Skip to main content
NCBI home page
Science Advances logoLink to Science Advances
. 2025 Aug 22;11(34):eady0937. doi: 10.1126/sciadv.ady0937

Naturally high fatigue performance of a 3D printing titanium alloy across all stress ratios

Zhan Qu 1,2, Zhenjun Zhang 1,2,*, Rui Liu 1,2,*, Zhefeng Zhang 1,2,*
PMCID: PMC12372868  PMID: 40845107

Abstract

Three-dimensional printing of structural materials, namely, additive manufacturing (AM), has notable advantages in fabricating structurally complex engineering components. These complex components usually endure comprehensive fatigue examination due to their complex stress distribution with varying stress ratios during service. Therefore, it is important to ensure the fatigue reliability of additive manufactured materials across all stress ratios. We found that the AM microstructure itself in a Ti-6Al-4V alloy successfully synthesizes the tripartite advantages of fine prior β grain boundaries, void-free, and fine α grains, which are respectively sensitive to the low, medium, and high stress ratio regions. Under this synergistic effect, the fatigue performance of the natural AM microstructure across all stress ratios not only outperforms all additive manufactured and forged Ti-6Al-4V alloys, but also surpasses other metallic materials. Our finding highlights the potential advantage of additive manufacturing technology in producing complex components with high fatigue resistance, substantially expanding its application scope.

A 3D printed titanium alloy has high fatigue resistance across all stress ratios via synergistic microstructure refinement.

INTRODUCTION

Additive manufacturing (AM) has revolutionized the notion and paradigms of traditional manufacturing for its high digitization, unrivaled forming freedom, and the rapid prototyping capacity of complex geometric components (1). In view of this competitive advantage, AM technology holds promise in overcoming the poor processability of titanium alloys and is expected to lead the way in replacing traditional manufacturing (2, 3), such as engine blades and disks in the aerospace area. However, this prospect has long been overshadowed by the generally poor fatigue performance of AM components. This is not only owing to the fact that fatigue is a prevalent catastrophic failure mode for structural components, but also because complex components normally endure comprehensive fatigue examination during service due to their complex stress distributions (2–5). This complexity is usually described by varying stress ratio R for cyclic loading, i.e., the ratio of minimum and maximum stress. Variations in stress ratio directly modulate the proportion of stress amplitude and maximum stress (6–8). The increase of stress amplitude will activate the persistent slip bands (PSBs) in extrusion, thereby tending to induce shear cracking, while the increase of the maximum stress will trigger the piling-up of dislocations, prone to causing cleavage cracking (7–17). That is to say, as the stress ratio changes, the suppression of one mechanism will inevitably promote the other (7–10). This seesaw relationship between them makes it extremely challenging to achieve high fatigue resistance across all stress ratios for a certain material. For example, in traditional forged titanium alloys, a specific microstructure usually exhibits its fatigue resistance advantage only within a certain stress ratio range (10, 16–20).

The unique physical metallurgy involved in AM, characterized by a small molten pool, rapid cooling, and dynamic thermal cycles, can effectively refine the initial microstructure and endow the alloys with exceptional tensile performance (1, 21–25). These advantages can not only provide high strength against shear damage but also prevent dislocation long-range piling-up, which can induce cleavage cracking of titanium alloys (4, 7–12, 26–29). On this ground, we propose that a void-free AM (Net-AM) microstructure might have naturally high fatigue damage immunity across all stress ratios, thereby achieving universally high fatigue resistance. Our previous work has demonstrated that Net-AM Ti-6Al-4V alloy exhibits ultrahigh fatigue resistance at a specific stress ratio (R = 0.1) (4), providing preliminary evidence for this hypothesis. Therefore, here, we systematically evaluate the fatigue performance of the Net-AM microstructure across all stress ratios. Clarifying this issue is crucial for the development of AM technology, as, in the future, with ongoing technological innovation for gradually eliminating microvoids, the concomitant high fatigue resistance across all stress ratios would greatly broaden the application of AM technology, especially for the components with complex structure in the aerospace area.

RESULTS

To investigate the fatigue resistance of the Net-AM microstructure, it is essential to first prepare a microstructure as close as possible to the ideal Net-AM state, that is, one with almost no microvoids and a microstructure similar to the as-printed state. To achieve this goal, an approximate void-free AM microstructure was successfully rebuilt using the previously developed posttreatment process (4), namely, the Net-AM preparation (NAMP) technique, integrating a high temperature–short time (HT-St) treatment after hot isostatic pressing (HIP). It is evident from Fig. 1 (A to C) that almost all microvoids are removed from the as-printed samples after the NAMP technique, except for a few tiny pores reopening during HT-St treatment (30, 31). Apart from the microvoids, a more important question is whether the original microstructures are preserved because the void healing procedures will invariably retrograde the original AM microstructures (Fig. 1B), as also commonly reported by others (32–34). The microstructure after NAMP still exhibits the ultrafine hierarchical α/α′ lath and variable lath variants within the fine prior β grains (PBGs) (Fig. 1C), very similar to the as-printed state in Fig. 1A. Next, systematic fatigue tests will be conducted on Ti-6Al-4V prepared by NAMP to examine whether the fatigue resistance of the Net-AM microstructure is naturally high at various stress ratios.

Fig. 1. Microvoid distribution and microstructure of Ti-6Al-4V alloy with as-printed, HIP, and NAMP states.

Fig. 1.

Open in a new tab

(A) The microvoid distribution and the microstructure characteristics of the as-printed state. (a) Three-dimensional (3D) visualization images of the quasi in situ x-ray tomography (XRT) test sample, where de is the equivalent diameter of microvoids. (b) Reconstructed orientation maps of PBGs based on the electron back-scattered diffraction (EBSD) results of the α phase. (c) Inverse pole figures (IPFs) of the α phase indicating the α lath size. (d) {0001} and {10-10} pole figures showing the orientation distribution of these lath variants. (e) Bright-field transmission electron microscopy (TEM) and (f) scanning TEM (STEM) micrographs showing the size and distribution of the α/α′ laths and β phases. (B) The microvoid distribution and the microstructure characteristics of HIP state. (C) The microvoid distribution and the microstructure characteristics of NAMP state.

The fatigue strengths of the NAMP Ti-6Al-4V at several typical stress ratios, including R = −1, R = −0.5, R = 0.1, and R = 0.5, were determined using the standard fatigue staircase method (35) to evaluate the fatigue resistance of Net-AM microstructure across all stress ratios, as shown in Fig. 2A. The data of maximum stress versus number of cycles are presented in fig. S1. Considering that R = 0.1 and R = −1 are the representative and commonly used stress ratios in various metallic materials, as can be seen from data S1, the fatigue performance at these two stress ratios should be paid particular attention. Our previous study has demonstrated that the NAMP microstructure exhibited the highest fatigue strength at R = 0.1 in Ti-6Al-4V (4). Here, the fatigue performances at R = −1 of NAMP, HIP, and as-printed Ti-6Al-4V alloy are given in Fig. 2B. It can be seen that after eliminating the microvoids of the as-printed alloys through the HIP technique, the fatigue lives at different stress amplitudes are improved significantly and the fatigue strength increased by 42%, which reveals the considerable detrimental effect of the microvoids on the fatigue resistance of the AM materials (5, 36, 37). After restoring the HIP microstructure to the AM microstructure using the NAMP technique, it was found that both fatigue life and fatigue strength are further improved, as displayed in Fig. 2B. Compared to the as-printed state, the fatigue strength of the NAMP state at R = −1 dramatically increased from 390 to 670 MPa (Fig. 2B), i.e., a 72% improvement. Furthermore, this value is compared with other AM and traditional forging microstructures, which shows that the NAMP state also exhibits the highest fatigue strength in Ti-6Al-4V at R = −1 (Fig. 2C). Similar results are also found for other R cases (Fig. 2C). Relevant data in Fig. 2C are summarized in data S1. Collectively, at various stress ratios, the NAMP microstructure presently exhibits the best fatigue performance in Ti-6Al-4V alloy.

Fig. 2. The fatigue properties of NAMP Ti-6Al-4V at various stress ratios.

Fig. 2.

Open in a new tab

(A) Standard staircase diagram for determining the fatigue strength at different stress ratios. (B) Maximum stress versus number of cycles (S-N) data at the representative stress ratio R = −1. (C) The tensile strength and fatigue strength of the Net-AM microstructure in comparison with other data of Ti-6Al-4V with different microstructures in the literature available, where (Ca) to (Cd) respectively correspond to the performance at stress ratios R = −1, −0.5, 0.1 and 0.5. The data for the R = 0.1 are cited from our previous study (6).

Then, does the fatigue performance of the NAMP microstructure represent that of the Net-AM microstructure? To address this, we must emphatically analyze the site of fatigue crack initiation, i.e., if fatigue cracks initiate from the inherent Net-AM microstructural constituents [e.g., ultrafine α laths and prior β grain boundaries (PBGBs) inherited from the original as-printed state], the answer is affirmative. Conversely, if cracks initiate from residual microvoids incompletely eliminated by NAMP, the Net-AM microstructure should exhibit higher fatigue performance, because fatigue cracks inherently initiate at the weakest link. For this purpose, a detailed investigation into the fatigue crack initiation behavior of the NAMP microstructure was conducted. As shown in Fig. 3, it demonstrates a pronounced stress ratio–dependent transition in the fatigue crack initiation site. We found the switch between microvoid cracking and microstructure cracking with changing stress ratios (6–15, 38–43), reconfirming the substantial impact of the aforementioned seesaw relationship in fatigue cracking mechanisms. Concretely, as the stress ratio decreases, the initiation sites shift gradually from microstructure-type PBGBs at R = 0.5 to defect-type microvoids at R = −0.5 and −1, with coexistence of both in between at R = 0.1, as illustrated in Fig. 3 (A to D). Consequently, for R = 0.5 where cracks initiate from the microstructural site, the performance of the NAMP microstructure can represent that of the Net-AM. In contrast, for R = −0.5 and −1 where cracks initiate from defective sites, the fatigue strength of the Net-AM microstructure should surpass that of the NAMP, as indicated by the red arrows in Fig. 3E. For R = 0.1 with dual initiation sites, that of NAMP should represent the lower limit of Net-AM. Therein, it can be concluded that the true fatigue performance of Net-AM is overall higher than that of NAMP, as illustrated by the red dashed line in Fig. 3E.

Fig. 3. Representative fatigue fractographies and corresponding crack initiation site of NAMP Ti-6Al-4V alloy at different stress ratios.

Fig. 3.

Open in a new tab

(A) R = −1: maximum stress σmax = 660 MPa, number of cycles to failure Nf = 9497661. (B) R = −0.5: σmax = 725 MPa, Nf = 6586112. (C) R = 0.1: left with σmax = 1100 MPa, Nf = 334538 and right with σmax = 1100 MPa, Nf = 1883048. (D) R = 0.5: σmax = 1175 MPa, Nf = 9759828. (E) Transition of fatigue crack initiation site and corresponding fatigue performance of the NAMP microstructure in Ti-6Al-4V alloy with stress ratio.

To objectively judge the fatigue performance of the Net-AM microstructure under various stress ratios, we plotted the classical Haigh maps by collecting the data from other Ti-6Al-4V microstructures, as well as other metallic structural materials, as displayed in Fig. 4. All these datasets are organized in data S1. As shown in Fig. 4A, at various stress ratios, the fatigue performance of Net-AM Ti-6Al-4V is overall higher than all the other AM and traditional forged counterparts. Moreover, in view of the lightweight design, the fatigue performance of Net-AM Ti-6Al-4V is further compared with other materials, normalized by their density, as displayed in Fig. 4B. It shows that, at various stress ratios, the Net-AM Ti-6Al-4V exhibits extraordinary specific fatigue performance over all the other materials, including steels, titanium alloys, aluminum alloys, magnesium alloys, copper alloys, superalloys, and high-entropy alloys. This indicates that the Net-AM microstructure has naturally high fatigue resistance across all stress ratios.

Fig. 4. Evaluation on fatigue properties of the Net-AM Ti-6Al-4V in comparison with other Ti-6Al-4V alloys and common metallic structural materials.

Fig. 4.

Open in a new tab

(A) Mean stress versus stress amplitudes of the Net-AM microstructure in comparison with other microstructures of Ti-6Al-4V alloys at various stress ratios. (B) Specific mean stress versus specific stress amplitudes of the Net-AM microstructure in comparison with other materials at various stress ratios.

DISCUSSION

The underlying mechanism of the excellent fatigue performance of Net-AM Ti-6Al-4V across all stress ratios will be expounded as follows. The fatigue failure of metallic materials generally arises from the localized irreversible damage accumulation at particular weaknesses, so these fatigue weaknesses are the key factors in controlling the fatigue resistance of materials (12, 27–29, 44–46). For AM and forged titanium alloys, there are three common types of fatigue weaknesses: two microstructure-type weaknesses, i.e., PBGBs (4, 17–19) and α grains (including primary α phases, α laths, and α colonies) (4, 8, 11, 34), and one defect-type weakness, i.e., microvoids (5, 15, 43), which are all observed in this study, as shown in Fig. 3 and figs. S2 and S3. As the stress ratio changes, the fatigue cracking mechanism undergoes a transformation, further leading to changes in the cracking sensitivity of the fatigue weaknesses. Therefore, the reason why the Net-AM microstructure has such high fatigue resistance across all stress ratios will be elucidated by analyzing the dominant mechanisms and the sensitive fatigue weaknesses at varying stress ratios.

1) At low R, commonly with high-stress amplitude, PSBs can be effectively activated; thus, the PSB in-extrusion cracking is the dominant mechanism (7–14). Under this mechanism, the PBGBs are the sensitive microstructural weakness, as the accumulation of PSB in extrusion along coarse PBGBs can lead to large intergranular cracks (17–19). Therefore, for the traditional forged lamellar/martensitic microstructures with coarse PBGBs, as shown by the blue-colored area in Fig. 5, the fatigue strengths exhibit a sharp decreasing trend at the low R region (region I). This also explains our recent finding that the lamellar and martensitic microstructures with PBGBs as their fatigue weakness show relatively lower fatigue strength and fast fatigue damage rate at low-stress ratios, as indicated by figs. S2 and S4. Accordingly, optimizing the fatigue performance at the low R region should particularly focus on controlling the size of PBGBs.

Fig. 5. Fatigue performance and cracking mechanisms for AM Ti-6Al-4V alloy with different types of microstructures at various stress ratios.

Fig. 5.

Open in a new tab

2) At high R, commonly with high maximum stress, the piling-up cleavage cracking becomes the dominant mechanism, especially for titanium alloys with limited operative slip systems (6–11, 13–16, 38–41). Under this mechanism, α grains become the sensitive microstructural weakness, as their dimension determines not only the dislocation piling-up distance, but also the initial crack size. Therefore, the traditional forged duplex/equiaxed microstructures with coarse primary α grains, as shown by the orange colored area in Fig. 5, in turn exhibit lower fatigue resistance and fast fatigue damage rate than the abovementioned lamellar/martensitic microstructures at high R (region III), which is also indicated by our experimental results in figs. S2 and S4. Thereupon, optimizing the fatigue performance of the high R region instead requires special attention in controlling the size of α grains.

3) At medium R with moderate stress amplitude and maximum stress, the effects of the above two fatigue cracking mechanisms are balanced, and thus, the microstructural PBGBs and α grains are comparable in affecting fatigue strength, as shown in region II of Fig. 5. However, for the microvoids, the inner surfaces accelerate the PSB in-extrusion accumulation, and meanwhile, the surrounding stress concentration facilitates the piling-up of dislocations (7–9, 15, 42), so the microvoids have large sensitivity at all R, which is especially reinforced at medium R. This is the reason why the fatigue performances of both as-printed state and NAMP state containing microvoids, as shown by the gray and red points in Fig. 5, deteriorate mostly at medium R (region II). Then, taking control of the void size is especially essential for optimizing the fatigue performance at the medium R region.

In view of the above varying cracking sensitivity of fatigue weaknesses with stress ratio, one microstructure with excellent fatigue properties at a certain R does not necessarily guarantee the same for others. Consequently, a high fatigue performance across the entire stress ratio range requires a simultaneous refinement of the three fatigue weaknesses. The unique solidification feature of the AM process can effectively refine the two microstructural weaknesses (21, 22, 47, 48). Therefore, through further eliminating the voids, the Net-AM microstructure successfully synthesizes the tripartite advantages, i.e., (i) fine PBGBs, (ii) void-free, and (iii) ultrafine α grains, thereby exhibiting comprehensive high fatigue resistance across all stress ratios, as shown in Fig. 5. Thus, it can be inferred that an overall microstructural refinement should be an effective strategy for enhancing the integrated fatigue performance of titanium alloys across all stress ratios.

In summary, the natural fatigue performances of Ti-6Al-4V alloy with the Net-AM microstructure at different stress ratios are revealed, using the approximate void-free AM microstructures prepared by our previously proposed NAMP technique. The results show that at various stress ratios, the Net-AM Ti-6Al-4V alloy not only has transcendent fatigue strength over the previously reported AM and forged Ti-6Al-4V alloys, but also yields the highest specific fatigue strength among all metallic materials. This outstanding fatigue performance across all stress ratios lies in the successful synergistic refinement of the three types of fatigue weaknesses respectively sensitive to the low R, medium R, and high R regions, i.e., PBGB, microvoid, and α grain. Our finding clarifies the naturally high fatigue resistance of AM microstructure across all stress ratios, thereby uncovering the untapped advantages of AM technology in producing anti-fatigue components with intricate topological structures or complex loading conditions. More broadly, this work also provides theoretical guidance and technical strategies for the anti-fatigue design of forged titanium alloys at various stress ratios.

MATERIALS AND METHODS

Sample preparation

Ti-6Al-4V, which is the mainstay of the titanium industry and the workhorse in the AM material family, was selected for this investigation and produced by laser powder bed fusion (LPBF) AM process in Xi’an Bright Laser Technologies Co., China. The LPBF printing equipment, the particle distribution and chemical composition of powder, and the control of oxygen content were consistent with our previous work (6). In addition, on the basis of our previous work (6), an optimal printing parameter with an energy density of 41.67 J/mm3 was chosen to reduce the presence of initial microvoids as much as possible. Rod-shaped samples with a length of 100 mm along the building direction (BD) and a diameter of 16 mm were deposited on the Ti-6Al-4V basal plate with a 200°C preheat temperature. After LPBF, all samples were subjected to a stress relief treatment at 550°C for 2 hours under a vacuum, which was termed the “as-printed state.” To further diminish the microvoids, the as-printed materials underwent the optimized HIP process at 920°C and 150 MPa for 3 hours using Quintus Technologies’ QIH48 URC equipment followed by air cooling in an argon atmosphere, which was termed the “HIP state.” As can be seen from Fig. 1Ba, the microvoids inside as-printed materials can be fully closed after the optimized HIP treatment via plasticity/creep-assisted and diffusion-based mass transport (31, 33). Although the microvoids were successfully removed, this initial approach caused a loss in the specific characteristics of the AM microstructure after the HIP treatment (Fig. 1, Bb to Bd). Consequently, to restore the beneficial AM microstructural features, an HT-St process combined with HIP was developed using the specific time node wherein the PBGs did not grow but the phase transformation was completed, which was termed as the NAMP technique. The detailed technological route of this technique was given in the previous work (6).

In addition, a forged Ti-6Al-4V alloy was chosen as the reference material for comparison. The as-received forged Ti-6Al-4V was supplied as a hot-rolled bar with a diameter of 53 mm and mill-annealed conditions, termed as the “forged original (F-O) state.” The F-O samples were subjected to a solution treatment at 950°C below the β-transus temperature for 1 hour followed by aging at 500°C for 2 hours, to obtain the duplex microstructure. The F-O samples were subjected to a solution treatment at 1050°C above the β-transus temperature for 3 hours followed by aging at 500°C for 2 hours to achieve a complete α→β phase transformation, to obtain a microstructure closest to the as-printed microstructure; this is termed the “forged martensite (F-M) state.” The F-M samples were further subjected to an annealing treatment at 900°C for 3 hours to promote the complete decomposition of martensite, to obtain a lamellar microstructure.

Mechanical tests

The fatigue specimens were machined into an hourglass shape along the BD with a gauge length of 15 mm and a diameter of 5 mm. In addition, a fatigue sample with a continuous radius of 30 mm and a minimum diameter of 3 mm was further designed based on the standard ISO (International Organization for Standardization) 1099:2006(E) to reduce the probability of microvoids as much as possible. The high-cycle fatigue (HCF) tests were carried out on a GPS100 high-frequency testing machine with a maximum loading capacity of 100 kN and a frequency range of 120 to 122 Hz. The fatigue tests were stopped when the specimens failed or reached 107 cycles. The fitting of S-N (maximum stress versus number of cycles) curves and the determination of fatigue strength were conducted according to the standard ISO 12107:2012: Metallic materials–Fatigue testing–Statistical planning and analysis of data (35).

Microstructural and microvoid examination

The microstructures were characterized by electron back-scattered diffraction (EBSD) using a ZEISS Sigma 500 field emission scanning electron microscope (SEM) with an operating voltage of 20 kV. Bright-field transmission electron microscopy (TEM) and scanning TEM (STEM) imaging were carried out using a Thermo Fisher Scientific Talos F200X TEM with an acceleration voltage of 200 kV. The information on microvoid defects inside the specimen was examined by the three-dimensional (3D) high-resolution transmission x-ray tomography (XRT) technique with a laboratory-based Xradia VersaXRM-500 system. The fatigue fracture morphology after the HCF test was observed by SEM. The image analysis software Image-Pro Plus was used to analyze the defect information on the SEM images of the fracture surfaces.

Acknowledgments

Funding: This work was supported by the National Natural Science Foundation of China (NSFC) under grant nos. 52322105 (Zhenjun Zhang), 52371084 (R.L.), 52321001 (Zhefeng Zhang), U2241245 (R.L.), 52130002 (Zhefeng Zhang), and 52261135634 (Zhefeng Zhang); the Strategic Priority Research Program of the Chinese Academy of Sciences under grant no. XDB1420000 (Zhefeng Zhang and Zhenjun Zhang); the Postdoctoral Fellowship Program and China Postdoctoral Science Foundation under grant no. BX20250307 (Z.Q.); National Key R&D Program of China grant 2023YFB4606600 (Zhenjun Zhang); Youth Innovation Promotion Association CAS grant no. 2021192 (Zhenjun Zhang); the Liaoning Science Foundation A (2025010059-JH6/1011) and XLYC2403113 (Zhenjun Zhang); International Joint Research Project of CAS grant no. 172GJHZ2022030MI (Zhefeng Zhang); IMR Innovation Fund 2023-ZD01 (Zhenjun Zhang); and the Shi Changxu Innovation Center for Advanced Materials (SCXKFJJ202212) (Zhenjun Zhang); the Natural Science Foundation of Liaoning Province under grant no. 2025-MS-076 (Z.Q.).

Author contributions: Conceptualization: Z.Q., Zhenjun Zhang, R.L., and Zhefeng Zhang. Methodology: Z.Q. and Zhenjun Zhang. Investigation: Z.Q. and Zhenjun Zhang. Visualization: Z.Q., Zhenjun Zhang, and R.L. Funding acquisition: Zhenjun Zhang and Zhefeng Zhang. Project administration: Zhenjun Zhang and Zhefeng Zhang. Supervision: Zhenjun Zhang and Zhefeng Zhang. Writing—original draft: Z.Q., Zhenjun Zhang, and R.L. Writing—review and editing: Z.Q., Zhenjun Zhang, R.L., and Zhefeng Zhang.

Competing interests: The authors declare that they have no competing interests.

Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.

Supplementary Materials

The PDF file includes:

Supplementary Text

Figs. S1 to S4

Legend for data S1

sciadv.ady0937_sm.pdf (989.4KB, pdf)

Other Supplementary Material for this manuscript includes the following:

Data S1

sciadv.ady0937_data_s1.zip (84.6KB, zip)

REFERENCES AND NOTES

  • 1.Zhang J. Q., Bermingham M. J., Otte J., Liu Y. G., Hou Z. Y., Yang N., Yin Y., Bayat M., Lin W. K., Huang X. X., StJohn D. H., Dargusch M. S., Ultrauniform, strong, and ductile 3D-printed titanium alloy through bifunctional alloy design. Science 383, 639–645 (2024). [DOI] [PubMed] [Google Scholar]
  • 2.Gu D. D., Shi X. Y., Poprawe R., Bourell D. L., Setchi R., Zhu J. H., Material-structure-performance integrated laser-metal additive manufacturing. Science 372, eabg1487 (2021). [DOI] [PubMed] [Google Scholar]
  • 3.Tan C. L., Weng F., Sui S., Chew Y. X., Bi G. J., Progress and perspectives in laser additive manufacturing of key aeroengine materials. Int. J. Mach. Tool Manuf. 170, 103804 (2021). [Google Scholar]
  • 4.Qu Z., Zhang Z. J., Liu R., Xu L., Zhang Y. N., Li X. T., Zhao Z. K., Duan Q. Q., Wang S. G., Li S. J., Ma Y. J., Shao X. H., Yang R., Eckert J., Ritchie R. O., Zhang Z. F., High fatigue resistance in a titanium alloy via near-void-free 3D printing. Nature 626, 999–1004 (2024). [DOI] [PubMed] [Google Scholar]
  • 5.Niu X., He C., Zhu S.-P., Foti P., Berto F., Wang L., Liao D., Wang Q., Defect sensitivity and fatigue design: Deterministic and probabilistic aspects in additively manufactured metallic materials. Prog. Mater Sci. 144, 101290 (2024). [Google Scholar]
  • 6.Oberwinkler B., On the anomalous mean stress sensitivity of Ti-6Al-4V and its consideration in high cycle fatigue lifetime analysis. Int. J. Fatigue 92, 368–381 (2016). [Google Scholar]
  • 7.Chi W., Wang W., Hou Z., Zhang M., Mechanism of mean stress sensitivity in Ti-6Al-4V under different stress ratios in high cycle fatigue. Int. J. Fatigue 183, 108268 (2024). [Google Scholar]
  • 8.Liu X., Sun C., Hong Y., Effects of stress ratio on high-cycle and very-high-cycle fatigue behavior of a Ti–6Al–4V alloy. Mater. Sci. Eng. A 622, 228–235 (2015). [Google Scholar]
  • 9.Chandran K. R., Duality of fatigue failures of materials caused by Poisson defect statistics of competing failure modes. Nat. Mater. 4, 303–308 (2005). [DOI] [PubMed] [Google Scholar]
  • 10.Ivanova S. G., Biederman R. R., Jr R. D. S., Investigation of fatigue crack initiation in Ti-6Al-4V during tensile-tensile fatigue. J. Mater. Eng. Perform. 11, 226–231 (2002). [Google Scholar]
  • 11.Chan K. S., Roles of microstructure in fatigue crack initiation. Int. J. Fatigue 32, 1428–1447 (2010). [Google Scholar]
  • 12.Stinville J. C., Charpagne M. A., Cervellon A., Hemery S., Wang F., Callahan P. G., Valle V., Pollock T. M., On the origins of fatigue strength in crystalline metallic material. Science 377, 1065–1071 (2022). [DOI] [PubMed] [Google Scholar]
  • 13.Wu Y., Liu J., Wang H., Guan S., Yang R., Xiang H., Effect of stress ratio on very high cycle fatigue properties of Ti-10V-2Fe-3Al alloy with duplex microstructure. J. Mater. Sci. Technol. 34, 1189–1195 (2018). [Google Scholar]
  • 14.Li W., Zhao H., Nehila A., Zhang Z., Sakai T., Very high cycle fatigue of TC4 titanium alloy under variable stress ratio: Failure mechanism and life prediction. Int. J. Fatigue 104, 342–354 (2017). [Google Scholar]
  • 15.Du L., Pan X., Qian G., Zheng L., Hong Y., Crack initiation mechanisms under two stress ratios up to very-high-cycle fatigue regime for a selective laser melted Ti-6Al-4V. Int. J. Fatigue 149, 106294 (2021). [Google Scholar]
  • 16.Tokaji K., Kariya H., Mean stress dependence of fatigue strength and subsurface crack initiation in Ti–15Mo–5Zr–3Al alloy. Mater. Sci. Eng. A 281, 268–274 (2000). [Google Scholar]
  • 17.Nalla R. K., Boyce B. L., Campbell J. P., Peters J. O., Ritchie R. O., Influence of microstructure on high-cycle fatigue of Ti-6Al-4V: Bimodal vs. lamellar structures. Metall. Mater. Trans. A 33, 899–918 (2002). [Google Scholar]
  • 18.R. Wanhill, S. Barter, Fatigue of beta processed and beta heat-treated titanium alloys (Springer Science & Business Media, 2012). [Google Scholar]
  • 19.Zuo J. H., Wang Z. G., Han E. H., Effect of microstructure on ultra-high cycle fatigue behavior of Ti–6Al–4V. Mater. Sci. Eng. A 473, 147–152 (2008). [Google Scholar]
  • 20.Chan K. S., Changes in fatigue life mechanism due to soft grains and hard particles. Int. J. Fatigue 32, 526–534 (2010). [Google Scholar]
  • 21.Zhang D., Qiu D., Gibson M. A., Zheng Y., Fraser H. L., StJohn D. H., Easton M. A., Additive manufacturing of ultrafine-grained high-strength titanium alloys. Nature 576, 91–95 (2019). [DOI] [PubMed] [Google Scholar]
  • 22.Song T., Chen Z., Cui X., Lu S., Chen H., Wang H., Dong T., Qin B., Chan K. C., Brandt M., Liao X., Ringer S. P., Qian M., Strong and ductile titanium-oxygen-iron alloys by additive manufacturing. Nature 618, 63–68 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Zhu Z. G., Nguyen Q. B., Ng F. L., An X. H., Liao X. Z., Liaw P. K., Nai S. M. L., Wei J., Hierarchical microstructure and strengthening mechanisms of a CoCrFeNiMn high entropy alloy additively manufactured by selective laser melting. Scr. Mater. 154, 20–24 (2018). [Google Scholar]
  • 24.Zhu Y., Wang B., Li H., Bi G., Xiao G., Chen C., An X., Weng F., Development of CoCrNi-based alloys: Insights into elemental doping and additive manufacturing. J. Mater. Sci. Technol. 245, 88–109 (2026). [Google Scholar]
  • 25.Zhang T., Huang Z., Yang T., Kong H., Luan J., Wang A., Wang D., Kuo W., Wang Y., Liu C.-T., In situ design of advanced titanium alloy with concentration modulations by additive manufacturing. Science 374, 478–482 (2021). [DOI] [PubMed] [Google Scholar]
  • 26.Dan C., Cui Y., Wu Y., Chen Z., Liu H., Ji G., Xiao Y., Chen H., Wang M., Liu J., Wang L., Li Y., Addad A., Zhou Y., Ma S., Shi Q., Wang H., Lu J., Achieving ultrahigh fatigue resistance in AlSi10Mg alloy by additive manufacturing. Nat. Mater. 22, 1182–1188 (2023). [DOI] [PubMed] [Google Scholar]
  • 27.An X. H., Wu S. D., Wang Z. G., Zhang Z. F., Significance of stacking fault energy in bulk nanostructured materials: Insights from Cu and its binary alloys as model systems. Prog. Mater Sci. 101, 1–45 (2019). [Google Scholar]
  • 28.Li P., Li S. X., Wang Z. G., Zhang Z. F., Fundamental factors on formation mechanism of dislocation arrangements in cyclically deformed fcc single crystals. Prog. Mater Sci. 56, 328–377 (2011). [Google Scholar]
  • 29.Zhang Z. F., Wang Z. G., Grain boundary effects on cyclic deformation and fatigue damage. Prog. Mater Sci. 53, 1025–1099 (2008). [Google Scholar]
  • 30.Tammas-Williams S., Withers P. J., Todd I., Prangnell P. B., Porosity regrowth during heat treatment of hot isostatically pressed additively manufactured titanium components. Scr. Mater. 122, 72–76 (2016). [Google Scholar]
  • 31.Shao S., Mahtabi M. J., Shamsaei N., Thompson S. M., Solubility of argon in laser additive manufactured α-titanium under hot isostatic pressing condition. Comput. Mater. Sci. 131, 209–219 (2017). [Google Scholar]
  • 32.du Plessis A., Macdonald E., Hot isostatic pressing in metal additive manufacturing: X-ray tomography reveals details of pore closure. Addit. Manuf. 34, 101191 (2020). [Google Scholar]
  • 33.Yan X., Yue S., Ge J., Chen C., Lupoi R., Yin S., Microstructural and mechanical optimization of selective laser melted Ti6Al4V lattices: Effect of hot isostatic pressing. J. Mater. Processes 77, 151–162 (2022). [Google Scholar]
  • 34.Alegre J. M., Díaz A., García R., Peral L. B., Cuesta I. I., Effect of HIP post-processing at 850 °C/200 MPa in the fatigue behavior of Ti-6Al-4V alloy fabricated by selective laser melting. Int. J. Fatigue 163, 107097 (2022). [Google Scholar]
  • 35.International Organization for Standardization, “Metallic materials-Fatigue testing-Statistical planning and analysis of data” (12107:2012(E), ISO, 2012).
  • 36.Wang S., Ning J., Zhu L., Yang Z., Yan W., Dun Y., Xue P., Xu P., Bose S., Bandyopadhyay A., Role of porosity defects in metal 3D printing: Formation mechanisms, impacts on properties and mitigation strategies. Mater. Today 59, 133–160 (2022). [Google Scholar]
  • 37.Sanaei N., Fatemi A., Defects in additive manufactured metals and their effect on fatigue performance: A state-of-the-art review. Prog. Mater Sci. 117, 100724 (2021). [Google Scholar]
  • 38.Sun C., Li Y., Huang R., Wang L., Liu J., Zhou L., Duan G., Crack initiation mechanism and fatigue life of titanium alloy Ti–6Al–2Sn–2Zr–3Mo-X: Effects of stress ratio and loading frequency. Mater. Sci. Eng. A 798, 140265 (2020). [Google Scholar]
  • 39.Li G., Ke L., Ren X., Sun C., High cycle and very high cycle fatigue of TC17 titanium alloy: Stress ratio effect and fatigue strength modeling. Int. J. Fatigue 166, 107299 (2023). [Google Scholar]
  • 40.Yang K., Zhong B., Huang Q., He C., Huang Z. Y., Wang Q., Liu Y. J., Stress ratio effect on notched fatigue behavior of a Ti-8Al-1Mo-1V alloy in the very high cycle fatigue regime. Int. J. Fatigue 116, 80–89 (2018). [Google Scholar]
  • 41.Furuya Y., Takeuchi E., Gigacycle fatigue properties of Ti–6Al–4V alloy under tensile mean stress. Mater. Sci. Eng. A 598, 135–140 (2014). [Google Scholar]
  • 42.Benedetti M., Fontanari V., Bandini M., Zanini F., Carmignato S., Low- and high-cycle fatigue resistance of Ti-6Al-4V ELI additively manufactured via selective laser melting: Mean stress and defect sensitivity. Int. J. Fatigue 107, 96–109 (2018). [Google Scholar]
  • 43.Fu R., Zheng L., Ling C., Zhong Z., Hong Y., An experimental investigation of fatigue performance and crack initiation characteristics for an SLMed Ti-6Al-4V under different stress ratios up to very-high-cycle regime. Int. J. Fatigue 164, 107119 (2022). [Google Scholar]
  • 44.Li L. L., Zhang Z. J., Zhang P., Zhang Z. F., A review on the fatigue cracking of twin boundaries: Crystallographic orientation and stacking fault energy. Prog. Mater Sci. 131, 101011 (2023). [Google Scholar]
  • 45.Liu R., Zhang P., Zhang Z. J., Wang B., Zhang Z. F., A practical model for efficient anti-fatigue design and selection of metallic materials: II. Parameter analysis and fatigue strength improvement. J. Mater. Sci. Technol. 70, 250–267 (2021). [Google Scholar]
  • 46.Liu R., Zhang P., Zhang Z. J., Wang B., Zhang Z. F., A practical model for efficient anti-fatigue design and selection of metallic materials: I. Model building and fatigue strength prediction. J. Mater. Sci. Technol. 70, 233–249 (2021). [Google Scholar]
  • 47.Davids W. J., Chen H., Nomoto K., Wang H., Babu S., Primig S., Liao X., Breen A., Ringer S. P., Phase transformation pathways in Ti-6Al-4V manufactured via electron beam powder bed fusion. Acta Mater. 215, 117131 (2021). [Google Scholar]
  • 48.Wang H., Chao Q., Chen H. S., Chen Z. B., Primig S., Xu W., Ringer S. P., Liao X. Z., Formation of a transition V-rich structure during the α′ to α+β phase transformation process in additively manufactured Ti-6Al-4V. Acta Mater. 235, 118104 (2022). [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Text

Figs. S1 to S4

Legend for data S1

sciadv.ady0937_sm.pdf (989.4KB, pdf)

Data S1

sciadv.ady0937_data_s1.zip (84.6KB, zip)

Articles from Science Advances are provided here courtesy of American Association for the Advancement of Science

RESOURCES