*3.4. Creep Testing*

Since the tensile mechanical properties of the ST and HIP specimens were found to be comparable across the entire temperature range of this study, only the ST-treated specimens were subjected to creep testing in accordance with ASTM E139−11 (2018), along with their SR and wrought annealed counterparts. Moreover, since the mechanical properties of the ST specimens over this temperature range were found to be build orientation-independent, only vertical ST specimens were subjected to this testing.

Figure 7 shows the typical creep diagrams for the SR, ST, and wrought annealed alloy specimens under stresses ranging from 0.5 to 0.9 YS (760 ◦C, 1400 ◦F). For the LPBF specimens subjected to low (0.5 YS) and intermediate (0.7 YS) stresses, three stages of creep behavior were clearly distinguished: primary, secondary (steady), and tertiary (fracture), with durations dependent on the levels of stress applied. In the case of the SR specimens (Figure 7a), a secondary stage of creep was less pronounced than in the case of the ST specimens (Figure 7b), and for the SR specimens at higher stresses (0.7 YS), the transition from the primary to the steady stage was hardly distinguishable. For the wrought annealed specimens, the primary stage of creep was almost skipped (Figure 7c); the steady stage started much more quickly: the start time was ≤ 0.003 h as compared to the SR and ST specimens, for which this time was 0.15–0.2 h.

**Figure 7.** Creep diagrams of the (**a**) SR, (**b**) ST, and (**c**) wrought annealed IN265 alloys at 760 ◦C (1400 ◦F).

Figure 8 collects the lifetime and rupture strain data for the normalized levels of creep stress (σ/YS). Note large data scattering in the case of LPBF specimens, both SR and ST, as compared to their wrought counterparts (Figure 8a). In the 0.3–0.9 σ/YS range, the ST-treated LPBF alloy manifested a longer life to rupture (note also a run-out at 48 h at σ/YS < 0.5), while the SR and wrought specimens exhibited similar rupture times in the entire 0.3–0.9 σ/YS range. Interestingly, the mean rupture strain in the case of the wrought specimens was significantly higher than that in the case of the SR- and ST-treated LPBF specimens (Figure 8b). The ST specimens showed the lowest mean rupture strain. Note also that the mean rupture strain values correlated well with (and did not exceed) the corresponding tensile ductility values at 760 ◦C (Figure 5).

**Figure 8.** Applied stress in relative values (σ/YS) as a function of the (**a**) creep rupture time and (**b**) rupture strain for the SR, ST, and wrought annealed IN625 alloy (760 ◦C).

After low-stress creep fracture (0.5 YS) at 760 ◦C, the fracture surfaces of the SR specimens had a mix of transgranular and intergranular patterns, while the ST specimens contained mostly transgranular patterns corresponding to a dimpled fracture after the same stress creep (Figure 9a,c). High stresses, however, led to mostly intergranular fractures for both the SR and ST specimens (Figure 9b,d).

The fracture surfaces of the wrought annealed alloy, containing numerous voids, dimples, and tearing ridges, indicated extremely ductile behavior of this alloy as compared to that of the ST-treated LPBF alloy, under creep conditions at 760 ◦C (Figure 9e,f).

Finally, it is worth mentioning that the SR specimens exhibited a build orientation dependency of creep properties. The vertical specimens showed a more pronounced steady stage (Figure 7a) and a higher lifetime and rupture strain (Figure 8) as compared to their horizontal counterparts. Note also that the horizontal SR specimens demonstrated significantly lower tensile and creep properties than their vertical counterparts (Figures 5 and 7).

**Figure 9.** Scanning electron microscope (SEM) images of the fracture zones of creep specimens: (**a**,**b**) SR, (**c**,**d**) ST, and (**e**,**f**) wrought annealed IN625 alloys.

#### **4. Discussion**

Grain size and grain orientation. It was shown that LPBF processing was responsible for the formation of an anisotropic microstructure. The SR specimens contained grains elongated in the build direction with the dominant [001] texture, inherited from the as-built material. The grain length was more than twice the layer thickness, while the average width was about 20 μm. During the ST and HIP treatments, grain growth caused the microstructure to become nearly equiaxed with a random texture. The average grain sizes after the ST and HIP treatments were about 20 and 45 μm, respectively [16].

Map of the mechanical properties. The mechanical properties obtained at a 10−3·s−<sup>1</sup> strain rate across a wide range of temperatures are shown in Figure 10a–c for the SR-, ST-, and HIP-treated LPBF IN625 alloys, respectively. The mechanical properties of the vertical and horizontal SR specimens have been presented here, while the properties of the ST and HIP specimens are limited to those of the vertical specimens. For reference, the mechanical properties of the wrought annealed alloy are also illustrated in Figure 10d.

**Figure 10.** Mechanical properties of the (**a**) SR, (**b**) ST, (**c**) HIP, and (**d**) wrought annealed IN625 alloys as a function of temperature.

Regarding the mechanical properties overall, all specimens exhibited a decrease in the YS and UTS values as the test temperature increased, due to enhanced plastic flow causing tensile deformation at relatively lower stresses (Figure 5). However, their elongations to failure evolved differently, as discussed further.

Tensile strength characteristics (YS, UTS): The results show that at room temperature, the SR specimens exhibited the highest mechanical strengths (YS and UTS) (Figure 10a) and their mechanical strength characteristics exceeded those reported in the ASTM F3056-14 standard (min. 275 MPa (YS) and min. 485 MPa (UTS)) and EOS datasheet (min. 414 MPa (YS) and min. 827 MPa (UTS)) for LPBF IN625 alloy. Regarding other studies, the mechanical properties at room temperature of the SR specimens obtained in this study were comparable to those reported in References [13,19,21,38] for the as-built LPBF IN625 alloy. However, the high mechanical strength of the SR specimens came with a significant orientation dependency in the mechanical characteristics due to a strong microstructure anisotropy, which was inherited from the as-built material [16]. Thus, a finer microstructure in the build plane resulted in higher YS and UTS values of the horizontal specimens as compared to their vertical counterparts. By contrast, the ST and HIP specimens containing equiaxed grains and a random texture manifested build-orientation-independent behavior, but lower mechanical strength characteristics. The equiaxed grain structure and reduced mechanical strength characteristics after post-treatments performed at temperatures higher than 1100 ◦C have also been reported in References [13,21].

At elevated temperatures of up to 593 ◦C, the YS and UTS values decreased continuously with increasing temperature for all the tested specimens. The orientation dependency and the high mechanical strength characteristics of the SR specimens were preserved at these temperatures. The YS and UTS values of the ST and HIP specimens were still slightly lower than those of the wrought annealed alloy.

With a further temperature increase to up to 760 ◦C, a rapid decrease in mechanical strength characteristics was observed for the SR specimens, which was caused by dynamic recrystallization [24]. As a result, no difference was observed between the YS and UTS values. The ST- and HIP-treated specimens were capable of maintaining their structural strength to up to 760 ◦C, and their UTS values became comparable to those of the SR specimens. For the ST- and HIP-treated specimens, dynamic recrystallization took place at 871 ◦C. At this temperature, the YS and UTS values of all the tested specimens were comparable.

Elongation to failure: From RT to intermediate temperatures (538 ◦C), the elongations to failure increased for all the specimens. However, the peak in elongations was observed at 593 ◦C for the SR specimens, and at 538 ◦C for the ST and HIP specimens. Note that for the wrought annealed alloy, the elongation always increased with an increase in testing temperature (Figure 10d). At higher temperatures, the LPBF alloy manifested a significant reduction in elongation. Such a significant decrease in elongation of the LPBF alloy at 760 ◦C was accompanied by changes in fracture pattern due to weakening of the grain boundaries at elevated temperatures (ductile/brittle transition) [24]. Moreover, the intergranular cracking mode led to a significant increase in the orientation dependency for the SR specimens in the 593–871 ◦C temperature range. The elongated grains oriented along the axis of testing were responsible for higher elongations of the vertical SR specimens as compared to their horizontal counterparts. After the ST and HIP treatments, the grains became equiaxed and their mean size increased, resulting in elongations to failure higher than for the horizontal SR, but lower than for the vertical SR specimens.

The elevated temperature embrittlement observed in this study is a known phenomenon for many Ni-based alloys. The reasons for this behavior are subject to discussion [22]. In particular, the embrittlement has been attributed to intergranular precipitates, grain boundary shearing or sliding, gas phase embrittlement, dynamic strain aging, grain boundary segregation, and glide plane decohesion [22]. For IN625 alloy, this phenomenon is mostly associated with precipitates (carbides, M23C6 and M6C, and δ phase) distributed along grain boundaries [2,24,39,40]. More specifically, according to Reference [24], cracks preferentially propagate along grain boundaries containing M6C carbides.

Finally, it should be mentioned that the mechanical behaviors of the ST- and HIP-treated specimens were similar (Figure 10b,c). While the YS and UTS values of the ST specimens were slightly higher, their elongations were slightly lower than those of the HIP specimens.

High-temperature creep properties. It was found that as well as the tensile properties, the SR-annealed alloy exhibited orientation dependency of its creep behavior: vertically built specimens showed higher lifetimes and lower rupture strains compared to their horizontally built counterparts (Figures 7 and 8). Despite the high lifetimes of the vertical SR specimens as compared to the vertical ST and wrought specimens (Figure 11), the anisotropy in mechanical properties of the SR LPBF IN625 alloy makes its behavior unpredictable. Thus, heat treatment including only stress-relief annealing is not recommended for the practical use of LPBF IN625 alloy.

Regarding the ST-treated LPBF alloy, at the same creep-to-yield stress ratio, the ST specimens exhibited significantly improved creep properties as compared to the reference wrought annealed alloy. However, in the "absolute creep stress-rupture time" diagram (Figure 11), at first sight, the ST specimens still had an advantage over the wrought annealed alloy, especially at intermediate and low stresses (≤200 MPa). Using the "strain–time" creep curves, the steady creep rates were measured and are collected in Figure 12. It was clearly observed that the higher the applied stress, the higher the steady creep rate, which led to shorter lifetimes for the SR, ST, and wrought alloys, and a higher fraction of the intergranular fracture areas for the SR and ST alloys. However, it was seen that under the same creep stress of ≥150 MPa, the steady creep rates for the wrought and SR-treated LPBF alloys were higher than that for the ST-treated LPBF alloy.

**Figure 11.** Applied stress versus creep rupture time at 760 ◦C for the SR, ST, and wrought annealed IN625 alloys.

**Figure 12.** Creep stresses (σ) and mean rupture times (τ) vs. mean steady creep rates (. ε) for the (**a**) SR, (**b**) ST, and (**c**) wrought annealed IN625 alloys.

In general, creep behavior depends strongly on the metallurgical structure, i.e., the grain and particle sizes, the concentration of the alloying elements, and the creep conditions: stress and temperature. All the structure characteristics mentioned will affect the structural stability against vacancies and dislocations mobility, grain boundary diffusion, and, consequently, crack initiation and propagation at high temperatures under creep stresses [41]. In this study, for example, concentrations of Fe and C in the LPBF alloy were lower than in the wrought alloy. Note that according to the data of Heubner [42], elevated contents of Fe and C in IN625 negatively affect the creep behavior, especially at high temperatures.

Ni-based superalloys are polycrystalline and multiphase materials, and as such, correlating their properties with structural features is not straightforward. In particular, regarding the grain size influence, some data for IN718 demonstrate that an increase in grain size reduced the total creep rupture time, regardless of whether their grain boundaries are clean or intensively decorated with δ precipitates (67%) [43]. However, in classic cases, the trend is just the opposite: the coarser the structure, the greater the creep resistance [41]. A possible explanation for these discrepancies is probably related to the testing methodology: the same stress would create more severe creep conditions for coarse-grained structures with lower strength characteristics (YS, UTS) than for fine-grained structures with higher strength characteristics.

It has also been shown that the greater the density of δ precipitates at IN718 alloy grain boundaries, the shorter the total creep time, but when this density exceeds 45%, the creep time increases rapidly [43]. It has been suggested that with an increase in the density of precipitates at grain boundaries, the formation of wedge cracks at triple points can be delayed to the stage where creep voids around precipitates situated far from the triple edges are able to grow into unstable cracks. Thus, the beginning of the tertiary stage, characterized by rapid crack propagation, is delayed. Note, however, that in this

study, all the creep tests were carried out at the same stresses, and thus, the variations in the mechanical strength of alloys with different precipitation densities were not taken into account.

A correlation between the creep rate (Figure 7) and the fractography (Figure 9) observations was be noted. For example, the specimens of the wrought annealed alloy accumulated a significant elongation to failure during their final (tertiary) creep stage, which was reflected by a dimpled fracture surface. At the same time, the ST specimens manifested a long, steady creep stage related to grain boundary sliding, void formation, and crack initiation, which was reflected by fast intergranular/transgranular crack propagation to failure.

Since the aim of the present article was mainly to present the tensile and creep properties of LPBF IN625 alloy, the authors will concentrate more efforts on establishing a correlation between the structural features and the functional properties in the next publication. However, to summarize, the SR-treated IN625 alloy contained needle-like δ phase and globular precipitates of M6C carbides on grain boundaries [24], whereas the grain structures and phase states of the ST- and HIP-treated alloys were comparable. It has been previously shown that HIP dissolves the δ phase precipitates, forms MC, and homogenizes initially anisotropic SR-structure [24]. Thus, the presence of a significant quantity (that should be evaluated) of intergranular δ phase precipitates in vertical SR specimens reinforces grain boundaries against the sliding and formation of voids under stresses. After ST, in the case of an equiaxed grain structure with inter- and transgranular carbides, the creep lifetime is less significant, but the ST structure still has advantages over the wrought annealed specimen.

Summary. It was seen that the SR-treated LPBF alloy exhibited the highest anisotropy, i.e., build orientation dependency, and the least predictable mechanical behavior, as compared to its STand HIP-treated counterparts. Therefore, the SR alloy is the least safe material for practical use, especially at elevated temperatures. The ST and HIP treatments improved the alloy homogeneity and provided isotropic properties, thus making these alloys more application-safe. The mechanical strength characteristics of the LPBF alloy after the ST and HIP treatments satisfied the ASTM B444 standard requiring high strength in the 25–593 ◦C (68–1100 ◦F) temperature range.

Note that while the HIP alloy can be seen as a material with more uniform mechanical characteristics and improved ductility over the ST-treated alloy, this advantage came at the expense of the lower strength characteristics.

Furthermore, at T ≥ 650 ◦C, special attention must be paid to the time-dependent properties (creep). It was seen that although the LPBF alloy manifested much lower static ductility at these temperatures, it offered significantly longer rupture times under stresses of <200 MPa, as compared to its wrought annealed counterpart (Figure 13).

**Figure 13.** Comparison of the high temperature mechanical properties (760 ◦C, 1400◦F) of the ST LPBF and wrought annealed IN625 alloys: (**a**) strength (YS, UTS) and elongation to failure (ε), and (**b**) creep rupture time (τ) under a creep stress of 185 ± 5 MPa.
