The elemental distributions in the solidification microstructure of the AM ingot of the TiNbTaZrMo bio-HEA were explained by a thermodynamic calculation that was focused on the formation of the BCC phase through the solidification of the single liquid phase while using FactSage (version 6.4, FactSage, Ecole Polytechnique, Montreal, Canada) and SGTE2007 [
17]. We discuss the elemental distributions of the dendrite and interdendrite regions in the TiNbTaZr MEA, TiNbTaZrMo bio-HEA, and TiNbTaZrX (X = V and W) HEAs based on the distribution coefficients at
TL, as evaluated by the thermodynamic calculation only assuming the liquid and BCC phases using FactSage (version 7.2) and SGTE2017. The thermodynamic calculation results for the solidification in the TiNbTaZrMo bio-HEA [
16] are updated in this study while using SGTE2017.
Table 6 summarizes the thermodynamic calculation results of the distribution coefficients at
TL of the TiNbTaZr MEA, TiNbTaZrMo bio-HEA, and TiNbTaZrX (X = V and W) HEAs. It should be noted that no significant differences in the distribution coefficients at
TL in the TiNbTaZrMo bio-HEA were observed between the previous study (SGTE2007) [
16] and this study (SGTE2017). The distribution coefficients of Ti and Zr were significantly smaller than unity, while that of Ta was significantly higher than unity, regardless of the alloy system, which indicates the formation of a Ta-rich dendrite region and the movement of Ti and Zr from the dendrite region to the residual liquid, which led to the formation of Ti–Zr-rich interdendrite regions. The distribution coefficient of V in the TiNbTaZrV HEA was smaller than unity, while those of Mo in the TiNbTaZrMo bio-HEA and W in TiNbTaZrW were larger than unity. The EPMA-WDS analysis results of the AM ingots (
Figure 6 and
Table 4) show the following tendency in the TiNbTaZrMo bio-HEA and the TiNbTaZrX (X = V, and W) HEA. The interdendrite region in the TiNbTaZrV HEA (
Figure 6b) was enriched in V, while the interdendrite regions in the TiNbTaZrMo bio-HEA (
Figure 6c) and TiNbTaZrW HEA (
Figure 6d) were enriched in Mo and W, respectively. No significant differences in the elemental distributions at the dendrite and interdendrite regions were observed between AM (
Figure 6c and
Table 3c) and CCLM (
Figure 9 and
Table 5) ingots in the TiNbTaZrMo bio-HEA. The elemental distributions of the X elements in the TiNbTaZrX (X = V, Mo, and W) HEA AM ingots and those in the CCLM ingot of the TiNbTaZrMo bio-HEA can be explained by the distribution coefficients that were evaluated by the thermodynamic calculation without any discrepancy. The control of
TL is important for further development of bio-HEAs, particularly for material processing to suppress fabrication cost. A lowering of the process temperature during the casting process is significantly effective for energy savings. From an engineering viewpoint, the casting processes were strictly limited when the process temperature was over 2273 K (2000 °C), especially for the alloys that contain Ti and Zr. The significantly high
TL of RHEAs and TiNbTaZrMo bio-HEA limits the casting process, and the arc melting process is the main route in the fabrication of RHEAs [
12]. A lower T
L was also effective in suppressing the formation of cold shuts during the arc melting process; this will be discussed in a later section. The correspondence between the solidification microstructure analysis results (
Figure 3,
Figure 4,
Figure 5,
Figure 6,
Figure 7,
Figure 8 and
Figure 9 and
Table 2,
Table 3,
Table 4 and
Table 5) and the thermodynamic calculation results (
Table 6) implies that the thermodynamic calculation is effective in predicting
TL.
Figure 10 shows pseudobinary phase diagrams that focus on the solidification in the TiNbTaZr MEA, TiNbTaZrMo bio-HEA, and TiNbTaZrX (X = V and W) HEAs. The pseudobinary phase diagrams were constructed while only using the Gibbs free energy of the single liquid and single BCC phases, which were used to determine the
TL and
TS of the BCC phase.
TL and
TS are represented by the solid and broken red lines in
Figure 10, respectively. In the TiZr–TiNb
2Ta
2Zr alloy system containing the TiNbTaZr MEA (
Figure 10a),
TL and
TS monotonically increased with the Nb and Ta concentrations of the TiNb
xTa
xZr (
x = 0 − 2) alloy.
TL monotonically decreased with an increase in the V concentration of the TiNbTaZrV
x alloy in the TiNbTaZr–TiNbTaZrV
2 alloy system containing the TiNbTaZrV HEA (
Figure 10b). In contrast,
TL was monotonically increased with the Mo concentration of the TiNbTaZrMo
x alloy in the TiNbTaZr–TiNbTaZrMo
2 alloy system containing the TiNbTaZrMo bio-MEA (
Figure 10c).
Figure 10d shows the pseudobinary phase diagram of the TiNbTaZr–TiNbTaZrW
2 alloy system containing the TiNbTaZrW HEA. The increase in the W concentration of the TiNbTaZrW
x alloy led to a significant increase in
TL and a significant decrease in
TS. The changes in
TL and
TS with the concentration of the X element of the TiNbTaZrX (X = V, Mo, and W) HEA strongly depend on the X element. It was observed that an increase in the concentration of the low melting-temperature V (2183 K) led to a monotonic decrease in
TL and that increases in the concentrations of high melting-temperature Mo (2896 K) and W (3695 K) led to a monotonic increase in
TL.
Figure 11 shows the average melting temperature (
) vs. liquidus temperature that was estimated by a thermodynamic calculation (
TL) plot. The parameter (
) is expressed as
where
xi is the mole fraction of the
i-th element and (
Tm)
i is the melting temperature of the
i-th element. The black open circle (○) represents the TiNbTaZr MEA and the TiNb
xTa
xZr (
x = 0 − 2), the blue open square represents TiNbTaZrV
x (
x = 0 − 2), the red-filled circles represent TiNbTaZrMo
x (
x = 0 − 2), and the green-filled square represents TiNbTaZrW
x (
x = 0 − 2). The arrows indicate the direction of the increase in the value of
x. The inset is the magnified image. The
TL values of TiNb
xTa
xZr (
x = 0 − 2), TiNbTaZrV
x (
x = 0 − 2), TiNbTaZrMo
x (
x = 0 − 2), and TiNbTaZrW
x (
x = 0 − 2) were lower than
, regardless of the alloy system and the value of
x. An increase in the value of
led to an increase in
TL, regardless of the alloy system. This indicates that
can be used as a rough indicator for alloy design in decreasing
TL in TiNb
xTa
xZr and TiNbTaZrX
x (X = V, Mo, and W) HEAs. The experimental measurements of
TL in RHEAs and bio-HEAs are very challenging, due to the very high
TL and reactivity of the molten state of the constituent elements; moreover, conventional solidification behavior analysis using differential thermal analysis (DTA) is not applicable. The thermodynamic calculation of
TL is helpful in the design of bio-HEAs with low
TL values.
Cold shuts were observed in the AM ingots, as shown in
Figure 4d in the TiNbTaZrMo bio-HEA and
Figure 5b2 in the TiNbTaZrV HEA.
Figure 12 shows a possible mechanism for the formation of a cold shut during the arc melting process that was observed in the present study.
Figure 12a shows the ingot on the water-cooled Cu hearth, and
Figure 12b shows the ingot after being turned over. During arc melting, the ingot in the upper part melted, while the one in the bottom part, which made contact with the water-cooled copper hearth, remained in its solid state as the non-melting zone (
Figure 12c1). The flow of molten metal occurred, as shown in
Figure 12c2, resulting in the formation of a cold shut during arc melting, as shown in
Figure 12c3. The significantly high
TL in the TiNbTaZr MEA, TiNbTaZrMo bio-HEA, and TiNbTaZrX (X = V and W) HEAs led to the existence of the large non-melting zone in the specimens during the arc melting process, which resulted in the formation of cold shuts in the AM ingots. Based on the mechanism in
Figure 12, lowering T
L is considered to be effective in suppressing the formation of a cold shut during the arc melting process. The formation of a cold shut was not observed in the CCLM ingots in the present study, which implied that cold crucible levitation melting processes can be effectively applied in the fabrication of bio-HEAs, and this is to be the topic of a future work.
Finally, the annealing induced structural change in the TiNbTaZrMo bio-HEA was briefly mentioned to consider the possibility of the formation of a single BCC phase. As denoted in the literature review of RHEAs, the formation of dual BCC phases has been reported in a number of RHEAs, which is similar to the present study. One may consider that a single BCC phase can be obtained after annealing. The microstructure of annealed AM ingots is shown in
Figure 13 as a typical example of the annealed structure in the TiNbTaZrMo bio-HEA. From an engineering viewpoint, the annealing temperature was strictly limited when the process temperature was over 1273 K (1000 °C), because of potential damage of the annealing furnace, the limitation of the furnace materials, and the energy cost. In the present study, the annealing temperature was set at 1273 K.
Figure 13 shows an SEM-BSE image and EPMA elemental maps of the central region of the AM ingots of the TiNbTaZrMo bio-HEA annealed at 1273 K for 168 h (one week). The coarsening of the dendrite by annealing is detected in SEM-BSE images (
Figure 13a). The EPMA elemental maps (
Figure 13b) indicate that the Ti and Zr elements were still segregated in the interdendrite regions. In spite of the relatively long time of annealing, a single BCC phase formation was not observed in the AM ingots of the TiNbTaZrMo bio-HEA. The solidification microstructure affected the structure that was annealed at 1273 K for 168 h. The present study did not focus on the possibility of a single BCC phase structure formation. However, it can be concluded that control of the solidification microstructure and prediction of the segregation are important, not only for the ingots of bio-HEAs, but also for the annealed products of bio-HEAs.