Phase Equilibria, Thermodynamics and Solidified Microstructure in the Copper–Zirconium–Yttrium System

A copper alloy with the addition of zirconium and yttrium is an attractive high strength and high conductivity (HSHC) copper alloy. The study of the solidified microstructure, thermodynamics and phase equilibria in the ternary Cu–Zr–Y system is expected to provide new insight into designing an HSHC copper alloy. In this work, the solidified and equilibrium microstructure and phase transition temperatures in the Cu–Zr–Y ternary system were studied by X-ray diffraction (XRD), electron probe microanalysis (EPMA) and differential scanning calorimeter (DSC). The isothermal section at 973 K was experimentally constructed. No ternary compound was found, while the Cu6Y, Cu4Y, Cu7Y2, Cu5Zr, Cu51Zr14 and CuZr phases substantially extended into the ternary system. According to the experimental phase diagram data from the present work and the literature, the Cu–Zr–Y ternary system was assessed using the CALPHAD (CALculation of PHAse diagrams) method. The isothermal sections, vertical section and liquidus projection calculated by the present thermodynamic description agree well with the experimental results. This study not only establishes a thermodynamic description of the Cu–Zr–Y system, but also contributes to the design of a copper alloy with the required microstructure.


Introduction
Copper alloy has excellent electrical and thermal conductivity, good ductility and moderate strength. It is widely used in new energy, electrical and electronics, rail transit and so on [1][2][3]. With the continuous iterative development of products, the requirements for the properties of copper alloy are gradually increasing [4][5][6][7][8]. The high strength and high conductivity (HSHC) copper alloy is drawing growing interest. The influence of zirconium (Zr) on the properties of copper alloy has been systematically studied by many researchers [9][10][11]. Peng et al. [9] studied phase transition for the Cu-0.12 wt.% Zr alloy in the course of aging at 723 K and concluded that the Cu 5 Zr precipitation contributes to the strength. Du et al. [10] investigated the Zr-containing precipitate evolution of the copperchromium-zirconium alloy and noted that the yield strength was mainly influenced by the Cu 5 Zr phase, Cr-rich precipitation and Zr-rich atomic clusters. To further improve the performance of copper alloy, more and more studies [12][13][14][15][16] have been conducted on the microstructure and properties influenced by the rare earth elements added to copper alloy. The properties for the copper-zirconium-yttrium alloy after cold-rolling and aging were studied systematically by Gao et al. [12] using selected area electron diffraction (SAED) and transmission electron microscopy (TEM), and they concluded that secondphase precipitation can be promoted by adding yttrium (Y). Wang et al. [15] noted that the addition of Y into a Cu-Cr alloy will inhibit the growth of Cr precipitation and decrease the dislocation density, thus increasing the ultimate tensile strength and hardness after cold-rolling and aging. As depicted above, the addition of Zr and Y can promote strength The purities of the elements used were Cu-99.99 wt.%, Zr-99.95 wt.% and Y-99.99 wt.%. The alloys were prepared in an arc furnace (WKDHL-1, Opto-electronics Co. Ltd., Beijing, China) with a water-cooled copper crucible in a highly pure argon environment. The alloys were flipped and remelted at least four times to ensure uniformity. The weight of each sample was usually between 4-6 g. The weight loss of every alloy was lower than 2.5%. The real overall composition of the alloys was determined through inductively coupled plasma-optical emission spectrometry (ICP-OES, Thermo Fisher Scientific Inc., Waltham, MA, USA). Each alloy was cut into two parts using wire electrical discharge machining (EDM). One part was analyzed as an as-cast alloy. The other part was encapsulated in a vacuum quartz pipe. A total of 12 samples were melted and numbered sequentially as A1-A12, and their compositions are listed in Table 1. Alloys A1-A8 were annealed at 973 K for 40 days, then water quenched. Alloys A9-A12 were annealed at 1073 K for 40 h, and then annealed at 973 K for 30 days and finally water quenched. XRD (Bruker-AXS D8) and EPMA with wavelength dispersive X-ray spectroscopy (WDX) (JXA-8230, JEOL, Japan) were then used to determine the microstructure and components of these alloys. Afterwards, differential scanning calorimeter (DSC, Netzsch, Germany) measurements of the phase change temperatures of the annealed alloys were performed. The DSC device was performed in an Al 2 O 3 crucible under a continuous flow of argon (99.998 wt.% purity) with a heating rate of 5 K/min from 303 K to 1323 K. The temperatures for the invariant reactions and liquidus were defined by the initial and peak temperatures, respectively.

Thermodynamic Models
The phase diagram of the Cu-Y system was minutely adjusted in the present work based on the work of Fries et al. [20]. The phase diagrams of the Cu-Zr system by Liu et al. [21] and the Zr-Y system by Bu et al. [22] are accepted in this work due to the consistency of the thermodynamic databases of copper alloy [23][24][25][26]. The crystallographic data for stable phases in the Cu-Zr-Y system are shown in Table S1 [20,[27][28][29][30][31][32][33][34][35] in the supporting information. The calculated binary sub-systems phase diagrams are shown in Figures 1-3.

Pure Elements
The Gibbs energy function for the pure element i ( =Cu, Zr and Y i ) is taken from the Scientific Group Thermodata Europe (SGTE) by Dinsdale [36] and described in the form of: (1) where means the mole enthalpy of element i relative to the stable element reference (SER) at 298.15 K and 1 bar, and T is the absolute temperature in Kelvin.

The Solution Phases
In the Cu-Y and Cu-Zr-Y systems, (Cu) is the fcc phase, (αZr) and (αY) are the hcp phases (βZr) and (βY) are the bcc phases. The liquid, fcc, hcp and bcc phases serve as a substitutional solution model. The molar Gibbs free energy of the solution phase is described as follows: (2) where is the mole fraction of element i ( =Cu, Zr and Y i ); denotes the molar Gibbs energy of component i in the state phase ϕ ; and R represents the gas constant.
represents the excess Gibbs energy, which is described by the Redlich-Kister (R-K) [37] polynomial. In the Cu-Zr-Y system, is represented as follows. (3)

Pure Elements
The Gibbs energy function for the pure element i (i= Cu, Zr and Y) is taken from the Scientific Group Thermodata Europe (SGTE) by Dinsdale [36] and described in the form of: where H SER i means the mole enthalpy of element i relative to the stable element reference (SER) at 298.15 K and 1 bar, and T is the absolute temperature in Kelvin.

The Solution Phases
In the Cu-Y and Cu-Zr-Y systems, (Cu) is the fcc phase, (αZr) and (αY) are the hcp phases (βZr) and (βY) are the bcc phases. The liquid, fcc, hcp and bcc phases serve as a substitutional solution model. The molar Gibbs free energy of the solution phase is described as follows: where x i is the mole fraction of element i (i= Cu, Zr and Y); 0 G ϕ i denotes the molar Gibbs energy of component i in the state phase ϕ; and R represents the gas constant. ex G ϕ m represents the excess Gibbs energy, which is described by the Redlich-Kister (R-K) [37] polynomial. In the Cu-Zr-Y system, ex G ϕ m is represented as follows.
where m L ϕ i,j is the mth binary interaction parameter, which can be shown by: where L ϕ Cu,Y,Zr is ternary interaction parameter, which can be shown by: The interaction parameters were optimized in this work based on the available experimental data.

Intermetallic Compounds
There are 11 intermetallic compounds in the system. The Cu 2 Y, CuY, Cu 8 Zr 3 , Cu 10 Zr 7 and CuZr 2 phases with ignorable homogeneity are considered as stoichiometric compounds according to the literature and current experimental data, whose Gibbs energy description is given as follows: where m and n stand for the ratios of stoichiometry and 0 G HSER A and 0 G HSER B represent the molar Gibbs energy referring to the SER states of A and B, respectively.
The Cu 4 Y, Cu 6 Y, Cu 7 Y 2 , Cu 5 Zr, Cu 51 Zr 14 and CuZr phases with solubilities of a third element were described by a two-sublattice model. Take a phase ϕ modeled as (A, B) p (A, C) q for example. The molar Gibbs energy formula is given as Equation (7): where y A and y A are the fraction of the constituent A in the first and second sublattice, respectively. i L represents the ith optimized interaction parameter in this work.

Microstructure and Phase Transition Temperatures Analysis
In this work, twelve alloys were prepared to determine the primary phase and phase equilibria of the Cu-Zr-Y system at 973 K. Table 1 summarizes the results measured by EPMA and XRD, including the primary phase and solidification paths of the as-cast alloy, as well as the equilibria phases of the annealed alloy. In this work, eight primary phases, i.e., (αY), (αZr), CuZr 2 , Cu 51 Zr 14 , Cu 5 Zr, CuY, Cu 7 Y 2 and Cu 6 Y, were found. The representative alloys are discussed in detail below. Table 2 lists the temperatures of the invariant reactions and liquidus measured by DSC. ,d presents the back-scattered electron (BSE) micrograph and XRD pattern of the as-cast alloy A1 Cu 90 Zr 8 Y 2 (at.%). The primary phase in the images is shown in bold. The analysis shows that the gray phase of Cu 5 Zr is the primary phase, and there are a lot of eutectic structures around it. During the arc-melting process, the side close to the copper crucible cools faster and the solidified microstructure is coarser, while the microstructure further away from crucible is finer. As shown in Figure 4a, the microstructure on the right is coarser than that on the left side. The coarser and finer microstructures in one alloy were also found in our previous Ag-Cr-Zr alloy [24]. Combined with the XRD results, the eutectic structure should be composed of (Cu) + Cu 5 Zr + Cu 6 Y. It can be postulated that the Cu 5 Zr phase solidifies first during the solidification process. Then the liquidus transformation component point soon contacts L → (Cu) + Cu 5 Zr eutectic reaction. Finally, the eutectic equilibrium reaction of L → (Cu) + Cu 5 Zr + Cu 6 Y occurs quickly, therefore a large number of eutectic structures are formed. Figure 5 shows a representative DSC curve for the A1 alloy with a heating rate of 5K/min. There are two visible peaks on this curve. Combined with the previous solidification analysis, the onset temperature of 1123 K of the first peak corresponds to the reaction of L → (Cu) + Cu 5 Zr. The peak temperature of 1153 K of the second peak corresponds to the liquidus temperature of the alloy.

Microstructure of Annealed Alloys
The experimental results of 12 alloys annealed at 973 K are summarized in Table 1. No ternary compound exists. Four two-phase regions, i.e., (Cu) + Cu 6 Y, (Cu) + Cu 5 Zr, Cu 51 Zr 14 + Cu 4 Y, Cu 51 Zr 14 + Cu 10 Zr 7 , and six three-phase regions, i.e., (Cu) + Cu 6 Y +Cu 5 Zr, Cu 10 Zr 7 + Cu 2 Y + Cu 7 Y 2 , (αY) + (αZr) + CuY, Cu 2 Y + CuZr 2 + CuZr, Cu 2 Y + CuZr 2 + CuY, (αZr) + CuZr 2 + CuY, are determined experimentally. The relationships are discussed as follows. Figure 6a,d shows the BSE micrograph and XRD pattern for alloy A1 Cu 90 Zr 8 Y 2 (at.%), respectively. The annealed microstructure of alloy A1 is similar to that of the cast state shown in Figure 4a. The XRD result indicates the presence of the Cu 6 Y phase and the BSE micrograph illustrates the presence of the Cu 6 Y phase in the eutectic microstructure. Therefore, in conjunction with the analysis above, alloy A1 consists of white Cu 5 Zr, gray Cu 6 Y and black (Cu). The solubility of Y in Cu 5 Zr is 2.03 at.%. The solubility of Zr in Cu 6 Y is 6.61 at.%. Figure 6b,e shows the BSE micrograph and XRD pattern for alloy A3 Cu 69 Zr 10 Y 21 (at.%), respectively. The annealing alloy A3 is composed of Cu 10 Zr 7 (white), Cu 2 Y (gray) and Cu 7 Y 2 (dark). The concave-convex morphology of the Cu 7 Y 2 phase and Cu 2 Y phase is the same as the result of He et al. [18]. The solubility of Y in Cu 10 Zr 7 is 1.53 at.%. The solubility of Zr in Cu 7 Y 2 is 7.83 at.%. Figure 6c,f shows the BSE micrograph and XRD pattern for alloy A8 Cu 48 Zr 41 Y 11 (at.%), respectively. According to the results, alloy A8 is located in the three-phase region, i.e., bright CuZr 2 , grey CuZr and black Cu 2 Y. The solubility of Y in CuZr is 3.38 at.%. Cu69Zr10Y21 (at.%), respectively. The annealing alloy A3 is composed of Cu10Zr7 (white), Cu2Y (gray) and Cu7Y2 (dark). The concave-convex morphology of the Cu7Y2 phase and Cu2Y phase is the same as the result of He et al. [18]. The solubility of Y in Cu10Zr7 is 1.53 at.%. The solubility of Zr in Cu7Y2 is 7.83 at.%. Figure 6c,f shows the BSE micrograph and XRD pattern for alloy A8 Cu48Zr41Y11 (at.%), respectively. According to the results, alloy A8 is located in the three-phase region, i.e., bright CuZr2, grey CuZr and black Cu2Y. The solubility of Y in CuZr is 3.38 at.%.    Figure 7c,f show the BSE micrographs and XRD patterns for alloys A11 Cu78Zr18Y4 (at.%) and A12 Cu66Zr32Y2 (at.%), respectively. In the standard PDF cards for the Cu51Zr14 and Cu10Zr7 phases, there is no standard peak after 77.757° of 2θ, therefore the peak larger than 78° is not calibrated in the XRD pattern. According to the results, alloys A11 and A12 are both located in the grey Cu10Zr7 and dark gray Cu51Zr14 two-phase region, although their microstructures differ considerably. More Cu51Zr14 phases are present in alloy A11. The measured maximum solubility of Y in Cu51Zr14 is about 4.45 at.%. The solubilities of the 3rd element are determined in this work for the Cu-Zr-Y system. Noticeable solubilities were also measured in the Al-Zr-Y system by Liu et al. [38] and Bao et al. [39].    Figure 7c,f show the BSE micrographs and XRD patterns for alloys A11 Cu 78 Zr 18 Y 4 (at.%) and A12 Cu 66 Zr 32 Y 2 (at.%), respectively. In the standard PDF cards for the Cu 51 Zr 14 and Cu 10 Zr 7 phases, there is no standard peak after 77.757 • of 2θ, therefore the peak larger than 78 • is not calibrated in the XRD pattern. According to the results, alloys A11 and A12 are both located in the grey Cu 10 Zr 7 and dark gray Cu 51 Zr 14 twophase region, although their microstructures differ considerably. More Cu 51 Zr 14 phases are present in alloy A11. The measured maximum solubility of Y in Cu 51 Zr 14 is about 4.45 at.%. The solubilities of the 3rd element are determined in this work for the Cu-Zr-Y system. Noticeable solubilities were also measured in the Al-Zr-Y system by Liu et al. [38] and Bao et al. [39]. Figure 8a,d shows the BSE micrograph and XRD pattern for alloy A4 Cu 33 Zr 7 Y 60 (at.%), respectively. The results show that alloy A4 is in the three-phase region for white αZr, grey αY and dark CuY. The rare earth Y is chemically active [40]. The oxidation of Y can be observed during the metallographic polishing process, which does not affect the phase equilibrium after vacuum annealing at 973 K. Combined with the XRD result, the black Y-oxide is Y 2 O 3 . The solubility of Cu in the hcp phase is about 2 at.%. Figure 8b,e shows the BSE micrograph and XRD pattern for alloy A9 Cu 53 Zr 12 Y 35 (at.%), respectively. The diagrams show that alloy A9 is in the three-phase region for white CuZr 2 , grey CuY and black Cu 2 Y. Figure 8c,f shows the BSE micrograph and XRD pattern for alloy A10 Cu 22 Zr 71 Y 7 (at.%), respectively. The diagrams show that alloy A10 is in the three-phase region for white αZr, grey CuZr 2 and dark CuY.
Based on the present experiments, the isothermal section at 973 K is constructed in Figure 9. Three two-phase regions and four three-phase regions agree well with those of He et al. [18]. The red dashed line in the experimental phase diagram indicates the phase regions that were not directly measured by the experiment. The alloys designed in the three-phase region of Cu 4 Y + Cu 6 Y + Cu 5 Zr tend to shift into the three-phase region of (Cu) + Cu 6 Y + Cu 5 Zr due to the small range of the phase region and the easy volatilization of Y during melting. This part has not been experimentally determined. Figure 8a,d shows the BSE micrograph and XRD pattern for alloy A4 Cu33Zr7Y60 (at.%), respectively. The results show that alloy A4 is in the three-phase region for white αZr, grey αY and dark CuY. The rare earth Y is chemically active [40]. The oxidation of Y can be observed during the metallographic polishing process, which does not affect the phase equilibrium after vacuum annealing at 973 K. Combined with the XRD result, the black Y-oxide is Y2O3. The solubility of Cu in the hcp phase is about 2 at.%. Figure 8b,e shows the BSE micrograph and XRD pattern for alloy A9 Cu53Zr12Y35 (at.%), respectively. The diagrams show that alloy A9 is in the three-phase region for white CuZr2, grey CuY and black Cu2Y. Figure 8c,f shows the BSE micrograph and XRD pattern for alloy A10 Cu22Zr71Y7 (at.%), respectively. The diagrams show that alloy A10 is in the three-phase region for white αZr, grey CuZr2 and dark CuY.  Figure 8a,d shows the BSE micrograph and XRD pattern for alloy A4 Cu33Zr7Y60 (at.%), respectively. The results show that alloy A4 is in the three-phase region for white αZr, grey αY and dark CuY. The rare earth Y is chemically active [40]. The oxidation of Y can be observed during the metallographic polishing process, which does not affect the phase equilibrium after vacuum annealing at 973 K. Combined with the XRD result, the black Y-oxide is Y2O3. The solubility of Cu in the hcp phase is about 2 at.%. Figure 8b,e shows the BSE micrograph and XRD pattern for alloy A9 Cu53Zr12Y35 (at.%), respectively. The diagrams show that alloy A9 is in the three-phase region for white CuZr2, grey CuY and black Cu2Y. Figure 8c,f shows the BSE micrograph and XRD pattern for alloy A10 Cu22Zr71Y7 (at.%), respectively. The diagrams show that alloy A10 is in the three-phase region for white αZr, grey CuZr2 and dark CuY. Based on the present experiments, the isothermal section at 973 K is constructed in Figure 9. Three two-phase regions and four three-phase regions agree well with those of He et al. [18]. The red dashed line in the experimental phase diagram indicates the phase regions that were not directly measured by the experiment. The alloys designed in the three-phase region of Cu4Y + Cu6Y + Cu5Zr tend to shift into the three-phase region of (Cu) + Cu6Y + Cu5Zr due to the small range of the phase region and the easy volatilization of Y during melting. This part has not been experimentally determined.

Thermodynamic Assessment
The PARROT module in Thermo-Calc software [41] was used for optimization based on the least square method. The thermodynamic descriptions for the Cu-Y [20], Cu-Zr [21] and Zr-Y [22] systems in the literature were combined to form a basis for the Cu-Zr-Y system assessment. Before optimizing the ternary system, the thermodynamic parameters of Cu−Y were optimized in order to satisfy the phase relations of (αY) + (αZr) + CuY and (αZr) + CuZr2 + CuY at 870K of the Cu-Zr-Y system. After optimization the temperature of the invariant reaction (αZr) + CuY → (αY) + CuZr2 was adjusted downward to 860 K. The presently calculated phase diagram of the Cu-Y system is consistent with the experimental data [20,[42][43][44] and the calculated date [45], which is shown in Figure 1. For the Cu-Zr-Y system, the experimental data measured by alloy-sampling from the literature [18] are also used in the optimization.
Then the measured ternary solubilities of Cu6Y, Cu4Y, Cu7Y2, Cu5Zr and Cu51Zr14 were considered. Optimization was carried out sequentially from the copper-rich end to the copper-poor end. Afterward, all the parameters were optimized simultaneously to achieve reasonable thermodynamic parameters. Finally, Table 3 provides the reasonable parameters of the Cu-Zr-Y system obtained in this work where the adjusted parameters are bolded. According to the current thermodynamic parameters, the calculated isothermal sections at 870 K, 973 K and 978 K are shown in Figure 10 compared with the reported experimental data [18,19] and the present measurement results. These calculations are in good agreement with experimental data. The calculated vertical section of the Cu-Zr-Y Figure 9. Experimental isothermal section of Cu-Zr-Y system with experimental data at 973 K.

Thermodynamic Assessment
The PARROT module in Thermo-Calc software [41] was used for optimization based on the least square method. The thermodynamic descriptions for the Cu-Y [20], Cu-Zr [21] and Zr-Y [22] systems in the literature were combined to form a basis for the Cu-Zr-Y system assessment. Before optimizing the ternary system, the thermodynamic parameters of Cu−Y were optimized in order to satisfy the phase relations of (αY) + (αZr) + CuY and (αZr) + CuZr 2 + CuY at 870K of the Cu-Zr-Y system. After optimization the temperature of the invariant reaction (αZr) + CuY → (αY) + CuZr 2 was adjusted downward to 860 K. The presently calculated phase diagram of the Cu-Y system is consistent with the experimental data [20,[42][43][44] and the calculated date [45], which is shown in Figure 1. For the Cu-Zr-Y system, the experimental data measured by alloy-sampling from the literature [18] are also used in the optimization.
Then the measured ternary solubilities of Cu 6 Y, Cu 4 Y, Cu 7 Y 2 , Cu 5 Zr and Cu 51 Zr 14 were considered. Optimization was carried out sequentially from the copper-rich end to the copper-poor end. Afterward, all the parameters were optimized simultaneously to achieve reasonable thermodynamic parameters. Finally, Table 3 provides the reasonable parameters of the Cu-Zr-Y system obtained in this work where the adjusted parameters are bolded. According to the current thermodynamic parameters, the calculated isothermal sections at 870 K, 973 K and 978 K are shown in Figure 10 compared with the reported experimental data [18,19] and the present measurement results. These calculations are in good agreement with experimental data. The calculated vertical section of the Cu-Zr-Y system at 10 at.% Zr, compared with the measured temperatures by DSC, are shown in Figure 11. It is noteworthy that only alloy A3 is located at the 10 at.% Zr vertical section. Alloy A2 is labeled in the vertical section by L → (Cu) + Cu 5 Zr + Cu 6 Y invariant reactions. Figure 12 shows the calculated liquidus projection in comparison with the experimental primary phases. The present thermodynamic parameters are used for Scheil solidification simulation to guide the solidification behavior. Figure 13 shows the Scheil solidification simulation results of alloys A1 and A9. It can be seen that the primary phases of alloys A1 and A9 are Cu 5 Zr and CuY, respectively. The ternary eutectic structure of (Cu) + Cu 5 Zr + Cu 6 Y and CuZr 2 + Cu 2 Y + CuZr are formed. These Scheil solidification simulations are consistent with the experimental result discussed in Section 4.1.1.
system at 10 at.% Zr, compared with the measured temperatures by DSC, are shown in Figure 11. It is noteworthy that only alloy A3 is located at the 10 at.% Zr vertical section. Alloy A2 is labeled in the vertical section by L → (Cu) + Cu5Zr + Cu6Y invariant reactions.  Figure 10. Calculated isothermal sections of Cu-Zr-Y system at (a) 870 K with experimental data [19], (b) 973 K with experimental data from this work and (c) 978 K with experimental data [18]. Figure 10. Calculated isothermal sections of Cu-Zr-Y system at (a) 870 K with experimental data [19], (b) 973 K with experimental data from this work and (c) 978 K with experimental data [18].
Materials 2023, 16, x FOR PEER REVIEW 13 of 17 Figure 11. Calculated vertical section of the Cu-Zr-Y system at 10 at.% Zr with the experimental data. Figure 12 shows the calculated liquidus projection in comparison with the experimental primary phases. The present thermodynamic parameters are used for Scheil solidification simulation to guide the solidification behavior. Figure 13 shows the Scheil solid-  Based on the CALPHAD method and key experiments, a reasonable set of thermodynamic parameters for the Cu-Zr-Y system was obtained. The thermodynamic parameters were used to simulate phase equilibrium and Scheil solidification, which can guide the design of composition and microstructure of HSHC copper alloy.

Conclusions and Summary
The solidification microstructure, phase equilibria and phase transition temperatures in the Cu-Zr-Y ternary system were investigated by XRD, EPMA and DSC technologies. Furthermore, the thermodynamic optimization and a sequence of calculations and simulations using the present thermodynamic parameters were carried out. The major conclusions are as follows:  Based on the CALPHAD method and key experiments, a reasonable set of thermodynamic parameters for the Cu-Zr-Y system was obtained. The thermodynamic parameters were used to simulate phase equilibrium and Scheil solidification, which can guide the design of composition and microstructure of HSHC copper alloy.

Conclusions and Summary
The solidification microstructure, phase equilibria and phase transition temperatures in the Cu-Zr-Y ternary system were investigated by XRD, EPMA and DSC technologies. Furthermore, the thermodynamic optimization and a sequence of calculations and simulations using the present thermodynamic parameters were carried out. The major conclusions are as follows: Based on the CALPHAD method and key experiments, a reasonable set of thermodynamic parameters for the Cu-Zr-Y system was obtained. The thermodynamic parameters were used to simulate phase equilibrium and Scheil solidification, which can guide the design of composition and microstructure of HSHC copper alloy.

Conclusions and Summary
The solidification microstructure, phase equilibria and phase transition temperatures in the Cu-Zr-Y ternary system were investigated by XRD, EPMA and DSC technologies. Furthermore, the thermodynamic optimization and a sequence of calculations and simulations using the present thermodynamic parameters were carried out. The major conclusions are as follows:

•
The solid solubility in the ternary system is determined. The maximum solubility of Zr in Cu 6 Y, Cu 4 Y and Cu 7 Y 2 are about 6.61, 6.27 and 7.83 at.% Zr, respectively. The solubility of Y in Cu 5 Zr, Cu 51 Zr 14 and CuZr are about 2.57, 4.45 and 3.38 at.% Y, respectively. The solubility of Cu in the hcp phase is about 2 at.%.

•
The Cu-Y system and the Cu-Zr-Y system were optimized by the CALPHAD method. The calculated isothermal sections, liquidus projection and vertical section are consistent with the experimental data.

•
The observed solidified microstructure agrees with the result of the Scheil solidification simulations using the thermodynamic parameters. The presently obtained thermodynamic description for the Cu-Zr-Y system can be used to guide the composition and microstructure design of Cu-Zr-Y alloys.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding author.