Role of Stabilization Heat Treatment Inducing γ′-γ″ Co-Precipitates and η Phase on Tensile Behaviors of Inconel 706

Abstract

Inconel 706 (IN706) alloy is commonly used in aircraft engines and power plant components that must meet very high performance requirements. The stabilization treatment has a significant effect on the precipitation and evolution of the reinforcing phases of the alloy, favoring the creep properties and adversely affecting the room-temperature tensile properties. However, the mechanism of the effect of the stabilization treatment on the mechanical properties of the alloys remains unclear. In this study, the effect of stabilization treatment time on the microstructure and tensile properties of IN706 alloy was investigated. The results showed that as the stabilization time gradually increased, the tensile strength remained basically unchanged (about 1250 MPa), the yield strength decreased from 1031 MPa to 985 MPa, and the plasticity decreased from 28.2% to 20.2%. The stabilization treatment induces the precipitation of granular, rod-shaped, and needle-like η phases at grain boundaries, accompanied by the appearance of a precipitate free zone (PFZ). Since the η phase is enriched with Ti and Nb, its precipitation along the grain boundary results in the depletion of Ti and Nb in the surrounding regions, thereby constraining the precipitation of the γ′ and γ″ phases, resulting in the appearance of PFZ. With increasing stabilization time, the size increase and volume fraction decrease in γ′-γ″ co-precipitates due to the precipitation of η-phase precipitates, leading to a decrease in their yield strength. Combined with in situ tensile tests, it was found that the decrease in the elongation of the stabilization treatment samples was due to the presence of η phase at the grain boundaries, which induced stress concentration and cracking at the grain boundaries. The results show that the mechanical properties of the material were gradually enhanced as the stabilization time decreased. This means it can help to choose the suitable process for IN706 alloy in different service conditions.

1. Introduction

IN706 is a nickel–iron alloy with origins in the Inconel 718 (IN718) alloy. Compared to IN718, IN706 is free of Mo and has increased Ti and Fe content, thereby reducing the content of the expensive elements Nb, Cr, and Ni. IN706 alloy is widely used in aerospace and energy applications, such as turbine disks, shafts, motor mounts and diffuser cases [1,2,3]. The IN706 alloy undergoes precipitation strengthening by changing the chemical composition of Mo, Nb, and Ti. This process involves the formation of two coherent Ni₃X-type compounds: Ni₃Nb and Ni₃ (Al, Ti), with reduced segregation within the alloy. Due to the importance of IN706 alloy, many researchers have studied its microstructural evolution and mechanical properties over the years. Typically, the microstructures of IN706 alloy consists of a face-centered cubic (FCC) γ matrix, a cubic Ni₃ (Al, Ti)-based γ′ phase, a tetragonal Ni₃Nb-based γ″ phase, and Ni₃Ti-based η-phase precipitates at the grain boundaries [4,5,6]. These precipitates are produced after the heat treatment of IN706 alloy and have a significant impact on the properties of the IN706 alloy [7,8,9]. The forging process is a crucial stage in the production of large metal parts as it involves the refinement of the initial as-cast grain structure. This is especially vital for large forged IN706 parts, where achieving a refined grain structure necessitates meticulous planning of multiple upsetting and shaping operations, along with carefully timed heating and reheating treatments.

The main strengthening phases of IN706 alloy are the γ′-γ″ co-precipitates and the η phase. The γ′-γ″ co-precipitates are coherent with the γ matrix and require additional deformation energy to pass through them [10,11,12]. The strength of the IN706 alloy increases as the size of the precipitates increases to 25 nm due to the interruption of the dislocation motion by the coherent γ′-γ″ co-precipitates through a shearing mechanism. Second, if the size of the precipitates exceeds 25 nm, it may cause a loss of precipitation coherence in the matrix, leading to a diminution of the precipitation hardening effect. At the same time, stabilization treatment leads to the distribution of the η phase along the grain boundaries, leading to a reduction in the impact toughness due to a brittle intergranular fracture. In addition, a moderate amount of η phase leads to excellent creep properties of the alloy. Given that precipitation in IN706 alloy is intricately linked to the heat treatment conditions, considerable attention has been dedicated to extensively studying the impact of heat treatment on the microstructures of IN706 alloy [13,14,15,16]. Specifically, the heat treatment process includes three steps: solid solution, stabilization, and aging to promote the precipitation and formation of the second phase in IN706 alloy [17,18,19]. The effect of the stabilization treatment on alloys is particularly important, and many research projects have investigated the effect of stabilization treatment on mechanical properties. Shibata et al. [20] focused on the effect of the stabilization phase (700 °C to 900 °C) on the creep performance and showed that IN706 had the best creep performance at stabilization temperatures of 810 °C and 840 °C. Kim et al. [21] observed that during the stabilization stage from 800 °C to 840 °C, the deformation of samples with a stabilization temperature of 840 °C was mainly influenced by antiphase boundary shear as well as the formation of isolated stacking faults. The dense γ′-γ″ co-precipitates contribute to the antiphase boundary shear mechanism, resulting in the longest creep rupture life of the 840 °C sample. Zhang et al. [22] found that direct aging has superior mechanical properties compared to standard heat treatment, attributed to smaller grain size, finer η precipitates and γ″ grains.

These studies have shown that stabilization treatments have a significant impact on the precipitation and evolution of the reinforcing phases of alloys, which is directly related to the mechanical properties, but the mechanism of the effect of the stabilization treatment on the mechanical properties of the alloys remains unclear. In this study, the effect of stabilization treatment time on the microstructure and tensile properties of IN706 alloy was investigated in detail. By developing different stabilization treatments, a process is selected that is both clean and mechanically sound. Different stabilization-time treatments were carried out on IN706 alloy to study the variation rule of heat treatment on the content and morphology of precipitated phases and to investigate the influence of the precipitation phase on tensile properties.

2. Experimental Procedures

2.1. Materials and Heat Treatments

The material was forged, and its chemical composition is as follows: Ni 44.2, Cr 17.4, Ti 2.2, Nb 3.3, Al 0.3, C 0.01 and Fe balance, in wt.%. The heat treatment process was carried out in a ceramic fiber muffle furnace, model TM-0914D. The heat treatment process was as follows: solid solution at 980 °C (1 h), then air-cooled; stabilization at 845 °C (0 h, 1 h, 3 h and 5 h), then air-cooled; and aging at 720 °C (8 h), then 55 °C/h furnace-cooled to 620 °C (8 h), then air-cooled, with a rate of increase in temperature of 5 °C/min. Depending on the different stabilization times (0 h, 1 h, 3 h, and 5 h), the samples are referred to as “S0”, “S1”, “S3”, and “S5”.

2.2. Mechanical Testing

In order to study the effect of different microstructure states on mechanical properties, tensile experiments were carried out on samples with different morphological organizations after heat treatment. The tensile tests were conducted using the ASTM E8M standard, and the strain rate of the dog bone tensile specimens was 10⁻³ s⁻¹. Uniaxial tensile tests were conducted using an MTS C43.504 tensile testing machine and a matching extensometer. Three tensile tests were carried out, and the results were averaged to determine the tensile properties. In situ tensile tests were performed on a Mini-MTS tensile bench mounted in a TESCAN Clara chamber. The in situ tensile tests were interrupted at yield with strain levels of 0.2%, 8% and 16% after yielding to obtain scanning electron microscopy data.

2.3. Microstructure Characterization

The microstructures of the samples were analyzed using a TESCAN S8000 scanning electron microscope (SEM). The samples were prepared by mechanical grinding and polishing and then etched with 2 g FeCl₃ + 40 mL HCl + 8 mL C₂H₅OH solution for scanning electron microscopy observation. In order to evaluate the degree of intracrystalline and grain-boundary local deformation, the kernel average orientation difference (KAM) was obtained using the backscattered electron diffraction (EBSD) technique. EBSD-analyzed samples were electrochemically polished after mechanical grinding using a solution containing 10% perchloric acid and 90% ethanol. A scanning step size of 3 μm and an accelerating voltage of 20 kV were selected throughout the EBSD data acquisition process. Transmission electron microscopy observations were carried out on a FEI Tecnai F20 field emission microscope operating at 200 keV. Transmission electron microscopy samples were prepared by grinding the films to 50 μm and then twin-jet polishing at 24 V and −20 °C using a solution of 10% perchloric acid and 90% ethanol. The size and volume fractions of the precipitated phases were determined through the statistical analysis of scanning electron microscope images using a grayscale threshold facilitated by ImageJ 1.54f software. More than 100 individual precipitates were measured to estimate the size and volume fraction of the precipitates in three micrographs.

3. Results

3.1. Microstructure Characterization

In Figure 1, the microstructures of different stabilization times at 845 °C are displayed. The average grain size of the four samples is about 50 µm. As indicated by the red framework in Figure 1a, almost no observations were made on the grain boundaries of the S0 sample η-phase precipitation. As indicated by the blue framework in Figure 1b, the S1 sample had the presence of η phases on the grain boundaries, and the morphology of these η phases was granular.

As indicated by the yellow framework in Figure 1c, in the S3 sample, there are rod-like and needle-like η phases precipitating at grain boundaries, accompanied by the appearance of PFZ. As indicated by the green framework in Figure 1d, in the S5 sample, the amount of rod-like and needle-like η phases precipitated on the grain boundaries further increased, and there was a tendency for the η phases to precipitate into the grain. In addition, the grain boundaries show a coarsening and zigzagging trend, and the PFZ was enlarged.

As shown by arrows in Figure 1g,h, the needle-like η phase possesses a PFZ in the vicinity. Since the η phase is enriched with Ti and Nb, its precipitation along the grain boundary results in the depletion of Ti and Nb in the surrounding regions, thereby constraining the precipitation of the γ′ and γ″ phases, resulting in the appearance of PFZ. Figure 1e shows the morphology of the needle-like η phase precipitated in the crystal under a bright field, and a certain size of PFZ also exists near the η phase (as shown by the red arrow in Figure 1e). In Figure 1f, the η phase is distributed on the grain boundary at an angle to the grain boundary and grows into the grain. The white area on the left side of the figure is the grain boundary, and a lot of tiny black γ′-γ″ co-precipitates can be seen distributed around it; it belongs to the intracrystalline region, and a bright white zone of PFZ still exists between the η phases (as shown by the red arrows in Figure 1f).

The γ′-γ″ co-precipitates’ morphology of IN706 after different stabilization treatment times is shown in Figure 2. In Figure 2a–d, it is found that the size of γ′-γ″ co-precipitates increases with the increase in stabilization time, and the content of γ′-γ″ co-precipitates decreases with the increase in stabilization time. TEM analysis of IN706 alloy intracrystalline precipitation phases shows (Figure 2e,f) that the large-size spherical phase in the specimen in the stabilization treatment specimen was the γ′ phase (Figure 2f), while the γ′-γ″ co-precipitates were in the form of a sandwich, with γ′ as the nucleus and γ″ as the shell envelope.

Figure 3a portrays the temporal evolution of η-phase content and size during the stabilization process. Evidently, both the η-phase content and size exhibit a gradual increase with prolonged stabilization time (S1: 0.99%, 1.23 μm; S3: 1.92%, 2.66 μm; S5: 2.32%, 2.74 μm). The η-phase content and size suddenly increased much higher at 3 h and 5 h of stabilization, which is due to the precipitation of a large amount of needle-like η phases precipitating at grain boundaries. Figure 3b represents the size of the γ′-γ″ co-precipitates’ increases with the increase in stabilization time (S0: 15.96 nm; S1: 18.67 nm; S3: 20.05 nm; S5: 20.36 nm). This is due to the fact that the η phase is enriched with Ni, Ti, and Nb, and while the γ′-γ″ co-precipitates also require these elements, as the η phase precipitates, this leads to a decrease in their content and a larger size.

3.2. Tensile Properties

Figure 4 shows the typical engineering stress–strain curves after tensile testing at room temperature. The yield strength and elongation decrease with increasing stabilization time (S0: 1031 MPa, 28.2%; S1: 982 MPa, 24.1%; S3: 997 MPa, 22.9%; S5: 985 MPa, 20.2%), and the tensile strength of the sample was maintained at 1250 MPa while the stabilization time increased.

Overall, it can be found that the yield strength decreases by 4.8% and elongation decreases by 28.4% as the stabilization time increases, and the tensile strength remains constant at around 1250 MPa. As shown in

Table 1. Comparison of the results of the current research with other similar works (S0-1 and S3-1) [1].

Samples	Heat Treatment	Yield Strength (MPa)	Tensile Strength (MPa)	Elongation (%)	References
S0	Stabilization 0 h	1031	1258	28	Current study
S1	Stabilization 1 h	982	1242	24	Current study
S3	Stabilization 3 h	997	1256	23	Current study
S5	Stabilization 5 h	980	1240	20	Current study
S0-1	Stabilization 0 h	1085	1262	22	[1]
S3-1	Stabilization 3 h	979	1245	17	[1]

, the material properties in this study showed a small difference in strength and substantially increased elongation compared to those in other reports.

Figure 5 shows the fracture morphology of IN706 alloy after undergoing room-temperature stretching. As can be observed from the enlarged red area in Figure 5a, the fracture of the S0 sample has a large number of dimples, which indicates that the alloy has undergone mainly ductile fracture. As can be observed from the enlarged blue area in Figure 5b, the tensile fracture morphology of the S1 sample is different from that of the S0 sample. Under this heat treatment condition, the alloy fracture is not entirely composed of dimples. Instead, it exhibits mixed fracture characteristics consisting mainly of transcrystalline fracture and intercrystalline fracture, but mainly transcrystalline fracture. As can be observed from the enlarged yellow area in Figure 5c, the S3 sample has a large number of rock candy patterns, which is typical of intercrystalline fracture. The fracture characteristics of the S3 sample mainly show a mixture of fracture characteristics (transcrystalline fracture and intercrystalline fracture), but intercrystalline fracture dominates. Some cracks can also be observed on the grain boundaries, as shown by the arrows in Figure 5c, which is due to the presence of the needle-like η phase at the grain boundaries of the S3 sample, and the brittle η phase will lead to cracks during the tensile process. As can be observed from the enlarged green area in Figure 5d, the fracture behavior of the S5 sample is similar to that of the S3 sample, but the fracture along the grain is more pronounced, which explains the relatively low plasticity of the S3 and S5 samples. The fracture characteristics of the alloys change with the stabilization time, and it can be observed that when the stabilization time is shorter, the samples mainly show transcrystalline fracture characteristics and less intercrystalline fracture characteristics; with the increase in stabilization time, the transcrystalline fracture characteristics decrease, while the intercrystalline fracture characteristics increase. This is due to the increase in η-phase precipitation on the grain boundaries with the increase in stabilization time, which leads to the generation of cracks and intercrystalline fracture.

4. Discussion

4.1. Effect of Stabilization Treatment on the Microstructures

It is widely acknowledged that both γ′ and γ″ represent metastable phases, poised for transformation into equilibrium phases, γ′ transitioning into η-Ni₃Ti [23] and γ″ transitioning into δ-Ni₃Nb [24] following prolonged exposure to elevated temperatures. As indicated by the arrows in Figure 1, PFZ was found in the S3 and S5 samples only near the region where the η phase is enriched. This is because Ni, Ti and Nb are consumed for the precipitation of the η phase [1,5], and γ′-γ″ co-precipitates’ precipitation is inhibited in the vicinity of the η phase due to the high amount of Ni, Ti and Nb required to produce γ′-γ″ co-precipitates. Therefore, the content of this region is relatively low compared to Ni, Ti and Nb in the γ matrix, leading to the appearance of PFZ. The S5 sample has a larger PFZ width than the S3 sample, which is due to the fact that the S5 sample has more η phases at the grain boundaries compared to S3, which leads to an excessive consumption of Ni, Ti, and Nb, widening the PFZ [25].

According to the experimental results, among the four stabilization treatment samples, the average size of the precipitates in the S0 sample was significantly smaller than the other samples due to different precipitation kinetics. During the stabilization treatment, the precipitation of the η phase leads to the depletion of Ti, Nb, and Al, resulting in a decrease in the supersaturation of the matrix. In addition, the increase in the η volume fraction implies a decrease in the presence of Ti, Nb, and Al in the γ matrix, which hinders the formation of γ′-γ″ co-precipitates during the stabilization heat treatment. Consequently, the nucleation rate of γ′-γ″ co-precipitates experience a decline. This is substantiated by the observation that the γ′-γ″ co-precipitates refrain from forming in the proximity of the needle-like η phases (as shown in Figure 1g,h). Moreover, owing to the elevated degree of supersaturation preceding the aging treatment, the nucleation driving force for the γ′-γ″ co-precipitates in the S0 sample surpasses that of the other samples. Consequently, a substantial quantity of γ′-γ″ co-precipitates undergo rapid nucleation. Additionally, the heightened number density of these co-precipitates restricts the spacing between individual precipitates. As a result, the γ′-γ″ co-precipitates in the S0 sample are notably smaller.

4.2. Effect of Stabilization Treatment on Tensile Properties

There are four main strengthening mechanisms for IN706 alloy: grain boundary strengthening, solid solution strengthening, precipitation strengthening and work hardening, all of which affect the yield strength [25,26,27]. Therefore, in order to understand the relationship between the process-microstructure mechanical properties of IN706 alloy under different heat treatment schemes, the strengthening mechanisms are discussed in detail in the following sections.

The enhancement of mechanical strength in metals and alloys is achieved through grain refinement; this scientific phenomenon can be described by the Hall–Petch relationship [28,29]. As shown in Figure 1, the grain size of the treatment samples was about 50 µm with no significant changes. Ti and Nb are the main strengthening elements of IN706 alloy. Due to the low volume fraction of the η phase in the stabilized heat treatment conditions, the difference between Ti and Nb contents in the γ matrix is not significant. According to Figure 6, The kernel average misorientations (KAMs) of the four samples are 0.380°, 0.382°, 0.367° and 0.363°. We find that their dislocations are similar despite different heat treatments. Therefore, the level of work hardening is similar under different heat treatments. Therefore, the three types of strengthening—grain boundary strengthening, solid solution strengthening and work hardening—are similar in stabilization treatments, and there is no significant difference in the contribution to the strength of the alloy.

The precipitation strengthening of IN706 alloy mainly originates from the formation of γ′-γ″ co-precipitates, so the variation of the size and volume fraction of γ′-γ″ co-precipitates has a great influence on the yield strength of IN706 alloy [30]. According to the results of the study, when the size of γ′-γ″ co-precipitates is within 25 nm, the yield strength increases with the increase in the size. However, in this study, the γ′-γ″ co-precipitation size increased with increasing stabilization time and remained within 25 nm. The actual yield strengths of S1, S3, and 5 h samples were lower than that of the S0 sample. Given that the volume fraction of the precipitated phase plays a role in coherent hardening, the stabilization treatment leads to the precipitation of the η phase, which results in a decrease in the volume fraction of the γ′-γ″ co-precipitates, and this may be one of the reasons for the decrease in yield strength.

Selecting S0 and S3 samples, the presence or absence of the η phase relative to the plastic deformation stage in IN706 alloy was investigated in detail by in situ tensile experiments. As shown in Figure 7a,d, the sample surface is flat when ε = 0.2%. As shown in Figure 7b,e, the slip systems of some grains are activated at ε = 8%, and interlaced slip traces are observed on the grain surfaces, which indicate that the multi-slip system is activated within most of the grains.

As shown in Figure 7c,f, when ε = 16%, the deformation of the grains is more severe, and most grains are unevenly deformed, as can be seen from the clear steps and steep morphology near the boundaries. In addition, it can be seen from the red framework in Figure 7f that cracks appeared at the grain boundaries of the S3 sample but not in the S0 sample, which is caused by the precipitation of the η phase, the presence of which tends to induce stress concentration and cracks.

Figure 8 illustrates the longitudinal microstructures near the fracture tip observed in the tensile direction. Figure 8a shows that the holes appeared near the carbides, and Figure 8b–d demonstrates that cracks sprouted at the grain boundaries and then expanded along the intergranularity, ultimately leading to perforated fracture. The presence of these cracks and holes indicates that localized stress concentrations and deformations occurred inside the sample during the tensile experiments, leading to the formation and expansion of cracks. Also, the cracks become larger as the stabilization time increases. This finding suggests that grain boundaries are the sites of crack nucleation and extension, which accelerates the fracture of the alloy samples and leads to the lower plasticity of samples possessing the η phase.

5. Conclusions

In this experiment, the effect of stabilization treatment time on the microstructures and tensile properties of IN706 was examined. According to the experimental results, the following conclusions can be drawn:

(1)

The yield strength and elongation of alloy IN706 decrease with increasing stabilization time.

(2)

As the stabilization time increases (from 0 h to 5 h), the η phase undergoes a transition from granular to rod-shaped and needle-like morphology, accompanied by an increase in both size and quantity. Simultaneously, the size of γ′-γ″ co-precipitates increases while their overall number decreases, and the presence of PFZ was found in S3 and S5 samples.

(3)

The yield strength of the samples decreases gradually with increasing stabilization time. This phenomenon is due to the influence of the size and volume fraction of γ′-γ″ co-precipitates on the alloy-strengthening effect. With the increase in stabilization time, the size of γ′-γ″ co-precipitates becomes larger and the volume fraction becomes smaller, leading to the decrease in their yield strength.

(4)

During room-temperature and in situ tensile tests, it was found that the decrease in elongation during stabilization treatment could be attributed to the presence of η phase at the grain boundaries, as the presence of the η phase on the grain boundaries tends to induce stress concentration, which leads to cracks at the grain boundaries, resulting in a decrease in the plasticity of the samples.

(5)

These results can provide theoretical guidance for the development and application of high-performance IN706 alloys.