Recent Advances in Additive Manufacturing of High Entropy Alloys and Their Nuclear and Wear-Resistant Applications

: Alloying has been very common practice in materials engineering to fabricate metals of desirable properties for speciﬁc applications. Traditionally, a small amount of the desired material is added to the principal metal. However, a new alloying technique emerged in 2004 with the concept of adding several principal elements in or near equi-atomic concentrations. These are popularly known as high entropy alloys (HEAs) which can have a wide composition range. A vast area of this composition range is still unexplored. The HEAs research community is still trying to identify and characterize the behaviors of these alloys under different scenarios to develop high-performance materials with desired properties and make the next class of advanced materials. Over the years, understanding of the thermodynamics theories, phase stability and manufacturing methods of HEAs has improved. Moreover, HEAs have also shown retention of strength and relevant properties under extreme tribological conditions and radiation. Recent progresses in these ﬁelds are surveyed and discussed in this review with a focus on HEAs for use under extreme environments (i.e., wear and irradiation) and their fabrication using additive manufacturing. properties, wide composition ranges and higher probability to ﬁnd simpler microstructures enabled HEAs to gain rapidly growing attention from researchers. HEAs are considered one of three innovations in the alloying techniques along with bulk metallic glasses and metal rubbers [2].


The History of High Entropy Alloys
Since the first copper-based alloy was developed around 7000 years ago, numerous metallic alloys have been utilized in various applications [1]. In traditional alloying engineering, the principal metal is used as a matrix to incorporate other alloying elements as solute. In most cases, alloying has been done to improve the strength and hardness of ductile metals. Until now, around 30 alloy systems have been introduced, based on the principal element alloying concept [2]. Increasing demands for advanced materials under harsher environments led to the innovative alloying strategies which improved the performance of existing materials against high temperatures, impact, fatigue fracture, corrosion, or wear. Heat treatments have also been used along with alloying to tailor the properties of materials for desired applications. In the 1970s, a new class of materials, named intermetallics, were developed to increase the specific hot hardness. In 1980s, another class of materials named super-alloys were developed. Inconel, Waspalloy, Hastelloy, MP35N, MP98T, Rene alloys, TMS alloys and CMSX single crystal alloys are widely used commercial superalloys. Figure 1 shows how engineering materials evolved over human history. In the beginning of the 21st century, when the alloying technology reached maturity and so did the capability of materials for more advanced applications, a new alloying concept emerged. These alloys were initially called by several different names, such as multi-principal elements alloys, equi-molar alloys, equi-atomic ratio alloys, substitutional alloys and multicomponent alloys. The most common name of these alloys is high entropy alloys (HEAs) given by J. W. Yeh [3], because these alloys have higher mixing entropy in their liquid or solid solution states than any other alloying systems. Attractive structural entropy alloys (HEAs) given by J. W. Yeh [3], because these alloys have higher mixi entropy in their liquid or solid solution states than any other alloying systems. Attract structural properties, wide composition ranges and higher probability to find simpler m crostructures enabled HEAs to gain rapidly growing attention from researchers. HE are considered one of three innovations in the alloying techniques along with bulk met lic glasses and metal rubbers [2]. The Germen scientist Franz Karl Archard could be called the predecessor for HE research [2]. In the 18th century, Archard studied equi-mass multicomponent alloys co taining five to seven different elements from Fe, Sn, Pb, Zn, Bi, Ag, Co, Sb, As and Cu. tested these compositions for ductility, hardness, impact resistance, wear and density, e He published his work in a French book entitled, Recherches sur les propriétés des allia métalliques [4,5]. In 1981, Brian Cantor and his student Alain Vincent started to work Archard's idea again at University of Sussex in Sussex, England [6]. They explored vario equi-molar combinations out of 20 different elements and found that the CoCrFeMn alloy formed a single face-centered cubic (FCC) structure. In another independent search (MS thesis of National Tsing Hua University, Taiwan, 1996 [7]), J. W. Yeh dev oped different HEAs based on a concept that high entropy of an alloy system reduces number of phases that appear in the final product. S. Ranganathan is another notable oneering researcher in the field who wrote a review paper on the concept of HEAs a talked about the possibility of fabricating HEAs in 2003 [8]. In 2004, two independ studies by Cantor et al. [9] and Yeh et al. [3] introduced the concept of HEAs prope with the experimental results. They developed metallic alloys having nearly equi-atom composition of more than five elements.
The basic idea of HEAs is to use five or more principal elements in or near eq atomic compositions. According to the Gibbs phase rule, the number of phases increa with the increasing number of elements. Most of these phases are expected to be interm tallics due to their strong negative enthalpies. Binary/ternary phase diagrams also indic The Germen scientist Franz Karl Archard could be called the predecessor for HEAs research [2]. In the 18th century, Archard studied equi-mass multicomponent alloys containing five to seven different elements from Fe, Sn, Pb, Zn, Bi, Ag, Co, Sb, As and Cu. He tested these compositions for ductility, hardness, impact resistance, wear and density, etc. He published his work in a French book entitled, Recherches sur les propriétés des alliages métalliques [4,5]. In 1981, Brian Cantor and his student Alain Vincent started to work on Archard's idea again at University of Sussex in Sussex, England [6]. They explored various equi-molar combinations out of 20 different elements and found that the CoCrFeMnNi alloy formed a single face-centered cubic (FCC) structure. In another independent research (MS thesis of National Tsing Hua University, Taiwan, 1996 [7]), J. W. Yeh developed different HEAs based on a concept that high entropy of an alloy system reduces the number of phases that appear in the final product. S. Ranganathan is another notable pioneering researcher in the field who wrote a review paper on the concept of HEAs and talked about the possibility of fabricating HEAs in 2003 [8]. In 2004, two independent studies by Cantor et al. [9] and Yeh et al. [3] introduced the concept of HEAs properly with the experimental results. They developed metallic alloys having nearly equi-atomic composition of more than five elements.
The basic idea of HEAs is to use five or more principal elements in or near equi-atomic compositions. According to the Gibbs phase rule, the number of phases increases with the increasing number of elements. Most of these phases are expected to be intermetallics due to their strong negative enthalpies. Binary/ternary phase diagrams also indicate In this context, this paper discusses recent updates on the fabrication of HEAs by additive manufacturing (AM) and the HEAs for applications under extreme environments (i.e., wear behavior and nuclear applications). Moreover, unlike previous reviews on these topics, this review would provide more convenience to readers who have just stepped into this field as well, since the reviewed research publications on AM, wear behavior and nuclear applications of HEAs are listed in a detailed tabular form with their results. Section 2 discusses additive manufactured HEAs in terms of their composition, microstructure and their mechanical properties, such as ultimate tensile strength (UTS), tensile elongation (ε), yield strength (YS), hardness (H), compressive strength (CS), compressive yield strength (CYS) and the amount of compression (C). In Section 3, the behaviors of HEAs under ion irradiation are analyzed in terms of dislocation, microstructure, irradiation resistance, hardness, phase stability, swelling resistance and self-healing. Furthermore, the tribological studies of HEAs are surveyed in terms of HEAs content variation, particle reinforcement, media and nitriding/carburizing/sulfurizing, in comparison with conventional materials. The wear behaviors of HEAs at higher temperatures and oxides formation are also reviewed.

The Definitions of High Entropy Alloys
The first ever definition of HEA was given by Yeh et al. [3] as a class of alloys composed of five or more principal elements having concentration between 5% to 35% for each element. The second definition was also proposed by the same group [13]. In the second definition, the three categories of alloys were introduced on the basis of the configurational entropy: low entropy alloys (configurational entropy alloys (ΔSconf) ≤ 0.69R), medium entropy alloys (0.69R ≤ ΔSconf ≤ 1.61R) and high entropy alloys (ΔSconf ≥ 1.61R) [30], where R is the universal gas constant. Here, the low entropy alloys are mostly conventional alloys with one or two major elements and the medium entropy alloys have two to four major elements. The high entropy alloys contain five or more major elements. The second definition does not require equi-atomic composition. For example, Ti2ZrHfV0.5Mo0.2 [80], FeCoNiCrTi0.2 [81] and Al0.1CoCrFeNi [82,83] are categorized as In this context, this paper discusses recent updates on the fabrication of HEAs by additive manufacturing (AM) and the HEAs for applications under extreme environments (i.e., wear behavior and nuclear applications). Moreover, unlike previous reviews on these topics, this review would provide more convenience to readers who have just stepped into this field as well, since the reviewed research publications on AM, wear behavior and nuclear applications of HEAs are listed in a detailed tabular form with their results. Section 2 discusses additive manufactured HEAs in terms of their composition, microstructure and their mechanical properties, such as ultimate tensile strength (UTS), tensile elongation (ε), yield strength (YS), hardness (H), compressive strength (CS), compressive yield strength (CYS) and the amount of compression (C). In Section 3, the behaviors of HEAs under ion irradiation are analyzed in terms of dislocation, microstructure, irradiation resistance, hardness, phase stability, swelling resistance and self-healing. Furthermore, the tribological studies of HEAs are surveyed in terms of HEAs content variation, particle reinforcement, media and nitriding/carburizing/sulfurizing, in comparison with conventional materials. The wear behaviors of HEAs at higher temperatures and oxides formation are also reviewed.

The Definitions of High Entropy Alloys
The first ever definition of HEA was given by Yeh et al. [3] as a class of alloys composed of five or more principal elements having concentration between 5% to 35% for each element. The second definition was also proposed by the same group [13]. In the second definition, the three categories of alloys were introduced on the basis of the configurational entropy: low entropy alloys (configurational entropy alloys (∆S conf ) ≤ 0.69R), medium entropy alloys (0.69R ≤ ∆S conf ≤ 1.61R) and high entropy alloys (∆S conf ≥ 1.61R) [30], where R is the universal gas constant. Here, the low entropy alloys are mostly conventional alloys with one or two major elements and the medium entropy alloys have two to four major elements. The high entropy alloys contain five or more major elements. The second definition does not require equi-atomic composition. For example, Ti 2 ZrHfV 0.5 Mo 0.2 [80], FeCoNiCrTi 0.2 [81] and Al 0.1 CoCrFeNi [82,83] are categorized as HEAs according to the second definition.
Metals 2021, 11, 1980 5 of 47 Moreover, these definitions are not strict, and it is not clarified which one should be used to categorize an alloy. For example, an alloy having composition of 5% A, 5% B, 20% C, 35% D and 35% E has the configuration entropy of 1.36R according to Equation (1) derived from Boltzmann's entropy formula [30].
where c n is the atomic fraction of the nth element. In case of equi-atomic composition, Equation (1) reduces to [30]: For example, an alloy having 25 components with equi-atomic concentration has ∆S conf = Rln(n = 25) = 3.22R. This material has the concentration of each element out of the range suggested by the first definition (between 5% to 35%), but it has sufficiently high entropy according to the second definition [15].
Considering both definitions together may often be confusing. In the past, researchers have also limited HEAs to have equi-atomic compositions or single-phase microstructures [26]. HEAs lack a standard definition that embraces all possible conditions. Both definitions are used frequently but neither clarifies the conditions of its usage. Generally, a metallic alloy with multiple principal elements and high configurational entropy is considered as a HEA.

Background and Conventional Methods
Brian Cantor estimated the total number of possible metallic alloys with different compositions to be up to around 1078 [12]. This means many new alloys are yet to be discovered. For the manufacturing of HEAs, the initial synthesis strategy was to choose equi-atomic concentration of principle elements to maximize the entropy of the system. However, later, HEAs in non-equi-molar ratios were also developed for various applications. Arc melting was mostly preferred to produce HEAs thanks to its convenience, availability and simplicity. Furthermore, developing a HEA became more complex as more non-equi-atomic compositions were considered and several other manufacturing techniques were used. Alshataif et al. [84] covered almost all kinds of processing techniques used so far for HEAs synthesis. They detailed solid state processing (i.e., powder atomization methods, ball milling, cold/hot pressing, sintering, spark plasma sintering), liquid state processing (i.e., arc melting, vacuum induction melting, directional solidification, infiltration, electromagnetic stirring), thin film deposition (i.e., magnetron sputtering, pulsed laser deposition, plasma spray deposition) and additive manufacturing. Most of these manufacturing techniques are commercially available. That means most HEAs would not require a special manufacturing process and mass-producing HEAs would be possible with the existing alloying technologies and facilities.

Additive Manufacturing of HEAs
3-D printing in manufacturing industries, when properly applied, not only makes a design phase more efficient and economic but also brings thoughtful impacts on product design. Recent advances in additive manufacturing (AM) made it more influential throughout the supply chain which generates revenue as well [152]. The additive manufactured HEAs showed improvement in their mechanical properties in comparison to as-cast HEAs [153][154][155][156][157][158][159][160]. Higher cooling rates in AM processes help suppress diffusional phase transformation and increase the chemical homogeneity of HEAs [161]. Under certain circumstances, AM gives a better control over the material processing and helps tailor application-specific microstructures which become more important for the parts for applications under extreme environments. For example, it was demonstrated that fine and tailorable microstructures in HEAs were obtained using AM techniques [162][163][164][165][166][167][168][169], which implies AM can improve the mechanical performance of at least some HEAs. However, this may not be a trivial task as a good understanding of the AM technique and material behavior during the AM process is required [170].
AM of HEAs has been discussed briefly in a few review papers [51,161,171,172] and books [2,173]. Xiaopeng Li [161] discussed the requirements and challenges of AM of HEAs and bulk metallic glasses. Chen et al. [51] examined the microstructural evolution and mechanical properties of AM-processed CoCrFeNi, Al x CoCrFeNi, CoCrFeMnNi and Ti 25 Zr 50 Nb 50 Ta 25 . Fabricating HEAs by spark plasma sintering (SPS) and their property analyses were discussed in the book chapter "Spark Plasma Sintering of High Entropy Alloys" of [174]. SPS followed by mechanical alloying has largely been used to develop HEAs, which was reviewed in detail by Vaidya et al. [175]. Therefore, SPS studies are not included here.
The microstructures and mechanical behaviors of the HEAs produced by different AM processes are still under investigation by several research groups. The HEAs listed in Tables 1-3 mainly have either FCC or BCC microstructures except Co 20 Cr 15 Fe 40 Mn 20 Si 5 which has HCP. Improvement in mechanical properties was reported when HEAs were fabricated with AM [158,169,177,180,188,193,268]. These improvements are mostly attributed to grain refinement. Grain refinement in HEAs is claimed to be due to the high cooling rates as it happens in various other materials [158,200,230]. Moreover, the wear behavior [189,249], thermo-mechanical analysis [199,246], effect of annealing [98], creep behavior [183], residual stresses [232], corrosion behavior [176,198,226,228,231,241,249], strengthening mechanisms [153] and deformation mechanism [237] of additive manufactured HEAs have also been reported.
Particle reinforcement in a HEA matrix with AM has been an area of interest for many researchers lately who expect microstructure refinement and mechanical properties enhancement [123,177,208,212,220,221,[269][270][271][272][273][274][275][276][277][278][279][280]. Li et al. [177] introduced nano TiN ceramic particles in a CoCrFeMnNi matrix, which led to equiaxed grains of 5 µm. The same group [179] also fabricated the same composition with SLM followed by laser remelting and obtained ultrafine grains (80% grains less than 2 µm and 90% grains less than 3.5 µm). Song et al. [188] showed that the YS and ductility of CoCrFeNi increased by 25% and 34%, respectively, when doped with 1.8 at% nitrogen. Fu et al. [279] noticed that adding Ti-C-O particles into NbTaTiV increased the UTS, YS and fracture strain up to 2270 MPa, 1760 MPa and 11%, respectively. Amar et al. [239] added TiC into CoCrFeNiMn and found the YS and UTS increased from 300 MPa to 385 MPa and from 550 MPa to 723 MPa, respectively. Similarly, Li et al. [238] embedded WC particles into CoCrFeNiMn alloy and observed improvement in YS from 300 to 675 MPa and UTS from 550 to 845 MPa due to the formation of Cr 23 C 6 precipitates. Li et al. [181] noticed that TiC reinforcement CoCrFeNiMn gave the UTS of around 1100 MPa. Rogal et al. [271] increased the UTS of CoCrFeNiMn up to 1600 MPa by introducing nano-Al 2 O 3 particles. Carbon doping was attempted [154,155,180,186] to enhance the mechanical properties of HEAs. Peng et al. [208] added diamond particles into CoCrFeNi and found out the bending strength was 925MPa. Park et al. [212] added carbon into CoCrFeNiMn ((CoCrFeNiMn) 99 C 1 ) and noticed that the YS and UTS were~741 MPa and~874 MPa, respectively. Similarly, Kim et al. [221] also added carbon into CoCrFeNiMn in a ratio (CoCrFeNiMn) 100−x C x (x = 0.5-1.5). The YS for x = 0.5, 1, 1.5 was measured to be 653, 752 and 753 MPa respectively. The UTS for x = 0.5, 1, 1.5 was found to be 766 ± 318.5, 895 ± 22.3 and 911 ± 125.1 MPa, respectively. Shen et al. [258] discussed the effect of SiC particles added to CoCrFeNi. They noticed that adding SiC particles changed the microstructure from the FCC phase to the FCC/Cr 2 C 7 dual phase. The hardness and YS improved significantly from~139 HV to~310 HV and 142 MPa to~713 MPa, respectively.
Various HEAs have exhibited significant improvement in their mechanical properties after AM synthesis as compared to the as-cast structures of the same compositions [153,154]. Zhou et al. [155] reported that arc-melted CoCrFeNi had the YS of 225 MPa whereas SLM-manufactured CoCrFeNi had the YS of 656 MPa. Brif et al. [156] observed that SLMmanufactured CoCrFeNi showed noticeable improvement in YS from 188 MPa (as-cast) Metals 2021, 11, 1980 11 of 47 to 600 MPa and in UTS from 457 MPa (as-cast) to 745 MPa. Peyrouzet et al. [157] showed that the YS of Al 0.3 CoCrFeNi increased from 275 MPa (as-cast) to 730 MPa and the UTS from 502 MPa (as-cast) to 896 MPa when manufactured with SLM. The UTS of as-cast Al 0.3 CoCrFeNi was 522 MPa and it was increased to 878 MPa with SLM processing [158]. Arc-melted Al 0.5 CoCrFeNi had the YS of 334 MPa and the UTS of 709 MPa [159]. SLM increased the YS up to 579 MPa and the UTS up to 721 MPa [160].
Moreover, the CS of AlCrCuFeNi was 2052 MPa when fabricated with SLM and 1750 ± 15 MPa [281] with arc-melting. The hardness of AlCoCrCuFeNi improved from 500 to 710 Hv [194] by using SLM. The YS of AlMgScZrMn manufactured with arc melting, SPS, and SLM is 188 ± 2.3 MPa, 231 ± 3 Mpa and 394 Mpa respectively [195]. Agrawal et al. [197] reported that the YS of as-cast and SLM-printed Fe 40 Mn 20 Co 20 Cr 15 Si 5 was 420 ± 20 Mpa and 530 ± 40 Mpa, respectively. The YS of CoCrFeNiMn was 2.5 times higher (around 518 Mpa) [236] with DED in comparison to that of cast parts (209 Mpa) [282] at room temperature (RT). Furthermore, the as-cast AlCoCrFeNi had the UTS of 956 MPa, and the EBM specimen had the UTS of 1073 MPa [226]. Similarly, Fujieda et al. [228] reported that EBM-synthesized CoCrFeNiTi showed the improved tensile strength of around 1178 MPa, which is much stronger than various commercial high corrosion resistant materials such as duplex stainless steel: 655 MPa, super duplex stainless steel: 750-800 MPa and Ni-based super alloys (i.e., Alloy C276: 690 MPa, Alloy 718: 1275 MPa).
Refractory HEA NbMoTaW has shown a drastic reduction in grain size when made with AM. The average grain size of BCC phase was 200 µm in as-cast sample [283] and 13.4 µm in SLM-processed sample. Additionally, this alloy did not follow the rule of mixtures. Instead, it showed the cocktail effect for the hardness of the final structure. The hardness of Nb, Mo, Ta and W was in the range of 85-410 HV but the final hardness of SLM processed NbMoTaW was measured to be 826 HV [198]. Senkov et al. [284] commented that NbMoTaW did not have any abrupt hardness changes at high temperatures, consistently exhibiting better hardness properties than superalloys. Moreover, SLM-processed Ni 6 Cr 4 WFe 9 Ti (UTS = 972 MPa, YS = 742 MPa, ε = 12.2%) had~93% increase in YS,~50% increase in UTS, and~77% increase in tensile ductility as compared to the vacuum arc melted samples (UTS = 649 MPa, YS = 385 MPa, ε = 6.9%) [200,201].

Nuclear Applications
Nuclear energy is contributing to around 13% of electricity demand worldwide [285] with negligible carbon emission. The safety, reliability and economy of these nuclear power plants depends heavily on the performances of advanced structural materials under highenergy irradiation and elevated temperatures [286,287]. Radioactive waste handling units also require radiation-tolerant materials. Not to mention nuclear applications, radiationresistant materials are in great demand in medical and aerospace fields as well.
The typical range of operating temperatures of nuclear reactors spans from 350 to 900 • C as listed in Table 4 [288]. At high temperatures, several effects come into play such as thermal expansion, vacancy concentration, diffusion rate, phase transformation, precipitation, recovery, recrystallization, dislocation climb, creep, grain weakening/migration/growth, oxidation and intergranular oxygen dispersion. With conventional alloys, design strategies for nuclear reactor materials were mostly concerned with tuning the microstructures by various heat treatments, precipitation, cold working and solute atoms to get desired properties. HEAs, though, introduce the concept of modifying compositional complexity of the structural materials to make them suitable for nuclear applications. Currently, reduced activation ferritic/martensitic steels (RAFM) (e.g., F82H, EURO-FER 97), are the most popular option for irradiation-resistant structural materials. Oxide dispersion strengthened (ODS) RAFM steels (i.e., EUROFER 97 reinforced with 0.3 wt.% Y 2 O 3 particles), C/C, SiC/C, SiC/SiC, refractory metals/alloys (W, Cr), V and Ti-based alloys are also being used [289,290]. HEAs are considered to be potential candidates for nuclear applications [2,[291][292][293]. Yeh et al. [294] mentioned that HEAs are potential candidates for structural materials of the 4th generation nuclear reactor. Previously, the irradiation responses and defect behaviors [65,66], intrinsic transport properties [66], irradiation induced structural changes [295] of HEAs were reviewed. Building upon these reviews, this section mainly focuses on ion irradiation resistance of HEAs.
The majority of the previous ion irradiation studies on HEAs are listed in Table 5 where phases, irradiation conditions and important findings are summarized. These HEAs were studied under Ni, Au, Ag, Ar, He, Kr, or Xe ions irradiation. The most popular strategy to design single-phase HEAs of high irradiation resistance used elements having low activation or thermal neutron absorption cross section [296][297][298][299][300].   Neutrons generated from fission/fusion reactors induce atomic displacements in structural materials and may introduce point defects. Development of interstitial and vacancy defects will change local lattice parameters of the original phase. This will lead to deterioration of structural materials, namely, hardening, phase instability, irradiationinduced segregation, irradiation-induced creep, volumetric swelling and H/He embrittlement [290,348,349]. For instance, Yang et al. [332] noticed various defects (i.e., dislocation loops, long dislocations and stacking-fault tetrahedra) were induced by irradiation of 3 MeV Au ion on Al 0.1 CoCrFeNi but they did not observe void formation. They noticed that defect density decreased and defect size increased with the increasing temperature. Chen et al. [81] investigated the effect of irradiation of He + 275 KeV on FeCoNiCrTi 0.2 at 400 • C. They reported that high pressure He bubbles generated at the peak damage region and faulted dislocation loops (1/3<111>) formed. Perfect loops were rarely noticed and the size of the faulted loops was observed to be abnormally large. They also determined stalking fault energy (SFE) and reported the upper limit for SFE at 400 • C for FeCoNiCr and FeCoNiCrTi 0.2 was estimated approximately to be~80 mJ·m −2 for the largest radius of a faulted loop of~15 nm and~30 mJ·m −2 for the largest radius of a faulted loop of~55 nm, respectively. This showed that adding Ti significantly reduced the SFE.
Yang et al. [314] studied the irradiation behavior of CoCrNiFe with 3 MeV Ni 2+ ions at 500 • C. They investigated the defects (dislocation loops and void distribution) as a function of depth. The defects were found at the depth of 200-600 nm and 1100-2000 from the surface. Overall, the average defect diameter was measured to be <10 nm. Lu et al. [302] observed the irradiation behavior of FeNi, CoFeNi, CoCrFeNi and CoCrFeNiMn under 3 MeV Ni 2+ ions at 500 • C. In SEM images, they noticed faulted (1/3<111>) dislocation loops in all of these alloys as a result of irradiation. These loops increased with the increasing number of principal elements. Yang et al. [298] studied Al 0.3 CoCrFeNi under 3 MeV Au ion irradiation. Average dislocation loop size was found to be~12 nm at 250 • C and~32 nm at 500 • C. Dislocation loop density was~18 × 10 21 m −3 at 250 • C and~2 × 10 21 m −3 at 500 • C. Hence, the average defect size increased and defect density decreased with irradiation temperature.

Hardness
Various researchers claimed that ion irradiation increased the hardness of the HEAs [306,307,326,327,329,337,343] although there are a few cases that reported some deviant behaviors [80,336,341]. HEAs showed better resistance to hardening by irradiation than stainless steels [326,329]. For instance, Sadeghilaridjani et al. [337] reported that the hardness of HfTaTiVZr increased by 20% under Ni 2+ ion irradiation, but under the same condition, the hardness of SS304 increased by around 50%. Such superior resistance to irradiation hardening was attributed to the reduced mobility of the point defects due to sluggish diffusion and self-healing ability. Similarly, Tolstolutskayaet et al. [326] irra-diated CrFeNiMn, Cr 0.18 Fe 0.4 Ni 0.28 Mn 0.14 and Cr 0.18 Fe 0.28 Ni 0.27 Mn 0.28 with 1.4 MeV and found out that the hardness increased by 22-45%. They also reported that the hardness of austenitic steels, such as X18H10T and SS316, almost doubled under similar irradiation conditions.

Phase Stability
The microstructure of a material under irradiation could be affected by formation of defect clusters, dislocation loops, stacking faults, precipitates, voids, or He bubbles. These phenomena may even alter local chemical compositions. Such microstructural changes could lead to deterioration in properties such as conductivity, ductility, fracture toughness or creep strength. Irradiation damage also can cause material swelling, irradiation-induced creep (IIC), irradiation-assisted stress corrosion cracking or irradiation growth. The attractive properties of the HEAs typically come from specific phases acquired with suitable compositions. Moreover, there are various potential intermetallics compositions present in HEAs. Therefore, the phase stability analysis becomes of utmost importance. Taking advantage of recent advancement in microstructure characterization technologies, most of the studies mentioned in Table 5 focused on the phase stability and microstructural changes after ion irradiation. Phase stability after irradiation was observed for several HEAs such as CoCrFeNiMn, CoCrFeNiPd and high entropy metallic glasses such as ZrTiHfCuBe and ZrTiHfCuBeNi [340] and Al x CoCrFeNi (x = 0.1-1.5) [82,83]. Such high phase stability was mainly attributed to sluggish diffusion and high configurational entropy that kinetically restrains precipitation by reducing thermodynamic driving force. Moreover, Kumar et al. [327] reported that Fe 0.27 Ni 0.28 Mn 0.27 Cr 0.18 alloy exhibited higher phase stability and higher resistance to radiation swelling and void formation in comparison to austenitic stainless steels. No micro-void formed in CoCrFeNi irradiated by 1.5 MeV Ni 2+ ion. Thanks to its superior interface stability during irradiation, AlCrMoNbZr/ (AlCrMoNbZr)N was considered a promising candidate for an accident-tolerant fuel cladding material [299].
On the other hand, Wang et al. [308] noticed a reduction in crystallinity resulted from He ion irradiation on CoCrFeCuNi alloy. Yang et al. [298] observed that precipitation of the L12 phase was suppressed due to irradiation-induced ballistic mixing at temperatures less than or equal to 500 • C. Additionally, precipitation of the B2 phase was favored at 650 • C due to improved diffusion by irradiation. They found that the swelling resistance of CoCrFeNiMn was less than 0.5% for temperatures up to 680 • C. They attributed the improved resistance to the void-induced swelling of these alloys to the complex arrangements of different atoms in the lattice structure. Gandy et al. [343] reported irradiation-induced phase transformation in SiFeVCrMo from the tetragonal sigma phase to BCC. This BCC phase formed at high temperatures as well as after ion irradiation. Under irradiation of 3 MeV Ni 2+ ion, NiCoFeCr maintained its phase stability; however, Al 0.12 NiCoFeCr showed phase change. Al 0.12 NiCoFeCr microstructure changed from single FCC to FCC matrix, nanoprecipitates (i.e., Ni 3 Al) and ordered structure (i.e., L12) [318]. Irradiation of 3 MeV Ni 2+ ion on the CoCrFeNiMn alloy induced Mn depletion and Co/Ni enrichment at grain boundaries [303]. He et al. [309] studied the phase stability of CoCrFeNi, CoCrFeN-iMn and CoCrFeNiPd under electron radiation. They concluded that Cr, Fe, Mn and Pd elements were most likely to deplete and Co/Ni preferred to accumulate at defect clusters (i.e., dislocation loops). Atwani et al. [341] irradiated CrTaVW HEA with 1 MeV Kr 2+ ions at 800 • C. This HEA showed the segregation of Cr and V at the triple junction and grain boundaries after irradiation. Yang et al. [332] and Lu et al. [302] also noticed that Ni and Co tended to enrich, but Cr, Fe and Mn preferred to deplete at defect clusters in Al 0.1 CoCrFeNi under 3 MeV Au ion irradiation.

Irradiation-Induced Creep (IIC)
Irradiation-induced creep is another important aspect since nuclear reactors operate at high temperatures as shown in Table 5. Jawaharram et al. [301] studied both thermal creep and irradiation-induced creep (IIC) of Cantor alloy (CoCrFeMnNi). For IIC measurement, CoCrFeNiMn was irradiated with 2.6 Mev Ag 3+ ion in the temperature range of 23-650 • C. They reported that over the temperature range investigated, ICC was more dominant that thermal creep.

Swelling Resistance
Density reduction by volume increase due to formation of voids and defects is the mechanism behind material swelling. In order to maintain the structural integrity and mechanical strength, reasonably high swelling resistance is required for structural materials for nuclear applications. Several HEAs showed adequate swelling resistance under irradiation. Yang et al. [332] reported negligible void formation in Al 0.1 CoCrFeNi under 3 MeV Au ion irradiation. Moreover, Yang et al. [314] noticed that dislocation loops and void distribution varied as a function of depth in CoCrFeNi under 3 MeV Ni 2+ ion irradiation. Jin et al. [306] suggested that compositional complexity, including the number and type of components, be taken into consideration to improve swelling resistance. They concluded that adding Fe and Mn would be more effective to reduce swelling than adding Co and Cr. Their results showed that NiCoFeCrMn exhibited 40 times higher swelling resistance than Ni.

Self-Healing
HEAs showed the ability to absorb and heal radiation-induced damages. Self-healing of HEAs was explained by Egami et al. [350] in detail. Xia et al. [83] reported that defect clustering in disordered FCC or BCC happened in smaller size than in the ordered B2 phase. They attributed this effect to the reduced defect mobility and large atomic stress in the disordered phases which might have led to self-healing. Patel et al. [344] investigated the phase stability of V 2.5 Cr 1.2 WMoCo 0.04 under 5 MeV Au + ion irradiation. They noticed that 96% of the BCC phase of the as-cast alloy remained intact up to the irradiation dose of 42 displacement per atom (dpa). The remaining 4% converted to another BCC phase with a little larger lattice parameter. Their energy dispersive X-ray (EDX) analysis detected no element segregation. Such phase stability was attributed to the self-healing capability [350]. Tong et al. [305] concluded that the local lattice distortion could be relaxed by lattice expansion with low dose of irradiation. Sellami et al. [316] irradiated CoCrFeNi with 1.5 MeV and 21 MeV Ni Ions. They reported elastic strain values of~0.035% and 0% for 1.5 MeV Ni ion and 21 MeV Ni ion irradiations, respectively. They suggested that the complex composition of HEAs induced higher chemical disorder that reduced the mobility of defects generated by low energy (i.e., 1.5 MeV Ni) ion irradiation; therefore, a small elastic strain of~0.035% was obtained. Furthermore, when low energy irradiation CoCrFeNi alloy was further subjected to high-energy (i.e., 21 MeV Ni) irradiation, elastic strain relaxation (resulted into~0% elastic strain) was observed which could be attributed to the mechanisms such as defect annealing, recombination and rearrangement.

Miscellaneous
Lu et al. [80] reported an unexpected decrease in lattice parameter by 0.676% after He ion irradiation on Ti 2 ZrHfV 0.5 Mo 0.2 . Similarly, TiZrNbHfTa showed insignificant changes in hardness but a considerable increase in UTS and YS without loss of ductility [336]. Li et al. [328] noticed an increase in electrical resistivity of Fe 0.27 Mn 0.27 Ni 0.28 Cr 0.18 after neutron irradiation.

Wear Behavior
The wear properties of HEAs were studied mostly with pin/ball on a disc set up with antagonist materials such as Al 2 O 3 , steels (i.e., SKH51, GCr15, 100Cr6), Si 3 N 4 , SiC, ZrO 2 , 1Cr18Ni9Ti, BN, inconel-718 and WC. For lubrication, mostly dry conditions were used but some studies also used H 2 O 2 , deionized water and acid rain (pH = 2). Previously, Tsai and Yeh et al. [351], Kasar et al. [352], Senkov et al. [67], Sharma et al. [16], Zhang et al. [37], Li et al. [42], Menghani et al. [353] and Ayyagari et al. [354] discussed the wear behaviors of HEAs. In this review, we will analyze the tribological studies of HEAs in terms of HEAs content variation, particle reinforcement, media and nitriding/carburizing/sulfurizing, temperature effects and oxide formation. Table 6 provides the details of the compositions, microstructures, methods and results (i.e., wear rate or wear resistance, hardness, friction coefficient) of the wear studies performed so far on HEAs.

Content Variation
Tribological studies on HEAs are mostly conducted with variations in content of one element and finding the optimum concentration of the element for minimum wear rate [360,[363][364][365]371,[377][378][379][380]383,384,386,390,394,395,397,398,407,408,412,421,422,424,427,428,431,440,454,455,463,470,472,474]. Furthermore, the effect of Al content on wear properties and hardness has been studied more than any other metal. For instance, the wear resistance of CoCrFeNiCu [440], CoCrFeNiTi 0.5 [386], CoCrFeNiSi [421] and CoCrFeN-iMn [364] was improved with Al addition. This was primarily ascribed to the oxide formation and hardness increase due to phase transformation, grain refinement and precipitation. Similarly, Cui et al. [398] studied the effect of Al x CoCrFeNiMn (x = 0-0.75) coatings on 4Cr5MoSiV alloys. This alloy experienced FCC to FCC + BCC transition and grain refinement when Al content increased, which in turn raised the hardness from 224 HV to 344 HV. The wear weight loss of 4Cr5MoSiV coated with Al x CoCrFeNiMn with x = 0, 0.25, 0.5 and 0.75 was measured to be 6.0 mg, 4.1 mg, 3.2 mg and 1.1 mg, respectively. Under the similar wear condition, the amount of wear weight loss of the uncoated 4Cr5MoSiV was around 10 mg. Meanwhile, Gu et al. [455] analyzed Al x Mo 0.5 NbFeTiMn 2 (x = 1-2) coating on Q235 steel and found that the microhardness and wear were positively correlated to the Al content. As the Al content increased, the grain size reduced and the hardness and wear resistance increased. Kumar et al. [412] studied the wear behavior of Al 0.4 FeCrNiCo x (x = 0-1) under oil lubricated conditions. The hardness of this alloy reduced with the increasing Co content from 377 HV (x = 0) to 199 HV (x = 1). The wear resistance of Al 0.4 FeCrNiCo was found to show a minimum among all other conditions. This alloy did not show any phase transformation until 1000 • C.
Erdogan et al. [428] investigated the influence of the Al and Ti content on the hardness and wear resistance of Al x CoCrFeNiTi y (x = 0-0.5, y = 0-0.5). Increasing Al and Ti improved the hardness from 210 HV (for CoCrFeNi) to 859 HV (for Al 0.5 CoCrFeNiTi 0.5 ). The wear rate also reduced from 2.1 × 10 −13 mm 3 ·N −1 ·m −1 (for CoCrFeNi) to 0.25 × 10 −13 mm 3 ·N −1 ·m −1 (for Al 0.5 CoCrFeNiTi 0.5 ). CoCrFeNi initially had the FCC phase. The added Al induced and stabilized the BCC phase in the FCC matrix. Additionally, Al and Ti formed some beneficial intermetallic phases such as AlNi and AlNiTi, which improved the hardness and wear resistance. Similarly, Chuang et al. [431] examined the influence of the Al and Ti content on the wear performance of the Al x Co 1.5 CrFeNi 1.5 Ti y (x = 0-0.2, y = 0.5-1.0) alloy. They reported that the wear resistance of Co 1.5 CrFeNi 1.5 Ti and Al 0.2 Co 1.5 CrFeNi 1.5 Ti was at least twice that of SUJ2 and SKH51 with the similar hardness values. Li et al. [390] studied the wear resistance of CoCrFeNi 2 V 0.5 Ti x (x = 0.5-1.25). They observed that the wear resistance and the hardness improved with the Ti content until x = 0.75 and started to decrease thereafter. CoCrFeNi 2 V 0.5 Ti 0.75 had the maximum hardness of~960 HV and the minimum wear rate of 4.43 × 10 −5 mm 3 ·N −1 ·m −1 . Chen et al. [424] reported that the wear resistance of Al 0.5 CoCrFeNiCuTi x (x = 0-2) increased with the Ti content until x = 1, and thereafter, it started to decrease. Here, a small content of Ti did not have much impact on wear rate, when Ti content reached x = 0.4, the BCC phase began to form which improved the hardness and wear resistance. With a Ni content of x = 0.8-1.2, the CoCr-like phases formed, which improved the wear resistance. However, after x = 1, the Ti 2 Ni-like and ordered BCC phases started to appear and the alloy became harder with the increasing Ti content, which reduced the wear resistance.
Increasing the Cu content from x = 0 to x = 1 reduced the wear rate of CoCrFeNiCu x (x = 0-1) alloy from 2.3 × 10 −5 to 1.7 × 10 −5 mm 3 ·N −1 ·m −1 at RT and from 2.5 × 10 −5 to 1.3 × 10 −5 mm 3 ·N −1 ·m −1 at 600 • C [377]. The wear rate of CoCrFeNi (x = 0) increased with the increasing temperature. This resulted from the enhanced plastic deformation and abrasive wear due to thermal softening at higher temperatures. In contrast, the wear rate of CoCrFeNiCu (x = 1) decreased at higher temperatures. This was attributed to the self-lubricating effect of Cu. Cu also tended to form protective oxide layer which prevented direct metal to metal contact [481]. Hsu et al. [423] showed the improved wear resistance of Al 0.5 CoCrFeNiCuB x (x = 0-1) with boron addition. Al 0.5 CoCrFeNiCuB showed the hardness of 736 HV and the maximum wear resistance of 1.76 mm 3 ·N −1 ·m −1 which was higher than that of SKD61 cold-work mold (1.39 m·mm 3 ) and SUJ2 bearing (1.52 m·mm 3 ). The improved wear resistance was attributed to the boride formation. Similarly, Liu et al. [378] analyzed the wear resistance of CoCrFeNiB x (x = 0.5-1.5). They noticed that the increased boron reduced the wear volume from 19 down to 4.5 mm 3 . The hardness increased from 550 to 1025 HV 0.2 as well with increasing boron, which was attributed to the increase in the fraction of the hard boride phase with B content. The maximum hardness was measured to be 1025 HV 0.2 for FeCoCrNiB 1.5 which was higher than that of Q245R steels. Kumar et al. [408] reported that the wear rate of Al 0.4 FeCrNiCo x (x = 0-1 mol) increased with the increasing Co content from 0.81 × 10 −4 mm 3 ·N −1 ·m −1 (x = 0) to 1.86 × 10 −4 mm 3 ·N −1 ·m −1 (x = 1). This was attributed to the hardness decrease form 377 HV (x = 0) to 199 HV (x = 1). As hardness decreases the wear resistance also decreases according to the Archard's law [482]. Chen et al. [454] reported how the variation of V in A l0.5 CoCrFeNiCuV x (x = 0-1) affected the wear performance. Here, increasing V from x = 0.6 to x = 1.2 improved the wear resistance by 20%. However, increasing V beyond x = 1.2 did not show a considerable effect. This was in line with the hardness results of this HEA. The hardness increased while x increased from x = 0.4 to x = 1 exhibiting its maximum value at x = 1. The optimal V addition was suggested to be in the range of x = 1-1.2 to improve the wear resistance.
Furthermore, improvements in the wear resistance of CoCrFeNiNb x (x = 0-1.2) [379], (CuCrFeTiZn) 1-x Pb x (x = 0.05-0.2) [395], CoCrFeNiCu x (x = 0-1) [377], MoTa x NbVTi (x = 0.25-1) [474] were reported. Wu et al. [410] analyzed the wear resistance of the AlCoCrFeNiTi x (x = 0, 0.5) alloy. They noticed that AlCoCrFeNi had the plate-like disordered Fe-Cr rich BCC phase embedded in the ordered Al-Ni rich BCC phase. AlCoCrFeNiTi 0.5 had the similar microstructure, but the Ti introduced honeycomb-like interdendrites which could play an important role in hindering dislocation motion. As a result, the hardness of this alloy increased from 887 HV 0.5 to 1147 HV 0.5 . Additionally, the atomic radius of Ti was 1.76 Å which was larger than those of Ni (1.49 Å) and Al (1.18 Å). Hence, adding Ti to the Al-Ni-Ti rich phase might have resulted in lattice expansion or distortion.
Liu et al. [422] varied the concentration of Si in AlCoCrFeNiSi x (x = 0-0.5) coatings and found out that the microhardness of the coatings was linearly proportional to the Si concentration. The wear rate reduced significantly from~5.2 × 10 −4 to~1.3 × 10 −4 mm 3 ·N −1 ·m −1 with the increasing Si content. This was attributed to the formation of wear-resistant SiO and SiO 2 layers on the wear surface. Huang et al. [373] studied FeCoCrNiSi x (x = 0-1) and found out that increasing Si content improved the hardness and wear resistance by promoting the transformation of FCC to BCC. Its microhardness increased significantly from 89.52 HV (x = 0) to 653.71 HV (x = 1) and the wear track depth decreased from 22.14 µm (x = 0) to 5.29 µm (x = 1).
Hsu et al. [456] reported that in AlCoCrFe x NiMo 0.5 (x = 0.6-2), the wear resistance (at 500 • C) of x = 2 was much lower than that of x = 1.5 although both had similar hardness values. This was attributed to the fact that the x = 2 alloy formed a thicker oxide layer than x = 1.5 and more loose oxides abraded away from the surface. By this reasoning, the Fe content of x = 0.6-1 was recommended as over x = 1.5-2 for optimum wear resistance. Qiu et al. [458] described that the wear resistance of Al 2 CoCrFeCuTiNi x (x = 0-2) first increased with the Ni content (i.e., x = 0-1) and then decreased for x = 1-2. However, the hardness monotonically increased with the increasing Ni content from~900 HV (x = 0) tõ 1100 HV (x = 2). They said that toughness and brittleness also affect the wear resistance of this alloy apart from the hardness. In this alloy, when Ni is added from x = 0-1 at% then the wear resistance increases because hardness increases. For Ni content x = 1-2 at%, the wear resistance starts to reduce because the material becomes brittle and wear happens through small brittle fracture on the surface. Beng et al. [360] analyzed the CoCrFeNiMo x (x = 0-0.3) alloy and noticed that CoCrFeNi had the FCC phase. The FCC phase was maintained even after Mo addition, but the lattice parameter (denoted by a) altered (a CoCrFeNi = 3.5733 Å, a CoCrFeNiMo0.1 = 3.6016 Å, a CoCrFeNiMo0.3 = 3.5854 Å) due to atomic size misfit between Mo and other constituents. The atomic size of Mo (1.363 Å) is distinctively larger than those of Co (1.251 Å), Cr (1.249 Å), Fe (1.241 Å) and Ni (1.246 Å) [93]. This Mo-induced lattice distortion increased the hardness from 415.2 to 465.9 HV and reduced the specific wear rate from 0.59 × 10 −3 to 0.33 × 10 −3 mm 3 ·N −1 ·m −1 .
Yadav et al. [397] reported the influence of Pb and Bi content in (CuCrFeTiZn) 100-x Pb x , (CuCrFeTiZn) 100-x Bi x on their wear properties. Here, reductions in wear rate achieved by adding Pb and Bi were 21% and 25%, respectively. This was attributed to the fact that the soft dispersoids acted as lubricants [483]. Li et al. [413] studied Al 0.8 CoCrFeNiCu 0.5 Si x (x = 0-0.5) coatings. They found the specific wear rate in the range of 1.19 × 10 −6 -8.99 × 10 −6 mm 3 ·N −1 ·m −1 , the hardness in the range of 479-592 HV which was almost 5 to 7 times higher than the hardness of substrate 5083 Al alloy (79 HV 0.2 ).
Cheng et al. [394] noticed that the wear resistance and hardness of (Fe 0.25 Co 0.25 Ni 0.25 (B 0.7 Si 0.3 ) 0.25 ) 100−x Nb x (x = 0-2) coating (on Q235 steel) increased with the Nb content from x = 0 to x = 2. The measured hardness values were 587.1 HV 0.1 and 821.5 HV 0.1 for Nb content of x = 0 and x = 1, respectively. This HEA had the FCC microstructure and the grain size was refined from 3.5 µm (x = 0) to 1 µm (x = 2). This grain refinement helped to increase the hardness and wear resistance. The wear rates were measured to be 3.64 × 10 −6 and 1.42 × 10 −6 mm 3 ·N −1 ·m −1 for Nb content of x = 0 and x = 2, respectively. Jones et al. [358] reported that CoCrFeNiMn showed a remarkably low wear rate of 10 −6 mm 3 ·N −1 ·m −1 with the hardness of 1.6 GPa. Zhu et al. [359] added V and Nb in equiatomic ratio into the CoCrFeNiMn (FCC phase) coating and studied the wear resistance. The measured wear rate ranged 1.85 × 10 −5 -6.39 × 10 −5 mm 3 ·N −1 ·m −1 and the hardness varied in the range of 145-948.5 HV. The measured hardness values of CoCrFeNiMn, CoCrFeNiMnNb, CoCrFeNiMnV and CoCrFeNiMnNbV was 145 HV, 609 HV, 621 HV and 948 HV, respectively. Additionally, the wear rate of CoCrFeNiMn reduced from 6.39 × 10 −5 to 1.85 × 10 −5 mm 3 ·N −1 ·m −1 (for CoCrFeNiMnNbV) when both Nb and V were added. These results could be attributed to the fact that Nb and V promoted the precipitation of the Laves phase (FeNb intermetallic) and the sigma phase (CoFeV intermetallics) into the FCC matrix, which increased the hardness. Moreover, Nb and V promoted the formation of protective oxide layer on the worn surface and reduced the amount of adhesive and abrasive wear.

Particle Reinforcement
Hard particle reinforcement into HEAs matrix is another approach employed to improve the tribological properties. For example, adding WC into Co 10 Cr 10 Fe 40 Mn 40 [382], adding TiN-Al 2 O 3 into CoCrFeNiMn [277], adding NbC into AlCoCrFeNi [272], adding TiC into CoCrCuFeNiSi 0.2 [384] and AlCoCrFeNi 2.1 [416] improved the tribological performances. Such improved wear resistance was largely attributed to the increased hardness, the restrained effect of adherence abrasion, solution strengthening or microstructure refinement. Adding 10 wt% TiC particles into AlCoFeNiCu reduced the wear rate by the factor of 8 [439]. Adding TiB 2 into (AlCrFeMnV) 90 Bi 10 resulted in~95% decrease in wear rate due to the cumulative effect of matrix, reinforcement, refined grains and improved hardness [463]. Gou et al. [396] used CoCrFeNi (reinforced with NbC) as a metal binder for Ti (C,N)-based cermet. The hardness and fracture toughness were measured to be 1853 HV and 9.93 MPa·m 1/2 , respectively. The minimum wear rate was as low as 4.14 × 10 −6 mm 3 ·N −1 ·m −1 which was attributed to the reduced grain size and improved hardness due to the solid solution strengthening effect of the HEA binder.
Ji et al. [376] improved the tribological behavior of CoCrFeNiCu by embedding MoS 2 (2-5 wt%) and WC (20-80 wt%) particles. Here, adding 2% MoS 2 improved the wear resistance and decreased the friction coefficient but adding 5% MoS 2 increased the wear rate. Adding WC continuously improved the wear resistance and 80% WC yielded the lowest wear rate. Likewise, the wear resistance of CrMnFeCoNi was improved considerably by adding 25 wt% Y 2 O 3 thanks to grain boundary strengthening, Orowan looping and load transfer effect [366]. Zhou et al. [189] reported that (FeCoCrNi) 1−x (WC) x (x = 3-11 wt%) alloy consisted of a FCC matrix and W/Cr rich carbides. The hardness increased with the WC content from 603 HV (for (FeCoCrNi) 0.97 (WC) 0.03 ) to 768 HV (for (FeCoCrNi) 0.89 (WC) 0.11 ). This was attributed to the hard WC particles and carbides precipitation into the FCC matrix. This material showed adhesive and abrasive wear. Most of the wear was caused by the debonding of WC particles.
Liu et al. [381] introduced graphene nanoplatelets into Fe 50 Mn 30 Co 10 Cr 10 . Adding 0.2% Graphene nanoparticles promoted self-lubricating properties because the coefficient of friction (COF) decreased by 62%, which in turn increased the wear resistance. Further addition of graphene did not have much impact on COF or self-lubricating properties. Similarly, Zhang et al. [370] added self-lubricating particles (i.e., nickel coated graphite and MoS 2 powder) into the CoCrFeNi matrix using SPS. This composite had four phases (i.e., FCC, graphite, MoS 2 and Nickel) and showed very low wear rates (~10 −5 mm 3 ·N −1 ·m −1 ) for the temperature range of RT-800 • C. After wear testing the worn surfaces were analyzed by Raman and EDS. They observed that graphite and MoS 2 particles were accumulated on the worn surface for temperatures RT-800 • C. They indicated that these particles helped reduce abrasive/adhesive wear. Moreover, at 600 and 800 • C, the significantly increased oxygen concentration on the worn surface facilitated the formation of a smooth glaze layer of oxide (i.e., Cr 2 O 3 , Fe 3 O 4 etc.) and reduced the wear rate. Likewise, Zhang et al. [368] added solid lubricants (i.e., Ag, BaF 2 /CaF 2 eutectic) into CoCrFeNi. This composite showed the better wear resistance (wear rates~10 −5 mm 3 ·N −1 ·m −1 ) than PM304 (wear rate~10 −4 mm 3 ·N −1 ·m −1 ) [484] for the temperature range of RT-800 • C. The improved wear resistance was attributed to the solid lubricants (help to minimize friction and wear on the surface) and oxide layer (Cr 2 O 3 , Fe 3 O 4 , etc.) formation on the worn surface.

Use of Media and Heat Treatment
Media used during wear experiments have been proved to have considerable impact on the mechanism and rate of material removal. For example, Duan et al. [453] reported that AlCoCrFeNiCu showed more wear weight loss of~1.2 mg when in H 2 O 2 lubricant and~0.8 mg when oil was used as media. This was because friction coefficient was lower when oil was used as a media. FeCoCrAlNiTi 2 showed the best wear resistance in distilled water and the worst in NaCl solutions [427]. Xiao et al. [421] noticed that the wear rates of Al x CoCrFeNiSi (x = 0.5-1.5 mol) reduced significantly from 5.5 × 10 −5 to 1.6 × 10 −6 mm 3 ·N −1 ·m −1 with the increasing Al content under dry conditions. However, when water was used as a lubricant, the addition of Al content did not have a considerable impact on the wear rate. Geng et al. [369] studied the wear behavior of CoCrFeNiMn in vacuum and air in the temperature range of RT-800 • C. They found the wear rate varied in the range of 1.3 × 10 −4 -8 × 10 −4 mm 3 ·N −1 ·m −1 . The elements of this HEA oxidized as temperature increased. At elevated temperatures, oxides (i.e., CoO 3 , CoO, CoCrO 4 , Co 2 O 3 , Fe 2 O 3 , Cr 2 O 3 and MoO 3 ) formed on the worn surface. From RT to 400 • C, the wear rate was higher in air than in vacuum due to peeling of the formed loose oxides. However, at 600 • C and 800 • C, robust oxides were formed, and these oxides served as a protective layer and improved the wear resistance resulting in the lower wear rate both in air than in vacuum.
The effect of heat treatment on the wear behavior of Al 0.5 CoCrFeNi was studied by Gwalani et al. [399]. They hot rolled (at 700 • C), annealed (at 1150 • C) and isothermally aged their alloy at 700 • C for 1, 4, 20, 40, 80 h. As the aging time increased, the B2 phase precipitated in the FCC matrix. These precipitates increased the hardness from~250HV tõ 302  Consequently, the wear rate reduced from~0.95 to~0.65 mg·mm 2 . However, when the current was further increased up to 550 A, the hardness reduced to~580 HV and the wear rate increased to~0.72 mg·mm 2 . Here, the wear resistance of the HEA was found to be directly proportional to the hardness. Malatji et al. [263] improved the wear resistance of AlCrCuFeNi by heat treating at temperatures 800, 950 and 1100 • C. The microhardness increased from 310 to 381 HV when heat treated at 800 • C but decreased to 365 HV when heat treated at 1100 • C. Heat treatment at 1100 • C homogenized the microstructure and promoted grain coarsening. This alloy initially had the FCC phase alone. Upon heat treatment at 800 and 950 • C, the B2 phase precipitated and the hardness increased. When the alloy was heat treated at 1100 • C, some of the B2 precipitates decomposed and grain coarsened, which decreased the hardness. This explains why the wear resistant first increased and later reduced with the heat-treating temperatures. The observed wear weight was 0.002 g without heat treatment, and 0.001 g, 0.0007 g and 0.0016 g with heat treatment at 800 • C, 950 • C and 1100 • C respectively.
Cui et al. [374] sulfurized CoCrFeMoNi at 260 • C for 2 h and improved the wear resistance. After sulfurization, the weight loss by wear reduced from 15.1 mg to 4.25 mg. This was attributed to the FeS/MoS 2 lubricant phases and sulfides boundary lubricant films.
Wu et al. [406] used the pack-boronizing method to improve the wear resistance of Al 0.1 CoCrFeNi alloy. A 17.3-57.9 µm boronized layer, composed of (Co, Fe, Ni) 2  Xiao et al. [363] investigated the effects of carbon content in CoCrFeNiMnC x (x = 0-1.2). The hardness was increased monotonically from 327.8 HV (x = 0) to 566.4 HV (x = 1.2) with carbon addition. They found out that the wear rates of CoCrFeNiMn and CoCrFeNiMnC 0.6 were 6.5 × 10 −3 mm 3 ·N −1 ·m −1 and 0.47 × 10 −3 mm 3 ·N −1 ·m −1 , respectively. This alloy was comprised of the FCC (Co, Ni rich) phase and the Cr-and C-rich M7C3 carbide phases. The porosity increased with the increasing carbon content. Initially, the hardness increased until the volume fraction reached x = 0.6 due to introduction of M7C3 carbide and solid solutions of C atoms in the FCC phase. As the amount of carbon further increased, the hardness started to fall due to a significant increase in porosity. Meanwhile, Zhang et al. [385] found out that the wear rate and hardness of (CoCrFeNiTi 0.5 )C x (x = 3-12 wt%) coating were in the range of 12-24 mg·N −1 ·m −1 and 300-950 HV, respectively. This coating microstructure was made of BCC solid solution, Cr 23 C 6 and TiC phases. Increase in carbon content raised the hardness due to carbides precipitates, but the wear resistance with x = 12 wt% was inferior to that with x = 6 wt% due to reduced ductility.
Many researchers formed coatings of HEAs on several commercial materials to improve their wear resistance. Zhang et al. [461] coated Ti-6Al-4V with AlTiSiVNi. They discovered that the wear resistance of AlTiSiVNi was 4 to 5 times higher than that of Ti-6Al-4V at RT and 800 • C. The hardness of AlTiSiVNi was also found to be in the range of 1151-1357 HV which was 4-6 times that of Ti-6Al-4V. The increase in hardness and wear resistance of AlTiSiVNi was attributed to the dispersion strengthening with (Ti,V) 5 Si 3 and solution strengthening with NiAl. Huang et al. [460] coated Ti-6Al-4V with AlTiSiVCr. The microstructure of AlTiSiVCr had hard silicides (Ti,V) 5 Si 3 dispersed into the BCC matrix. The dispersed silicides improved the wear resistance of this HEA by lowering abrasive/adhesive wear. The hard BCC phase of the silicides also resisted crack propagation. While the specific wear rate for Ti-6Al-4V was~6.5 × 10 −5 to 9 × 10 −5 mm 3 ·N −1 ·m −1 , that for TiVCrAlSi ranged 2 × 10 −5 to 2.6 × 10 −5 mm 3 ·N −1 ·m −1 . This showed that TiVCrAlSi could become a promising anti-wear coating material for Ti-6Al-4V.
Islak et al. [391] improved the wear resistance of AISI 1040 steel with CrFeNiMoTi coating. They found that the hardness and wear rate of CrFeNiMoTi coating were~450 HV 0.3 and~2.732 × 10 −3 -3.952 × 10 −3 mm 3 ·N −1 ·m −1 , respectively. Meanwhile, the hardness and wear rate for AISI 1040 were~200 HV 0.3 and~8.125 × 10 −3 -9.455 × 10 −3 mm 3 ·N −1 ·m −1 , respectively. Gu et al. [455] analyzed Al x Mo 0.5 NbFeTiMn 2 (x = 1-2) coating and found that the microhardness and wear resistance were positively related to the Al content. The increasing Al content reduced the grain size and increased both the hardness and wear resistance. The hardness of Al 2 Mo 0.5 NbFeTiMn 2 was measured to be 1098.5 HV 0.2 which was 5 times higher than that of Q235 steel (~200 HV 0.2 ). The high hardness was attributed to the microstructure composed of the BCC solid solution and (Nb,Ti)C carbides. The wear rates of Al 2 Mo 0.5 NbFeTiMn 2 and Q235 steel were measured to be~0. 3

Higher Temperatures Wear Resistance
HEAs have also exhibited promising tribological behaviors and thermal stability at elevated temperatures. Jin et al. [438] studied the characteristics of AlCoCuFeNi coating at temperatures up to 800 • C. This coating was mainly composed of Fe rich FCC and Cu rich BCC phases and showed good thermal stability without any phase transformation until 780 • C. Mainly Al 2 O 3 , Cr 2 C 3 , Fe 2 O 3 and CuO were present in the oxide layer. Meanwhile, the weight losses of NiCrCoTiV at RT and 600 • C were measured to be 3.7 ± 0.1 mg and 3.5 ± 0.1 mg respectively [392]. The wear weight loss of 304L stainless steel was 7.7 ± 0.2 mg at RT. Fang et al. [405] used Al 0.3 CoCrFeNi as a metal binder for Ti (C,N)-based cermet. The hardness, fracture toughness and flexural strength of this cermet were 1137 HV, 6.46 MPa·m 1/2 and 761 MPa at 1000 • C respectively. This superior hightemperature performance was attributed to the hindrance of the slip system and higher oxidation resistance of the HEA binder. The wear resistance of this Al 0.3 CoCrFeNi (wear groove width = 49 µm) was also better than that of the conventional Ni-Co (wear groove width = 172 µm) metal binder.
Yadav et al. [395] concluded that the wear resistance of (CuCrFeTiZn) 100−x Pb x improved due to oxide formation at the surface at elevated temperatures (800-1000 • C). Because these oxides made a layer on the surfaces of the HEA which helped to avoid metal to metal contact, hence reduced the material removal rate. Moreover, Chen et al. [400] said that annealed Al 0.6 CoCrFeNi showed the wear resistance higher than GCr15 by a factor of three at 600 • C. They also attributed this improved wear resistance to the formation of oxides on the surface. In this case, Fe 2 O 3 , Cr 2 O 3 , Al 2 O 3 and Al(OH) 3 , formed on the worn surfaces of Al x FeCrNiCo x and reduced the wear rate. Alvi et al. [478] studied wear behavior of AlCoCrFeNi in the temperature range of RT-600 • C. They noticed that the oxidation of Cu (into CuO) causes wear rate decrease at 400 • C. Joseph et al. [365] analyzed wear behavior of Al x CoCrFeNi in the temperature range of RT-900 • C. The wear resistance of this HEA surpassed that of Inconel 718 at 900 • C. Here again, the wear resistance was enhanced at the higher temperatures thanks to the oxide layer formed at the contact interface.
Researchers also showed that the wear rate of HEAs increased initially and then decreased with temperature [380,401,434,467]. It was claimed that initially loose oxides formed on the surface and lower the wear rate up to moderately high temperatures. As temperature increased higher, such loose oxides were damaged by thermal softening and therefore, the wear rate increased. In contrast, other studies reported the opposite behaviors. The wear resistance initially decreased up to moderately high temperatures and then increased at higher temperatures. This behavior was attributed to the formation of a thicker and more robust oxide layer on the worn surface which reduced the area of direct metal to metal contact. For instance, Pole et al. [467] studied the wear resistance of refractory HEAs, TiZrHfTaV and TiZrTaVW in the temperature range of RT-500 • C. The hardness of these HEAs (6-8.1 GPa) was found to be larger than two times that of SS304. The measured wear rate of these alloy was in the range of 0.5 × 10 −5 -8 × 10 −4 mm 3 ·N −1 ·m −1 . Various oxides, such as ZrO 2 , TiO 2 , Ta 2 O 5 , V 2 O 5 , HfO 2 and WO 3 , formed on the worn surface. The wear rate increased until 150 • C due to the formation of a delicate oxide layer at the worn surface. However, as the temperature increased further from 150 to 500 • C, a strong protective oxide layer formed on the worn surface and the wear rate reduced for both refractory HEAs. Similar findings were reported by Lobel et al. [434] for the wear resistance of AlCoCrFeNiTi 0.5 coating in the temperature range of RT-900 • C. The depth of wear increased from~62 µm at RT to~82 µm at 500 • C, and afterwards it decreased down to~50 µm at 900 • C. As the temperature increased, a loose oxides layer formed on the worn surface at lower temperatures. As temperature increased, a stronger oxides protective layer formed and it reduced material loss. Similarly, CoCrFeNiNb x showed good resistance to thermal softening by showing less reduction in hardness (i.e., 35% from RT to 1000 • C) [380]. Here, the wear resistance decreased from RT to 400 • C and then increased. This improvement above 400 • C was attributed to the oxide layer formation at higher temperatures which lead to reduction in wear rates of the CoCrFeNiNb 0.65 and CoCrFeNiNb 0.8 . By a similar mechanism, the wear rate of Al 0.25 CoCrFeNi increased from RT to 300 • C due to thermal softening but it decreased thereafter due to oxides formation at the contact surface [401].
Some HEAs showed fluctuation in wear resistance due to microstructural transformations at elevated temperatures. The wear rate of Al 0.6 TiCrFeCoNi increased from RT to 300 • C and reduced from 300 • C to 500 • C, mainly due to phase transformation (i.e., the formation of sigma-CrFe) at higher temperatures [435]. Such phase transformation at higher temperatures could also significantly reduce fracture toughness. The wear resistance of (CoCrFeMnNi) 85 Ti 15 increased from RT to 400 • C and then decreased thereafter [367]. Miao et al. [415] studied the wear behavior of AlCoCrFeNi 2.1 with different antagonist materials (i.e., Al 2 O 3 , Si 3 N 4 and GCr15). They also analyzed wear rate of the same composition for a temperature range of RT-900 • C against SiC. The wear rate of AlCoCrFeNi 2.1 against Al 2 O 3 , Si 3 N 4 and GCr15 is~42 × 10 −5 ,~37 × 10 −5 , 32.5 × 10 −5 mm 3 ·N −1 ·m −1 . The hardness of Al 2 O 3 , Si 3 N 4 and GCr15 is 2300, 1500 and 680 HV, and the wear rate of AlCoCrFeNi 2.1 increased with the hardness of the antagonist material. Moreover, the wear rate of AlCoCrFeNi 2.1 (against SiC) increased from~75 × 10 −6 to~140 × 10 −6 mm 3 ·N −1 ·m −1 when temperature increased from RT to 900 • C. This increased wear rate with temperature was attributed to the thermal softening. Joseph et al. [355] noticed that when the wear resistance of CoCrFeNiMn was examined at high temperatures, ultrafine grains and the sigma phase formed at the contact surface, which resulted in the improved wear resistance.

Summery and Future Direction
This review covers recent advances in the development and manufacturing of HEAs and their performances under extreme environments such as nuclear and wear applications. The HEAs were tabulated based on manufacturing methods, irradiation responses and wear performances.
The most widely used method for HEAs manufacturing was arc melting due to its simplicity, when the idea of HEAs was conceptualized. However, recently AM processes (SLM, EBM and DED) have gained interest since they may potentially provide more freedom in shape and in properties by changing process parameters. However, there are some issues that need to be resolved. For instance, low productivity, formation of micro-level defects such as pores or unfused particle boundaries, residual stresses, composition shift due to selective evaporation of constituents with lower vapor pressure, lack of standards for quality evaluation, high initial investment cost and more. With these issues being gradually resolved, AM could make a powerful and versatile manufacturing method to fabricate application-specific HEAs with desired properties for some compositions. Most of the AM HEAs are studied in the as-cast state (after AM). In terms of HEAs characterization, most studies focused on the microstructure using SEM, tensile behavior and hardness. Structural characterization at a smaller length scale, using transmission electron microscope, would be helpful to better understand the structural evolution under various loadings. In order to identify the effect of manufacturing methods and explore more applications for HEAs, more research is needed on creep properties, dislocation behaviors, deformation microstructures, compressive strength, fatigue and more.
Structural materials for next generation nuclear reactors must survive high energy irradiation at high temperatures with reasonable service life. Similar to other structural metals, high energy irradiation on HEAs often induces microstructural changes which in turn deteriorated their mechanical properties including hardness, swelling or embrittlement. For a number of cases of HEAs, their compositional complexity hinders such microstructural degradation and results in superior irradiation resistance compared to other conventional alloys. This makes them promising candidate materials for nuclear applications. However, more studies must be conducted on the irradiation behaviors of HEAs to better understand their applicability to the next generation nuclear reactors. To date, very little is known about HEAs phase diagrams and equilibrium phases. In addition, the defect generation and movement mechanism as a result of irradiation are not clearly understood.
HEAs have also demonstrated superior tribological performances over a wide range of temperatures from RT to high temperatures in comparison to the commercial materials (i.e., Steel, Inconel, Ti-6Al-4V, Q235, SUS304 etc.). Moreover, the wear behavior of HEAs is affected by composition, particle reinforcement, media, nitriding/carburizaing/sulfurizing treatments, temperature and oxides formation. Most of the wear studies are on cantor alloy or its derivatives; therefore, more elements and combinations are needed to be explored to further understand potential candidates for wear applications. For the wear resistant applications where the weight is not a critical factor, refractory high entropy alloys would make a good option thanks to their high hardness [466][467][468][469][470][471][472][473][474][475][476][477][478][479][480]. Most of these alloys are equimolar. The equimolar ratios are probably a good point to start but they might not be the best to get the highest potential out of the particular element composition. Non-equimolar refractory HEAs are worth more exploration.
HEAs are being researched for more than a decade and are not yet commercially available. One reason could be because HEAs could not be manufactured with most widely used processes suitable for mass production (i.e., casting, molding etc.). Arc melting (most popular manufacturing method for HEAs) is limited to manufacture laboratory-scale samples for testing. Recently, a number of attempts have been made to fabricate HEAs using AM techniques. Some AM HEAs showed the improved mechanical properties. However, there are still many issues that need to be addressed for AM to be used for mass production of HEAs.
Overall, there are still a great number of possible HEAs compositions that are to be studied. Apart from cantor alloy, the characteristics of the majority of other HEAs have not been investigated well enough for safe practical applications. These studies cannot be used to generalize characteristics for HEAs but we can take them as screening efforts. Moreover, it is not practical to perform all the characterization studies on all these compositions. Therefore, it would be more reasonable to use simulations and material informatics to screen compositions before experimental studies instead of using trial and error.

Data Availability Statement:
The authors declare that the data supporting the findings of this study are available within the article.

Conflicts of Interest:
The authors declare no conflict of interest.