2025 Volume 66 Issue 5 Pages 590-599
The production of ultra-thick Ti-6Al-4V (Ti-64) alloy plates is critical for aerospace applications, however, meeting the demanding mechanical and microstructural standards presents significant challenges. This study focuses on optimizing the manufacturing process of 75 mm ultra-thick Ti-64 plates to meet the stringent requirements of the AMS4905E specification. Initial attempts revealed that the presence of the primary hcp-α phase, even after β-annealing, severely limited ductility and work-hardening capacity. The mechanical performance, particularly elongation, failed to meet the standard despite achieving adequate yield strength (YS) and ultimate tensile strength (UTS). To overcome these limitations, an optimized process was developed, incorporating a homogenization step prior to hot forging and a globularization step after hot rolling. This approach was aimed at achieving a more uniform microstructure and enhancing the α-to-β phase transformation during heat treatment. Microstructural characterization using optical microscopy (OM), scanning electron microscopy (SEM), and electron backscatter diffraction (EBSD) confirmed the elimination of the primary hcp-α phase, with the formation of a refined and homogenous Widmanstatten structure. These changes resulted in a significant improvement in mechanical properties, particularly elongation, which increased from 4–7.5% in the preliminary trials to 10–11.5%, fully meeting the AMS4905E requirements. In addition to mechanical properties, the size of the prior β grains was carefully controlled to remain within the specified limits of the standard, ensuring full compliance with both microstructural and performance criteria. This optimized process not only enhances the mechanical performance of 75 mm ultra-thick plates but also establishes a reliable foundation for the future production of 100 mm thick Ti-64 plates, which are currently under development. The advancements demonstrated in this study contribute significantly to the field of high-performance titanium alloys for aerospace applications.
Titanium alloys, particularly Ti-6Al-4V (Ti-64), play a crucial role in the aerospace, ocean, and chemical industries due to their excellent balance of mechanical properties, including high strength-to-weight ratio, corrosion resistance, and thermal stability [1–7]. Among the various forms of titanium alloy products, plates are extensively used in the production of key components such as pressure vessels, shells, and structural elements in aerospace applications [8–13]. In modern aerospace engineering, Ti-64 is favored for its unique combination of lightweight properties and durability, making it an ideal material for critical components subjected to high stress and temperature extremes.
In the aerospace sector, thick titanium alloy plates have emerged as indispensable materials for the manufacture of components such as aircraft pylons, end frames, and bases in hypersonic vehicles. The pylon, in particular, is a structural component that connects the aircraft engine to the wing, bearing significant loads during flight while exposed to fluctuating temperatures. These components demand materials that can not only withstand high mechanical stresses but also resist fatigue under cyclical loading conditions. Ti-64 meets these demands with its high strength, fatigue resistance, and ability to retain mechanical properties at elevated temperatures, making it a preferred choice for such applications.
Ti-64 has become one of the most widely produced titanium alloys globally, accounting for over 80% of titanium alloy usage in the aerospace industry in regions such as the United States and Europe. Its superior properties, including an ultimate tensile strength of around 900 MPa and impressive fracture toughness [9], have established it as a staple material for critical aerospace structures. The alloy’s ability to maintain performance at temperatures exceeding 350°C [14, 15] while offering low-density benefits has made it essential for components exposed to harsh operational conditions.
Despite the widespread use of thin titanium alloy plates, there is limited research on the production and optimization of ultra-thick titanium alloy plates. The demand for thick plates, especially in aerospace and defense, is increasing due to their superior resistance to fatigue and their capacity to withstand more severe operational conditions. However, the production of such ultra-thick plates presents significant challenges, primarily due to the limitations of current hot-rolling technologies. These limitations often lead to inhomogeneous microstructure and mechanical properties, resulting in variability in performance across the plate. This issue necessitates in-depth research and technological advancements to optimize the microstructure and mechanical characteristics of thick plates, ensuring the production of reliable and high-performance components.
Developing ultra-thick Ti-64 plates, particularly those up to 75 mm in thickness, offers significant opportunities for enhancing the performance of critical aerospace components. By improving the manufacturing processes, including heat treatment and rolling techniques, it is possible to overcome the current challenges associated with producing such thick plates. Moreover, advancements in this field are essential to reducing the lead times and costs associated with manufacturing large-scale titanium structures while simultaneously boosting the competitiveness of industries involved in aerospace and defense.
This study aims to address these challenges by exploring the optimization of the production processes for 75 mm ultra-thick Ti-64 alloy plates. Through the development of new techniques, this research seeks to contribute to the broader goal of enhancing the global competitiveness of titanium alloy production and ensuring the reliable performance of critical aerospace components.
The aviation-grade 23 Ti-64 extra-low interstitials (ELI) ingots were manufactured into forged slabs at SeAH Changwon Integrated Special Steel, where they were subjected to upsetting and cogging processes to break down the initial cast structure. These processes ensured the refinement of the microstructure, promoting enhanced mechanical uniformity. The forged slabs were subsequently processed at POSCO using thick plate rolling equipment, where they were transformed into ultra-thick plates with a final thickness of 75 mm.
The 75 mm thick plates were subjected to further thermal treatments, with the aim of deriving heat treatment conditions applicable to aerospace industry standards. Various heat treatment scenarios were designated and applied to optimize the material’s microstructure and mechanical properties. These heat treatments are intended to replicate practical industrial conditions and will be detailed in the Results and Discussion section.
Microstructural analysis of the thick plates was performed on the transverse section of the specimens extracted from both the central and near-edge regions of the plates. Optical microscopy (OM; Olympus OLS5100) and scanning electron microscopy (SEM; SU5000, Hitachi) coupled with electron backscatter diffraction (EBSD) were employed to characterize the grain structure and phase composition. The investigations were conducted at an accelerating voltage of 20 kV with a working distance of ∼14 mm. A step size of 0.3 µm was employed for acquiring the EBSD maps. OIM Analysis 8.6™ software was used to process the EBSD data. This combination of techniques enabled a comprehensive understanding of the microstructural evolution within the material as a result of the applied processing and heat treatments.
The mechanical properties of the plates were assessed using uniaxial tensile tests performed on cylindrical dog-bone tensile samples extracted as per the ASTM E370 standard from the same central and near-edge regions. The tensile tests were conducted to determine key mechanical properties such as yield strength (YS), ultimate tensile strength (UTS), and elongation, which are critical for meeting the aerospace material specification AMS4905E [16], using an MTS E43 universal testing system at room temperature and a strain rate of 10−3 s−1 for all samples.
In the course of attempts to produce the 75 mm ultra-thick Ti-64 plates, various scenarios were explored. This paper discusses the major challenge that led to the failure of the preliminary attempts and provides a detailed analysis of the underlying reasons for this setback. Following this discussion, the paper presents the successful approach that ultimately enabled the authors to overcome the challenges encountered.
3.1 Preliminary attemptThe schematic representation of the manufacturing procedure for the 75 mm ultra-thick Ti-64 plates is illustrated in Fig. 1. During this process, the ingots underwent hot forging at a temperature of 1200°C, followed by hot rolling at 921°C, with a total reduction ratio of at least 70%. The resulting plates were subsequently subjected to β-annealing at 1010°C, followed by a secondary annealing at 730°C for stress relief.
Schematic representation showing the manufacturing process of the initial steps for the 75 mm ultra-thick Ti-64 plate.
The average mechanical properties, including YS, UTS, and elongation, were measured at both the center and near-edge of the produced plate, as shown in Fig. 2. It is important to highlight that the black, red, and blue dashed lines represent the minimum requirements of the AMS4905E standard for YS (745 MPa), UTS (841 MPa), and elongation (8%), respectively. While the YS (black squares) and UTS (red circles) values significantly exceeded the minimum requirements, falling within the ranges of 880–900 MPa and 960–980 MPa, respectively, all the examined samples exhibited elongation below the required 8%, ranging from 4% to 7.5%, thus failing to meet the standard specification.
Average mechanical properties of the 75 mm ultra-thick Ti-64 plate; result from the preliminary attempt, displaying YS, UTS, and elongation average values of both the center and near-edge regions, with reference to AMS4905E standard. (online color)
To better understand the cause of this poor elongation, microstructural analyses were performed using SEM and EBSD on the center and near-edge regions of the ultra-thick plates, as shown in Fig. 3. The SEM images, presented in Fig. 3(a) and (b), reveal a high fraction of large, elongated phases within and along the boundaries of the prior β grains. EBSD analysis (Fig. 3(c)) confirmed the presence of these phases, while the phase map (Fig. 3(d)) identified them as the hcp-α phase. The KAM map (Fig. 3(e)) further indicated that these large hcp-α phases (indicated by white arrows) exhibit a much lower dislocation density compared to the surrounding Widmanstätten morphology, suggesting that these are primary hcp-α phases that did not transform during the β-annealing process.
Microstructural characterization of the 75 mm ultra-thick Ti-64 plate after β-annealing heat treatment; preliminary attempt. SEM images reveal the microstructure at the center (a) and near-edge (b) regions. EBSD maps (c–e) illustrate the inverse pole figure (IPF), phase distribution, and kernel average misorientation (KAM) in the center region. (online color)
To investigate the origin of this observation, further microstructural examination was conducted after the hot forging and hot rolling stages, as shown in Fig. 4. OM images taken from the near-edge and center regions of the slab after hot forging, Fig. 4(a) and (b), reveal a non-homogeneous microstructure characterized by α-colonies, in which coarse equiaxed α grains are primarily surrounded by finely transformed β phase. In the subsequent hot-rolling step at the (α + β)-regime, OM images (Fig. 4(c) and (d)) continued to show an inhomogeneous microstructure. After the hot-rolling step, two distinct morphologies of α-colonies emerge: the majority are aligned parallel to the rolling direction (RD) and consist of thick, straight α lamellae, while a smaller proportion exhibit large equiaxed hcp-α phases. Both morphologies are encased by fine-transformed β. The EBSD analysis of the samples post-hot rolling confirmed that the large, elongated hcp-α phases, exhibiting strong crystallographic orientation, dominated the microstructure at both the center and near-edge of the ultra-thick plate. As shown in Fig. 4(c) and (d) the fraction of spheroidized hcp-α phases is minimal, even though rolling was conducted at a relatively high temperature of 921°C. In contrast, some regions of the microstructure display very coarse equiaxed hcp-α phases. These variations in dynamic spheroidization can be attributed to the differing responses of various α-colonies to spheroidization [17], with the initial microstructural inhomogeneity following hot-forging likely playing an even more significant role in producing the observed non-uniform microstructure.
Microstructural analysis following hot forging and hot rolling in the initial attempt to produce the 75 mm ultra-thick Ti-64 plate. Optical micrographs depict the near-edge (a, c) and center (b, d) regions, while EBSD maps (e-i, f-i) show phase distribution, IPF maps (e-ii, f-ii), and KAM maps (e-iii, f-iii) for both regions. (online color)
These observations suggest that, although the presence of the primary hcp-α phase after β-annealing contributed to the high YS, it severely limited the material’s hardening capacity (Hc = σUTS/σy − 1), which remained around 0.09. This can be attributed to the coexistence of primary hcp-α and Widmanstätten structure, which caused an inhomogeneous stress distribution, with stress concentrations forming at phase interfaces, which acted as preferential sites for crack initiation. Furthermore, the large, elongated morphology of the primary hcp-α phases observed in Fig. 3 provided an easy path for crack propagation along these interfaces, thus significantly facilitating crack growth. Therefore, it can be inferred that the primary cause of the poor elongation observed in the preliminary attempt is the presence of the coarse, elongated primary hcp-α phase. Future efforts must focus on eliminating this phase, as its removal is crucial not only for improving ductility but also for compliance with the AMS4905E standard, which explicitly prohibits the presence of primary or equiaxed hcp-α phases in the microstructure. This issue will be addressed in detail in the following section.
3.2 Optimized attemptThe findings from the initial attempts indicated that the presence of the primary hcp-α phase within the microstructure was a significant factor contributing to the poor elongation results. This large-sized hcp-α phase likely results from elemental distribution inhomogeneity in the as-received ingot, leading to its stabilization during hot-forging and rolling steps. Furthermore, EBSD microstructural analysis following hot rolling reveals a pronounced texture in the elongated, large hcp-α phase, which impedes sufficient dynamic globularization during rolling and preserves microstructural inhomogeneity, even after annealing above the β-transus temperature. Consequently, the β-annealing duration was insufficient for completing the α-to-β transformation. In this scenario, extending the β-annealing time above the β-transus temperature could potentially resolve the issue. However, the AMS4905E imposes strict requirements not only on mechanical properties but also on microstructural features, such as the prior β grain size. Therefore, extending the β-annealing step could significantly increase the prior β grain size [18], potentially compromising compliance with required standards. Therefore, to effectively eliminate the primary hcp-α phase, both the stability of this phase must be reduced, and the kinetics of the α-to-β transformation must be accelerated.
To address these challenges, the first objective was to achieve a homogeneous microstructure at each deformation stage of processing, in contrast to the heterogeneous structure observed in the preliminary attempts (refer to Fig. 4). A homogenous microstructure promotes more uniform mechanical properties throughout different regions of the ultra-thick plate, as the even distribution of deformation during hot-forging and -rolling ensures consistent properties across the entire plate. Additionally, this uniformity facilitates the subsequent heat treatments, allowing them to achieve the desired results in all areas of the plate within an appropriate time frame. Another crucial step in destabilizing the primary hcp-α phase is to reduce the size of the coarse elongated hcp-α phase before the β-annealing step. By decreasing the size of these phases, the interface area within the microstructure is significantly increased, thereby promoting the α-to-β transformation [19–21]. This allows for shorter β-annealing time while ensuring a fully transformed microstructure, thus overcoming the limitations imposed by the prior β grain size.
To achieve the desired homogeneous structure, the manufacturing process was optimized by introducing two additional steps: (i) Since compositional inhomogeneity across the as-received ingot was identified as a primary cause of the non-uniform microstructure observed after hot forging, a homogenization treatment at 1200°C was introduced prior to hot forging to eliminate this inhomogeneity; (ii) to further refine the microstructure and facilitate the α-to-β transformation during the β-annealing step, a sub-β-transus-static annealing (commonly known as globularization treatment) at 900°C was added following hot rolling. The optimized manufacturing process is depicted in Fig. 5. The microstructural evolution after hot forging, hot rolling, and the globularization treatment at the center and near-edge parts of the 75 mm ultra-thick plate was analyzed, with the results presented in Fig. 6.
Schematic representation showing the manufacturing process of the optimized attempt for producing the 75 mm ultra-thick Ti-64 plate.
Microstructural analysis following hot forging, hot rolling, and globularization step in the optimized attempt to produce the 75 mm ultra-thick Ti-64 plate. Optical micrographs depict the near-edge (a, c, and e) and center (b, d, and f) regions, while EBSD maps (g-i, h-i) show phase distribution IPF maps (g-ii, h-ii), and KAM maps (g-iii, h-iii) for both regions. (online color)
The OM images of the center and near-edge regions after hot forging, shown in Fig. 6(a) and (b), where a homogenization step was performed prior to forging, reveal a significantly more uniform microstructure compared to the preliminary attempt (see Fig. 4(a) and (b)). The post-hot-forging microstructure is characterized by α-colonies with a low fraction of transformed β and a uniformly distributed hcp-α phase, consisting of smaller and thinner lamellae of α-phase. Following the hot-rolling process (Fig. 6(c) and (d)), although the near-edge region of the plate shows a slightly higher fraction of transformed β-phase and elongated hcp-α plates compared to the center, the microstructure is notably more homogeneous, with considerable improvement in uniformity compared to the preliminary results (Fig. 4(c) and (d)). Notably, the microstructure after hot rolling reveals a substantially higher fraction of refined hcp-α plates, which can be attributed to dynamic spheroidization occurring during the hot-rolling stage [22] compared to the preliminary attempt. This pronounced occurrence of dynamic spheroidization during hot rolling is likely due to the more homogeneous microstructure achieved through the homogenization treatment.
The subsequent globularization treatment at 900°C, shown in Fig. 6(e) and (f), reveals that, while the near-edge region contains a higher fraction of the transformed β phase compared to the center, the overall microstructure exhibits further refinement. These findings are further corroborated by EBSD analysis (Fig. 6(g) and (h)), which shows a high degree of spheroidization in both the near-edge and center regions, with a more pronounced effect at the center. This can be explained by the slower cooling rate at the center of the slab, which allows for a longer retention of high temperatures, resulting in a greater degree of spheroidization from the center toward the outer edge. Variations in the degree of spheroidization in certain α-colonies are attributed to the slower rate of spheroidization due to differing orientations of the α-colonies [23]. Although spheroidization has been initiated in these colonies, the process remains incomplete due to insufficient treatment time. It is noteworthy that, based on the IPF maps, the microstructure in both the center and near-edge regions exhibits a significantly more random texture compared to the texture observed after rolling in the preliminary attempt (refer to Fig. 4(f-ii)).
Following these steps, β-annealing and secondary annealing were performed to evaluate the impact of the additional steps on the final microstructure. Figure 7 presents the SEM and EBSD analyses of the center and near-edge regions of the 75 mm ultra-thick plate after β-annealing at 1010°C, in the optimized attempt. The SEM images of both regions (Fig. 7(a) and (b)) indicate the formation of a nearly complete Widmanstätten structure of the hcp-α phase. This observation is supported by the IPF maps (Fig. 7(c-i) and (d-i)). Furthermore, the phase maps (Fig. 7(c-ii) and (d-ii)) and KAM maps (Fig. 7(c-iii) and (d-iii)) show no evidence of the primary hcp-α phase, despite the β-annealing being performed for the same duration as in the preliminary attempt. Therefore, it is evident that the addition of the homogenization step before hot forging and the globularization step after hot rolling successfully eliminated the primary hcp-α phase, leading to a fully transformed and optimized microstructure. For enhanced clarity, the evolution of the microstructure during both the preliminary and optimized attempts is schematically illustrated in Fig. 8.
Microstructural characterization of the 75 mm ultra-thick Ti-64 plate after β-annealing heat treatment; optimized attempt. SEM images reveal the microstructure at the center (a) and near-edge (b) regions. EBSD maps (c-i)–(c-iii) and (d-i)–(d-iii) illustrate the IPF, phase distribution, and KAM maps in the center and near edge regions, respectively. (online color)
Schematic illustration of microstructure evolution during preliminary and optimized attempts, highlighting the effects of the homogenization and globularization steps in the optimized process. (online color)
The mechanical properties of various sections of the 75 mm ultra-thick Ti-64 plate, produced using the optimized manufacturing process, were evaluated through tensile testing of samples taken from different regions of the plate. The average results are presented in Fig. 9. A slight reduction in YS is observed, with values decreasing from 880–900 MPa in the preliminary attempt to 810–850 MPa in the optimized attempt, which still meets the AMS4905 specifications. The observed decrease in YS is attributed to the elimination of the primary hcp-α phase. Meanwhile, the UTS remains in a range similar to that of the preliminary attempt. This indicates that the optimized process significantly improved the material’s hardening capacity (Hc) across the ultra-thick plate, increasing from 0.09 in the preliminary attempt to 0.17. Moreover, this enhanced work-hardening behavior resulted in a marked improvement in elongation, increasing from 4–7.5% in the preliminary attempt to 10–11.5% in the optimized process, fully meeting the AMS4905 requirements.
Average mechanical properties of the 75 mm ultra-thick Ti-64 plate; result from the optimized attempt, displaying YS, UTS, and elongation average values of both the center and near-edge regions, with reference to AMS4905E standard. (online color)
In addition to mechanical properties, AMS4905E also imposes regulations on specific microstructural features, including the previously mentioned prohibition of the primary hcp-α phase. Another critical microstructural criterion set by AMS4905E is the size of the prior β grains. According to this standard, prior β grains exceeding 1.27 mm in width and 2.54 mm in length must not constitute more than 10% of the microstructure [16]. Therefore, an OM analysis was conducted on the 75 mm ultra-thick plate to evaluate the prior β grain size. A total area of 66.15 mm2 from the plate thickness was examined, assessing the lengths and widths of over 200 grains, as presented in Fig. 10.
Analysis of the prior β grain size characteristics of the 75 mm ultra-thick Ti-64 plate to ensure compliance with AMS4905E; results from the optimized procedure. (a) OM image of the investigated area, (b) measured width and length of the prior β grains in reference to the AMS4905E limitations, (c) prior β grain size distribution of over 200 examined grains. (online color)
Figure 10(a) provides the OM image of the analyzed region, while Fig. 10(b) details the width and length of the prior β grains. The black and blue dashed lines represent the AMS4905E limitations on grain width and length, respectively. As shown, only one grain out of over 200 exceeded both the width and length limits. When the area of this grain is calculated, it constitutes only about 4% of the total investigated area, thus fully meeting the standard’s requirements. The distribution of prior β grain sizes, shown in Fig. 10(c), reveals that over 55% of the grains fall within the 300–600 µm range, with 25% measuring between 600–900 µm. Notably, only around 6% and 3% of the grains were found to be in the 900–1200 µm range and beyond, respectively. These findings demonstrate that the optimized manufacturing procedure, designed to eliminate the primary hcp-α phase, not only achieved superior mechanical properties well above the stringent requirements of AMS4905E but also successfully met the microstructural limitations outlined by the standard.
Given the considerable thickness and the associated challenges of rolling and heat treatment procedures in producing such high-thickness plates, the successful production of the 75 mm ultra-thick Ti-64 plates that fully comply with aerospace specifications represents a significant accomplishment. The next step toward realizing the ultimate goal of producing 100 mm thick Ti-64 plates is currently underway.
This study focused on optimizing the manufacturing process of 75 mm ultra-thick Ti-64 alloy plates to meet the stringent mechanical and microstructural requirements set forth by the AMS4905E aerospace standard. Through a series of initial and optimized attempts, the influence of microstructure on mechanical properties was carefully analyzed, leading to significant improvements in mechanical properties. The following are the key findings and achievements of the project:
This work was supported by the Technology Innovation Program [grant number 20016092], Development of Ti-64 alloy plate with 100 mm thickness by rolling process satisfying aerospace material specification, funded by the Ministry of Trade, Industry & Energy (MOTIE, Korea). This work was also supported by the Korea Institute for Advancement of Technology (KIAT) by the Korea Government (MOTIE) (RS-2024-00406598, HRD program for Industrial Innovation).
