8.2.1 Room-Temperature Mechanical Properties

Different alloys including refractory HEAs are described in this section. Cast and wrought processes, and powder metallurgy route including mechanical alloying and hot isostatic pressing, are used for synthesis. Hardness, compressive, tensile, fatigue properties, and deformation behavior are discussed.

The first system that has been studied extensively is the Al_xCoCrCuFeNi alloys (Tong et al., 2005a, 2005b; Tung et al., 2007; Tsai et al., 2009b). The hardness of the system increases from 133 HV for x=0–0.5 to 655 HV for x=3.0 (Figure 8.1). This can be attributed to the increase in lattice distortion as Al is the largest atom amongst the constituents of the alloy (Callister, 2003). In addition, Al forms strong bonds with other elements in the alloy as reflected in the enthalpy of mixing (Miedema et al., 1980; Takeuchi and Inoue, 2010). Hence the solid solution strengthening effect increases with higher Al content. It was also observed that as the Al content increases the phase changes from FCC to BCC (including disordered BCC and ordered BCC (B2)). The BCC and B2 phases are stronger than the FCC phase. Apart from this, nanoprecipitates formed due to slow diffusion kinetics also strengthen the material. The SD BCC+B2 modulated structure occurring in the BCC DR regions further strengthens the material (Yeh et al., 2004b; Tong et al., 2005a, 2005b). The cracks due to the indents, whose total length is an indication of toughness, are not present for the compositions where FCC phase is dominant (Figure 8.1). But, as the x value increases such that a BCC phase starts to dominate, the cracks start to appear. The length of the crack increases with increased amount of BCC phase, which indicates increased brittleness. Li et al. (2009) reported similar effect of Al on hardness. Addition of V to the Al_0.5CoCrCuFeNi alloy to synthesize Al_0.5CoCrCuFeNiV_x alloy increases the hardness (Chen et al., 2006a). After x=0.4, the BCC phase becomes the DR phase and increases in volume fraction from thereon. Also a σ phase forms after x=0.6, which further contributes to the hardness of the alloy.

Alloying elements with larger atomic sizes have a tendency to form secondary phases and cause precipitation strengthening. For example, among the 10 systems studied by Li et al. (2009), the highest hardness (566 HV) was achieved with the addition of Zr and Ti, which have larger atomic size and cause precipitation of secondary phases. Ma and Zhang (2012) reported that Nb addition from x=0 to 0.5 increases the hardness of AlCoCrFeNb_xNi alloy from 500 to 750 HV. Combined additions of Ti and Nb increase the hardness to 797 HV, which is higher than that achieved by adding only one of these two elements (Razuan et al., 2013). In another report (Hsu et al., 2010a), AlCoCrFe_xMo_0.5Ni alloy’s hardness was reported to decrease with an increase in Fe, because the alloy has a soft BCC phase and a harder σ phase in the microstructure. As the amount of Fe increases, the amount of BCC phase increases at the expense of σ phase causing a drop in hardness and wear resistance as shown in Figure 8.2. Similar to Fe, increase in Co content in AlCo_xCrFeMo_0.5Ni decreases the hardness from 788 HV at x=0.5 to 596 HV at x=2.0 (Hsu et al., 2010b).

The AlCoCrFeNiTi_x alloys showed good mechanical properties in compression (Zhou et al., 2007b, 2008a). The composition with x=0.5 had a BCC structure and was found to give optimum properties. The yield strength and fracture strength were found to be 2.26 and 3.14 GPa, respectively, which are much higher than that of BMGs. The alloy also had a 23.3% plastic elongation. The high strength was attributed to the spinodal decomposition in DR region, while there was precipitation of BCC phase particles in the ID region. Thus apart from solid solution strengthening, nanoparticle and precipitation strengthening also play an important role.

The major deformation mechanism in the Al_0.5CoCrCuFeNi alloy has been found to be twinning but not slip (Tsai et al., 2009b). This is different from that in conventional FCC alloys in which twinning is not commonly observed. One evidence is that etch pits that were observed in the presence of slip bands were absent in the microstructure and TEM did not reveal dislocation tangles. Figure 8.3 shows the dark-field and bright-field images of the nanotwins formed between the Widmanstätten precipitates and their plane and direction for the 5% cold-rolled alloy (Tsai et al., 2009b). This suggests that nanotwinning in this HEA starts in the initial stage of plastic deformation itself. This is because there are nanoscale precipitates distributed throughout the matrix of DR and slip movement is easily arrested. One more reason is that presence of large number of elements in HEA alloys causes Suzuki interaction, which significantly reduces the stacking fault energy (SFE), thus making it easier for twinning. Hence, nanotwinning occurs because the twin nucleation is initiated at nanoscale region between the Widmanstätten precipitates. Similar nanotwins have also been observed in CoCrFeNi HEA obtained by mechanical alloying followed by SPS. AlCoCrCuFeNi alloy made by splat quenching and casting was compared by Singh et al. (2011b). The hardness values of splat-quenched (539 HV) and as-cast (534 HV) alloys are almost equal. However, the indentation elastic modulus of the as-cast material (182 GPa) is markedly higher than that of the splat-quenched alloy (105 GPa).

In a single-phase FCC CoCrFeMnNi alloy (Otto et al., 2013a), it was found that the deformation is governed by planar slip of 1/2110 type dislocations on 111 planes at strains <2.4%, which also indicates low SFE. At a strain above 20%, the prevalence of deformation twinning was observed at 77 K, but the formation of dislocation cell structures was observed at 293 K. The evolution of microstructure and texture after heavy cold rolling and annealing in the same CoCrFeMnNi HEA was also reported (Bhattacharjee et al., 2014). A predominantly brass-type texture after 90% cold rolling was observed indicating the low SFE of this alloy. They proposed that the combined effect of high energy level of the distorted matrix and the strain energy relief of stacking fault by in situ atom position adjustment results in very low SFE as compared with conventional alloys featured by the small-distorted matrix with a host element.

Using refractory elements (melting points higher than 1650°C) as constituents, the American Air Force Research Laboratory developed refractory HEAs in order to overcome the temperature limitation associated with conventional superalloys. The room-temperature compressive yield stress of the first reported alloys MoNbTaW and MoNbTaVW was found to be 1058 and 1246 MPa, respectively. The quaternary MoNbTaW alloy showed a maximum strength of 1211 MPa and failed after 2.1% strain by a splitting mechanism (Senkov et al., 2010, 2011b). The Young’s modulus of this alloy was 220±20 GPa. On the other hand, the quinary alloy MoNbTaVW showed a maximum strength of 1270 MPa and failed after 1.7% strain by the same mechanism. The Young’s modulus of this alloy was 180±15 GPa. The failure occurred in a quasi-cleavage fashion parallel to the stress direction and hence in tensile mode.

Another refractory HEA HfNbTaTiZr, was also synthesized by arc melting (Senkov et al., 2011a). This alloy has a slightly lower yield stress of 929 MPa than the two refractory HEAs discussed above but deforms in compression up to 50% strain without any sign of fracture. This alloy showed good work hardening at a steady rate. Figure 8.4 shows uniform deformation behavior and evidences of twins and flow lines. Thus, the alloy can be strained to required strength and still has good ductility. Because most BCC alloys and ordered BCC (B2-type) intermetallic phases are brittle and tend to fracture by cleavage, this finding provides a good basis for developing ductile BCC alloys by modifying the composition.

Most of the studies measure only compressive properties. Yet, there are a few studies done exclusively to find out the tensile behavior of HEAs. Tensile behavior of the alloy Al_0.5CoCrCuFeNi was studied on the cold-rolled sample with 80% reduction (Tsai et al., 2010a). At room temperature, the alloy showed yield strength of 1292 MPa, ultimate tensile strength of 1406 MPa, and elongation of 6%. Thus, this alloy has a combination of strength and ductility better than cold-rolled 304 stainless steels. These cold-rolled samples were further annealed at 900°C and their tensile properties were studied. The combination of yield strength (656 MPa), ultimate tensile strength (796 MPa), and elongation (29%) of this alloy is better than that of 304 stainless steel (310 and 620 MPa, and 30%, respectively). The alloy in the as-annealed state thus has remarkable mechanical properties for engineering applications.

Hemphill et al. (2012) were the first to study the fatigue behavior of an Al_0.5CoCrCuFeNi HEA. The arc-melted samples were annealed at 1000°C for 6 h, water quenched, and cold rolled. The rolled sheets were subsequently machined to fatigue samples for four-point bending fatigue test. The fatigue investigation indicates that the fatigue behavior of HEAs compares favorably with that of many conventional alloys, such as steels, titanium alloys, and advanced BMGs with a fatigue endurance limit of between 540 and 945 MPa and a fatigue endurance limit to ultimate tensile strength ratio of between 0.402 and 0.703. The encouraging fatigue resistance demonstrates that Al_0.5CoCrCuFeNi HEAs without surface defects may be useful in future applications where fatigue is a factor.

All the above studies on room-temperature mechanical properties are carried out on cast or wrought alloys. As discussed previously, when the alloys are manufactured by MA and sintering, usually better strength and hardness are achieved. The nanocrystalline nature of the structure causes the increase in hardness along with solid solution strengthening. Hardness of AlCrCuFeTiZn solid solution was 2 GPa in the MA-sintered condition. The same alloy when processed using MA-vacuum hot pressing (VHP) had a hardness of 9.50 GPa and compressive strength of 2.19 GPa and those after MA-hot isostatic pressing (HIP) were 10.04 and 2.83 GPa, respectively. Wear resistance of the alloy was found to be higher than that of the commercially used materials such as Ni-hard-faced alloy. Similarly, hardness and compressive strength of the AlCoCuNiTiZn HEA processed by MA-VHP were found to be 7.55 and 2.36 GPa, respectively. When HIP is used for the same alloys, hardness and compressive strength were 8.79 and 2.76 GPa, respectively (Varalakshmi et al., 2008, 2010a–c).

Al_xCoCrCuFeNi (x=0.45, 1, 2.5, and 5) alloys synthesized by MA and SPS were studied by Sriharitha et al. (2014). Highest specific hardness of 960 HV was achieved in the sintered Al₅CoCrCuFeNi alloy. Hall–Petch analysis based on hardness measurements carried out on sintered samples reveals that the contribution of solid solution strengthening and order strengthening, in comparison to grain size strengthening, increases with the increase in Al content (Figure 8.5).

Oxide-dispersed AlCoCrFe HEA synthesized by MA and SPS was studied by Praveen et al. (2013a). High hardness values of 1050 and 1070 HV were observed without and with oxide dispersion, respectively. It appears that solid solution strengthening effect in HEAs has superseded the effect of oxide dispersion. A hardness of 570 HV was achieved in CoCrFeNi alloy by the same synthesis technique (Praveen et al., 2013b).

8.2.2 High-Temperature Mechanical Properties

Due to sluggish diffusion effect and second-phase strengthening, HEAs might exhibit high strength at elevated temperatures. For example, AlCo_xCrFeMo_0.5Ni alloy (x=1) shows a hardness of 347 HV at 1273 K, which is higher than that of Ni-based superalloys IN 718/IN 718H by 220 HV (Hsu et al., 2010b). AlCoCrFeMo_0.5Ni_x (x=0–1.5) also shows that hot hardness remains higher than that of superalloys IN 718 and IN 718H up to 1273 K (Figure 8.6). These alloys also possess a less negative value of softening coefficient (B_II) at high-temperature regime compared with the same superalloys (Juan et al., 2013). Besides, AlCoCuNiTiZn alloy produced by MA-VHP was found to be stable at elevated temperature of about 800°C as it retained its nanostructure (Varalakshmi et al., 2010c).

MoNbTaW and MoNbTaVW HEAs exhibit high yield strength of 405 and 477 MPa, respectively, at 1600°C which is higher than the melting point of superalloys (Senkov et al., 2010, 2011a). Senkov et al. (2013a, 2013b) also developed notable low-density refractory HEAs including NbTiVZr, NbTiV₂Zr, CrNbTiZr, and CrNbTiVZr having densities, 6.52, 6.34, 6.67, and 6.57 g/cc, respectively. Figure 8.7 compares the compression curves of these alloys from 298 to 1273 K, some of which reached 50% strain without fracturing. It is interesting to note that although the room-temperature ductility of Cr-containing HEAs is somewhat low, their high-temperature ductility is remarkably high. In addition, the specific strength of CrNbTiVZr alloys is far superior than that of the other three HEA alloys and compared to In718 and Haynes 230 superalloys. Therefore, CrNbTiVZr alloy shows the most attractive properties, such as considerably improved elevated-temperature strength, reduced density, and much higher melting point. Based on this, better combination of strength and ductility with low density could be expected if compositions and microstructure are further adjusted. As a result, Senkov et al. (2013a, 2013b) recommended a reasonable approach for the development of HEAs with both solid solution and ordered phases as candidates for the next generation of high-temperature structural materials.

Superplastic property was also studied for AlCoCrCuFeNi HEA (Kuznetsov et al., 2013). A fine equiaxed duplex structure with an average grain size of ~1.5 μm was formed after hot multidirectional forging. The forged alloy exhibited superplastic behavior in the temperature range of 800–1000°C at a strain rate of 10⁻³/s. Elongation to failure approached 604% at 800°C, decreased to 405% at 900°C, and increased again to 860% at 1000°C. Under the strain rate from 10⁻⁴ to 10⁻²/s at 1000°C, the superplastic elongation was in the range of 753–864%. An increase in the strain rate to 10⁻¹/s still gave an elongation of 442%. This study demonstrates that HEAs also have the potential in obtaining high-strain rate superplasticity at strain rates higher than 10⁻²/s.