of 1 mm in the EUROfusion Work Package “Divertor” was set as a reference case for comparison. The single W monoblock has a dimension of 23 × 22 × 4 mm3, and the monoblock in ITER-like baseline design has a dimension of 25 × 23 × 4 mm3 [8]. The cooling tube has a thickness of 1 mm (baseline design: 1.5 mm) and

Temperature distribution

Fig. 4 shows the temperature distribution of the divertor target mockup without interlayer. The maximum temperature of W monoblock at 10 MW/m2 (805 °C) is well below the recrystallization temperature of W. At 20 MW/m2 the maximum temperature at the edge of the W monoblock is 1610 °C, while at the mid line of the top surface along the axial direction the maximum temperature is 1391 °C. The temperature at the top surface of W armor is more than 100 °C (50 °C at 10 MW/m2) lower than that (1723 °C at the

Summary and conclusions

The thick Cu interlayer which has been used as stress-relaxing soft buffer between the stiff armor block and tube in an ITER-type W monoblock PFC is predicted to undergo severe plastic fatigue damage under cyclic HHF loads. Furthermore, the pure Cu interlayer is anticipated to be heavily embrittled under neutron irradiation due to transmuted helium. Thus, the presence (or absence) of the thick Cu interlayer is expected to bring forth trade-off effects on the structural integrity of a W

Acknowledgments

This work has been carried out within the framework of the EUROfusion Consortium and has received funding from the Euratom research and training program 2014–2018 under grant agreement No 633053. The views and opinions expressed herein do not necessarily reflect those of the European Commission.

Cited by (11)

  • Crack formation in the tungsten armour of divertor targets under high heat flux loads: A computational fracture mechanics study

    2022, Fusion Engineering and Design
    Citation Excerpt :

    The shortest distance from the plasma-facing surface to this Cu interlayer is 8 mm. Comparing to the previous design [5], the current design has this 8 mm armor thickness instead of 5 mm to assure a longer erosion lifetime [6]. The boundary conditions are the same as in the previous research [6], where the coolant temperature is 150 °C, hydraulic pressure 5 MPa with velocity 16 m/s.

  • Structural lifetime assessment for the DEMO divertor targets: Design-by-analysis approach and outstanding issues

    2021, Fusion Engineering and Design
    Citation Excerpt :

    Although the structural part (pipe) is hardly affected by embrittlement, the higher tensile stress intensity in the armor may cause a crack formation at or near the bond interface leading to a local reduction of heat conduction. Cracking of armor blocks in the absence of a stress-relieving soft interlayer was indeed observed in previous HHF tests and explained by simulation [50,51]. The production and accumulation of helium bubbles increases (approximately linearly) with neutron fluence.

  • Preventive failure assessment of divertor armor blocks with a postulated flaw under slow transient conditions

    2020, Fusion Engineering and Design
    Citation Excerpt :

    Engineering application studies were also conducted on tungsten mono-blocks to evaluate weakness against cracking [9], to predict crack growth in use of stress intensity factors [10], to estimate possibilities of crack initiation and propagation via elastic-plastic fracture mechanics analyses [11] under high heat flux loading. Recently, a few expanded analyses were carried out beyond normal transient conditions like the heat sink tube of mono-block in a high confinement mode operation [12] as well as dealing with the structure-mechanical effect of Cu interlayer under cyclic high heat flux loading [13] and relatively huge geometry like the outer vertical target(OVT) with a postulated flaw under low heat flux loading [14]. However, attention is needed to reflect these valuable results in ongoing designs such as DEMO [15,16] because the previous data are limited and their details are somewhat different.

  • Neutron diffraction measurement of residual stresses in an ITER-like tungsten-monoblock type plasma-facing component

    2019, Fusion Engineering and Design
    Citation Excerpt :

    3D quarter geometry model was used with kinematic symmetry boundary conditions. The thin Cu interlayer was included in the model [9]. J2 plasticity with the von-Mises type yield criterion was assumed, which allows at least one of the three principal stress components exceed the yield stress (see hoop stress in Fig. 4 below).

View all citing articles on Scopus
View full text