Dual‐Array Nano Configuration for High‐Performance Metastable β Titanium Alloys
Tianle Li, Renhao Wu, Jiabao Liu, Sang‐Ho Oh, Xiang Wu, Hyojin Park, Byeong‐Joo Lee, Hidemi Kato, Hyoung Seop Kim, Xiaochun Liu, Xifeng Li
TL;DR
A new titanium alloy design achieves high strength and ductility at high temperatures through a unique nanostructure formed by dislocation and phase interactions.
Contribution
A novel dislocation-phase coupling mechanism enables dynamic microstructure control in metastable titanium alloys.
Findings
The alloy achieves 863 MPa ultimate tensile strength and 78.3% elongation at 500°C.
Dual-array nano α grains form through interlaced and parallel configurations.
Phase-field simulations and strain mapping validate the new mechanism.
Abstract
Catastrophic failures in engineering metallics frequently occur at high temperatures. A fundamental understanding of plastic deformation and the mechanisms governing the strength‐ductility trade‐off is essential for developing titanium alloys exhibiting superior properties at elevated temperatures. Herein, a metastable β titanium alloy (Ti‐15.1Mo‐3.1Nb‐2.77Al‐0.21Si, wt.%) exhibits unexpected mechanical properties, including an ultimate tensile strength of 863 MPa and a total elongation of 78.3% at 500 °C, accompanied by a continuous and strong work hardening rate (2000–3100 MPa). Dislocation slip and heating play pivotal roles in interlaced parallel α nucleation, and thermal activation promotes interlaced α nucleation. Finally, the dual‐array nano configuration of dense (≈68%) and thin (≈10 nm in width) α phase forms. Hierarchical microstructural evolutions, including β to α phase…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsTitanium Alloys Microstructure and Properties · Intermetallics and Advanced Alloy Properties · Magnesium Alloys: Properties and Applications
Introduction
1
The strength‐ductility trade‐off dilemma in metallic alloys dictates that enhancing these properties simultaneously remains a formidable challenge due to their intrinsic inverse correlation.^[^ 1, 2, 3, 4 ^]^ Consequently, achieving synergistic improvements in strength and ductility is regarded as a pivotal advancement in materials science.^[^ 5, 6, 7, 8 ^]^ Recent innovations in metallurgical design, thermomechanical processing, and composite engineering have enabled the development of heterostructured and nanoscale materials with exceptional room‐temperature mechanical performance.^[^ 9, 10, 11 ^]^ During plastic deformation, strain hardening in metallic systems arises from mechanisms, such as dislocation strengthening,^[^ 12 ^]^ transformation‐induced plasticity (TRIP),^[^ 13 ^]^ twinning‐induced plasticity (TWIP),^[^ 14 ^]^ hetero‐deformation induced (HDI) strengthening^15,16^ and multi‐principal elements theory.^[^ 17, 18, 19 ^]^ However, the role of spatially arranged nanograins with high thermal stability in governing elevated‐temperature strength and ductility remains underexplored.
For ultra‐high‐strength alloys, limited plastic formability poses significant challenges for industrial processing at room temperature.^[^ 20 ^]^ A promising strategy involves the initial fabrication of alloy sheets with low yield strength and high ductility, akin to TWIP steels,^[^ 21 ^]^ followed by post‐forming treatments to achieve ultra‐high strength and retained plasticity. Such an approach enhances component performance in applications demanding intrusion resistance and energy absorption.^[^ 22, 23 ^]^ Strain‐aging processes, combining plastic deformation and bake hardening, have been utilized to modulate precipitation behavior and dislocation substructures, enabling formability and property enhancement in high strength steels.^[^ 22, 24 ^]^ It is significant to achieve a possible yield strength, high work hardening rate, and large elongation in metastable β titanium alloys by adjusting deformation mechanisms, such as precipitation strengthening. Nevertheless, precise control over precipitate characteristics (e.g., morphology, size, volume fraction, and spatial distribution) for optimal high‐temperature performance remains elusive.
Titanium (Ti) alloys achieving exceptional strength‐ductility synergy at elevated temperatures are critical for aerospace and other high‐performance sectors.^[^ 25, 26 ^]^ Current research on high‐temperature Ti alloys predominantly emphasizes near α alloys, TiAl intermetallic, Ti_2_AlNb‐based systems, and Ti‐matrix composites.^[^ 27, 28 ^]^ Nevertheless, metastable β Ti alloys strengthened by α‐phase precipitation have received significantly less attention.^[^ 29 ^]^ The plastic deformation process of metastable β phase at high temperatures leads to a complex microstructural evolution, involving recovery, phase transformation, recrystallization, dislocation multiplication, and dislocation annihilation. Recovery and dislocation annihilation can enhance plastic deformation and decrease flow stress, leading to decreased strength. Moreover, metastable β Ti alloys typically attain high strength and ductility through solution treatment followed by single‐ or dual‐step aging processes, often requiring prolonged aging times ranging from hours to hundreds of hours. Inspired by the strain‐aging principles, we propose a hot‐straining protocol involving short‐term and simple stretching to tailor nanoscale α precipitates and dislocation architectures in β‐21S Ti alloy. This approach facilitates significant plastic formability while circumventing prolonged aging treatments,^[^ 30, 31 ^]^ ultimately achieving unprecedented strength‐ductility balance at 500 °C. We further comprehensively elucidated the precipitation mechanisms of the dual‐array α phase and mechanisms of the strength‐ductility synergy.
Results
2
Figure 1a presents a fully recrystallized β single‐phase microstructure with a random crystallographic orientation, as evidenced by the inverse pole figure (IPF) map. The equiaxed β grains, with an equivalent diameter of several micrometers, exhibit a nearly uniform size distribution. The geometrically necessary dislocations (GNDs) density of the initial microstructure is low at 0.12 × 10^14^ m^−2^, as shown in Figure 1b. According to a selected area electron diffraction (SAED) pattern under the zone axis of [011] in Figure 1c, there are no precipitated ω particles within the body‐centered cubic (BCC) β grains.^[^ 32 ^]^ EBSD and TEM results show that the initial Ti sample is fully composed of recrystallized β grains.
Initial microstructure, thermodynamic calculation of phase transformation, and mechanical properties of the β‐21S alloy at elevated temperatures: a) inverse pole figure (IPF) map of the β phase before tension, b) geometrically necessary dislocations (GNDs) density image and 1D characteristics, c) TEM bright field (BF) map under the zone axis of [110]BCC, d) calculated driving force of α precipitate nucleation from a metastable β matrix, e) calculated diffusion distance of each element for 10 min (expected time for a tensile test) as a function of temperature, (f) calculated composition of HCP phases at temperatures ranging from 300 to 600 °C.
According to the thermodynamic calculation in Figure 1d, the driving force of α phase nucleation increases from 0.7 to 2.7 kJ mol^−1^ when the temperature decreases from 600 to 300 °C. Here, the positive driving force indicates that nucleation occurs spontaneously. In addition, the calculated diffusion distance of each element does not exceed 1 µm at temperatures ranging from 300 to 600 °C (Figure 1e; Figure S1a, Supporting Information). Given that the typical size of a nucleus in a metallic system is in the order of 10^−9^–10^−8^ m,^[^ 33, 34 ^]^ nucleation would not be easy to achieve within a short time at temperatures below 500 °C due to sluggish diffusion kinetics, despite the large thermodynamic driving force. Even if α precipitates form at higher temperatures, their size is not expected to exceed 1 µm.
We further examined the alloying effects, which are not expected to significantly influence the driving force of nucleation (Figure S2c, Supporting Information). According to the calculated phase equilibria at 300–600 °C, Al mostly exists in the Ti‐rich HCP phase (Figure 1f), whereas Mo and Nb mainly exist in the Mo‐rich BCC phase (Figure S2b, Supporting Information). The Ti‐rich HCP phase shows limited solubility for Mo and Nb,^[^ 34, 35 ^]^ while exhibiting a certain solubility for Al up to 14 at% at 600 °C.^[^ 36 ^]^ Consequently, the concentration of Mo and Nb in the β matrix is expected to increase along with the nucleation and growth of α precipitates since these elements will not tend to diffuse into α precipitates. The slowest elements Mo and Nb would segregate at the α/β interface, contributing to delaying the growth of α precipitate.^[^ 37 ^]^
Figure 2a displays the engineering stress‐strain curves of this Ti alloy tested at elevated temperatures ranging from 300 to 600 °C. Notably, this alloy deformed at 500 °C exhibits excellent mechanical properties, with a yield strength (YS) of 511 MPa, ultimate tensile strength (UTS) of 863 MPa, and total elongation (TE) of 78.3%, respectively. The superior strength‐ductility can be attributed to the strong work hardening ability, as depicted in Figure 2b. The work hardening rate in stage II, corresponding to the true strains from 0.032 to 0.54, initially increases with a serrated behavior at a true strain below 0.26, and then gradually rises from ≈2000 to 3100 MPa. This exceptionally superb work hardening rate has not been reported previously for Ti alloys deformed at elevated temperatures. Figure 2c plots the UTS versus TE of our metastable β Ti alloy and other reported Ti alloys at 500 °C.^[^ 38, 39, 40, 41, 42, 43, 44, 45 ^]^ Notably, the TE increases by about three times without compromising high strength compared to other high‐temperature Ti alloys with UTS ranging from 800 to 900 MPa. It is rare for Ti alloys to achieve an excellent UTS and TE with a strong and continuous work hardening rate simultaneously at 500 °C. Moreover, the mechanical properties, such as specific UTS‐TE (Figure 2d), are outstanding compared to other high‐temperature alloys at 500 °C. These high‐temperature alloys are also shown in the Supporting Information. In addition, by designing a load of ≈47% pre‐strain and unloading, the Ti alloy can achieve an excellent combination of performance metrics, with a YS of 1152 MPa, UTS of 1224 MPa, and TE of 15.1% upon reloading at 500 °C, as shown in Figure S2 (Supporting Information).
*Mechanical properties of the Ti alloy: a) engineering stress‐strain curves at temperatures ranging from 300 to 600 °C, b) work hardening rate curve at 500 °C, c) comparison of UTS‐TE data between the present Ti alloy and other high‐temperature Ti alloys tested at 500 °C from Refs,[
38 , 39 , 40 , 41 , 42 , 43 , 44 , 45
] d) comparison of specific UTS‐TE data between the present Ti alloy and other high‐temperature alloys tested at 500 °C, and the references are present in the Supporting Information.*
The high‐resolution STEM image in Figure S3 (Supporting Information) presents some β′ particles in the sample held at 500 °C before tension. The β′ phase, with barren β stabilizers, shows a consistent crystal structure and a coherent interface relationship with the β matrix.^[^ 46 ^]^ When the engineering strain reaches up to ≈5%, several α grains with hexagonal close‐packed (HCP) structure, rodlike in size of ≈6 nm, transform from the β′ phase or/and β matrix, as displayed in Figure S3a (Supporting Information). According to the FFT pattern, these α precipitates comply with the Burgers orientation relationship (BOR) with the β parent, i.e., {110}β∥{0001}α, <111>β∥<112_0>α.^[^ 29, 47 ^]^
The morphology and distribution of α precipitates and transitional ω particles, with further increasing engineering strains of ≈28% (Figure S4, Supporting Information) and 64% (Figure 3a,b) with increasing strains at 500 °C, are also characterized using TEM and STEM. From the initial growth stage, α/β phase boundaries extend in a specific direction. DF images of the sample with an engineering strain of 64.4% at 500 °C reveal that thin acicular α precipitates present two distribution methods: an interleaved array extending in various directions (Type I) and a parallel array with an extension along the same direction (Type II). The interleaved array of α grains is a common pattern,^[^ 7, 25 ^]^ while the parallel array has not been reported in previous studies. On average, the width and length of α grains in Type I are 8.4 and 108.5 nm, respectively, while those of α grains in Type II are 12.3 and 74.2 nm in Figure 3c, respectively. In addition, the area fraction of fine α precipitates reaches up to 68% in the sample with an engineering strain of 64%, as shown in Figure S5 (Supporting Information). XRD spectra in the samples after tensile deformation at 500 °C also confirm the β to α phase transformation in Figure S6 (Supporting Information).
TEM images showing the α precipitates in the samples deformed at a–c) 500 °C, d) 400 °C, and 600 °C and near fracture: (a,b) two types of α precipitates (Types I and II) from β parent at an engineering strain of ≈64% with a SAED pattern at 500 °C, (c) the grain sizes of the two types of α precipitates, (d) a dark field (DF) image showing ω particles from the β parent at 400 °C, e,f) two types of α precipitates (Types I and II) from the β parent at 600 °C.
The comparative study on precipitation behaviors in the sample during tension at 400 and 600 °C was conducted to reveal the logical essence of the precipitation process. Figure 3d shows ω precipitates after tension at 400 °C, and no α grains are found. This agrees well with our theoretical expectations. Figure 3e,f presents acicular α precipitates with interleaved and parallel arrays, which is consistent with those in the sample deformed at 500 °C. In contrast, the α precipitates can actively grow, and the length and width of acicular α precipitates in the sample deformed at 600 °C are ≈2 and 3.6 times greater than those in the sample deformed at 500 °C, respectively. This is comparable to the relative differences in diffusion distance (≈4.3 times larger at 600 °C compared to 500 °C).
To explore the deformation mechanisms for the sample deformed at 500 °C, Figure 4 illustrates the dislocation characteristics with increasing engineering strain using STEM and TEM. From the bright field (BF) images in Figure 4a,b and Figure S7 (Supporting Information), the overall dislocation density increases with the engineering strain, ranging from ≈28% to 64%. Detailed characterization of local regions reveals a change in the dislocation motion pattern from planar slip in the β matrix (marked by red arrows in Figure S7a, Supporting Information) to multiple slip and cross slip (Figure 4a,b) with increasing plastic deformation. This also leads to a dislocation tangle near α/β phase boundaries, inducing strain hardening. The active slip and proliferation of dislocations facilitate the large plastic deformation of the alloy. Moreover, the GNDs densities in the regions with interleaved and parallel nano α precipitation arrays after a strain of 49% at 500 °C were measured using 4D‐STEM in Figure 4c and Figure S8 (Supporting Information). The interleaved α array region and nano α phase can hinder dislocation slip and promote dislocation proliferation, inducing increased work hardening. Moreover, the parallel α array region and β matrix relatively provide conditions for dislocation slip, inducing continuous plastic deformation. Compared with the strain mechanisms of the Ti alloy at 500 °C, the deformation mechanisms change at 400 °C (Figure 4d–f; Figure S9, Supporting Information) and 600 °C (Figure 4g–i; Figure S10, Supporting Information). The dislocation planar slip within the coarse β matrix plays a leading role in plastic deformation at 400 °C. Meanwhile, the brittle ω precipitates accelerate the damage at ω/β boundaries, resulting in common elongation. When the temperature reaches 600 °C, the dislocation slip resistance becomes weak due to α grain growth, and dislocations tend to annihilate at elevated temperatures. Therefore, the sample presents high elongation and common strength during deformation at 600 °C.
Bright field (BF) images of the sample deformed at a–c) 500 °C, d–f) 400 °C, and g–i) 600 °C using TEM and 4D‐STEM showing the dislocation distribution and motion under the two‐beam mode: (a–c) the sample with engineering strains of 49% and 28%, (a,b) BF images, (c) GNDs densities in the regions with interleaved and parallel nano α precipitate arrays, (d–f) dislocation planar slip and cross slip, (g–i) a low density of dislocations.
Discussion
3
According to the results above, we should clarify the precipitation mechanisms of α grains and all strengthening mechanisms to support the excellent mechanical properties. In general, the α phase formation, including its nucleation and growth from the β parent, is sluggish in near and metastable β Ti alloys during the static aging process.^[^ 46, 48, 49 ^]^ However, deformation or dislocation slip also plays a critical role in α phase nucleation apart from element partitioning under the action of elevated temperatures.^[^ 50, 51 ^]^ TEM (AC‐TEM) and EDS images in Figure 5 are utilized to analyze the precipitation mechanism of α grains under the strain at 500 °C. Figure 5a also shows that the dense and acicular α grains with a width of only ≈2 nm extend toward the crystal direction <111> within the β parent. It is well known that the favorable slip direction of the β matrix is the crystal direction <111>.^[^ 52, 53 ^]^ According to the IPF image (Figure 5c1,c3) and the Schmid factor image for slip systems {110} <111> (Figure 5c2), both the slip planes and the α precipitation are crystal planes {110}. We also note that a high density of geometrically necessary dislocations and subgrain boundaries (misorientation angle below 2°) form along the crystal planes {110} in Figure 5d. These results significantly indicate the relevance of α precipitation and dislocation formation and slip. Moreover, the α/β phase boundaries in Figure 5b maintain a semi‐coherent relationship based on the lattice misfit δ in Equation S1 (Supporting Information).
a) TEM, b) STEM, c,d) EBSD, and e,f) EDS observations displaying precipitation mechanism of α grains (Type II) under strains of ≈28% (c1–c4, d) and 49% (a,b,e,f) at 500 °C: (a) HR‐TEM image showing the fine α precipitates, (b) phase boundary under the zone axes of [110]BCC and [0001]HCP from (a), (c1,c3) IPF images of the β matrix, (c2) Schmid factor for slip systems {110} <111>, (c4) GNDs density image, (d) misorientation angle and KAM across some slip traces in (c3), 3D crystal cells illustrating the crystal orientation, (e) element distribution mapping and α grains are marked by white lines, (f) line scanning map along a yellow line in (e), (g) schematic sketch describing the two types of nucleation and growth mechanisms of nano α grains during tensile deformation at 500 °C.
Fine α grains are observed within localized regions exhibiting high dislocation density and accumulation, as depicted in Figure S11 (Supporting Information). Elemental mapping in Figure 5e,f reveals the distribution and concentrations of Mo, Al, and Nb within α grains and the β matrix. Notably, Mo concentrations in certain α grains approximate those in the β matrix, indicating that compositional partitioning is not the only factor affecting precipitation. Moreover, the element interdiffusion due to thermal activation effect contributes to the formation and growth of α in the α region, as shown in Figure S12 (Supporting Information). Therefore, the profuse α precipitation is primarily induced by local crystal defects such as dislocations,^[^ 54 ^]^ rather than active atom diffusion at 500 °C. The limited atomic diffusivity at this temperature, combined with the short processing duration, insufficiently supports α‐phase coarsening, resulting in fine and dense α characteristics.^[^ 46, 55 ^]^
According to the results and discussion mentioned above, Figure 5g presents a schematic sketch that describes two types of nucleation and growth of α grains, as well as multiple strength and ductility mechanisms during tensile deformation at 500 °C. Except for the thermal activation effect, deformation or dislocation slip plays a pivotal role in the second phase precipitation.^[^ 56, 57 ^]^ Crystal defects along slip lines, originating from plastic deformation, provide numerous sites and energy for α grain nucleation. The spacing of slip lines is ≈100 atoms (≈20 nm in width),^[^ 58 ^]^ which is consistent with the combination of width from the α grain and the adjacent and retained β matrix (Figure 3b). The slip bands and traces associated with dislocation slip can be noted within β grains in the sample after tension at 500 °C in Figure S13 (Supporting Information). Crucially, the density of slip lines governs the α‐grain nucleation rate, while the orientation of slip traces establishes favorable pathways for α‐grain growth. This defect‐mediated process results in a parallel array of nano α precipitates (Type II). Nano‐scale α grains can be obtained due to a low growth rate, which can be attributed to two aspects: insufficient high temperature and andante diffusion ability of Mo atoms.^[^ 59 ^]^ In addition, the α grains precipitate from the β parent following three available patterns^[^ 26 ^]^: direct β → α, β → β´ → α, and β → ω → α according to the maps in Figures S3 and S4 (Supporting Information), respectively.
To further elucidate the strength and ductility mechanisms of this Ti alloy at 500 °C, we explored the dislocation‐precipitate interactions by high‐resolution TEM and STEM images, as shown in Figure 6. For parallel‐patterned α precipitates with a (Type II) in Figure 6a,b, plastic deformation primarily occurs through the mechanism of dislocation planar slip within the β matrix channels between acicular α precipitates. Here, α/β phase interfaces act as pinning points, impeding the motion of dislocations. This pattern facilitates more channels for dislocation movement, resulting in significant plastic deformation and enhanced ductility. Conversely, Figure 6c depicts a high density of dislocation pile‐up and entanglement at α/β phase boundaries for α precipitates with an interleaved pattern (Type I) based on the α distribution in the DF map in Figure 6d. These phase interfaces function as impenetrable barriers that severely restrict dislocation transmission, significantly augmenting the deformation resistance.^[^ 60, 61 ^]^
a,c–g) TEM and b) STEM micrographs showing dislocation motion associated with α precipitate morphology at 500 °C with different strains of ≈28% (a,c,d,f,g) and ≈49% (b,e): (a,b) BF images showing dislocation slip within β matrix along parallel α precipitates (Type II), (c,d) BF and DF maps showing dislocation pile‐up and tangle near α/β boundaries for α precipitates type I, e) weak beam DF image presenting dislocations within fine α grains, f) HR‐STEM image showing α precipitates and the β matrix, g) lattice strain distribution of β matrix and α precipitates calculated from lattice images with zone axes [110]BCC and [0001]HCP, h) schematic sketch describing the strength‐ductility mechanisms including β to α phase transformation, the interaction between dislocations and nano α precipitates with dual‐array configurations, and dislocation interactions during tensile deformation at 500 °C.
The phase boundaries also serve as dislocation sources to continuously emit dislocations toward α and β grains (Figure S14, Supporting Information). Subsequently, dislocation slip invariably follows a favorable slip system in a specific direction. However, the random distribution of α grains with a high‐volume fraction can obstruct dislocation slip and induce strengthening. Phase and grain boundaries can act as dislocation sources, promoting plastic deformation by stimulating dislocation emission. Meanwhile, these boundaries can also act as strong barriers against dislocation slip, inducing work hardening.^[^ 62, 63 ^]^ In addition, dislocation slip and tangle are further resolved via weak‐beam dark‐field imaging in Figure 6e, which means that dislocation motion within α grains can also promote the high strength and ductility strategy. Furthermore, the HAADF image (Figure 6f) and lattice strain distribution map (Figure 6g) directly visualize lattice distortions arising from dislocation interactions and precipitate coherency strains. The entire zone, including the β matrix, α precipitates, and α/β boundaries, presents strong distortion. This strong distortion further indicates the severe dislocation action. Figure S15 (Supporting Information) indicates the ductile fracture. Figure S16 (Supporting Information) reveal weak α precipitates along the prior β grain boundaries in the regions near the fracture surface of the cross‐sectioned tensile samples at 500 °C. Prior β grain boundaries with a low fraction of α_GB_ precipitates can contribute to continuous and stable deformation with a large strain.^[^ 64, 65 ^]^
Figure 6h presents a schematic sketch that describes the mechanisms of the high strength and ductility strategy, mainly originating from three aspects: intrinsic slip behaviors of the HCP structure compared to the BCC structure (i.e., the β to α phase transformation effect);^[^ 26 ^]^ hindrance of dislocation motion due to numerous α precipitates with an interleaved array; and dislocation interactions, especially entanglement within the β matrix. Meanwhile, α precipitates with a parallel array provide conditions for long‐range slip of dislocations, and the initial β matrix (BCC structure) also contributes to large plastic strains, which results in high ductility. The heterogeneous deformation in this case is primarily due to the differential action of two types of nano α array with dislocation slip and the ability of α lattice and β matrix to accommodate dislocations. Additionally, the non‐uniform distribution of dislocations is a result of this strengthening mechanism induced by heterogeneous deformation. In short, these findings extend the understanding and application of the strain‐aging process, replacing the conventional strain and long‐time aging process in metastable β Ti alloys.
Conclusion
4
In this work, the strategy of harnessing nano precipitate heterogeneities in the studied metastable β Ti alloy, as opposed to avoiding them, expands the possibilities for microstructure engineering through straightforward thermomechanical processing. Herein, a metastable β titanium alloy (Ti‐15.1Mo‐3.1Nb‐2.77Al‐0.21Si, wt.%) exhibits unexpected mechanical properties, including an ultimate tensile strength of 863 MPa and a total elongation of 78.3% at 500 °C, accompanied by a continuous and strong work hardening rate (2000–3100 MPa). Dislocation slip and heating play pivotal roles in interlaced parallel α nucleation, and thermal activation promotes interlaced α nucleation. Finally, the dual‐array nano configuration of dense (≈68%) and thin (≈10 nm in width) α phase forms. Hierarchical microstructural evolutions mechanisms, including β to α phase transformation, nano α grains with dual‐array configurations of interleaved and parallel arrays, and dislocation interaction, are synergistically achieved during plastic strain. Compared to conventional sequential mechanical processing and aging treatment that span a few hours, this thermo‐mechanical forming method emerges as an excellent candidate for manufacturing parts of metastable β Ti alloys with full β grains for cost‐effective applications with superior properties.
Experimental Section
5
Alloy Treatment
Sheets of β‐21S Ti alloy with a thickness of 1.5 mm and a nominal composition of Ti‐15.1Mo‐3.1Nb‐2.77Al‐0.21Si wt.% were supplied by Baoti Group Co., Ltd. of China. A β‐21S sheet underwent a solution treatment in the β‐region at 830 °C/0.5 h in a muffle furnace, followed by water quenching to obtain complete β grains. The Ti alloy plates were heated within a furnace with a heating rate of 10 °C min^−1^.
Mechanical‐Property Tests
Flat dog‐bone specimens, with a gauge length of 8 mm, a width of 6 mm, and a thickness of 1.4 mm, were machined from the solution‐treated sheets. Tensile tests were performed at room and elevated temperatures under an initial strain rate of 0.002 s^−1^. The specimens were heated to specific temperatures (300–600 °C) inside an oven and held for 2 min before tensioning. The tensile specimens were heated when the furnace temperatures reached the set temperatures. The strains were measured via a video extensometer, and the measured surfaces of the tensile specimens were coated with a layer of high temperature resistant paint and dispersed black spots. Each sample was conducted for three repetitions of the tensile test.
Computational Approaches
Thermodynamic computation with diffusion coefficient calculation was conducted by CALPHAD (CALculation of PHAse Diagrams) software. For simplicity, the Ti‐Mo‐Nb‐Al quaternary system was assumed for the project, while Si was not considered, as it was present only in dilute amounts and was not expected to be critical for the analysis. A self‐consistent thermodynamic database and atomic mobility database were built based on literature data.
Thermodynamic description of a quaternary system requires thermodynamic descriptions of four subordinate ternary systems. Thermodynamic descriptions for Al‐Mo‐Ti and Al‐Nb‐Ti systems could be found in a reference, while those for Mo‐Nb‐Ti and Al‐Mo‐Nb systems were not available. For Mo‐Nb‐Ti system, only reliable experimental information was that the isothermal section at 1373 K solely consists of the BCC phase. For Al‐Mo‐Nb system, isothermal phase equilibria at 1273 and 1473 K were investigated. A full assessment of the parameters for the Al‐Mo‐Nb system would require significant efforts, while the Al‐Mo‐Nb ternary thermodynamic interaction was less significant compared to other interactions (Ti‐Al‐Mo, Ti‐Al‐Nb, and Ti‐Mo‐Nb) in the current Ti alloy. Hence, thermodynamic parameters only for BCC and M_3_Al phases were assessed to ensure adequate predictive performance for the BCC phase. The cited references are listed in Supporting Information. For Mo‐Nb‐Ti system, only reliable experimental information was that the isothermal section at 1373 K solely consists of the BCC phase. This could be fully described without additional ternary thermodynamic parameters. Most of atomic mobility information was taken from the references in the Supporting Information.
Microstructure Characterization
Electron back‐scattered diffraction (EBSD), transmission electron microscope (TEM), scanning TEM (STEM), and aberration correction‐TEM (AC‐TEM) observations were carried out to reveal the microstructural evolution in β‐21S Ti alloy before and after tensile deformation. EBSD measurements, with a 4 × 4 binning and step size of 0.7 µm, were conducted using a TSL system attached to a field emission gun scanning electron microscope (SEM). The samples were polished using a SiO_2_ suspension with a particle diameter of 0.03 µm, and the cross‐sectioned surfaces were etched using a solution of 2% hydrofluoric acid (HF), 4% nitric acid (HNO_3_), and 94% water (H_2_O) for SEM observation. TEM (STEM and AC‐TEM) slices were cut from the gauge sections of specimens and ground to a thickness of 50 µm. Thin foil samples with a diameter of 3 mm were prepared by twin‐jet‐electro‐polishing in a chemical solution of 6% perchloric acid and 34% butanol in methanol, followed by ion milling in an ion reducing instrument. Energy dispersive X‐ray spectroscopy (EDS) equipped in AC‐TEM was used to measure the element distribution. Dislocation density was measured using nano‐beam electron diffraction (NBED) in a STEM. 4D‐STEM measurements were performed on a ThermoFisher Scientific Spectra 200 microscope operating at 200 kV. 4D‐STEM generates a 4D dataset, collecting 2D images in real and reciprocal space at the same time. During scanning, the recorded 4D dataset includes information on local crystal orientation, structural deformation, and crystallinity. NBED data were collected with an EMPAD detector at a rate of 1000 frames per second. Each dataset consisted of electron diffraction patterns taken at each scan position with a probe step size of ≈1 nm. Around 300 × 300 scan positions were recorded in each region with a dwell time of 1 ms per frame. A convergence angle of 30 mrad, spot size of 8, and diffraction pixel size of 0.16 Å^−1^ were used in the micro‐probe lens configuration to balance real‐space resolution and strain resolution. Briefly, geometrically necessary dislocations (GNDs) were associated with the intragranular lattice rotation, which parameter could be measured by using high‐resolution AZtecCrystal software.
Conflict of Interest
The authors declare no conflict of interest.
Supporting information
Supporting Information
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1S. Z. Han , E. A. Choi , S. H. Lim , S. Kim , J. Lee , Prog. Mater. Sci. 2021, 117, 100720.
- 2J. Y. Sun , H. Q. Li , Y. J. Chen , X. H. An , Adv. Sci. 2024, 11, 2407283.
- 3K. Xie , J. Nie , C. Liu , W. Cha , G. Wu , X. Liu , S. Liu , Adv. Sci. 2023, 10, 2207208.10.1002/advs.202207208 PMC 1047787537431694 · doi ↗ · pubmed ↗
- 4H. Wu , G. H. Fan , Prog. Mater. Sci. 2020, 113, 100675.
- 5C. Zhang , X. Bao , M. Hao , W. Chen , D. Zhang , D. Wang , J. Zhang , G. Liu , J. Sun , Nat. Commun. 2022, 13, 5966.36216815 10.1038/s 41467-022-33710-1PMC 9550820 · doi ↗ · pubmed ↗
- 6H. Dai , W. Dai , Z. Hu , W. Zhang , G. Zhang , R. Guo , Adv. Sci. 2023, 10, 2207192.10.1002/advs.202207192 PMC 1019057236935371 · doi ↗ · pubmed ↗
- 7T. T. Song , Z. B. Chen , X. Y. Cui , S. L. Lu , H. S. Chen , H. Wang , T. Dong , B. L. Qin , K. C. Chan , M. Brandt , X. Z. Liao , S. P. Ringer , M. Qian , Nature 2023, 618, 63.37259002 10.1038/s 41586-023-05952-6PMC 10232360 · doi ↗ · pubmed ↗
- 8L. Wang , P. Xu , H. Yin , Y. Yue , W. Kang , J. Liu , Y. Fan , Adv. Sci. 2023, 10, 2303238.10.1002/advs.202303238 PMC 1052062837518855 · doi ↗ · pubmed ↗