With an ever-increasing demand for strong and ductile structural materials, substantial efforts have been devoted to developing new alloys [[1], [2], [3], [4], [5], [6]]. High- and medium-entropy alloys (HEAs and MEAs), e.g., CoCrFeMnNi and CrCoNi alloys, have drawn much interest due to their outstanding properties, particularly at cryogenic-temperature [1,4,[6], [7], [8], [9], [10]]. The substantial interaction between lattice dislocations and nano-twins, i.e. twinning induced ductility (TWIP) effect, is responsible for the cryogenic-temperature ductility of these alloys [4,6,11]. Recently, metastability engineering strategy, as a breakthrough of HEAs/MEAs design concept, enables the development of transformation-induced plasticity (TRIP) HEAs/MEAs [2,12]. Various HEAs/MEAs with improved mechanical properties were developed by reducing the face-centered cubic (fcc) phase stability [12,13]. Destabilizing fcc phase through reducing the Mn content in the Fe_80-xMn_xCo₁₀Cr₁₀ (at.%) HEA system, promotes deformation-induced phase transformation from fcc to hexagonal close-packed (hcp) phase, and then Fe₅₀Mn₃₀Co₁₀Cr₁₀ HEA overcome the strength-ductility trade-off at room-temperature [2]. An increase in Co content and a decrease in Fe and Ni contents reduced the fcc phase stability and stacking fault energy (SFE) in Co_xCr₂₅(FeNi)_70-xMo₅ HEA system, and then strong and ductile single-fcc-phase HEA was developed by promoting the fcc→hcp martensitic transformation at room-temperature [13].

However, despite such significant progress, the utilization of metastability-engineering strategy has mainly focused on how to reduce the stability of the parent phase [8,12,13]. Furthermore, the cryogenic deformation mechanism of these TRIP HEAs with fcc→hcp martensitic transformation has not been fully studied, especially for the Fe₅₀Mn₃₀Co₁₀Cr₁₀ HEA with engineering application potential. The rate of martensitic transformation is a critical parameter which determines strain hardening capability in metastable fcc alloys [3]. In addition, it is well known that the ductility of hcp crystal structure is worse than that of fcc, especially at cryogenic temperature [14,15]. If the fcc phase is very unstable, i.e., most of them are transformed into hcp phase under low strain level, and thus the ductility of the alloy will become poor. Therefore, it is of great significance to adjust transformation rate to maximize utilization of TRIP effect. It is found that fcc→hcp martensitic transformation is influenced by grain size and SFE [[13], [14], [15]]. The refined grain size and high SFE are both unfavorable for the phase transformation [2,14].

In this work, the effects of phase transformation rate on mechanical properties of Fe₅₀Mn₃₀Co₁₀Cr₁₀ HEAs were investigated at room and cryogenic temperature. The phase transformation rate in Fe₅₀Mn₃₀Co₁₀Cr₁₀ was reduced via refining the grain size. Interestingly, both the cryogenic ductility and strength are increased with the decrease of grain size, which originates from the retarded but more persistent TRIP effect. To further verify the effect of phase transformation rate on the ductility of HEAs, the mechanical properties of Fe₄₀Mn₄₀Co₁₀Cr₁₀ HEAs with higher SFE were also investigated at room and cryogenic temperature, respectively. Our results will provide a potent way for the future development of HEAs.

The Fe_xMn_80-xCo₁₀Cr₁₀ (x = 50 and 40 at.%) HEAs were fabricated by arc melting under Ar atmosphere in a water-cooled copper hearth, hereinafter denoted by Fe50 and Fe40, respectively. To ensure chemical homogeneity, the ingots was flipped and remelted 5 times with electromagnetic stirring and then drop casted into a water-cooled copper mold with dimension of 40 mm × 40 mm × 10 mm. These as-cast plates are coarse-grained (CG) samples. In order to obtain grain-refined (FG) samples, the Fe 50 and Fe40 plates was cold-rolled to a thickness reduction of 45% and 80% respectively, and subsequently annealed at 900 °C for 10 min followed by water-quenching. Flat dog-bone shape specimens with gauge dimensions of 10 mm × 2 mm × 1 mm were cut from these plates. The tensile tests were conducted using SUNS CMT5105 testing machine with crosshead speed of 0.5 mm/min at 298 K and 77 K, respectively. The structures of the specimens were examined by X-ray diffraction (XRD) using CuKα radiation and electron backscatter diffraction (EBSD). A step size of ∼300 nm was used for EBSD scans. The elemental distribution in the samples was investigated using energy-dispersive X-ray spectroscopy (EDS). Three samples were also performed XRD measurements at 77 K.

The EBSD IPF and phase maps in Fig. 1 show the grain morphologies and phase constituent of the CG and FG Fe50 HEAs before tensile tests, respectively. These data reveal that the alloy consists of fcc and hcp phases. All constitute elements in the CG and FG HEA are uniformly distributed, confirming the compositional homogeneity of the sample (see Figs. S1 and S2 in Supporting Information). The high angle grain boundaries (HAGB) and twin boundaries were marked in the phase maps in terms of black and red lines, respectively. The grain size of fcc matrix is larger than 70 μm in the CG Fe50 HEA, and the hcp phase with dendrite morphology is formed within the fcc matrix (Fig. 1(a)). From EBSD analyses, the average fractions of the fcc and hcp phases are approximately about 42% and 58%, respectively. Although the volume fraction of the deformation-induced hcp phase will further increases after rolling with thickness reduction of 45%, the alloy will occur thermally induced abundantly reverse transformation from the hcp phase to the fcc phase after annealing at 900 °C for 10 min [16]. Therefore, after rolling and heat treatment, the hcp fraction is largely reduced to be about 25% and the grain size of fcc matrix was refined to less than 10 μm, as shown in the FG Fe50 HEA (Fig. 1(b)). These experimental results are consistent with the results reported in Ref. [14]. Thus, we have successfully prepared CG and FG Fe50 HEAs in anticipation.

The engineering stress-strain curves for CG and FG Fe50 HEAs at 298 K and 77 K are presented in Fig. 2(a). The mechanical properties of the HEAs were listed in Table 1. The yield strength and ultimate tensile strength (UTS) of the Fe50 HEAs increases significantly with the reduction of grain size at 298 K, but its ductility basically remains unchanged. This is mainly due to the Hall-Petch effect. In addition, the mechanical responses of the Fe50 HEAs show two distinct features as the temperature decreases. Firstly, the strength of the Fe50 HEAs increased dramatically with decreasing the testing temperatures from 298 K to 77 K. This phenomenon is associated thermally-activated process of metallic alloys [9]. In comparison to tensile properties at 298 K, the cryogenic UTS of all the CG and FG Fe50 HEAs were increased by about 61% and 57%, respectively. The second feature is that the ductility of CG Fe50 HEA decreases greatly from 42% to 17% as the temperature decreases, but that of FG Fe50 HEA increases slightly from 40% to 44%. In addition, the strain hardening rate of FG Fe50 HEA is higher than that of CG Fe50 HEA at both 298 K and 77 K (Fig. 2(b)). Obviously, the CG Fe50 HEA exhibits ductile-brittle transition at cryogenic-temperature. As mentioned above, HEAs generally have higher ductility at cryogenic-temperature. The significant ductility change at 77 K is related to the phase transformation rate, which will be discussed in detail later.

Although TRIP effect is utilized to obtain good mechanical properties of Fe50 HEAs at room-temperature, cryogenic-temperature makes the fcc phase very unstable, because the SFE decreases with the decrease of temperature and then promotes the fcc→hcp phase transformation [17,18]. In addition, the flow stress increases at 77 K, which also enhances the phase transformation [18]. It can be speculated that the phase transformation needs to be retarded but more persistent to improve the cryogenic-temperature ductility of these HEAs. Tuning grain size is an effective method to retard phase transformation [14,18]. The ductility of FG Fe50 HEAs at 77 K exhibits a slightly increase rather than a significant decrease, compared to that at 298 K, which should be due to the retard phase transformation.

Fig.3 shows the EBSD IPF and phase maps of the CG and FG Fe50 HEA samples after fracture with different local strain at room and cryogenic temperature, respectively. For comparison, we define the phase transformation rate as the average increment of the volume fraction of hcp phase per strain (ΔV_hcp/strain), which is listed in Table 1. Note that this value only represents the relative speed of the phase transformation. It can be seen from the IPF maps that the grain size of Fe50 HEA is refined after fracture, compared with the original structure (Fig. 1), which is probably due to the phase transformation and the interaction between hcp and fcc phases [16]. It can be seen from the phase map that ∼19% of fcc phase in the CG Fe50 HEA was transformed into hcp phase after ∼15% strain at room-temperature (Figs. 1(a) and 3 (a)). For FG Fe50 HEA, about 65% of fcc phase was transformed into hcp phase after 40% strain at room-temperature (Figs. 1(b) and 3 (b)). The phase transformation rate of the CG and FG HEA at room temperature is related to the amount of plastic strain [2,14,19]. The larger degree of plastic deformation, the slower phase transformation rate [14,19]. Both the CG and FG HEA have larger ductility (∼40%) at room temperature, which has little to do with the phase transformation rate. This is because the phase transformation products (hcp phase) in the HEA is ductile at room-temperature [19]. This phase transformation is the main reason for the good ductility of Fe50 HEA at room-temperature [2].

At cryogenic-temperature, even though the CG HEA underwent only ∼15% strain, most of the fcc phases were transformed into hcp phase (Fig. 3(c)). The fast phase transformation rate makes the CG HEA exhibits poor ductility, because the phase transformation products (hcp phase) is brittle at cryogenic temperature. Interestingly, FG Fe50 HEA has slower phase transition rates at cryogenic temperature (Fig. 3(d), Table 1), and thus exhibits larger ductility. Therefore, slower phase transformation rates and more persistent phase transformation is very important for HEAs to achieve higher ductility. The findings obtained from these experimental phenomena may be apply to other alloy systems. Specifically, if the phase transformation product is ductile, the phase transition rate has little influence on the ductility; if the phase transition product is brittle, the ductility of the alloy will be significantly affected.

High SFE can also inhibit fcc→hcp phase transformation, and may also properly reduce phase transformation rate [2,20]. To further verify the effect of phase transformation rate on the ductility of HEAs, the Fe40 HEA was prepared and its mechanical behavior was also investigated. All chemical elements in the Fe40 HEA are uniformly distributed (see Fig. S3). The EBSD IPF and phase maps for the Fe40 HEA specimens before tensile tests are shown in Fig. 4(a). The Fe40 HEA consists of fcc phase and a small amount (3%, area fraction) of hcp phase. He et al. [17] also found a small amount of hcp phase in the undeformed Fe40 HEA. Note that the hcp phase fraction in the Fe40 HEA is much less than that in FG Fe50 HEA (Fig. 1(b)), indicating that the Fe40 HEA has higher SFE. Both the ductility and train hardening rate of the Fe40 HEA are greatly improved when the temperature decreased from 298 K to 77 K (Fig. 2), which is mainly due to the TRIP effect [17,21]. In addition, the ductility of the Fe40 HEA at 77 K is also higher than that of FG Fe50 HEA (Fig. 2(a)).

To quantitatively evaluate the phase compositions and microstructural features of the Fe40 HEA deformed at room and cryogenic temperature, EBSD maps for the fractured specimens are shown in Fig. 4(b) and (c), respectively. It is found that the grains have been severely elongated after deformation. The number of twins increases greatly after the Fe40 HEA fracture at room-temperature, while only trace amounts (3%) of fcc→hcp phase transformation are identified (Fig. 4(b)). Therefore, its ductility at room-temperature is mainly attributed to TWIP effect [22]. However, ∼37% of fcc phase in the Fe40 HEA was transformed into hcp phase after ∼50% cryogenic strain (Fig. 4). The phase transformation rate in the Fe40 HEA is lower than that of FG Fe50 HEA at 77 K, which accounts for the fact that the Fe40 HEA has higher ductility than that of FG Fe50 HEA at 77 K. In conclusion, tuning SFE is also an effective method to increase the ductility of HEAs by decreasing its fcc→hcp phase transformation rate at cryogenic temperature.

Although the mechanism of deformation-induced phase transformation in HEAs/MEAs has been investigated, the effect of pure temperature on the phase transformation has not been considered yet [12,17,18]. Similarly, one interesting question in this work is still remained, i.e., how about the pure temperature affects the fcc→hcp phase transformation in studied HEAs. To address this question, we performed low temperature XRD measurements for CG and FG Fe50 HEAs before tensile tests at both 298 K and 77 K together with the fractured area of the CG Fe50 HEA after tensile test at both 298 K and 77 K in Fig. 5. From XRD patterns recorded at 298 K with 77 K for three samples, it is found that the pure temperature change from 298 K to 77 K does not obviously alter the relative fractions of hcp phase in all three studied samples. This means that the fcc→hcp phase transformation in the present HEAs is mainly caused by stress. Consequently, the stress-induced fcc→hcp phase transformation improves the strength and the ductility of FG Fe50 HEA sample in Fig. 2.

We studied the mechanical properties and phase transformation rate-dependent deformation mechanisms of Fe_xMn_80-xCo₁₀Cr₁₀ HEAs at 298 K and 77 K, respectively. The conclusions are as follows:

(1) The CG Fe₅₀Mn₃₀Co₁₀Cr₁₀ HEA has good ductility of 42% at 298 K, but its ductility sharply decreased to 17% at 77 K. Because the hcp phase is brittle at cryogenic temperature and its fcc→hcp phase transformation rate is too fast to maximize utilization of TRIP effect.

(2) The ductility of the FG Fe₅₀Mn₃₀Co₁₀Cr₁₀ HEA increases slightly from 40% to 44% as the temperatures decrease from 298 K to 77 K, which is mainly due to slowly phase transformation rate and more sustained TRIP effect caused by grain refinement.

(3) The ductility of Fe₄₀Mn₄₀Co₁₀Cr₁₀ HEA at 77 K (77%) is much larger than (54%) that at 298 K, because higher SFE also postpones phase transformation and makes the TRIP effect more persistent.

(4) The fcc→hcp phase transformation in these HEAs is mainly caused by stress.