High-Pressure Torsion for Synthesis of High-Entropy Alloys

High-pressure torsion (HPT) is widely used not only as a severe plastic deformation (SPD) method to produce ultrafine-grained metals but also as a mechanical alloying technique to synthesize different alloys. In recent years, there have been several attempts to synthesize functional high-entropy alloys using the HPT method. In this paper, the application of HPT to synthesize high-entropy materials including metallic alloys, hydrides, oxides and oxynitrides for enhanced mechanical and hydrogen storage properties, photocatalytic hydrogen production and high light absorbance is reviewed.


Introduction
High-pressure torsion (HPT) is a severe plastic deformation (SPD) method [1], which is widely used to generate nanograins or ultrafine grains (UFG) in metallic materials [2]. In this method, as schematically shown in Figure 1, a small disc sample is compressed between two anvils under a pressure of several giga-Pascals and, concurrently, shear strain is introduced in the sample by rotating one of the anvils with respect to the other one (γ = 2πrN/h, γ: shear strain, r: distance from disc center, N: anvil turns, h: sample thickness) [1,2]. The method was originally introduced by Bridgman in 1935 [3] to examine polymorphic phase transformations under high pressure (see the historical development of the HPT method in [4,5]). In 1988, Valiev et al. [6] introduced the HPT method as an SPD technique to achieve nanograined microstructures in metallic materials. This study resulted in the development of the Nano-SPD field [7] and accelerated the research activities on the HPT method by many other research groups [8][9][10][11].
It has been shown that the HPT method can introduce extremely large shear strains of over 1000 (up to 100,0000), so that the microscopic phases can be geometrically thinned to the subnanometer range [20,32]. The shear strain of 1000 is the nominal strain that is geometrically required to reduce the size of a phase with 1 µm to less than 1 nm; however, the required shear strain for mechanical alloying can be lower or much higher than this level depending on the size of phases, their miscibility, their co-deformation, and their recrystallization [32]. In general, the strains required for mechanical alloying are a few orders of magnitude higher than those required for only grain refinement by HPT processing [2,4]. The introduction of such large strains, which has received the terminology of ultra-SPD [20,32], is quite effective for the synthesis of new materials even from immiscible systems such as Mg-Ti and Mg-Zr with hydrogen storage capability [33,34] or Al-Zr and Al-La-Ce with high strength [35,36]. Although high pressure is a key parameter of HPT to deform any kind of materials, shear strain plays the most important role in mechanical alloying. Mechanical alloying by HPT is usually conducted at ambient temperature with a low deformation speed (usually 1 rpm), and the temperature rise during the process is not significant [37]. All these HPT processing conditions provide an opportunity for the simultaneous synthesis and cold consolidation of bulk nanostructured alloys from their elemental powders [20,32]. Compared to ball milling as a popular mechanical alloying technique [38], in the HPT method, the synthesis time is shorter, the rotation speed is lower, the contamination is minor, and the final product is in the bulk form, which does not require further consolidation processes [2,4]. However, the main disadvantage of HPT is the low quantity of the synthesized product; thus, scaling up the sample [39,40] and making the process continuous [41,42] are currently major challenges for the commercialization of HPT.

Synthesis of High-Entropy Alloys with High Hardness
Nanostructured materials show not only enhanced functional properties but also high hardness due to the Hall-Petch strengthening mechanism [1,7]. In an attempt to produce nanostructured HEAs, Kilmametov et al. [64] synthesized a CoCrFeMnNi alloy with a face-centered cubic (FCC) structure starting from the elemental powders, applying HPT with 100 turns. They selected this alloy (also known as the Cantor alloy [38]) because it is one of the first designated HEAs and there have been numerous studies to synthesize and process the alloy by different methods. They used atom probe tomography analysis and reported a successful atomic-scale mixture of elements, as shown in Figure 2. They also showed that, besides mechanical alloying, good consolidation to a bulk form and nanograin formation occurred. These microstructural features resulted in a high hardness level of 700 HV in the alloy. They also reported that such high hardness is not only due to microstructural refinement but also due to the formation of small amounts of chromium oxide particles, also shown in Figure 2 [64]. such as high-entropy hydride MgTiVCrFe-H for hydrogen storage [75], high-entropy oxide TiZrHfNbTaO11 for photocatalytic hydrogen production [76] and high-entropy oxynitride TiZrHfNbTaO6N3 as a photocatalyst with large light absorbance [77]. In this article, the main achievements in the synthesis of HEAs by ultra-SPD via the HPT method are reviewed and the significance of the process for attaining advanced functionalities is discussed.

Synthesis of High-Entropy Alloys with High Hardness
Nanostructured materials show not only enhanced functional properties but also high hardness due to the Hall-Petch strengthening mechanism [1,7]. In an attempt to produce nanostructured HEAs, Kilmametov et al. [64] synthesized a CoCrFeMnNi alloy with a face-centered cubic (FCC) structure starting from the elemental powders, applying HPT with 100 turns. They selected this alloy (also known as the Cantor alloy [38]) because it is one of the first designated HEAs and there have been numerous studies to synthesize and process the alloy by different methods. They used atom probe tomography analysis and reported a successful atomic-scale mixture of elements, as shown in Figure 2. They also showed that, besides mechanical alloying, good consolidation to a bulk form and nanograin formation occurred. These microstructural features resulted in a high hardness level of 700 HV in the alloy. They also reported that such high hardness is not only due to microstructural refinement but also due to the formation of small amounts of chromium oxide particles, also shown in Figure 2   In another attempt, González-Masís et al. [78] synthesized TiNb, TiNbZr, TiNbZrTa and TiNbZrTaHf alloys from the elemental powders using the HPT process. These alloys were selected because of their biocompatibility and their tendency to form single bodycentered cubic (BCC) phases. All materials showed the formation of fully BCC structures with the presence of nanograins. In addition to the successful synthesis of materials by the HPT method, they reported that there are increases in hardness when increasing the configurational entropy or the number of elements, as shown in Figure 3. Detailed analyses confirmed that these hardness changes cannot be justified only by conventional strengthening mechanisms such as solution hardening, dislocation hardening, grain-boundary hardening and precipitation hardening, because the ternary, quaternary and quinary alloys in Figure 3 had similar nanograin sizes, similar dislocation densities and similar atomic-size-mismatch parameters [78]. Since the steady-state hardness of HPT-processed metals is their intrinsic property and depends on the atomic bonding energy [79], and since this energy can be reduced by increasing the entropy [43,44], it was hypothesized by In another attempt, González-Masís et al. [78] synthesized TiNb, TiNbZr, TiNbZrTa and TiNbZrTaHf alloys from the elemental powders using the HPT process. These alloys were selected because of their biocompatibility and their tendency to form single bodycentered cubic (BCC) phases. All materials showed the formation of fully BCC structures with the presence of nanograins. In addition to the successful synthesis of materials by the HPT method, they reported that there are increases in hardness when increasing the configurational entropy or the number of elements, as shown in Figure 3. Detailed analyses confirmed that these hardness changes cannot be justified only by conventional strengthening mechanisms such as solution hardening, dislocation hardening, grain-boundary hardening and precipitation hardening, because the ternary, quaternary and quinary alloys in Figure 3 had similar nanograin sizes, similar dislocation densities and similar atomicsize-mismatch parameters [78]. Since the steady-state hardness of HPT-processed metals is their intrinsic property and depends on the atomic bonding energy [79], and since this energy can be reduced by increasing the entropy [43,44], it was hypothesized by González-Masís et al. that high configurational entropy and a decrease in the internal energy of the system can also influence the steady-state hardness of HEAs [78]. González-Masís et al. that high configurational entropy and a decrease in the internal energy of the system can also influence the steady-state hardness of HEAs [78].

Synthesis of High-Entropy Hydrides for Hydrogen Storage
The safe and high-density storage of hydrogen as a clean fuel is still a challenging issue for the future development of a hydrogen-based society [80]. Hydrides provide the most compact technology for the storage of hydrogen, although there are still drawbacks regarding the kinetics and thermodynamics of hydrogenation and dehydrogenation in hydrides [81,82]. The synthesis of hydrides is a critical issue because the performance of hydrides can be significantly influenced by the processing route [81,82]. Mechanochemistry routes such as HPT are considered effective to produce hydrides with promising hydrogen storage properties [83,84]. It has been shown that the HPT method is not only effective to synthesize hydrogen storage materials but also to enhance their activity, kinetics, and air resistance [85]. The enhancement of the activity and kinetics of hydrogen storage materials by HPT processing is mainly attributed to the formation of lattice defects and grain boundaries, which can act as quick pathways for hydrogen transport from surface to bulk [84,85].
In one of the first attempts to synthesize high-entropy hydrogen storage materials, de Marco et al. [75] synthesized MgVCr-H and MgVTiCrFe-H from the elemental powders using ball milling, followed by the HPT method and hydrogenation. The alloys after HPT processing showed a mixture of BCC and amorphous phases, and energy dispersive X-ray spectroscopy confirmed good mechanical mixing of the elements. This first study on the synthesis of high-entropy hydrogen storage materials by the HPT method confirmed the potential of the method, although the reported hydrogen storage capacity was not high, as shown in Figure 4. Since recent studies suggested the potential of high-entropy hydrogen storage materials (HfNbTiVZr [86], TiZrNbHfTa [87], MgZr-TiFe0.5Co0.5Ni0.5 [88], TiZrNbMoV [89], TiZrHfScMo [90], CoFeMnTiVZr [91], ZrTiVCrFeNi [92], (VFe)60(TiCrCo)40−xZrx [93], TiVZrNbTa [94], AlCrFeMnNiW [95]) and particularly those materials designated by inputs from the theoretical calculations (TiZrCrMnFeNi [59], TiZrNbFeNi [96], and TiZrNbCrFe [97]), a combination of theoretical design and mechanical synthesis by HPT processing should be an effective strategy to explore new hydrogen storage materials. Edalati et al. [59] and Floriano et al. [96] suggested three criteria to explore high-entropy materials for room-temperature hydrogen storage: (i) selection of an AB2 or AB system, where A represents the hydride-forming elements such as Mg, Ti, Zr, V, Nb, etc., and B represents elements with low chemical Figure 3. Variation of hardness versus configurational entropy for TiNb, TiNbZr, TiNbZrTa and high-entropy alloy TiNbZrTaHf, synthesized by high-pressure torsion, reproduced from [78], with permission from ELSEVIER, 2021. Grain size, dislocation density and atomic-size-mismatch parameters were similar for ternary, quaternary, and quinary alloys.

Synthesis of High-Entropy Hydrides for Hydrogen Storage
The safe and high-density storage of hydrogen as a clean fuel is still a challenging issue for the future development of a hydrogen-based society [80]. Hydrides provide the most compact technology for the storage of hydrogen, although there are still drawbacks regarding the kinetics and thermodynamics of hydrogenation and dehydrogenation in hydrides [81,82]. The synthesis of hydrides is a critical issue because the performance of hydrides can be significantly influenced by the processing route [81,82]. Mechanochemistry routes such as HPT are considered effective to produce hydrides with promising hydrogen storage properties [83,84]. It has been shown that the HPT method is not only effective to synthesize hydrogen storage materials but also to enhance their activity, kinetics, and air resistance [85]. The enhancement of the activity and kinetics of hydrogen storage materials by HPT processing is mainly attributed to the formation of lattice defects and grain boundaries, which can act as quick pathways for hydrogen transport from surface to bulk [84,85].
In one of the first attempts to synthesize high-entropy hydrogen storage materials, de Marco et al. [75] synthesized MgVCr-H and MgVTiCrFe-H from the elemental powders using ball milling, followed by the HPT method and hydrogenation. The alloys after HPT processing showed a mixture of BCC and amorphous phases, and energy dispersive X-ray spectroscopy confirmed good mechanical mixing of the elements. This first study on the synthesis of high-entropy hydrogen storage materials by the HPT method confirmed the potential of the method, although the reported hydrogen storage capacity was not high, as shown in Figure 4. Since recent studies suggested the potential of high-entropy hydrogen storage materials (HfNbTiVZr [86], TiZrNbHfTa [87], MgZrTiFe 0.5 Co 0.5 Ni 0.5 [88], TiZrNbMoV [89], TiZrHfScMo [90], CoFeMnTiVZr [91], ZrTiVCrFeNi [92], (VFe) 60 (TiCrCo) 40−x Zr x [93], TiVZrNbTa [94], AlCrFeMnNiW [95]) and particularly those materials designated by inputs from the theoretical calculations (TiZrCrMnFeNi [59], TiZrNbFeNi [96], and TiZrN-bCrFe [97]), a combination of theoretical design and mechanical synthesis by HPT processing should be an effective strategy to explore new hydrogen storage materials. Edalati et al. [59] and Floriano et al. [96] suggested three criteria to explore high-entropy materials for roomtemperature hydrogen storage: (i) selection of an AB 2 or AB system, where A represents the hydride-forming elements such as Mg, Ti, Zr, V, Nb, etc., and B represents elements with low chemical affinity with hydrogen, such as Cr, Mn, Fe, Co, Ni, etc.; (ii) valence electron concentration of 6.4-6.5; (iii) Laves phase stability, which should be examined by thermodynamic calculations using the CALPHAD (calculation of phase diagram) method. Although these suggested criteria were used successfully to design room-temperature hydrogen storage materials [59,96], there are some less successful attempts to design HEAs with higher A/B ratio with the Laves or BCC structures [75,97]. affinity with hydrogen, such as Cr, Mn, Fe, Co, Ni, etc.; (ii) valence electron concentration of 6.4-6.5; (iii) Laves phase stability, which should be examined by thermodynamic calculations using the CALPHAD (calculation of phase diagram) method. Although these suggested criteria were used successfully to design room-temperature hydrogen storage materials [59,96], there are some less successful attempts to design HEAs with higher A/B ratio with the Laves or BCC structures [75,97].

Synthesis of High-Entropy Oxides for Photocatalytic Hydrogen Production
The production of hydrogen without CO2 emission is one major challenge in realizing hydrogen as a clean fuel for the future [80]. The cleanest technology to produce hydrogen is photocatalytic water splitting under sunlight [98]. In photocatalysis, light absorption by a semiconductor photocatalyst leads to the excitation of electrons to the conduction band and the formation of holes in the valence band. These electrons and holes on the surface of the catalyst contribute to the oxidation and reduction of water to produce O2 and H2, respectively. Despite the significant scientific interest in photocatalysis, the efficiency of this technology for hydrogen production is still far from commercialization [99,100]. There are now significant efforts being made around the world to develop new photocatalysts that can produce large amounts of hydrogen from water. Large light absorption, long electron/hole recombination time and high chemical stability are some parameters that are important in designing new photocatalysts [99,100]. High-entropy photocatalysts are one of the newest family of photocatalysts developed recently for hydrogen production [76,77].
In one of the first attempts to produce high-entropy photocatalysts [76], powders of Ti, Zr, Hf, Nb and Ta were mixed by HPT processing with 200 turns to produce a BCC TiZrHfNbTa alloy. The Ti, Zr, Hf, Nb and Ta elements were selected because their cations generate a d 0 electronic configuration and it is empirically known that oxides with the d 0 electronic configuration can usually exhibit good photocatalytic properties for hydrogen production [99,100]. The BCC alloy was further oxidized at 1443 K for 24 h to produce a dual-phase high-entropy oxide with an orthorhombic and a monoclinic phase and a general composition of TiZrHfNbTaO11. The synthesized dual-phase oxide had an orange color (Figure 5a), which corresponds to the absorbance of blue light based on the Itten color ring (Figure 5b). It had large particle sizes (Figure 5c) compared to its grain sizes, which were at the nanometer level (Figure 5d). Examination of the photocatalytic performance of the material, as shown in Figure 6a, confirmed that the material could produce hydrogen from water splitting under UV light. Moreover, examination of the material by X-ray diffraction before and after three cycles of photocatalytic testing (Figure 6b) confirmed that the material is chemically stable. High stability is a general feature of many

Synthesis of High-Entropy Oxides for Photocatalytic Hydrogen Production
The production of hydrogen without CO 2 emission is one major challenge in realizing hydrogen as a clean fuel for the future [80]. The cleanest technology to produce hydrogen is photocatalytic water splitting under sunlight [98]. In photocatalysis, light absorption by a semiconductor photocatalyst leads to the excitation of electrons to the conduction band and the formation of holes in the valence band. These electrons and holes on the surface of the catalyst contribute to the oxidation and reduction of water to produce O 2 and H 2 , respectively. Despite the significant scientific interest in photocatalysis, the efficiency of this technology for hydrogen production is still far from commercialization [99,100]. There are now significant efforts being made around the world to develop new photocatalysts that can produce large amounts of hydrogen from water. Large light absorption, long electron/hole recombination time and high chemical stability are some parameters that are important in designing new photocatalysts [99,100]. High-entropy photocatalysts are one of the newest family of photocatalysts developed recently for hydrogen production [76,77].
In one of the first attempts to produce high-entropy photocatalysts [76], powders of Ti, Zr, Hf, Nb and Ta were mixed by HPT processing with 200 turns to produce a BCC TiZrHfNbTa alloy. The Ti, Zr, Hf, Nb and Ta elements were selected because their cations generate a d 0 electronic configuration and it is empirically known that oxides with the d 0 electronic configuration can usually exhibit good photocatalytic properties for hydrogen production [99,100]. The BCC alloy was further oxidized at 1443 K for 24 h to produce a dual-phase high-entropy oxide with an orthorhombic and a monoclinic phase and a general composition of TiZrHfNbTaO 11 . The synthesized dual-phase oxide had an orange color (Figure 5a), which corresponds to the absorbance of blue light based on the Itten color ring (Figure 5b). It had large particle sizes (Figure 5c) compared to its grain sizes, which were at the nanometer level (Figure 5d). Examination of the photocatalytic performance of the material, as shown in Figure 6a, confirmed that the material could produce hydrogen from water splitting under UV light. Moreover, examination of the material by X-ray diffraction before and after three cycles of photocatalytic testing (Figure 6b) confirmed that the material is chemically stable. High stability is a general feature of many metallic [47][48][49][50] and non-metallic [65][66][67][68][69][70][71][72][73][74] high-entropy materials, which results from the high entropy of mixing [43,44]. Although the hydrogen production rate on TiZrHfNbTaO 11 was low, similar to other available photocatalysts [99,100], the study introduced the HPT method as a potential pathway to produce new materials for energy-related applications such as photocatalysis [76]. It should be noted that the HPT method as a processing route can also enhance the photocatalytic hydrogen production through three main mechanisms [24][25][26]: (i) bandgap narrowing by the introduction of vacancies; (ii) the introduction of lattice defects as active sites for photocatalytic water splitting; (iii) the stabilization of highpressure phases such as TiO 2 -II with higher photocatalytic activity. metallic [47][48][49][50] and non-metallic [65][66][67][68][69][70][71][72][73][74] high-entropy materials, which results from the high entropy of mixing [43,44]. Although the hydrogen production rate on TiZrHfNbTaO11 was low, similar to other available photocatalysts [99,100], the study introduced the HPT method as a potential pathway to produce new materials for energyrelated applications such as photocatalysis [76]. It should be noted that the HPT method as a processing route can also enhance the photocatalytic hydrogen production through three main mechanisms [24][25][26]: (i) bandgap narrowing by the introduction of vacancies; (ii) the introduction of lattice defects as active sites for photocatalytic water splitting; (iii) the stabilization of high-pressure phases such as TiO2-II with higher photocatalytic activity.   metallic [47][48][49][50] and non-metallic [65][66][67][68][69][70][71][72][73][74] high-entropy materials, which results from the high entropy of mixing [43,44]. Although the hydrogen production rate on TiZrHfNbTaO11 was low, similar to other available photocatalysts [99,100], the study introduced the HPT method as a potential pathway to produce new materials for energyrelated applications such as photocatalysis [76]. It should be noted that the HPT method as a processing route can also enhance the photocatalytic hydrogen production through three main mechanisms [24][25][26]: (i) bandgap narrowing by the introduction of vacancies; (ii) the introduction of lattice defects as active sites for photocatalytic water splitting; (iii) the stabilization of high-pressure phases such as TiO2-II with higher photocatalytic activity.   (a) Photocatalytic hydrogen production for three cycles and (b) X-ray diffraction patterns before and after photocatalysis for dual-phase high-entropy oxide TiZrHfNbTaO 11 , synthesized by high-pressure torsion and high-temperature oxidation, reproduced from [76], with permission from ROYAL SOCIETY OF CHEMISTRY, 2020.

Synthesis of High-Entropy Oxynitrides with Large Light Absorbance
Large light absorbance is a major requirement in some applications, such as photocatalysis [99,100]. Photocatalysts are mainly oxides, but most of the oxides can only absorb UV light (due to their large bandgap), and UV light accounts for only 6% of the sunlight on Earth [101]. Nitrides have a lower bandgap than oxides due to the higher energy level of the nitrogen 2p orbital as compared to the oxygen 2p orbital; however, nitrides suffer from poor stability [102]. Therefore, there are now significant attempts to develop oxynitrides that possess both large light absorbance and high stability.
To develop oxynitrides with low bandgap and high stability for photocatalytic application, Edalati et al. [77] synthesized a high-entropy oxynitride based on the mixture of Ti, Zr, Hf, Nb and Ta cations, which can generate a d 0 electronic configuration that is appropriate for photocatalytic activity [101,102]. They developed a three-step synthesis technique: (i) mechanical alloying of elemental powders by HPT processing for 100 turns to produce a BCC TiZrHfNbTa alloy; (ii) oxidation of the BCC alloy at 1373 K for 24 h to produce TiZrHfNbTaO 11 oxide; (iii) nitriding the oxide in ammonia at 1373 K for 7 h to produce an oxynitride. The final product was a dual-phase high-entropy oxynitride with monoclinic and FCC structures and a general composition of TiZrHfNbTaO 6 N 3 . Its grain sizes were small, in the range of nanometers, as shown in Figure 7. The material showed a narrow bandgap of 1.6 eV, which is one of smallest bandgaps reported for oxynitrides [101,102]. Moreover, its valence band and conduction band satisfied the energy requirements for photocatalytic water splitting, as shown in Figure 8a (conduction band bottom was more negative than 0 eV vs. NHE for H 2 O to H 2 reduction and valence band top was more positive than 1.23 eV vs. NHE for H 2 O to O 2 oxidation [98]). Furthermore, the light absorbance of this oxynitride was better than all binary oxides formed in the Ti-Zr-Hf-Nb-Ta system as well the corresponding high-entropy oxide, as shown in Figure 8. This high-entropy oxynitride with high chemical stability could produce hydrogen from photocatalytic water splitting, introducing high-entropy oxynitrides as new low-bandgap and stable photocatalysts [77].  absorbance compared with binary and high-entropy oxides for nanograined dual-phase highentropy oxynitride TiZrHfNbTaO6N3, synthesized by high-pressure torsion followed by hightemperature oxidation and nitriding, reproduced from [77], with permission from ROYAL SOCI-ETY OF CHEMISTRY, 2021.

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