1. Introduction
Inconel 718, a Ni-Fe-Cr based precipitation-hardened superalloy, is renowned for its high strength, corrosion resistance, and excellent creep and fatigue properties, making it essential in aerospace and various industrial sectors1-4). The alloy’s versatility has fueled extensive research and led to its use in diverse manufacturing methods to meet modern engineering demands. Traditional techniques like casting and forging have long been foundational in producing Inconel 718 components, enhancing material properties through controlled deformation and enabling the production of intricate shapes1). Recently, advancements in 3D printing have revolutionized these traditional methods by offering unprecedented design freedom and efficiency. These technologies enable the fabrication of complex geometries with reduced waste and shorter lead times, and also allow for the creation of tailored microstructures, presenting a robust alternative to conventional production processes for Inconel 7185).
Furthermore, this superalloy is a complex, multiphase mixture predominantly consisting of Ni, Cr, and Fe, enriched with substantial quantities of Nb, Ti, Al, and Mo. The inclusion of Nb facilitates the precipitation of the primary strengthening phase, γ″(Ni3Nb), which displays a body-centered tetragonal (BCT) D022 crystal structure during the aging process5,6). Concurrently, the δ phase, chemically identical to γ″(Ni3Nb), forms with an orthorhombic D0a crystal structure. Additionally, the presence of Al and Ti contributes to the emergence of the γ′(Ni3(Al,Ti)) phase, which is characterized by a face-centered cubic (FCC) L12 crystal structure. Despite its superior attributes, Inconel 718 is prone to the development of unwanted phases, such as the Laves phase ([Ni,Cr,Fe]2[Nb,Mo,Ti]), under certain processing and operational conditions.
Welding processes introduce localized thermal cycles, which can induce microstructural changes, including the formation of undesirable phases. The weldability issues in joining of Inconel 718 superalloy includes segregation of alloying elements and consequential Laves phase development in fusion zone (FZ), heat-affected zone (HAZ) liquation cracking and solidification cracking. Even with meticulous control of welding parameters, the inherent complexity of the microstructure remains susceptible to phase transformations. These transformations can occur under varying thermal conditions. Despite the completion of post-heat treatment (HT) processes to yield finished components, the welding of Inconel 718 components may inadvertently reintroduce laves phase formation, compromising the integrity and performance of welded joints. Consequently, extensive studies have been conducted on this alloy from the perspectives of solidification segregation and thermal cycle variations7-14). Similar to the welding process, 3D printing fundamentally involves the melting and solidification of materials. A quintessential example comparing conventional manufacturing methods with 3D printing is the study by G. Meng et al.13), which analyzes the microstructures and associated machinability of Inconel 718 fabricated through laser directed energy deposition, laser powder bed fusion, and forging processes. The research highlights how the grain size, dislocation density, and texture obtained from each process could influence mechanical properties and underscores the necessity for designs that consider deformation mechanisms. In terms of Inconel 718 3D printing, research has particularly focused on the high- productivity method of Wire Arc Additive Manufacturing (WAAM) using wire. Although numerous studies discuss the need for HT in Inconel 718 manufactured via WAAM, similar to conventional processes9,14-16), changes in the HAZs due to subsequent thermal events in finished specimens have not yet been thoroughly examined. Considering the need for research from the perspectives of repair welding or cladding, it is imperative to comprehensively investigate the transformations of various precipitate phases, such as γ′, γ″, δ, Laves, and NbC, and the matrix in the heat-affected zones facilitated by this study.
In this context, the present study seeks to explore the detailed microstructural changes induced by welding heat cycles in Inconel 718 superalloy, specifically comparing materials fabricated using traditional manufacturing methods-such as casting, forging, and HT- with those produced by 3D printing technologies. This investigation meticulously examines how different stages of conventional processing influence the microstructure before and after welding, and contrasts these findings with the characteristics observed in 3D printed Inconel 718. The goal is to understand the differential effects that manufacturing techniques have on the alloy’s response to welding, particularly focusing on the development of phases that may affect mechanical properties and structural integrity.
2. Experimental procedure
2.1 Sample preparation
The chemical composition of the Inconel 718 used in this study is 52.5 wt.% Ni, balance Fe, 19% Cr, 5.1% Nb, 3% Mo, 0.5% Al, 1.0% Ti, 0.08% C, among others. For this research, three distinct base metal (BM) specimens were prepared using different manufacturing processes, designated as as-cast, as-forged, and as- printed BMs. These classifications are detailed in Table 1. Firstly, the casting process utilized vacuum induction melting as an essential step, followed by vacuum arc remelting to further eliminate impurities and segregations, resulting in a highly purified ingot. The resulting as-cast BM was partially sectioned to fabricate billets through the forging process. Each of these processes adhered to the conventional manufacturing conditions typical for Inconel 718 production. The WAAM process utilized a Cold-Metal Transfer (CMT) welding system and a six-axis robot to construct bulk specimens. The three types of BM specimens produced were subjected to subsequent HT processes to induce appropriate precipitation-hardened microstructures. Each specimen was designated with an ‘H’ suffix to denote HT, resulting in labels such as AC, AFH1/AFH2, and APH1/ APH2, corresponding to as-cast, as-forged heat-treated, and as-printed heat-treated, respectively. The procedures, consisting of solution HT followed by aging HT, are depicted in Fig. 1.
Table 1
Acronyms and stages of specimen treatment for Inconel 718 base metals. This table outlines the progression and designation of base metal specimens through different stages of treatment and processing. It includes their initial forms (as-cast, as-forged, as-printed), post-heat treatment identifiers, and acronyms assigned after welding
| Description | Processing phase | |||
|---|---|---|---|---|
| Base metal | After solution heat treatment | After aging heat treatment | After welding | |
| As-cast | AC | - | - | - |
| As-forged | AF | AFH1 | AFH2 | AFHW |
| As-printed | AP | APH1 | APH2 | APHW |
To examine the microstructural variations induced by the welding heat cycle, bead-on-plate welding experiments were performed using Gas Tungsten Arc Welding (GTAW) on the three types of heat-treated BM specimens. The welding was executed with a voltage of 12V and the current of 220A on plates that had been finish machined to a thickness of 3 mm. Argon gas was utilized as the shielding gas, with a flow rate set at 15 L/min. To achieve full penetration and the formation of a back bead, an appropriate travel speed was meticulously determined for the welding process. Subsequently, each specimen was labeled with a ‘W’ suffix added to its respective acronym, resulting in identifiers such as ACHW, AFHW, and APHW. The geometry of the specimens is presented in Fig. 2.
2.2 Microstructural analysis and hardness test
For microstructural analysis, specimens were prepared using standard metallography techniques, which included cutting cross-sections, hot mounting, grinding, polishing, and chemical etching. Subsequent analyses were conducted on select samples using a field emission scanning electron microscope (FE-SEM; JSM- 7001F, Jeol, Japan), equipped with energy-dispersive X-ray spectroscopy (EDX; X-Max 80, Oxford Instruments, UK). For detailed microscopic examination, Transmission Electron Microscope (TEM003B Talos F200X, Thermo Fisher Scientific, USA) samples were extracted using focused ion beam (FIB) techniques. Moreover, SEM analyses were enhanced by EDX and electron backscatter diffraction (EBSD; NordlysNano, Oxford Instruments, UK), providing comprehensive composition and phase analysis. EBSD was performed with a collection angle of 70° and a maximum step size of 200 nm in both x and y directions. Additionally, for the subsequent microstructure images, clear explanations of the specimen locations are provided, and to aid understanding, the built direction (BD), normal direction (ND), and transverse direction (TD) are indicated.
To evaluate the mechanical properties of the bead- on-plate welds, Vickers hardness testing was employed. Each test was meticulously conducted across the FZ, HAZ, and BM to establish a comprehensive profile of the Inconel 718 hardness variations. Utilizing a standardized Vickers hardness testing machine (AHM43, Leco, USA), an indentation load of 500 gf (gram-force) was applied for a dwell time of 10 s. The results, illustrated in a detailed hardness line profile, reveal significant variations that correlate with microstructural changes induced by the welding process. Values for each location were determined using three replicate measurements, which were subsequently plotted on a graph to illustrate the variations.
3. Results and discussion
3.1 Base metal microstructure in Inconel 718 superalloy fabricated by conventional manufacturing processes
Before exploring the substantive discussion, it is necessary to consider conventional manufacturing methods for Inconel 718. During the casting process, this material undergoes solidification, accompanied by segregation, nucleation and growth, solid-state transformations, and subsequent reheating during the forging process, as well as dynamic recrystallization and phase transformations during cooling. These microstructural changes are crucial for process optimization, and as such, extensive research has been conducted to understand and enhance these phenomena. The process conditions used in this study for preparing the BM specimens are based on these accumulated experiences, representing optimized procedures. Fig. 3 presents the results of SEM analysis of the base metal microstructures for each process. The as-cast microstructure was observed to consist of a fully austenitic matrix with grain sizes in the tens of micrometers. At the grain boundaries, particles such as NbC and Laves were detected due to segregation (Fig. 3a). After forging, most of the precipitates had dissolved, leaving δ phase at the grain boundaries, and some grain growth was noted (Fig. 3b). Following solution heat treatment and two-stage aging, δ phase was predominantly identified at the grain boundaries in the SEM analysis (Fig. 3c), and it is speculated that intragranular strengthening phases such as γ′, γ″ may have precipitated within the matrix, which will be further discussed in the results of the TEM analysis in Section 3.3. This supports the observed trend of increasing hardness values, which progress from 224.2 Hv0.5 in AC, to 303.8 Hv0.5 in AFH1, and finally to 405.8 Hv0.5 in AFH2.
3.2 Base metal microstructure in Inconel 718 superalloy fabricated by 3D printing
Conventional manufacturing methods and 3D printing processes fundamentally differ based on whether deformation processes such as rolling and forging are involved, and on the type of heat source used for layer-by-layer additive manufacturing, which can result in microstructural variations. Fig. 4 presents the results of an EBSD analysis evaluating the crystal orientation of base metal specimens. In the as-cast state, Inconel 718 typically exhibits some directional properties but predominantly forms equiaxed grains; however, abnormal grain growth driven by segregation can be observed in certain areas. Upon subsequent forging at elevated temperatures, even in the presence of dynamic recrystallization, there is generally an increase in grain size, leading to a more uniform grain distribution. In contrast, in the as-printed state, grain growth aligned with the build direction (BD) is evident, with grain sizes significantly larger by several orders of magnitude compared to the other specimens. The orientation of grain growth is generally opposite to the direction in which heat dissipates, particularly in 3D printing where it typically occurs away from the substrate. This phenomenon is especially pronounced in the WAAM process due to the relatively high heat input. Hence, to achieve uniform mechanical properties across different directions, it is essential to consider various process strategies that can partially mitigate this effect16). Fig. 5 presents SEM images that observe the microstructural changes of Inconel 718 fabricated via 3D printing following solution HT and two-stage aging HT. In the AP specimen, the dendritic boundaries of γ-austenite exhibit precipitates such as Laves and NbC (Fig. 5a). After undergoing solution HT, while some NbC remains, a significant portion of the Laves phase is dissolved (Fig. 5b). This trend generally mirrors that observed in conventionally manufactured Inconel 718, with the primary distinction being the predominant precipitation of δ phase at regions affected by solidification segregation. Following two-stage aging HT, no significant differences are noted (Fig. 5c); however, the hardness trend is similar to that observed in AC and AFH2 specimens, with hardness values increasing from 214.4 Hv0.5 in AP to 393.7 Hv0.5 in APH2 specimens.
Fig. 4
Electron backscatter diffraction analysis of phase and texture in the base metal, comparing specimens from different manufacturing processes: As-cast, as-forged and as-printed Inconel 718
Fig. 5
Microstructural analysis of inconel 718 base metal fabricated by the 3D printing processes, (a) AP, (b) APH1, and (b) APH2
The TEM analysis of the Inconel 718 matrix, prepared using the FIB method, reveals the presence of nanosized precipitates within the austenite matrix, as shown in corresponding Dark Field (DF) image of Fig. 6. Selected Area Electron Diffraction (SAED) pattern analysis along the [001] zone axis of the γ-austenite matrix confirms that these precipitates are composed of γ’ and γ’’. These two phases, intermetallic compounds, share a simple integer ratio chemical formula of A3B, yet differ structurally, possessing FCC and BCT crystal structures, respectively, and also vary in their alloy composition ratios. Typically, in Inconel 718, these phases are denoted as Ni3(Al,Ti) for γ’ and Ni3Nb for γ’’ respectively. Ultimately, the two-stage aging HT following solution HT is critical for reducing harmful Laves phase precipitates formed during solidification and effectively precipitating the γ’ and γ’’ phases, which are essential for enhancing the material’s properties. The HT process optimized for as-forged Inconel 718 (AF) may not be suitable for as-printed Inconel 718 (AP) produced via WAAM. This is because the HT temperature/time conditions required for the dissolution of harmful phase depend on a diffusion function, which can vary based on the size of precipitates, segregation, and grain size. This discrepancy arises due to the unique component geometries intended for 3D printing, which require optimized path planning during the layer-by-layer additive manufacturing process. Such planning significantly influences the orientation of grains solidified during the process. Consequently, the subsequent HT must be tailored to accommodate the microstructural characteristics developed during printing, necessitating the design of complex process parameters that consider these specific attributes.
3.3 Microstructures of the heat-affected zone and hardness variations after welding in Inconel 718 superalloy
Welding typically involves remelting, characterized by the formation of a FZ, and the dissipation of heat outward, leading to the creation of a HAZ. During this process, the characteristics of the FZ can be controlled through variables such as welding method, the use of filler materials, shielding gas, and the amount of heat input. However, the properties of the HAZ are predominantly influenced by the characteristics of the BM. In other words, the design of the material, or alloy design, should be thoughtfully considered during the manufacturing stages. To prevent degradation, the amount of heat input is meticulously regulated, and this principle applies to Inconel 718, where the properties of the HAZ undergo alterations accordingly.
Fig. 7 presents the inverse pole figure maps of Inconel 718 weldments, analyzed using EBSD. These maps compare two different specimens, AFHW and APHW, to highlight the crystallographic orientation differences between them, particularly in areas adjacent to the fusion line. Although the FZ of both specimens exhibit similar characteristics, notable differences in crystal orientations are observed in the HAZ adjacent to the fusion line for both AFHW and APHW specimens. Upon closer analysis, especially when correlating Fig. 7 with the crystal orientation results from Fig. 3c for AFH2, Fig. 5c for APH2, and the findings presented in Fig. 4, it can be concluded that the distribution of crystal grains in the BM remains relatively unchanged in the HAZ regions. This suggests that the thermal influence during the welding process does not significantly alter the crystallographic structure of the BM, maintaining a consistency across different thermal treatments and conditions. As illustrated in Fig. 8, a notable difference related to precipitate phases is the observed increase in δ phase at the dendritic boundaries. Given the thermodynamic metastability of the γ″ phase, it is prone to transitioning into the δ phase when subjected to elevated temperatures17). To quantitatively evaluate the degradation of specimens, specifically the changes in the γ′/γ″ phases, the hardness of two specimens was characterized using a Vickers hardness test, as shown in Fig. 9. In the AFHW specimen, the hardness of the as- forged BM is notably high, approximately 400 Hv0.5, due to the presence of strengthening phases developed through heat treatment. However, as the measurement approaches the HAZ and the FZ, there is a relative decrease in hardness. Similarly, the APHW specimens demonstrate the same trend, where the hardness significantly drops to approximately 200 Hv0.5 near the fusion line in the HAZ. This suggests that most of the strengthening phases formed by the two-stage aging heat treatment have likely dissolved due to the welding heat. To understand the microstructural changes induced by welding heat, equilibrium phase fractions were calculated using Calphad software. As shown in Fig. 10, the phases that induce intragranular strengthening and those that contribute to high-temperature cracking and degradation of creep properties exist within distinct temperature ranges. Considering the manufacturing process, NbC, which includes titanium, can persist through the as-cast or as-printed states and throughout the post-HT processes. However, when heated to temperatures near or below the lowest temperature limit of the liquidus-solids coexistence region, which is around 1265°C at the fusion line, most precipitate phases inevitably undergo dissolution. Consequently, the strengthening effects formed during the HT processes in both conventionally manufactured and 3D printed Inconel 718 are effectively nullified by the welding heat.
Fig. 7
Inverse pole figure maps of Inconel 718 weldments adjacent to the fusion line, (a) compares the AFHW and (b) the APHW specimens
Fig. 8
Microstructural analysis of the heat-affected zone in Inconel 718 fabricated by 3D printing process, (a) Low magnification, and (b) high magnification images
4. Conclusions
This study explores the microstructural evolution of Inconel 718, a precipitation-hardened superalloy, across various conventional manufacturing processes and 3D printing, coupled with different heat treatments. Significant findings include the microstructural variations from the as-cast, as-forged states through solution heat treatment and two-stage aging heat treatment. The hardness values were observed to increase from 224.2 Hv0.5 in the as-cast condition to 405.8 Hv0.5 after the second stage of heat treatment in conventionally manufactured specimens. Similarly, 3D printed Inconel 718 specimens exhibited an increase in hardness from 214.4 Hv0.5 in the as-printed state to 393.7 Hv0.5 in specimens after two-stage aging heat treatment, indicative of the intragranular strengthening effect from the γ′/γ″ phases. However, bead-on-plate welding experiments demonstrated that welding heat could negate these hardening effects, reducing hardness to levels around 200 Hv0.5, similar to those in the fusion zone. Consequently, when considering additional welding operations such as repair welding, it is crucial to design structures that minimize the thermal impact or to employ strategies that reduce the extent of the heat-affected zone.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
