Material and component developments for the DEMO divertor using fibre reinforcement and additive manufacturing

Within the research along the European Fusion Roadmap, water-cooled divertor PFCs are foreseen in the design of a first fusion demonstration power plant (DEMO) in order to provide reliable heat removal capability. In the frame of this concerted attempt, the Max Planck Institute for Plasma Physics is concentrating on the development and testing of composite materials based on tungsten (W, preferred armour material) and copper (Cu, preferred heat sink material). W fibres (Wf) as monofilaments and yarns as reinforcement play a central role in these investigations due to their extraordinary properties concerning ductility already at room temperature and high tensile strength. Recent investigations on the impact of radiation damage suggest that the fibres retain their ductility upon irradiation. W reinforced with W fibres (Wf/W) allows to overcome the intrinsic brittleness of W. Quantitative mechanical fracture tests of Wf /W confirm the basic mechanisms of fibre reinforcement and the increased resistance to mechanical fatigue. The good wettability of W with liquid Cu and the absence of any metallurgical solubility make up an ideal material pairing for composite production. W fibre-reinforced Cu (Wf/Cu) cooling tubes provide a rather high thermal conductivity (> 250 W mK−1) and at least twice the strength of CuCrZr in hoop direction in the temperature range up to at least 500 °C. Very recent neutron irradiation experiments confirm the sustainment of ductility of the Wf/Cu composite. Numerical simulations suggest that thermal stresses in W-Cu PFCs could be strongly reduced by tailoring the local W and Cu volume fraction. This ‘freely’ distributed material composition can be achieved by means of additively manufactured W skeletons consecutively infiltrated by Cu. Investigations with W preforms produced by Laser Beam Powder Bed Fusion and infiltrated by Cu demonstrate the feasibility of this approach while testing of specifically prepared specimen is ongoing.


Introduction
In view of the severe operating conditions for plasma facing components (PFCs) in future power producing fusion devices, the development of advanced components and materials is mandatory [1].The PFCs not only have to withstand high steady state power loads but also high number of thermal cycles and shocks.Moreover, the change of thermo-mechanical properties by lattice damage, activation and transmutation through fusion neutrons has to be considered when designing PFCs and selecting adequate armour and structural materials.Within the research along the European Fusion Roadmap, conceptional design work is performed for a DEMO fusion reactor.The European DEMO is planned to convert fusion heat into electricity (∼500 MWe), achieve tritium self-sufficiency (Tritium Breeding Ratio >1), to provide reasonable availability over several full power years, to minimize activation waste (no long-term storage) and to act as component test for advanced concepts of breeding blankets for a commercial Fusion Power Plant (FPP) [2].A recent special issue [3] provides a detailed overview on the recent status of the planning.Water-cooled PFCs are foreseen in a first DEMO design in order to provide reliable heat removal capability and to only moderately extrapolate the technologies developed and tested for ITER [4].As it turned out that the ITER target technology based on CuCrZr heat sink pipe and W mono-blocks is prone to cracking at cyclic loads of 20 MW m −2 (twice the steady state design load) [5] due to low cycle fatigue (crack initiation) and brittle behaviour during cool down [6], attempts were undertaken to optimize the design as well as the armour and heat sink materials of the DEMO divertor PFCs [7].Specifically, it could be shown by FEM analysis [8] as well as in experiments [9], that by reducing slightly the mono-block width the cracking could be eliminated for power loads of up to 25 MW m −2 in steady state.Another route for optimization is the development of materials which either have superior properties or reduce the thermal stresses in the PFC.The paper will report on recent results on the development of composite materials in this context at the Max Planck Institute for Plasma Physics.The paper is based on a preprint produced on the occasion of the 28th IAEA Fusion Energy Conference in 2021 [10].
To make best use of the water-cooled divertor PFC concept, copper (Cu) based alloys (as for example CuCrZr) are foreseen as heat sink whereas tungsten (W) based materials will be used as armour.This solution has been chosen in the current European DEMO concept [4] because at the moment the physics extrapolation seems to require the highest heat exhaust capabilities possible, which can only be realized by water cooling combined with a Cu (alloy) heatsink.In a later stage of the development of fusion reactors, more advanced solutions using a combination of W / ODS steel / He coolant might be envisaged if a safe operation with maximum power loads below 10-12 MW m −2 can be achieved [11,12].In this case higher operating temperatures could be used and issues arising from activation of Cu could be minimized.
Combining Cu and W in a high heat flux component bears the difficulty that their optimum operating temperatures do not overlap: W should be operated above 800 °C in order to be in a ductile state even under neutron irradiation and thus avoid brittle cracking under cyclic load, whereas CuCrZr should be operated below 300 °C to provide enough mechanical strength [13,14].A remedy for both issues could be the use of W fibres in W and Cu based composites.
With their high strength at high temperatures, the W fibres could help to extent the operating temperature range of a W f /Cu composite based heat sink.This concept bears also the possibility to reduce the mismatch in the coefficients of thermal expansion (CTE) of the armour (W) and the heatsink (Cu or Cu alloy) materials.The W fibres can be also incorporated in W f /W composites using their ductility and additional toughening effects (as in ceramic fibre reinforced composites, see below) to reduce the brittleness of the armour material.In principle other fibres, as for example silicon carbide (SiC) fibres could be used as has been shown in previous works [15].However, due to fact that W fibres are commercially available in many diameters and show a ductile behaviour in a very broad temperature range (see below), the work further concentrated on W fibres.Alternatively, the concept of functionally graded materials could be exploited by the use of additive manufacturing methods allowing an optimized materials distribution in view of the reduction of thermal gradients and stresses.
The paper is organized as follows: After the introduction, the basic properties of the tungsten fibres and the development of yarns and textiles will be shortly reviewed in section 2. Section 3 deals with tungsten fibrereinforced tungsten (W f /W) whereas section 4 will present the results on tungsten-fibre reinforced copper (W f /Cu).The concept of additively manufactured W-Cu PFCs will be described in section 5. Finally, the paper will be concluded by a summary and an outlook is given in section 6.

Tungsten fibres, yarns and preforms
The W fibres investigated with regard to PFC applications are drawn potassium doped tungsten wires as used in the lighting industry.They are characterized by high tensile strength (>2000 MPa), ductility already at room temperature and embrittlement by recrystallization and grain growth only above 2200 °C [16].Below this temperature, potassium bubbles at the grain boundaries inhibit grain boundary migration retaining the elongated grain structure established by the drawing process.The ultimate tensile strength increases with decreasing fibre diameter due to the increased deformation of the W grains reaching values above 4000 MPa for fibres with 16 μm diameter [17].
Industrial textile techniques have been successfully established to prepare W fibre preforms for the production of flat tungsten fibre reinforced tungsten (W f /W) samples [19] and tungsten fibre reinforced copper (W f /Cu) pipes [20].A further improvement in workability and strength is achieved by the use of W yarns instead of monofilaments, consisting of up to 20 thin individual W fibres with a diameter of up to 20 μm [18,21].
Figure 1 (left) shows a stress-strain diagram for a 150 μm diameter single W fibre and for a W yarn with similar nominal thickness consisting of fourteen 20 μm diameter fibres.It demonstrates the higher ultimate tensile strength of the thinner fibres also effective in the yarn, as well as the consecutive failure of the individual fibres (stepwise failure).The insert at the bottom shows a knotted W yarn exemplifying its extreme flexibility.In the right side of figure 1, micrographs of the single fibres of the W yarn after the tensile test are shown.The necking as well as the typical knife edge structure at the fracture surface are clear indications of their ductile behaviour already at room temperature.Experiments irradiating W fibres with W ions simulating the creation of defects originating from the strong n-irradiation in a fusion reactor revealed that up to a damage of 9 dpa (displacements per atom) only a 20% reduction of strength was observed but no signs of embrittlement were found [22].This provides good confidence that the strengthening and toughening by fibre reinforcement is preserved, which is also confirmed in very recent n-irradiation experiments ( [23], see also section 4).Some of the milestones of the work on W-fibres, yarns and textile preforms are summarized in table 1.

Tungsten fibre-reinforced tungsten
In brittle materials such as bulk tungsten there is no possibility of stress redistribution which leads to a scattering in strength and defect sensitivity.Extrinsic toughening is widely known in ceramic fibre-reinforced ceramics using mechanisms as crack deflection at interfaces, crack tip shielding by fibre bridging, energy dissipation by fibre pull-out and-where possible-by ductile deformation of the reinforcing fibres [26].By introducing W fibres, the toughness of W increased significantly already at room temperature [27].Tungsten fibre reinforced W (W f /W) can either be produced by chemical vapour deposition/infiltration (CVD/CVI) decomposing WF 6 at moderately elevated temperature (<600 °C) [26] or by powder metallurgy [28].Whereas in the latter case typically randomly oriented short W fibres are used, CVD allows the introduction of long fibres and textiles promising a more effective toughening albeit at a higher effort.
Initially samples with the size of 50 × 50 × 5 mm 3 were produced with a fibre fraction of 10% -30%.From the available material specimen for mechanical testing as well as mock-ups for high heat flux testing were produced.Figure 2 shows the fracture surface of a specimen (with W fibres with a diameter of 150 μm) after a Charpy impact test performed at room temperature.Whereas the W matrix fails in a brittle way, several typical  Tensile behaviour of drawn tungsten wire used in tungsten fibre-reinforced tungsten composites 2017 [17] The effect of heat treatments on pure and potassium doped drawn tungsten wires 2018 [24,25] Textile preforms for tungsten fibre-reinforced composites 2018 [19] The use of tungsten yarns in the production for W f /W 2020 [21] features of extrinsic toughening as crack bridging, fibre pull out and plastic deformation of the fibres are observed.In a series performing those tests on W f /W in the temperature range from −150°C to 1000 °C it turned out the fibres start to become ductile already between −100 °C and −50 °C, the matrix behaves brittle up to 900 °C [29].
The increased toughness of the W f /W was estimated in 3-point bending tests where crack bridging and ductile deformation of the fibres was identified to be the main mechanism contributing to the fracture toughness [27].In addition to a single overload leading to the cracking of the material, there are cyclic loads acting on a PFC caused by the change of thermal loads.Cyclic loading stress, even if they are below the maximum strength of the material can lead to fatigue fractures.Fatigue fractures can be explained by repetitive local overloading and thus crack initiation and growth within a material.This can be caused by microstructural flaws or inhomogeneities, such as pores or solid precipitates.In order to examine the behaviour of W f /W under those conditions cyclic  Advanced materials for a damage resilient divertor concept for DEMO: Powder-metallurgical tungstenfibre reinforced tungsten 2017 [28] Estimation of the fracture toughness of tungsten fibre-reinforced tungsten composites 2020 [27] Design of tungsten fiber-reinforced tungsten composites with porous matrix 2021 [33] Interlayer properties of tungsten fibre-reinforced composites and their determination by different methods 2021 [34] tensile tests were performed with the same setup which was used for static tensile tests previously [27].The unidirectional reinforced W f /W specimens with a fibre volume fraction of ∼11% and a matrix density of ∼99% were tested.The specimen had a dog bone shape with an overall length of 41 mm and a thickness of 3 mm.The thinned test section was 25.5 mm in length with a width of 2 mm.The specimens were loaded in fibre direction with 10000 cycle at 60%, 70%, 80%, 90% and 100% of the maximum load (defined as average ultimate strength of preceding tensile tests).
Figure 3 shows the stress / strain diagram of the W f /W during the tests at different stress levels.During the first cycles at a given stress level load drops occurred accompanied by a clicking noise, indicating matrix cracks.During the further cycles at constant stress, no further load drops were detected in the stress-strain curves.Between the different stress levels an increasing hysteresis can be seen, pointing to the progressing damage of the sample.This experiment demonstrates impressively that differently to a fully brittle material W f /W can cope with cyclic loads even under pre-damaged conditions as long as its maximum tensile strength is not exceeded.
As described above, small mock-ups for high heat flux testing were produced from the same material, consisting of a heat sink equipped with W f /W armour blocks of different thickness and orientation.Due to different thickness a variation of the maximum temperatures was reached for a given heat flux.The mock-ups have been loaded up to 15 MW m −2 leading to a surface temperature above 2500 °C.Beside surface modifications caused by the high temperatures most blocks survived in good condition.It was shown that nonoptimized fibre orientation can lead to delamination and premature failure.
Table 2 summarizes the major milestones in the development of W f /W composites.Currently the main emphasis of the development is put on the upscaling of the sample production in order to arrive at material samples suited for large mechanical test specimen and mono-block-like mock-ups.First tests with sample sizes up 84 × 24 × 10 mm 3 and a fibre fraction of about 10% confirm the pseudo-ductile behaviour originating from the fibre properties and yield a fracture toughness of K Rmax = 346 MPam 1/2 [30].CVD samples produced with yarn based preforms allowed a higher fibre fractions (14%-17%) and densities (97%) and increased reproducibility [31] compared to the 150 μm single fibre-based fabrics described above [30].

Tungsten fibre-reinforced copper
W-Cu composites can act as advanced PFC heat sink materials because the W reinforcement enhances the mechanical material properties especially at elevated temperatures, the Cu matrix leads to a high overall thermal conductivity and it provides an acceptably ductile material behaviour for sufficiently high Cu contents [15].Tailoring the Cu content can also reduce the mismatch of the macroscopic CTEs between armour and heatsink.Since the wettability of W with Cu melt is good, Cu melts already below the recrystallization temperature of W and there is no metallurgical solubility of W and Cu, the system is very suited for Cu melt infiltration.Using W preforms out of thin W fibres, i.e. the abovementioned yarn leads to a composite material with improved mechanical properties reaching high (high temperature) strength already at a rather low W content thereby reducing less the superior thermal conductivity of the Cu matrix.As stated already above for the fibres, a very recent n-irradiation campaign in the BR-2 reactor (Mol, Belgium) with 2.5 dpa (in Cu) revealed that W f /Cu composite retains benign mechanical behaviour under neutron irradiation [23].Some of the milestones of the work on W f /Cu are summarized in table 3.As for the currently favoured W mono-block design of high heat flux PFCs the heatsink consists of a Cu-alloy tube, it is expedient to produce W fibre-reinforced Cu tubes which could be operated at higher temperature and which would cause less stress at the bonding to the W mono-block due the adjusted CTE.The tubular geometry is well suited for applying textile techniques and therefore circular braided preforms were developed together with a partner (DITF, 73770 Denkendorf, Germany).The preforms consist of several braided layers of W fibre / yarn as shown in figure 4 with a high braiding angle in order to achieve a pronounced reinforcement in hoop direction [20].
Braided fibrous W preforms were infiltrated with Cu, machined to desired dimensions before being brazed to W mono-blocks. Figure 5 shows a quarter of a perpendicular cross section of an infiltrated W f /Cu pipe (a) and details of a longitudinal cross section proving the perfect Cu infiltration.
Mock-ups consisting of the W f /Cu tube and W mono-blocks were subjected to cyclic high heat flux (HHF) tests in IPP's neutral beam facility GLADIS.Figure 6 shows infrared and visible images of the mock-up taken during the first and the last pulse (1000 th ) under DEMO relevant power load (20 MW m −2 ) and cooling conditions (T cool = 130 °C, 4 MPa, 16 m s −1 ).During the cycling (10 s beam on, 60 s cool down) the W surface reached well above 2000 °C.
Figure 7 shows a photograph and a surface map of the mono-blocks after the high heat flux loading.Clear signs of surface modifications with distortion of the originally flat surfaces by up to 200 μm can be observed (bottom part of figure 7.).Metallographic cross section of mock-up with W mono-blocks and the W f /Cu heat sink pipe (figure 8) confirm the considerable plastic deformation also in other dimensions.According to FEM analysis, these deformations can be explained by accumulation of inelastic strain by creep which increases with increasing grain size at the surface due to the temperatures considerably above the recrystallization temperature of W [38].This interpretation is further confirmed by EBSD analysis of the cross section of the mono-block.As can be seen in the right part of figure 8, the original fine grain structure of W has vanished at the surface in favour of a mm-sized grain.However, despite this considerable deformation and the strong recrystallization, no cracking or degradation of the thermal properties is found, demonstrating the very high performance of the mock-up with W f /Cu composite pipes [20].

Tungsten copper composites produced by additive manufacturing
Additive manufacturing (AM) could be a powerful technique for application in fusion technology, because many of the components (not necessarily PFCs) are still at the level of prototypes.Due to its flexibility it can provide solutions that go beyond those of conventional manufacturing methods.However, experience on additive manufacturing of W was rather limited until recently.Therefore, Laser Beam Powder Bed Fusion (PBF-LB/M) of pure W was investigated and optimized together with Fraunhofer IGCV (Augsburg), allowing the manufacturing of W parts with a mass density as high as 98% [39].Preheating the build plate to up to 800 °C helps to reduce defects in the W bulk material, albeit a complete suppression could not be reached.Mechanical testing of the additively manufactured W showed that the presence of micro-cracks strongly reduce the tensile strength but also provide some pseudo-ductility.Since the joining of the armour (W) to the heat sink (Cu/ CuCrZr) is a crucial part for the functionality of the PFCs, the envisaged use of AM was concentrated on this subject.In the current design the topmost armour still consists of mechanically deformed W (rolling, forging)  due to its better mechanical performance (compared to AM W).The heat sink is produced by infiltrating the AM W structure with copper.Figure 9 illustrates the different process steps on the way to the final AM W-Cu PFC. Figure 9(a)) shows honeycombs with different cell sizes during the PBF-LB/M process.The honeycombs glow because they are in good thermal contact with the W building plate which is preheated to up to 600 °C and are heated up during exposure to the laser beam.The building plate will function as plasma facing armour and the AM structures are designed to provide a functional transition (see details below) to the Cu infiltrated heat sink.Figure 9(b)) shows the W honeycomb skeletons after cleaning and before infiltration.As in the case of the W-fibre braids (cf figures 5 and 7), a perfect infiltration of the AM W honeycombs with Cu could be achieved as can be deduced from figures 9(c)) and (d).
Combining the concept of functionally graded composite materials with the prospects of AM opens up the possibility to produce W-Cu composite structures with tailored material distribution and additionally minimizing thermally induced stresses.For this purpose, a numerical scheme based on the finite-element method was set up to computationally optimize the W-Cu material distributions in a PFC.As a result, the macroscopic von Mises stresses could be reduced in the simulations by a factor of 6 compared to the conventional W mono-block design at a heat load of 10 MW m −2 [40].In these calculations, the shape of the PFC was intentionally chosen to be of the typical W mono-block geometry in order to highlight the benefit of the optimized W-Cu materials distribution.Generally, exploiting the whole flexibility of the AM method, cooling channel geometries with even more beneficial properties can be designed.This optimisation is part of ongoing work.Figure 10(a) shows the CAD drawing of a cross-section of an optimized W-Cu structure in which a bcc W lattice is embedded.In figure 10(b) the AM counterpart produced by PBF-LB/M and ready for infiltration with Cu is presented.
Figure 10(c)) shows a readily Cu-infiltrated and machined PFC mock-up with (an earlier version of an optimized) W honeycomb skeleton and castellated W armour which was cut from the W build plate (corresponding to the top of figure 9(a))) via electrical discharge machining [41].Recently, heat flux tests successfully started in order to qualify W-Cu AM mock-ups for the use in PFCs.Up to now more than 100 pulses with a duration of 10 s have been performed.The pulse length was chosen to reach thermal equilibrium.The thermal behaviour was screened with heat fluxes starting from 6 MW m −2 ranging to 25 MW m −2 .The  Interfacial optimization of tungsten fibre-reinforced copper for high-temperature heat sink material for fusion application 2009 [36] Melt infiltrated tungsten-copper composites as advanced heat sink materials for plasma facing components of future nuclear fusion devices 2017 [37] Application of tungsten-copper composite heat sink materials to plasma-facing component mock-ups 2020 [20] Effect of neutron irradiation on tensile properties of advanced Cu-based alloys and composites developed for fusion applications 2023 [23] corresponding equilibrium surface temperatures ranged from 560 °C to 1800 °C, respectively, proving the effectiveness of the AM W-Cu heatsink.After further cycling at 10 MW m −2 with 90 pulses no deterioration of the W armour surface nor the W-Cu interface were observed.Further cycling at higher heat loads is planed in the near future.Table 4 summarizes the rather new devolopments on AM W-Cu composites performed by the group.Additive manufacturing of pure tungsten by means of selective laser beam melting with substrate preheating temperatures up to 1000°C 2019 [40] Progress in additive manufacturing of pure tungsten for plasma-facing component applications 2022 [41] 6. Summary and outlook In view of the severe operating conditions for PFCs in future power producing fusion devices, the further development of advanced components and materials is mandatory.The components not only have to withstand high steady state power loads but also high number of thermal cycles and shocks.Moreover, the change of thermo-mechanical properties by lattice damage, activation and transmutation through fusion neutrons has to be considered when designing PFCs and selecting the adequate armour and structural materials.Within the research along the European Fusion Roadmap, water cooled PFCs are foreseen in a first DEMO design in order to provide reliable heat removal capability and to only moderately extrapolate the technologies developed and tested for ITER.In the frame of this concerted attempt IPP is concentrating on the development and testing of composites based on tungsten as armour and Cu as heat sink materials.W fibres as reinforcement play a central role in these investigations due to their extraordinary properties concerning ductility already at room temperature, flexibility and very high ultimate tensile strength.Using yarns consisting of W fibres these properties can be further improved and harnessed in preforms produced by textile techniques.Recent investigations on the impact of radiation damage using heavy ions and neutrons suggest that the fibres retain their ductility up to several dpa.In order to produce yarns and preforms for the W fibre reinforced composites (see below) a batch of more than 1400 km of W fibre with a diameter of 16 μm was procured and the corresponding yarn has been produced for braiding and weaving preforms.
To overcome the intrinsic brittleness of W which gives rise to cracking of the PFC armour under cyclic load, W fibre reinforcement can play an important role to increase the fracture toughness and ameliorating the failure scenario.Quantitative mechanical fracture tests of W f /W confirm the basic mechanisms of fibre reinforcement and the increased resistance to mechanical fatigue.First high heat flux tests with small actively cooled mock-ups seem to support the results of the mechanical tests, albeit the currently available material sample size is too small to produce mock-ups of similar dimensions as the DEMO 'baseline' mono-blocks.In collaboration with FZ Jülich efforts are under way to prepare bulk W f /W material samples for mechanical testing with larger specimen (to reduce size effects) and to produce PFCs mock-ups allowing a one to one comparison in high heat flux experiments with standard mono-block DEMO PFCs.
In comparison to W f /W the development of W f /Cu is rather advanced and production of heat sinks from the composite material is already performed in collaboration with industrial partners.The high wettability of W with liquid Cu and the absence of any chemical reaction make up an ideal material pairing for composite production.The used W fibre/yarn preforms are well adapted to their application in cooling tubes and the textile technique has been optimized to achieve an improved strength in hoop direction.The about 50/50 weight composition of W fibres and Cu provides a still rather high thermal conductivity (>250 W mK −1 ) and at the same time at least twice the strength of CuCrZr in the temperature range up to at least 500 °C.Very recent neutron irradiation experiments reaching up to 2.5 dpa confirm the sustainment of some ductility as already indicated by the single fibre experiments.Future work will concentrate on further industrial upscaling of the cooling tubes and the use of CuCr alloy in order to confirm the role of W f reinforced heatsinks as a risk mitigation material for DEMO PFCs.
In contrast to the approaches described above, which aim at a strengthening and/or toughening of the composite material, the latest approach for components produced from W composites started at IPP aims at the reduction of intrinsic thermal stresses by an optimized material composition.Numerical simulations suggest that thermal stresses could be strongly reduced by tailoring the local W and Cu volume fractions.This 'freely' distributed material composition can be achieved by means of additively manufactured W skeletons consecutively infiltrated by Cu.Investigations with W preforms produced by PBF-LB/M and infiltrated by Cu demonstrate the feasibility of this ansatz and mechanical testing of specifically prepared specimen is ongoing.
Comparing the two different production methods for W-Cu composites for future applications in fusion it can be stated that the W fibre reinforcement of Cu or Cu-alloys has already reached the level of (small scale) industrial production and its application is very close to the established use of Cu/CuCrZr as structural elements.Therefore, the W f /Cu composite is foreseen as risk mitigation solution in the engineering the design of the European DEMO reactor [4].On the other hand, the AM based W-Cu heatsink is much less industrially established and substantial further qualification in terms of material properties and high heat flux behaviour of complete components has to achieved.However, if the predicted superior behaviour of the components is confirmed in further experiments, AM could provide a valuable alternative approach to the design and manufacturing of PFCs for high heat flux applications.

Figure 1 .
Figure 1.Left: Typical tensile behaviour of a twisted W yarn with 14 × 20 μm filaments tested at room temperature (blue) compared to a single W fibre with a diameter of 150 μm (orange).Adapted from [18].The insert shows a micrograph of a knotted W yarn to demonstrate its flexibility and ductility.Right: Micrographs of fibres after the tensile test.The necking as well as the knife edge structure at the fracture surface are clear indications of the ductile behaviour of the W fibres already at room temperatures.

Table 1 .
Selected milestones for the development of tungsten fibres, yarns and textile preforms.Topic Milestone Year Reference Fibres, Yarns & Textile Preforms Development of tungsten fibre-reinforced tungsten composites towards their use in DEMO -potassium doped tungsten wire 2016 [16]

Figure 2 .
Figure 2. SEM picture (top) and 3D picture (bottom) of the fracture surface of W f / W after Charpy impact testing at room temperature.The W matrix fractures brittle whereas the fibres show the same ductile behaviour in the single fibre tensile testing.Besides this, a typical feature of fibre reinforcement as the fibre pull out is clearly visible.The diameter of the reinforcing W fibres is 150 μm.The colour bar in the lower picture provides the elevation of the surface in μm.

Figure 3 .
Figure 3. Stress / strain diagram of a W f / W composite during cyclic test at different stress levels with 10000 cycles each.

Figure 4 .
Figure 4. Left: Multi-layered tungsten yarn braided preform for W f /Cu heat sink tube.Right: Detailed view of yarns comprising of 20 μm W fibres.

Figure 5 .
Figure 5. Cross section of a W f /Cu tube.(a) Quarter circumference of the tube.(b) Detailed cross section demonstrating the complete Cu infiltration.Reprinted from [37], Copyright (2017), with permission from Elsevier.

Figure 6 .
Figure 6.Top: PFC mock-up consisting of W mono-blocks and a W f /Cu heat sink pipe.Bottom: Infrared and optical images of the mock-up during cyclic HHF loading with 20 MW m −2 for pulses no. 1 and 1000.Clear signs of plastic deformation of the W monoblocks evolved during the testing, but no damage of the component was observed.Adapted from [20] © EUROfusion Max-Planck-Institut für Plasmaphysik.All rights reserved.

Figure 7 .
Figure 7. Top: Photograph of the surface of the mock-up with W mono-blocks and a W f /Cu heat sink pipe after high heat flux loading with 20 MW m −2 .Bottom: Confocal laser scanning microscopy of the surface of two of the W mono-blocks after HHF loading with 20 MW m −2 .

Figure 8 .
Figure 8. Left: Metallographic cross section of mock-up with W mono-blocks and a W f /Cu heat sink pipe after high heat flux loading at 20 MW m −2 .Right: EBSD measurements of one of the W mono-blocks showing strong grain growth at the highly loaded surface.

Figure 9 .
Figure 9. (a) PBF-LB/M of W honeycombs with different dimensions and a substrate preheating of up to 1000 °C.(b) Additively manufactured W honeycombs; (c) & (d) Different cross sections of a PFC Cu infiltrated W honeycomb.

Figure 10 .
Figure 10.(a) CAD model of a lattice structure (W: grey, Cu: orange) deduced from W-Cu material distribution optimisation.(b) Corresponding W structure additively manufactured by means of laser powder bed fusion.(c) PFC mock-up with W-Cu heat sink based on additively manufactured W (honeycomb structure) and bulk W armour tiles.

Table 2 .
Selected milestones for the development of W f /W composites.

Table 3 .
Selected milestones for the development of W f /Cu composites.

Table 4 .
Selected milestones for the development of additively manufactured W-Cu components.