_______________________________________________________________________________ author’s email: fantao@ctn.tecnico.ulisboa.pt

HEAs: High Entropy Alloys for advanced systems and engines

Francisco José Neto Antão

Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1049-001, Lisboa, Portugal,

In this study, WxTaCrNbV(x = 20 and 30 at. % W) and CuxCrFeTiV (x = 5, 10, 20 and 30 at. % Cu) high entropy alloys have been devised for thermal barriers between the plasma facing tungsten tiles and the copper-based heat sink in the first wall of nuclear fusion reactors. The high entropy alloys were produced by ball milling the elemental powders, followed by consolidation with pulse plasma sintering and spark plasma sintering. Heat treatments in both systems were performed to assess the thermal stability of the consolidated samples. Structural analysis was performed by scanning electron microscopy, coupled with energy dispersive X-ray spectroscopy and X-ray diffraction. Thermal diffusivity was performed on the CuxCrFeTiV (x = 5, 10 and 30 at%). Irradiation of the equiatomic CuCrFeTiV sample was carried out at room temperature with Ar+ (300 keV) beams with a fluence of 3 x 1018 at/cm2, to simulate conditions of implantation in the sample, associated with the operative conditions of a nuclear reactor. The results showed the presence of heterogenous and multiphasic microstructures in all samples. The diffractogram of the tungsten-based system suggests that it may be described as a majority-multielement BCC structure, with a secondary BCC structure of smaller lattice parameter however with similar composition. Moreover, in the copper-based system, with the increase of the Cu content it is possible to observe the formation of Cu-rich structures. The diffractogram of the CuCrFeTiV sample revealed major peaks of a BCC crystal structure and minor peaks of an FCC crystal structure. The CuxCrFeTiV (x = 5, 10 and 30 at. %) system presented stable thermal diffusivity values, in the range of 3-6 mm2/s. In addition, after irradiation roughly 2×1017 cm-2 Ar was retained in the equiatomic sample.

Keywords: high entropy alloys, microstructure, irradiation damage, X-ray diffraction

1. Introduction

The production of clean energy from the fusion of light mass particles, using nuclear fusion reaction facilities, has been under study for the past 50 years. These reactors can convert the energy released by controlled nuclear fusion into thermal energy for further conversion to mechanical or electrical forms, with no emission of harmful greenhouse gases and no long-lived radioactive waste. With the implementation of these machines the opportunity to produce clean energy with nearly unlimited energy reserves arises. However, numerous challenges come to light when engineering the power stations to handle these fusion reactions in a contained space. These challenges, in materials science and engineering, must be overcome to be able to achieve conditions for sustained plasma fusion reaction. One of the issues in the divertor, a region of high heat flux with the function of extracting the power from the conductive/convective heat flux, and radiation whilst maintaining the plasma purity, is the fact that its materials must endure high temperatures and irradiation damage. It is a fully water-cooled, tungsten armoured unit, bearing vertical targets, themselves constituted of a series of plasma-facing units made up of chains of W monoblocks bonded to a CuCrZr cooling tube. In Figure 1, it is possible to observe that, in between W and CuCrZr is an interlayer, which in ITER is pure copper.

1.1 Motivation

Tungsten is a promising material for plasma facing materials in fusion reactors (ITER and the future DEMO reactor) owing to its high melting point, high sputtering resistance, low deuterium/tritium retention, and good

Figure 1 - Schematic view of ITER divertor[55] consisting of a) inner vertical target and b) outer vertical target, c) dome

umbrella; (d) PFU, with monoblock geometry[6].

2

thermal–mechanical properties [1]. Also, it has a low production rates of transmutation elements (nuclear reactions initiated by neutrons leading to changes in the chemical composition of materials) comparing to others candidates to plasma facing components [2]. Nevertheless, it has drawbacks, namely its embrittlement at temperatures lower than its DBTT (350-400 oC)[3] [4], which makes the W tiles brittle at water cooling temperatures (150 oC) [8][6]. The heat sink material chosen with the task of draining the generated heat in the W armour is a CuCrZr alloy, due to its high conductivity, high strength, microstructural stability and ductility at the predicted operating temperatures[7][8][9]. It has a high fracture toughness, allied to a good availability [10]. The issue arises in a dissimilarity between both materials: the high ductile-brittle transition (DBTT) temperature of tungsten [11] demands higher temperature, while CuCrZr imposes a low service temperature [12] and suffers from radiation embrittlement [13]. This will lead to a less efficient heat extraction and shorter service lifetimes of the materials in use; therefore, a thermal barrier interlayer is necessary.

A proposed strategy to solve this issue is the use of copper-based and tungsten-based high entropy alloys to minimize the thermal mismatch between tungsten and the CuCrZr and is a new concept worth exploring.

1.2 Background

High Entropy Alloys (HEAs) are alloys composed of

five or more metallic elements, with concentrations between 5 at.% and 35 at.% [14], that can form random simple solid solutions with distorted lattices (mainly BCC or FCC) conferring them enhanced properties, instead of the complex and brittle intermetallics that are formed in conventional, non HEA, alloys. Interesting properties were obtained with HEAs that reveal their potential as thermal barriers: high hardness [15], enhanced wear [16], oxidation [17] and corrosion [18][19] resistances, high thermal stability [20], both under continuous and cyclic operation, and intrinsically low thermal conductivity [20][21]. Moreover, recent studies shown that the accumulated lattice strain might confer higher radiation resistance in HEAs [22]. However, the high energy neutrons from fusion reactions are expected to induce atomic displacement cascades in the first wall materials, but the response of these novel materials to irradiation damage is not known and requires elucidation.

Because of the unique multiprincipal element composition, HEAs can possess special properties, which are related to the collective behaviour of the constituent elements. To characterize this behaviour, as well as to predict and ascribe the structures governing each system, a few parameters have been taken into account for the past decade.

From statistical thermodynamics, the Boltzmann’s equation can be used for the calculation of configurational entropy of a system:

𝛥𝑆𝑐𝑜𝑛𝑓 = 𝑘 ln (𝑤)

Where 𝑘 is Boltzmann’s constant and 𝑤 is the number of ways in which the available energy can be shared among the particles of the system [23]. Thus, the configurational entropy change per mole for the formation of a solid solution from n elements with mole fractions 𝑐𝑖 is:

𝛥𝑆𝑚𝑖𝑥 = −𝑅 ∑ 𝑐𝑖 ln(𝑐𝑖)

𝑛

𝑖=1

where 𝑅 is the gas constant.

In what concerns the RMS average atomic size mismatch (parameter) δ used to characterize the atomic size difference, the equation is the following:

𝛿 = 100√∑ 𝑐𝑖(1 − 𝑟𝑖/�̅�)2

𝑛

𝑖=1

where �̅� = ∑ 𝑐𝑖𝑟𝑖

𝑛𝑖=1 , 𝑐𝑖and 𝑟𝑖 are the atomic fraction

and atomic radius of the ith element, respectively [24].

𝛥𝐻𝑚𝑖𝑥 = ∑ 𝛺𝑖𝑗𝑐𝑖𝑐𝑗

𝑛

𝑖,𝑗=1𝑖≠𝑗

Where 𝛺𝑖𝑗 = 4 𝛥𝑚𝑖𝑥𝐴𝐵 and 𝛥𝑚𝑖𝑥

𝐴𝐵 is the mixing enthalpy of

binary liquid AB alloys.

In addition, Zhang et al. [24] considered two more parameters. One is the RMS averaged electronegativity difference, 𝛥𝜒:

𝛥𝜒 = √∑ 𝑐𝑖(𝜒𝑖 − �̅�

𝑛

𝑖=1

)2

Where �̅� = ∑ 𝑐𝑖𝜒𝑖𝑛𝑖=1 , 𝜒𝑖 is the Pauling electronegativity

for the ith element. The other parameter is the valence electron concentration (VEC). According to Zhang study [24], it is believed that the VEC parameter strongly determines the phase stability of intermetallic compounds. VEC is defined by:

𝑉𝐸𝐶 = ∑ 𝑐𝑖(𝑉𝐸𝐶)𝑖

𝑛

𝑖=1

Sheng et al.[25], regarding the control of solid solution formation, concluded that the mixing entropy must match the range of 11 ≤ ΔSmix ≤ 19,5 J/(K.mol), and that

3

FCC phases are found to be stable at VEC ≥ 8 and BCC at a lower VEC (< 6.87). Between these values mixed FCC and BCC phases will co-exist.

2. State of the art This section gives an insight on the most common

materials proposed for the plasma facing components and in particular for thermal barriers.

2.1 Plasma Facing Components In the International Thermonuclear Experimental

Reactor (ITER), the material used for the first wall will be tungsten, due to its high melting point, thermal conductivity, and resistance to sputtering and erosion [10], [26], [27]. Research concerning novel materials such as composites to be suitable for first wall materials in future fusion power plants are being developed: Research conducted by M. Dias et al. about W-Ta composites [28], Yang et al. concerning W-ZrC-Re [29], and by Arshad et. al about W-V alloys [30]. Due to the extreme thermal, mechanical and physical impact on the components, leading to damage and degradation of the materials [31], the development of the plasma facing components of the divertor is still a serious, on-going engineering challenge.

2.1.1 Thermal Barriers Several thermal barrier materials have been

suggested, like diamond-Cu composites [32]; in this particular case, the irradiation, as well as high temperature of the fusion reactor tends to form an intermediate carbon phase with a structure between graphite and diamond, which changes its thermal expansion coefficient and mechanical properties, which is not a good feature. Pintsuk et al. [33] developed W-Cu functionally graded materials, which presented poor mechanical results at elevated temperatures. Schöbel et al., presented a tungsten monofilament reinforced Cu composite [34], which exhibited strong thermal fatigue during thermal cycling, due to the mismatch stress between the W-wires and the matrix.

Other proposals are also found in the literature, however, amongst them, none possesses the properties, nor the characteristics, needed to be considered a valid candidate for a thermal barrier. That is the motive why it is momentous to propose a material truly capable of functioning as a compatible thermal barrier interlayer between the high temperature plasma facing tungsten and the CuCrZr heat sink – the high-entropy alloys.

2.2 High-Entropy Alloys (HEAs) The first work on exploring this brave new world was

done in 1981 by Cantor with his undergraduate project student Alain Vincent. They studied many elements, but particularly transition metals.

In 2004, J.W. Yeh independently explored the multicomponent alloys and the effect of the entropy of mixing within these materials. He also evidenced the absence of complex phases or microstructures, which manifested the lowering of free energies by the high entropy of mixing and showed the significance of simple-structured HEAs, whose development is encouraged to this day. Later on, he “baptized” these alloys as “HEAs” by pointing out the known thermodynamic fact that configurational entropy of a binary alloy is a maximum when the elements are in equiatomic proportions [23].

To this day only about thirty HEA systems have been investigated, with the solid solution formation observed mainly in systems based on 3d transition elements.

3. Experimental Procedure

3.1. Alloys preparation and Consolidation

W-based high entropy alloys

Powders of W, Cr, Ta, Nb and V with compositions as shown in Table 2 were produced by pulse plasma sintering (PPS) from powders of 99.9% nominal purity with average particle sizes in the range 1 µm to 3 µm (AlfaAesar), and mixed in a glove box, under argon atmosphere to avoid oxidation of the components. The powders were mechanically alloyed in a planetary ball mill, PM 400 MA type, with WC balls and vials. The balls to powder mass ratio was 10:1, and the milling occurred for effective times up to 2 hours, at 380 rpm.

Consolidation was carried out at 1773 K in 5x10−3 mbar vacuum with an applied mechanical pressure of 100 MPa.

Cu-based high entropy alloys

Powders of Cr, Cu, Fe, Ti and V with purity greater than 99.5% and 45 μm particle size (Alpha Aesar) were mixed inside a glove box, with argon atmosphere to prevent oxidation, to obtain the CuxCrFeTiV (x = 0.21, 0.44, 1 and 1.7 molar ratio) compositions . The powders were mechanically alloyed in a planetary ball mill, PM 400 MA type, with stainless steel balls and vials. The balls to powder ratio was 10:1, and the milling occurred for an effective time up to 20 hours, at 380 rpm. A solution of ethanol anhydrous was used as process control agent, to prevent heating and sticking to the balls. The as-milled powders were consolidated by spark-plasma sintering

4

(SPS) in an FCT Systeme Gmbh sintering machine, at a pressure of 65 MPa and temperature of 1178 K for a holding time of 5 minutes.

3.2 Irradiation

The polished surface of one sintered alloy, the CuCrFeTiV equiatomic composition, was irradiated at room temperature with a 300 keV Ar+ beam to a fluence of 3 x 1018 at/cm2. The region affected by Ar+ irradiation is limited to ~300 nm in depth.

3.3 Annealing

W-based high entropy alloys

An annealing was performed on the W20TaCrNbV sample in a vacuum tube furnace at 1100 oC for 72 hours,

under a pressure of 3x10−6 mbar. Cu-based high entropy alloys

For the Cu20CrFeTiV sample, the same apparatus was used, in the same vacuum conditions, but for three different conditions: 2 h at 550 oC, 8 h at 550 oC, and 8 h at 700 oC.

3.4 Microstructure analysis

Metallographic preparation of the samples was performed by grinding with SiC paper and polishing with diamond suspensions (6 µm, 3 µm and 1 µm) and fine polished with colloidal silica suspension (0.2 µm). The microstructures were observed before and after irradiation by secondary electrons (SE) and backscattered electrons (BSE), using a JEOL JSM-7001F field emission gun scanning electron microscope equipped with an Oxford energy dispersive X-ray spectroscopy (EDS) system. After the irradiation and in order to observe the irradiation damage, samples were observed under a tilt of 50º. 3.5 RBS

The RBS measurements were performed in the Cu20CrFeTiV specimen with a 2 MeV 4He+ beam from a 2.5 MV Van de Graaff accelerator, at room temperature and a pressure of about 10−6 mbar.

3.6 Thermal Diffusivity

After consolidation, the Cu20CrFeTiV specimen with a 1cm diameter was investigated in terms of thermal diffusivity, using a Netzsch LFA457 Microflash apparatus in the range of 20 oC up to 1000 °C.

3.7 X-ray diffraction

3.7.1 Powder X-ray Diffraction

Powder X-ray diffraction was collected/performed at room temperature with monochromatic Cu Kα1 radiation

( = 0.15408 nm) using an Inel CPS 120 diffractometer. The powders were prepared in a glove box, under argon atmosphere, by deposition onto a silicon wafer and agglomeration with acetone; the analysis were performed in the 2θ range from 10° to 80° degrees, with 10 seconds scan steps of 0.04° width. Analysis and phase identification was simulated using Powder Cell software [44], and the lattice parameters were extracted from Pearson’s Crystal Data software [27].

W-based high entropy alloys

After 2 hours, the samples were analysed in the said apparatus to evaluate the degree of solid solution formation. Following the 2-hour milling, there was consolidation through PPS.

Cu-based high entropy alloys

The powder mixtures were analysed after 2 hours of milling, to ascertain the degree of solid solution formation. This process was repeated after 6 hours, 12 hours and 20 hours of milling time. After this period, the powder was consolidated through SPS.

3.7.2 Grazing-Incidence Diffraction

After consolidation the samples of both systems were studied in near grazing X-ray incidence, with a Bruker D8 AXS diffractometer using a G�̈�bel mirror. The incidence angle was 3o to the sample surface, the scan steps of 0.02° width and 6 seconds duration, in the 2θ range between 30o-100o.

4. Results and Discussions

4.1 W-based high entropy alloys

Table 1 displays the δ, VEC, ΔHmix, ΔSmix and χ value calculated for W20TaCrNbV and for W30TaCrNbV.

Table 1 - Values of δ, VEC, ΔHmix, ΔSmix and χ respectively calculated

for both alloys. HEA δ VEC ΔHmix/

/kJmol−1

ΔSmix/

/JK−1mol−1

Δχ

W20TaCrNbV 5,096 5,400 -5,280 13,381 0,310

W30TaCrNbV 4,768 5,475 -5,355 13,147 0,353

5

The VEC values for the W-based HEAs are around 5.4-5.5 which predict BCC phases [45] in the microstructure.

Figure 2 presents the experimental diffractograms of (a) the mixture of the raw powders, (b) as milled powder mixture, and finally (c) the consolidated W20TaCrNbV high entropy alloy, with a corresponding BCC simulation performed with the Powder Cell software code (as referred to in chapter 3.7.1).

Figure 3 shows the X-ray diffraction data for the consolidated W20TaCrNbV together with the two BCC simulations. The diffractogram reveals the presence of one BCC structure with lattice parameter a = 0.3156 nm and a second BCC with a = 0.3345 nm. Additional low-intensity diffraction peaks were identified as tantalum oxide (TaO2), tantalum carbide (Ta2C) and vanadium oxide (VO).

Figure 3 - Diffraction pattern of the (a) W20TaCrNbV consolidated alloy, with the respective simulations indexed. (a) consolidated

W20TaCrNbV alloy, (b) BCC_1 simulation with a = 0.3156 nm and (c) BCC_2 simulation with a = 0.3345 nm.

The results for the W30TaCrNbV are very similar to the results shown for the equiatomic sample as can be depicted below in Figure 4.

Figure 4 - Diffractograms of the W30TaCrNbV alloy: (a) mixture of raw

powders, (b) as-milled powders and (c) consolidated W30TaCrNbV alloy, together with (d) BCC simulation by the Powder Cell software

code. Carbide WC is identified as *.

Figure 5 displays the microstructure of the as-

sintered (a) and (c) W20TaCrNbV and (b) and (d) W30TaCrNbV, samples as seen by SEM.

The micrographs of the W20TaCrNbV microstructure,

depicted in Figures 5 (a) and (c), show the presence of three phases: a majority phase, identified as A and two minor phases identified as B and C. Moreover, some intergranular precipitates over intergranular regions were observed (see circles in Figure 5 (c)). A similar microstructure was found for W30TaCrNbV where four phases were identified, namely a majority phase D and

B

A

D

A

C

F

G

D

E

(d)

(b) (a)

(c)

1 µm

1 µm 1 µm

1 µm

B

Figure 5 - BSE images of the microstructure of the (a) low and (c) high magnification of W20TaCrNbV sample and (b) low and (d)

high magnification of W30TaCrNbV alloys.

Figure 2 - Diffractograms of the W20TaCrNbV alloy: (a) mixture of raw powders, (b) as-milled powders and (c) consolidated W20TaCrNbV

alloy, together with (d) BCC simulation by the Powder Cell software code. Carbide WC is identified as *.

6

three minor phases E, F and G, as can be seen in Figures 5 (b) and (d).

The corresponding EDS elemental maps for the equiatomic WTaCrNbV are shown in Figure 6. According to the EDS results, phase A phase shows the presence of the elements W, Ta, Nb (along with Cr and V), suggesting there is a random distribution of these elements in solid solution, alongside Nb and Ta richer inclusions ((clearly seen pointed at by the arrows in Figure 6). On the other hand, phase C reveals a depletion in W, appearing with less brightness, eluding that the bulk concentration of heavy elements is localized in the majority regions (A). Moreover, the darkness and the depletion in the heavier elements, through BSE, of phase B point to the presence of an eventual oxide and/or carbide contamination. Additionally, in the B phase, it is noticeable the presence of oxygen in the EDS map of Figure 5 (g), alongside with tantalum and vanadium.

Those results indicate the possible formation of tantalum oxides and/or carbides, consistent with the XRD analysis previously presented.

In order to attempt to increase the chemical homogeneity in the microstructure and its stability, the consolidated sample of W20TaCrNbV was heat treated at 1100 oC for 72 h, under vacuum conditions of around 1x10-4 mbar. After the annealing (Figure 20 (b)), the already mentioned precipitates from Figure 4 form an almost-continuous phase around the grain boundary. These fine precipitates are embedded in the intergranular zone, ranging in size from approximately 0.1 µm to 0.5 µm.

4.2 Cu-based high entropy alloys

The δ, VEC, ΔHmix, ΔSmix and χ were calculated for Cu5CrFeTiV, Cu10CrFeTiV Cu20CrFeTiV and for Cu30CrFeTiV and are listed below, in Table 2.

Table 2 - Values of δ, VEC, ΔHmix, ΔSmix and χ respectively calculated

for the four alloys.

The calculations for the HEA compositions of Table 2,

predict BCC solid solutions dominating all the compositions except for the one richer in Cu (with 30 at. % Cu) which is predicted to have both.

The diffractograms taken from Cu20CrFeTiV at different stages of milling, between 0 h to 20 h, are presented in Figure 7. The as-mixed powder (0 h of milling) shows the presence of the pure elements, Cu, V as well as TiO0,2, Cr3O and FeO. Oxidation occurred in the mixture since it is possible to identify three oxides TiO0,2,

FeO, and the presence of Cr3O. The results evidence a peak broadening and weakening which increases as the milling time increases (Figure 7 (a) and (e)). Moreover, after 6 h of milling the peaks of elemental powders have disappeared and new phases are formed.

Figure 7 - X-ray diffractograms of the Cu20CrFeTiV alloy: (a) raw powder mixture; as milled powders after (b) 2 hours, (c) 6 hours, (d) 12 hours, and (e) 20 hours, and (f) consolidated Cu20CrFeTiV alloy.

Figure 8 exhibits the diffractogram of the as-sintered Cu20CrFeTiV together with simulations of BCC and FCC crystal structure. The diffractogram reveals a major peak of a BCC crystal structure (Figure 8 (b)) assigned with a lattice parameter of 0.2879 nm, and minor peaks of a FCC

HEA δ VEC ΔHmix/

/kJmol−1

ΔSmix/

/JKmol−1

Δχ

Cu5CrFeTiV 6.646 6.012 -7.125 12.600 0.120

Cu10CrFeTiV 6.492 6.275 -5.400 13.076 0.131

Cu20CrFeTiV 6.172 6.800 -2.400 13.381 0.133

Cu30CrFeTiV 5.844 7.325 9.853×10-16 13.147 0.139

(a) Cr

(g) O (f) Ta (e) V

C

B

(b) Nb (c) W

A

(d)

Figure 6 – EDS map performed on the (d) W20TaCrNbV sample, and on the constitutional elements of this material (a) Cr-Lα, (b) Nb-Lα,

(c) W-Lα, (e) V-Lα, (d) Ta-Lα and (g) O-Kα.

7

crystal structure (Figure 8 (c)) although with a lattice parameter of 0.3626 nm.

The samples with lower Cu concentrations are very

similar and present porosity. Two different phases, noted as X and Y, can be readily identified in Figure 9. The corresponding EDS elemental maps for the equiatomic CuCrFeTiV composition are shown in Figure 10. According to the EDS results, region A is chromium rich while region Y is copper rich. Moreover, in the EDS elemental maps is possible to observe another phase, indicated as Z in Figure 10. The results evidence a smooth transition between region A and region C, which suggests a relation between both, as a solid solution with increasing atomic concentrations of Cr and Fe and decreasing of Cu. The presence of a copper rich phase (phase Y in Figure 10) can be attributed to the high enthalpy of mixture of copper with the other transition

elements (it is only negative, −9 kJ/mol, for the Cu-Ti pair), which indicates a weaker binding force between Cu and the other elements, favouring segregation as opposed to mixing.

Nevertheless, the high standard deviation results of

the EDS quantification, and the higher atomic percentage of Fe and Cr in X phases in detriment of the expected alloys’ composition, arise the possibility of a contamination. To investigate this possibility, a PIXE analysis, with 2 MeV protons, was performed in the equiatomic sample, which yielded the results presented in Table 3.

Aiming at evaluating the potential of the Cu-based HEAs under study for use as thermal barriers, the thermal diffusivity of three CuxCrFeTiV specimens/compositions (with x = 5, 10, and 30 at. % Cu) was measured in the temperature range 0 – 1020 oC. It is noteworthy that the diffusivity has a minimum for all samples, occurring in the range of 600-700 oC, as can be depicted in Figure 11.

The values obtained for the thermal diffusivities of the Cu-based HEA compositions examined are all significantly lower than those of either the divertor material, W (thermal diffusivity 50-30 mm2/s [46]), or the heat exhaust piping, CuCrZr (85-95 mm2/s [47]) , a requirement and indication of its potential usefulness as thermal barrier.

Elements PPM Atomic Percentage

Fe 580022,5 56.7

Cr 163190,6 17.1

V 81426.8 8.7

Cu 91142,5 7.8

Ti 84234.5 9.6

Figure 8 - Diffraction pattern of the (a) Cu20CrFeTiV consolidated alloy, with the respective (b) BCC and (c) FCC simulations.

(a) (b)

(d) (c)

X

Z

X X

Y

Y

Y

X

Z

1 µm 10 µm

1 µm 1 µm

Figure 9 - SEM images, in BSE mode, of the microstructure of the (a) Cu5CrFeTiV, (b) Cu10CrFeTiV (c) Cu20CrFeTiV (d) and Cu30CrFeTiV

alloys, in BSE mode, with the respective phase identification.

Table 3 - Atomic fractions of each element measured through PIXE, with 2 MeV protons.

(c)

(b) Cr

(d) Fe

(a) Cu

(e) Ti (f) V

X Y

Figure 10 - EDS map performed on the (c) Cu20CrFeTiV sample, and on the constitutional elements of this material (a) Cu-Kα, (b) Cr-Kα,

(d) Fe-Kα, (e)Ti-Kα, (f) V-Kα.

8

The annealings were conducted by increasing

temperature and/or time in order to assess the thermal stability of the system. The composition used was the equiatomic (Cu20CrFeTiV) and the annealing conditions – temperature and duration – were 550 oC for 2 hours (b), 550 oC for 8 hours (c), and 700 oC for 2 hours (d), as can be depicted in Figure 12, at a vacuum pressure of roughly 1×10-5 mbar.

However, annealing at 700 oC promoted significant changes in the microstructure; the material experienced a high degree of decomposition of the Cu-rich phase (Y), resulting in the dissolution of that phase into the matrix. Further TEM analysis will be necessary to enlighten the evolution of microstructure of these alloys.

In order to collect data about the irradiation performed on these specimens, in particular detect and profile argon implanted during irradiation, RBS spectra were collected from one as-sintered Cu20CrFeTiV, and

from samples irradiated with 31018 at/cm2 and 300 keV Ar+ ions. The superpositions of RBS spectra taken from samples in the above given conditions are displayed in Figure 13.

The correlation between the intrinsic properties of

these HEAs and the evolution of radiation-induced defects is still unresolved.

5. Conclusions

In this work, WxTaCrNbV (x = 20 and 30 for 20 and 30 at. % W) and CuxCrFeTiV (x = 5, 10, 20 and 30 for 5, 10, 20 and 30 at. %) high entropy alloys have been produced, characterized and tested with the aim at their potential use as thermal barriers.

W-based high entropy alloys

The VEC values for the W-based HEAs are around 5.4-5.5 which predict BCC phases [45], in accordance with the experimental findings.

X-ray diffraction analysis of the W-system suggests that it may be described as a majority-multielement BCC structure, with a secondary BCC structure of smaller lattice parameter but very similar composition. The the presence of sparse Ta2C, VO and TaO2 in the matrix, were disclosed.

The increase in tungsten contents (from 20 at.% to 30 at.%) did not promote allotropic transformations in the microstructure.

The alloy with equiatomic composition was found stable up to 1100 oC.

Cu-based high entropy alloys

The VEC values for the Cu-based HEAs with Cu contents up to the equiatomic composition (20 at.% in Cu) lay between 6.0 and 6.8 which predict [32] a

Figure 21 - Thermal diffusivity of the alloys identified in the inset, measured by the laser flash method.

(a)

(c) 500 oC, 8 hours

(b) 500 oC, 2 hours

X

Y

X

X

Z

X

Y

Z

Y

Z

(d) 700 oC, 2 hours

1 µm

1 µm

1 µm

1 µm

Y

X

Figure 12 - BSE images of the (a) as-cast Cu20CrFeTiV alloy and the alloys annealed at (b) 550 oC for 2 hours, (c) 550 oC for 8 hours and

(d) 700 oC for 2 hours.

(a) (b)

Figure 13 - RBS spectra taken from Cu20CrFeTiV samples with 2 MeV 4He+ ions, in as-cast unirradiated and irradiated conditions,

at scattering angles of (a) 165o and (b) 140o.

9

dominant phase of BCC nature. Only for the 30 at.% Cu composition BCC and FCC should coexist (VEC = 7.3).

The experimental results showed accordingly that the equiatomic system could be described as a majority-multielement BCC, with a minor FCC structure. The high standard deviations in EDS quantifications and higher contents of Fe and Cr in the EDS maps, led to the perception of contamination from the milling equipment, later confirmed by PIXE analysis.

The increase in copper contents in the samples led to a concomitant increase in the volume fraction of copper rich structures, promoting phase separation, which is likely to be associated with the high binary enthalpy of mixture of copper with the remaining elements.

The Cu-system’s equiatomic sample was heat treated at 550 oC for 2 hours and 8 hours, and at 700 oC for 2 hours. In the ones treated at 550 oC no microstructure changes occurred. However, despite the main phase showing microstructural stability up to 700 oC, an almost complete disintegration of the copper-rich phase into the matrix occurred.

Thermal diffusivity showed an almost constant value of approximately 3 mm2/s – 5.5 mm2/s, with noticeable minima in the range 600 oC to 700 oC. This suggests a phase transition (either of magnetic and/or of allotropic nature).

Irradiation of equiatomic Cu20CrFeTiV was carried out aiming at preparing these systems for assessing the role of radiation defects in the mechanical and thermal properties and stability of the alloys.

Future work

Future work concerns deeper analysis of particular mechanisms, new proposals to try different materials, or simply curiosity.

The following ideas and methods ought to be tested and performed, as they would complement this labour as well as introduce and strengthen the scientific knowledge around this new-born and still under development HEA world:

1) Concerning the irradiation with Ar+ ions: evaluation of the nature and distribution of defects and the associated microstructural transformations in the Cu20CrFeTiV and all other compositions produced.

2) Concerning the thermal diffusivity measurements: an extensive study should be carried out, extending to all the compositions produced, and seeking for the rationale behind the values that were measured.

3) Concerning the contaminations in the copper-based system: XRD and EDS examination after

all stages of milling should be done, along with evaluation of the degradation rate of the milling equipment under use.

4) Concerning high-heat flux tests: given the fact that these alloys must be placed in-between tungsten monoblocks and CuCrZr cooling tubes, an interesting idea would be to investigate the response behaviour of the three materials to an environment similar to that prevailing in operation of a nuclear fusion reactor.

5) Concerning tensile stress-strain and hardness tests: these methods would allow for a more complete evaluation of the mechanical properties and give a broader notion of its behaviour in extreme conditions which is still fairly unknown in the literature.

Despite all the challenges facing their production and characterization, HEAs were found very promising across the literature, due to their enhanced properties, and the findings in this work contribute to supporting this view. These properties foreshadow new discoveries and inventions, which may ultimately lead to new applications.

References

[1] R. Liu et al., “Microwave sintering of W/Cu functionally graded

materials,” J. Nucl. Mater., vol. 431, no. 1–3, pp. 196–201, 2012.

[2] A. Hasegawa, M. Fukuda, T. Tanno, and S. Nogami, “Neutron Irradiation Behavior of Tungsten,” Mater. Trans., vol. 54, no. 4, pp. 466–471, 2013.

[3] H. Search, C. Journals, A. Contact, M. Iopscience, and I. P. Address, “Status of R & D activities on materials for fusion power reactors,” vol. 696.

[4] P. Gumbsch, “Brittle fracture and the brittle-to-ductile transition of tungsten,” J. Nucl. Mater., vol. 323, no. 2–3, pp. 304–312, 2003.

[5] A. G. Magdalena Galatanu, Monica Enculescu, George Ruiu, Bogdan Popescu, “nao referenciar assim Cu-based composites as thermal barrier materials in DEMO divertor components,” --, no. 633053, pp. 3–6, 2018.

[6] V. Philipps, “Tungsten as material for plasma-facing components in fusion devices,” in Journal of Nuclear Materials, 2011, vol. 415, no. 1

SUPPL, p. S2.

[7] S. J. Zinkle, “Applicability of copper alloys for DEMO high heat flux components,” Phys. Scr., vol. T167, no. T167, p. 014004, 2016.

[8] J. H. You, “Copper matrix composites as heat sink materials for water-cooled divertor target,” Nucl. Mater. Energy, vol. 5, pp. 7–18, 2015.

[9] J. Boscary et al., “Development activities of the high heat flux scraper element,” Fusion Eng. Des., vol. 98–99, pp. 1231–1234, 2015.

[10] V. Barabash et al., “Materials challenges for ITER - Current status and future activities,” J. Nucl. Mater., vol. 367-370 A, no. SPEC. ISS., pp. 21–32, 2007.

[11] X. Zhang, Q. Yan, S. Lang, M. Xia, X. Liu, and C. Ge, “Thermal shock and fatigue resistance of tungsten materials under transient heat loading,” J. Nucl. Mater., vol. 455, no. 1, pp. 537–543, 2014.

[12] D. Stork et al., “Developing structural, high-heat flux and plasma facing materials for a near-term DEMO fusion power plant: The EU assessment,” J. Nucl. Mater., vol. 455, no. 1–3, pp. 277–291, 2014.

[13] A. Li-Puma, M. Richou, P. Magaud, M. Missirlian, E. Visca, and V. P.

10

Ridolfini, “Potential and limits of water cooled divertor concepts based on monoblock design as possible candidates for a DEMO reactor,” Fusion Eng. Des., vol. 88, no. 9–10, pp. 1836–1843, 2013.

[14] J. W. Yeh et al., “Nanostructured high-entropy alloys with multiple principal elements: Novel alloy design concepts and outcomes,” Adv.

Eng. Mater., vol. 6, no. 5, pp. 299-303+274, 2004.

[15] C.-J. Tong et al., “Microstructure characterization of Al x CoCrCuFeNi high-entropy alloy system with multiprincipal elements,” Metall. Mater. Trans. A, vol. 36, no. 4, pp. 881–893, Apr. 2005.

[16] C.-Y. Hsu, J.-W. Yeh, S.-K. Chen, and T.-T. Shun, “Wear resistance and high-temperature compression strength of Fcc CuCoNiCrAl0.5Fe alloy with boron addition,” Metall. Mater. Trans. A, vol. 35, no. 5, pp. 1465–1469, May 2004.

[17] P.-K. Huang, J.-W. Yeh, T.-T. Shun, and S.-K. Chen, “Multi-Principal-Element Alloys with Improved Oxidation and Wear Resistance for Thermal Spray Coating,” Adv. Eng. Mater., vol. 6, no. 12, pp. 74–78, Feb. 2004.

[18] Y. Y. Chen, T. Duval, U. D. Hung, J. W. Yeh, and H. C. Shih, “Microstructure and electrochemical properties of high entropy alloys—a comparison with type-304 stainless steel,” Corros. Sci., vol.

47, no. 9, pp. 2257–2279, Sep. 2005.

[19] Y. Y. Chen, U. T. Hong, H. C. Shih, J. W. Yeh, and T. Duval, “Electrochemical kinetics of the high entropy alloys in aqueous environments—a comparison with type 304 stainless steel,” Corros. Sci., vol. 47, no. 11, pp. 2679–2699, Nov. 2005.

[20] M.-H. Tsai et al., “Thermal Stability and Performance of NbSiTaTiZr High-Entropy Alloy Barrier for Copper Metallization,” J. Electrochem. Soc., vol. 158, no. 11, p. H1161, 2011.

[21] C.-L. Lu, S.-Y. Lu, J.-W. Yeh, and W.-K. Hsu, “Thermal expansion and enhanced heat transfer in high-entropy alloys,” J. Appl. Crystallogr., vol. 46, no. 3, pp. 736–739, Jun. 2013.

[22] T. Egami, W. Guo, P. D. Rack, and T. Nagase, “Irradiation resistance of multicomponent alloys,” Metall. Mater. Trans. A Phys. Metall. Mater. Sci., vol. 45, no. 1, pp. 180–183, 2014.

[23] B. S. Murty, <B.S. Murty, Jien-Wei Yeh and S. Ranganathan (Auth.)-High Entropy Alloys-Butterworth Heinemann, Butterworth-Heinemann (2014).pdf>. .

[24] Y. Zhang, Y. J. Zhou, J. P. Lin, G. L. Chen, and P. K. Liaw, “Solid-solution phase formation rules for multi-component alloys,” Adv.

Eng. Mater., vol. 10, no. 6, pp. 534–538, 2008.

[25] S. Guo and C. T. Liu, “Phase stability in high entropy alloys: Formation of solid-solution phase or amorphous phase,” Prog. Nat. Sci. Mater. Int., vol. 21, no. 6, pp. 433–446, 2011.

[26] T. Hirai et al., “Status of technology R&D for the ITER tungsten

divertor monoblock,” J. Nucl. Mater., vol. 463, pp. 1248–1251, 2015.

[27] “The International Centre for Diffraction Data - ICDD.” [Online]. Available: http://www.icdd.com/. [Accessed: 29-May-2019].

[28] M. Dias et al., “Consolidation of W-Ta composites: Hot isostatic pressing and spark and pulse plasma sintering,” Fusion Eng. Des., vol.

98–99, pp. 1950–1955, 2015.

[29] C. S. Liu et al., “Tungsten-zirconium carbide-rhenium alloys with extraordinary thermal stability,” Fusion Eng. Des., vol. 106, pp. 56–62, 2016.

[30] K. Arshad et al., “Influence of vanadium precursor powder size on microstructures and properties of W-V alloy,” Int. J. Refract. Met. Hard Mater., vol. 50, pp. 59–64, 2015.

[31] J. H. You, “A review on two previous divertor target concepts for DEMO: Mutual impact between structural design requirements and

materials performance,” Nucl. Fusion, vol. 55, no. 11, 2015.

[32] D. Hancock et al., “Testing candidate interlayers for an enhanced water-cooled divertor target,” Fusion Eng. Des., vol. 98–99, pp. 1323–1327, 2015.

[33] G. Pintsuk, S. E. Brünings, J. E. Döring, J. Linke, I. Smid, and L. Xue, “Development of W/Cu-functionally graded materials,” Fusion Eng. Des., vol. 66–68, pp. 237–240, 2003.

[34] M. Schöbel et al., “Thermal fatigue damage in monofilament reinforced copper for heat sink applications in divertor elements,” J.

Nucl. Mater., vol. 409, no. 3, pp. 225–234, 2011.

[35] O. N. Senkov, G. B. Wilks, D. B. Miracle, C. P. Chuang, and P. K. Liaw, “Refractory high-entropy alloys,” Intermetallics, vol. 18, no. 9, pp. 1758–1765, 2010.

[36] M. S. Anzorena et al., “Characterization and modeling of a MoTaVWZr high entropy alloy,” Mater. Des., vol. 111, pp. 382–388, 2016.

[37] X. J. Yao, X. F. Shi, Y. P. Wang, G. Y. Gan, and B. Y. Tang, “The mechanical properties of high entropy (-like) alloy Wx(TaTiVCr)1-xvia first-principles calculations,” Fusion Eng. Des., vol. 137, no. August,

pp. 35–42, 2018.

[38] Y. J. Hsu, W. C. Chiang, and J. K. Wu, “Corrosion behavior of FeCoNiCrCux high-entropy alloys in 3.5% sodium chloride solution,” Mater. Chem. Phys., vol. 92, no. 1, pp. 112–117, 2005.

[39] M. Galatanu, M. Enculescu, G. Ruiu, B. Popescu, and A. Galatanu, “Cu-based composites as thermal barrier materials in DEMO divertor components,” Fusion Eng. Des., vol. 124, pp. 1131–1134, 2017.

[40] M. Murali, S. P. K. Babu, B. J. Krishna, and A. Vallimanalan, “Synthesis and characterization of AlCoCrCuFeZnx high-entropy alloy by mechanical alloying,” Prog. Nat. Sci. Mater. Int., vol. 26, no. 4, pp.

380–384, 2016.

[41] C. M. Lin and H. L. Tsai, “Effect of annealing treatment on microstructure and properties of high-entropy FeCoNiCrCu0.5 alloy,” Mater. Chem. Phys., vol. 128, no. 1–2, pp. 50–56, 2011.

[42] M. Dias, “Materials for fusion applications,” Acta Polytech., vol. 53,

no. 2, pp. 197–212, 2013.

[43] C. Hsu, J. Yeh, S. Chen, and T. Shun, “Wear Resistance and High-Temperature Compression Strength of Fcc CuCoNiCrAl ^ s ...,” Metall. Mater. Transcations A, vol. 35A, no. May, pp. 1465–1469, 2004.

[44] G. Nolze and W. Kraus, “BAM Berlin PowderCell.” [Online]. Available: http://www.ccp14.ac.uk/ccp/web-mirrors/powdcell/a_v/v_1/powder/e_cell.html. [Accessed: 29-May-2019].

[45] S. Guo, C. Ng, J. Lu, and C. T. Liu, “Effect of valence electron concentration on stability of fcc or bcc phase in high entropy alloys,” J. Appl. Phys., vol. 109, no. 10, 2011.

[46] C. Pradère, J. M. Goyhénèche, J. C. Batsale, S. Dilhaire, and R. Pailler, “Thermal diffusivity measurements on a single fiber with microscale diameter at very high temperature,” Int. J. Therm. Sci., vol. 45, no. 5,

pp. 443–451, 2006.

[47] J. N. Brooks et al., “Plasma-surface interaction issues of an all-metal ITER,” Nucl. Fusion, vol. 49, no. 3, 2009.

[48] IPFN, “European Shock-Tube for High Enthalpy Research Atmospheric Entry Science : A Critical Technology for Access to

Space,” 2012.