Optimized Control of Hot-Working Parameters in Hot-Forged (CoCrNi)₉₄Al₃Ti₃ Medium-Entropy Alloy

Abstract

It is essential to develop the optimal hot-working process of the (CoCrNi)₉₄Al₃Ti₃ alloy, a recently developed precipitation-hardened medium-entropy alloy with promising mechanical properties, for its industrial application. In this study, the hot workability of the as-forged (CoCrNi)₉₄Al₃Ti₃ alloy was investigated over a temperature range of 1000 °C to 1150 °C and a strain rate ranging from 0.001 to 1 s⁻¹ using a Gleeble-1500D thermal simulation machine of Dynamic Systems Inc., USA. As a result, the constitutive relationship was established, and the hot deformation activation energy was calculated as 433.2 kJ/mol, suggesting its well-defined plastic flow behavior under low-energy-input conditions. Hot-processing maps were constructed to identify the stable hot-working regions. Microstructure analysis revealed that the hot-forged (CoCrNi)₉₄Al₃Ti₃ alloy exhibited continuous dynamic recrystallization (CDRX) behavior under optimal hot-working conditions. Considering the hot-processing maps and DRX characteristics, the optimal hot-working window of hot-forged (CoCrNi)₉₄Al₃Ti₃ alloy was identified as 1100 °C with a strain rate of 0.1 s⁻¹. This work offers valuable guidance for developing high-efficiency forming processes for (CoCrNi)₉₄Al₃Ti₃ medium-entropy alloy.

1. Introduction

The development and application of metallic materials have long been fundamental to industrial advancement. With the accelerated exploration of extreme environments such as deep space, conventional metallic materials are increasingly unable to meet significant performance requirements. In 2004, Cantor [1] and Yeh [2] proposed a novel alloy design strategy based on mixing entropy, later termed high-entropy alloys (HEAs). Due to their unique compositional design, HEAs have shown exceptional mechanical properties [3,4,5], excellent corrosion resistance [6,7], and outstanding electrocatalytic performance [8,9]. For example, HEA coatings recently developed by laser cladding exhibited outstanding wear, corrosion, and oxidation resistance [10]. As the compositional space of HEAs continues to expand, a variety of novel alloys are further developed, such as CoCrNi medium-entropy alloy (MEA), derived from the Cantor alloy [3]. Previous research has shown that, at cryogenic temperatures, CoCrNi MEA achieves an UTS of 1.3 GPa and a fracture toughness of 275 MPa·m1/2, surpassing state-of-the-art conventional alloys like stainless steels and nickel-based superalloys [11]. Nevertheless, their relatively low yield strength remains one of the key factors limiting the engineering application of both HEAs and MEAs.

Enhancing the yield strength of HEAs/MEAs can be effectively achieved through precipitation hardening [12,13,14,15]. For instance, Zhao et al. [11] demonstrated that minor additions of Al and Ti induced the formation of highly coherent nanosized precipitates within the CoCrNi matrix, thereby activating multiple strengthening effects. Consequently, the precipitation-hardened (CoCrNi)₉₄Al₃Ti₃ MEA exhibited an excellent balance between strength and ductility, with a yield strength of 750 MPa, a UTS of 1.3 GPa, and a ductility of 44%. More recently, by employing a combination of cryogenic rolling, high-temperature annealing, and aging treatments, a uniform dual-phase nanosized structure was achieved in (CoCrNi)₉₄Al₃Ti₃ MEA, yielding a UTS of up to 2.2 GPa and a uniform elongation of 13% [16]. Furthermore, the study by Pan et al. [17] on the dynamic deformation of (CoCrNi)₉₄Al₃Ti₃ MEA suggests that Al/Ti co-alloying results in a spall strength exceeding that of previously reported FCC-type MEAs/HEAs. Thus, compared with CoCrNi MEA, the precipitation-hardened (CoCrNi)₉₄Al₃Ti₃ MEA holds more promise for industrial use as a heavy-duty structural material. Nonetheless, the inherent complexity of the multicomponent system in HEAs/MEAs often leads to considerable segregation and cracking during cooling and solidification, thus necessitating appropriate thermomechanical processing for subsequent engineering applications.

Hot working offers significant advantages in reducing defects and transforming as-cast ingots into homogeneous and refined products, which makes it a widely used method in conventional metal forming [18,19]. Recently, Yi et al. [20] investigated hot deformation behavior and dynamic recrystallization characteristics of homogenized (CoCrNi)Al₃Ti₃ alloy, reporting a thermal activation energy of 566.123 kJ/mol and identifying an optimal hot-working condition of 0.001 s⁻¹ at 1000 °C. However, the hot deformation behavior and DRX mechanism of hot-forged (CoCrNi)₉₄Al₃Ti₃ alloy, crucial for subsequent hot-working processes such as hot rolling, have yet to be reported.

In this work, as-cast (CoCrNi)₉₄Al₃Ti₃ alloys were prepared using an induction skull melting furnace and hot-forged after homogenized treatment. Then, their hot deformation behavior was systematically investigated over a temperature range of 1000–1150 °C and strain rates ranging from 0.001 to 1 s⁻¹. The thermal activation energy for high-temperature plastic flow was calculated as 433.19 kJ/mol. Then, we constructed hot-processing maps and examined the dynamic recrystallization behavior for the hot-forged (CoCrNi)₉₄Al₃Ti₃ alloy. Based on the hot-processing maps and dynamic recrystallization characteristics, the optimal hot deformation condition was determined to be 0.1 s⁻¹ at 1100 °C.

2. Materials and Methods

2.1. Initial Material

(CoCrNi)₉₄Al₃Ti₃ MEAs, denoted as Al₃Ti₃, were prepared using high-purity raw materials of Co, Cr, Ni, Al, and Ti in a high-vacuum induction skull melting furnace. Prior to melting, the furnace was purged with argon three times to minimize oxidation of the raw materials. Next, the raw materials were heated into a fully homogeneous melt and then held for approximately 20 min. Then, the chemical composition of the as-cast (CoCrNi)₉₄Al₃Ti₃ MEAs was analyzed by energy dispersive spectroscopy (EDS) of Oxford Instruments, UK. The EDS results of dendritic and interdendritic regions in as-cast (CoCrNi)₉₄Al₃Ti₃ MEAs are summarized in

Table 1. Chemical compositions (at. %) of the as-cast (CoCrNi)₉₄Al₃Ti₃ MEAs analyzed by EDS.

Region	Co	Cr	Ni	Al	Ti
Nominal	31.33	31.33	31.33	3	3
DR	33.77	31.69	29.98	2.48	2.08
IR	20.80	10.45	42.24	13.82	12.68

. The as-cast ingot was homogenized at 1225 °C for 24 h followed by air cooling to eliminate micro-segregation. The homogenized ingot was subsequently hot-forged on a hot-forge machine, achieving a 50% reduction in thickness at 1150 °C, followed by air cooling. The cylindrical samples (with a diameter of 8 mm and a height of 12 mm), used for subsequent hot compression experiments, were cut from the hot-forged samples using an electric discharge wire-cutting machine.

2.2. Hot Compression Experiments

Prior to the hot compression experiments, the cylindrical samples were carefully polished using #2000 sandpaper. High-temperature graphite sheets were placed on both ends of the samples, serving as lubricants, to minimize friction during deformation. Hot compression tests were performed on a Gleeble-1500D thermal simulation machine of Dynamic Systems Inc., USA at deformation temperatures of 1000 °C, 1050 °C, 1100 °C, and 1150 °C, and under strain rates of 0.001 s⁻¹, 0.01 s⁻¹, 0.1 s⁻¹, and 1 s⁻¹, respectively. Each sample was heated to the target temperature and held for 3 min before compression. The specimens were then compressed to achieve a 60% reduction in height, corresponding to a true strain of approximately 0.9, and immediately water-quenched to preserve their deformed microstructure. The central region of each hot-deformed specimen was subsequently extracted for microstructure analysis.

2.3. Phase and Microstructure Analysis

An X-ray diffractometer (XRD) with a Cu-Kα ray source of Rigaku Corporation, Japan, was used to identify the phase structure of the hot-forged Al₃Ti₃ MEA, scanning from 10 to 90° at a scanning speed of 2°/min. For microstructure analysis, the specimens were mechanically polished and then electrochemically polished at 15 V in a solution consisting of 10 vol% perchloric acid and 90 vol% alcohol. The initial microstructure was observed using a SU8010 scanning electron microscope (SEM) of Hitachi Ltd., Japan, equipped with an Oxford energy dispersive spectroscopy (EDS). The electron backscatter diffraction (EBSD) characterization of the initial and hot compression microstructures was conducted with an EBSD detector of Oxford Instrument, UK. The EBSD data, including grain size, grain orientation, and the volume fractions of various grains, were analyzed using HKL Channel 5.12 and Aztech Crystal 2.0 software.

3. Results and Discussion

3.1. Initial Structure

The optical micrograph (OM) and SEM, XRD, and EBSD images of the hot-forged Al₃Ti₃ MEA before hot compression are shown in Figure 1. As shown in Figure 1a, the optical micrograph reveals a typical equiaxed grain structure with a few twins within the grains, and with an average grain size of about 200~300 μm. Higher-magnification SEM images, shown in Figure 1b, reveal the presence of spherical nano-precipitates within the grains. According to previous studies [12], these precipitates were identified as L1₂-(Al, Ti, Ni)-rich phase. In the XRD pattern shown in Figure 1c, only diffraction peaks corresponding to the FCC phase are observed, as the nano-precipitates were undetectable by XRD. The inverse pole figure (IPF) in Figure 1d indicates that the initial microstructure exhibited no significant texture and that DRX sufficiently occurred during hot forging.

3.2. High-Temperature Plastic Flow Behavior

The true stress–strain curves of the as-forged Al₃Ti₃ MEA compressed at different temperatures and strain rates show a similar trend, as shown in Figure 2a–d. In the initial deformation stage, the stress demonstrates a rapid increase with an increase in strain, primarily due to the work hardening (WH) caused by dislocation pile-up. With a further increase in the strain, the stress of this alloy gradually approaches a steady state without significant dynamic softening and a distinct peak, resembling the flow behavior previously reported for certain HEA/MEAs [21,22,23]. This steady state suggests a dynamic equilibrium between work hardening and dynamic softening caused by dynamic recovery and DRX. The peak stress values extracted from each stress–strain curve are plotted as functions of temperature and strain rate in Figure 2e, together with the data of the homogenized Al₃Ti₃ MEA for comparison [20]. It is evident that the peak stress decreases with an increase in temperature and a decrease in the strain rate, indicating that the Al₃Ti₃ MEA shows positive strain rate sensitivity. Compared with the homogenized Al₃Ti₃ alloy, the hot-forged alloy shows higher peak stresses, especially at high strain rates. This is mainly attributed to the finer grain structure of the hot-forged Al₃Ti₃ MEA compared to its homogenized counterpart.

3.3. Constitutive Law Construction

The flow behavior of metals during hot compression can be described mathematically by the constitutive relationship correlating flow stress, strain rate, and deformation temperature. The constitutive relationship is generally expressed using Arrhenius-type hyperbolic sinusoidal equations, originally proposed by Sellars [24] and Tegart [25].

$$Z=A·{[\sinh (\alpha \sigma \left)\right]}^n=\dot{\epsilon }·\exp (Q_{\mathrm{c}}/RT),$$

(1)

where Z is the short Zener–Holloman parameter [26], also known as the temperature-compensated strain rate; A is the material constant; α is the fitting coefficient; σ is the flow stress (MPa); n is the stress exponent; $\dot{\epsilon }$ is the strain rate (s⁻¹); Q_c is the thermal activation energy for hot deformation (kJ/mol); R is the gas molar constant; and T is the hot compression temperature (K). According to Equation (1), $\dot{\epsilon }$ can be expressed by a function of flow stress, activation energy, and hot deformation temperature as follows:

$$\dot{\epsilon }=A[\sinh \left(\alpha \sigma \right){]}^n\mathrm{e}\mathrm{x}\mathrm{p}(-{\displaystyle \frac{Q_{\mathrm{c}}}{RT}}) (\mathrm{for} \mathrm{all} \sigma ).$$

(2)

Following the general calculation method, the unknown material parameters in Equation (2) can be determined by linear fitting, as shown in Figure 3. The average values of the slopes of the fitting lines in Figure 3a,b yields an $\alpha$ of 0.0079. The apparent activation energy (Q_c) for high-temperature deformation can be derived from Equation (2) and expressed as follows:

$$Q_c=1000·R{[{\displaystyle \frac{\partial \ln \dot{\epsilon }}{\partial \ln \sinh \left(\alpha \sigma \right)}}]}_T·{[{\displaystyle \frac{\partial \ln \sinh \left(\alpha \sigma \right)}{\partial (1000/T)}}]}_{\dot{\epsilon }}$$

(3)

where ${\displaystyle \frac{\partial \ln \dot{\mathsf{\epsilon }}}{\partial \ln \sinh \left(\alpha \sigma \right)}}$ and ${\displaystyle \frac{\partial \ln \left[\sinh \left(\mathsf{\alpha }\mathsf{\sigma }\right)\right]}{\partial (1000/T)}}$ can be determined by the linear fitting of $\mathrm{l}\mathrm{n}\dot{\epsilon }$ vs. $\ln \sinh \left(\alpha \sigma \right)$ and $\ln \sinh \left(\alpha \sigma \right)$ vs. 1000/T. As shown in Figure 3c,d, the average slopes of the fitting lines were determined to be 3.5756 and 14.5722, respectively, yielding a $Q_{\mathrm{c}}$ of 433.195 kJ/mol. The values of $n$ and $\ln A$ can be evaluated by the slope and y-axis intercept of the fitting line of $\ln Z$ vs. $\ln \sinh \left(\alpha \sigma \right)$, respectively, as shown in Figure 3e. Ultimately, by substituting the calculated $A$, $\alpha$, $n$ and $Q_c$ into Equation (2), the constitutive equation of Al₃Ti₃ MEA is given as

$$\dot{\epsilon }=7.14\times {10}^{14}[\sinh \left(0.0079\sigma \right){]}^{3.5494}\mathrm{e}\mathrm{x}\mathrm{p}(-{\displaystyle \frac{433195}{RT}}).$$

(4)

3.4. Processing Maps

Processing maps serve as effective tools for optimizing the hot-working conditions of metallic materials. These maps are developed through dynamic material models (DMMs), which correlate microstructural changes (such as DRX) during hot deformation with energy conversion of materials. The construction of a processing map needs two key parameters: the efficiency of power dissipation ($\eta$) and the instability coefficient ($\xi$) [27]. $\eta$ reflects the proportion of energy dissipated for microstructural evolution during hot deformation, while $\xi$ identifies the processing domains that should be avoided during hot working. Accordingly, based on the DMM, we constructed three-dimensional (3D) contour maps of $\eta$ and $\xi$ as functions of deformation temperature and strain rate, as shown in Figure 4a,b, respectively. The processing maps of Al₃Ti₃ MEA were then constructed by superimposing the projection of the 3D map from Figure 4a and the partial map with $\xi$ < 0 from Figure 4b onto the Z = 0 plane, as shown in Figure 4c.

Region A (blue-shaded area) in Figure 4c corresponds to the hot deformation conditions of 1000–1050 °C/1 s⁻¹, where $\xi$ < 0. As defined by the instability criterion of Prasad et al. [27], this region is considered unstable and should therefore be avoided during the hot deformation process. As the temperature increases and the strain rate decreases, $\xi$ rises above 0, while $\eta$ gradually increases. As shown in Figure 4c, Region B (orange and red areas) is the high energy dissipation efficiency zone, with $\eta$ ranging between 0.36 and 0.4 [28], where the softening mechanism is identified as DRX. Consequently, the hot-working conditions of 1050–1150 °C/0.1–0.001 s⁻¹ corresponding to Region B are conducive to sufficient DRX. Besides processing maps, DRX characteristics are also essential for determining the optimal hot-working parameters.

3.5. Recrystallization Behavior

Figure 5 shows the optical micrographs of the hot-forged Al₃Ti₃ MEA at 1050–1150 °C/0.1–0.001 s⁻¹. Under low deformation temperatures (1050 °C and 1100 °C) and high strain rates (0.1 s⁻¹ and 0.01 s⁻¹), a distinct “necklace” microstructure is observed, characterized by fine grains surrounding elongated coarse grains. As the strain rate decreases to 0.001 s⁻¹, only equiaxed grains are evident, indicating complete DRX. As the deformation temperature continues to rise to 1150 °C, complete DRX is achieved at even lower strain rates, and fine grains quickly begin to form.

To further investigate the DRX behavior of the as-forged Al₃Ti₃ MEA, EBSD characterization was performed on the deformed structures processed at a constant strain rate but under different deformation temperatures. Figure 6 shows the EBSD results at a strain rate of 0.01 s⁻¹ under deformation temperatures of 1050 °C, 1100 °C, and 1150 °C. Figure 6a₁–c₁ present the IPF maps at these temperatures, indicating that the grain orientations are random with no obvious texture. Figure 6a₂–c₂ display the corresponding grain average misorientation (GAM) maps. As shown in Figure 6a₂, the microstructure after hot deformation at 1050 °C exhibits a fully equiaxed grain structure. As the deformation temperature rises to 1150 °C, the volume fraction of dynamically recrystallized grains remains almost unchanged, as shown in Figure 6a₃–c₃, indicating that the promoting effect of high temperature on DRX is not significant. In addition, the grain size distributions at different deformation temperatures were analyzed. As shown in Figure 6a₄–c₄, with an increase in temperature, the average grain size increases from 4.83 μm to 8.61 μm, suggesting that higher temperatures promote grain growth.

Figure 7 presents the EBSD results of the hot-deformed microstructures at a constant deformation temperature but under different strain rates of 0.001 s⁻¹, 0.01 s⁻¹, and 0.1 s⁻¹. As shown in Figure 7a₁–c₁, the grain orientations remain random with no distinct texture. As the strain rate increases, the volume fraction of recrystallized grains rises rapidly, while that of the substructure decreases, demonstrating that rapid deformation promotes DRX. Moreover, an increase in strain rate also leads to a significant change in the grain size distribution, with the average grain size quickly decreasing from 15.54 μm to 4.86 μm. Therefore, combined with the hot-processing maps, the optimal hot-working condition for the hot-forged Al₃Ti₃ MEA is 1100 °C with a strain rate of 0.1 s⁻¹ to achieve uniform and refined recrystallization grains. Notably, compared to deformation temperature, the strain rate has a more pronounced influence on DRX, likely because the deformation of Al₃Ti₃ MEA is mediated by stacking faults [12].

To better control the hot-forming process of the hot-forged Al₃Ti₃ MEA, its DRX mechanisms under optimal hot-working conditions were determined. According to Huang’s review [29], metallic materials mainly exhibit two DRX mechanisms: discontinuous (DDRX) and continuous dynamic recrystallization (CDRX). CDRX is typically characterized by the gradual misorientation accumulation within subgrains due to dislocation absorption, eventually leading to the formation of new grains with high-angle grain boundaries (LAGBs). This mechanism can be identified through the measurement of the local misorientation and cumulative misorientation within the grains [30]. As shown in Figure 8a, the point-to-point (local misorientation) and point-to-origin (cumulative misorientation) misorientation profiles along the paths a and b in the IPF of Al₃Ti₃ MEA deformed at 1100 °C under 0.1 s⁻¹ reveal that the local misorientation exceeds 3°, and the cumulative misorientation exceeds 10°. These results indicate that CDRX is the dominant recrystallization mechanism, consistent with findings in the CoCrNi MEA [22]. Moreover, Figure 8b shows the subgrains retained within grains, further supporting CDRX as the prevailing softening mechanism. Yi et al. reported that the addition of Al and Ti atoms promoted dislocation pile-up, enhancing DDRX under high temperatures and low strain rates [20]. However, Lu et al. found that in (CoCrNi)₉₄Al₃Ta₃ MEAs, high temperatures led to a complete dissolution of precipitates, facilitating easier dislocation slip [31]. Therefore, CDRX may be promoted by the dissolution of L1₂-(Al, Ti, Ni)-rich nano-precipitates in Al₃Ti₃ MEA at 1100 °C/0.1 s⁻¹.

4. Conclusions

In summary, the hot deformation behavior of an as-forged (CoCrNi)₉₄Al₃Ti₃ medium-entropy alloy was systematically studied at temperatures in the range of 1000 °C to 1150 °C and at strain rates of 0.001–1 s⁻¹. Based on stress–strain curves, the constitutive relationship was established to determine the activation energy. Processing maps were constructed to determine the stable hot-processing window. Moreover, the characteristics of the hot-deformed structures were carefully investigated using EBSD to reveal the DRX mechanism. The major conclusions can be summarized as follows:

(1)

The hot-forged (CoCrNi)₉₄Al₃Ti₃ MEA showed a fully equiaxed grain structure with an average grain size of about 200~300 μm. Additionally, the initial structure showed no obvious texture, indicating sufficient DRX during hot-forging.

(2)

Based on the stress–strain data, the peak stress of the as-forged (CoCrNi)₉₄Al₃Ti₃ MEA at high strain rates was found to be higher than that of its homogenized counterpart. The constitutive equation was established as $\dot{\mathsf{\epsilon }}=7.14\times {10}^{14}[\sinh \left(0.0079\mathsf{\sigma }\right){]}^{3.5494}\mathrm{e}\mathrm{x}\mathrm{p}(-{\displaystyle \frac{433195}{RT}})$, yielding a thermal activation energy of 433.19 kJ/mol.

(3)

Considering the hot-processing map and DRX characteristics, the optimal processing window for the hot-forged (CoCrNi)₉₄Al₃Ti₃ MEA was identified at 1100 °C with a strain rate of 0.1 s⁻¹. The dominant DRX mechanism at the optimal hot-working condition was confirmed to be CDRX. These findings offer valuable guidelines for developing efficient forming processes for (CoCrNi)₉₄Al₃Ti₃ MEA, which is essential for its engineering applications.