2,870
Views
2
CrossRef citations to date
0
Altmetric
Reviews

Recent progress in the high-cycle fatigue behaviour of γ-TiAl alloys

ORCID Icon
Pages 1919-1939 | Received 12 Jan 2018, Accepted 06 Apr 2018, Published online: 27 Apr 2018

Figures & data

Figure 1. Schematic of FCG curves. Adapted and redrawn from [Citation9] (reproduced with permission).

Figure 1. Schematic of FCG curves. Adapted and redrawn from [Citation9] (reproduced with permission).

Table 1. Characteristics of the three regimes of FCG, adapted from [Citation10].

Figure 2. FCG curves for various TiAl alloy microstructures; all have high gradients in Stage II. From [Citation13] (reproduced with permission).

Figure 2. FCG curves for various TiAl alloy microstructures; all have high gradients in Stage II. From [Citation13] (reproduced with permission).

Figure 3. Gigacycle fatigue S-N of Ti–45Al–10Nb. Redrawn from [Citation32] (reproduced with permission).

Figure 3. Gigacycle fatigue S-N of Ti–45Al–10Nb. Redrawn from [Citation32] (reproduced with permission).

Figure 4. Mechanisms for FCG in intermetallics, compared with those present in metals and ceramics. Extrinsic toughening in intermetallics may be achieved by grains bridging cracks, such as bridging lamellae in lamellar γ-TiAl alloys. In ceramics and intermetallic composites, crack bridging may additionally involve strong and stiff fibres, or elongated metallic grains that undergo ductile rupture [Citation16]. Ahead of the crack in γ-TiAl intermetallics, crack growth may occur by cleavage fracture along fatigue cycled slip planes and by coalescence of microcracks ahead of the crack tip; static fracture modes also occur. Microcrack/microvoid toughening occurs in the proximity of cracks. Adapted from [Citation16] (reproduced with permission).

Figure 4. Mechanisms for FCG in intermetallics, compared with those present in metals and ceramics. Extrinsic toughening in intermetallics may be achieved by grains bridging cracks, such as bridging lamellae in lamellar γ-TiAl alloys. In ceramics and intermetallic composites, crack bridging may additionally involve strong and stiff fibres, or elongated metallic grains that undergo ductile rupture [Citation16]. Ahead of the crack in γ-TiAl intermetallics, crack growth may occur by cleavage fracture along fatigue cycled slip planes and by coalescence of microcracks ahead of the crack tip; static fracture modes also occur. Microcrack/microvoid toughening occurs in the proximity of cracks. Adapted from [Citation16] (reproduced with permission).

Figure 5. Small crack growth rates in TiAl alloy TNM-B1, with the minimum in FCG rate at the first microstructural barrier arrowed in red. Adapted from [Citation46] (reproduced with permission).

Figure 5. Small crack growth rates in TiAl alloy TNM-B1, with the minimum in FCG rate at the first microstructural barrier arrowed in red. Adapted from [Citation46] (reproduced with permission).

Figure 6. Kitagawa–Takahashi diagram for EBM-processed TiAl alloys. Redrawn and adapted from [Citation47] (reproduced with permission).

Figure 6. Kitagawa–Takahashi diagram for EBM-processed TiAl alloys. Redrawn and adapted from [Citation47] (reproduced with permission).

Figure 7. Effect of temperature on FCG behaviour: (a) from [Citation39] and (b) from [Citation61] (reproduced with permission).

Figure 7. Effect of temperature on FCG behaviour: (a) from [Citation39] and (b) from [Citation61] (reproduced with permission).

Figure 8. Schematic of a cuboidal PST (single colony) lamellar TiAl specimen with the angle Φ of the lamellar planes to the vertical loading axis indicated.

Figure 8. Schematic of a cuboidal PST (single colony) lamellar TiAl specimen with the angle Φ of the lamellar planes to the vertical loading axis indicated.

Figure 9. Effect of the equiaxed γ content on the measured fatigue threshold. From [Citation83] (reproduced with permission).

Figure 9. Effect of the equiaxed γ content on the measured fatigue threshold. From [Citation83] (reproduced with permission).

Figure 10. Stress-life testing of PST crystals. From [Citation70] (reproduced with permission).

Figure 10. Stress-life testing of PST crystals. From [Citation70] (reproduced with permission).

Figure 11. Potential microstructural variety in directionally solidified crystals with the same Φ angle: (a) polysynthetically twinned crystal, PST, (b) columnar grains, approximately co-planar lamellar interfaces, (c) columnar grains, non-coplanar lamellar interfaces, but same Φ angle in every grain. In the present example, . Adapted from [Citation107] (reproduced with permission).

Figure 11. Potential microstructural variety in directionally solidified crystals with the same Φ angle: (a) polysynthetically twinned crystal, PST, (b) columnar grains, approximately co-planar lamellar interfaces, (c) columnar grains, non-coplanar lamellar interfaces, but same Φ angle in every grain. In the present example, . Adapted from [Citation107] (reproduced with permission).

Figure 12. (a) Yield stress of PST specimens as a function of the angle Φ at which the compression axis is inclined to the lamellar planes, from [Citation9], using data from Fujiwara et al. [Citation119] (black circles) and Nomura et al. [Citation120] (white circles). (b) Dependence of the fatigue threshold on the lamellar orientation, Φ, of the colony where the crack is propagating, from [Citation118] (reproduced with permission). Note that both the yield stress and the fatigue threshold present a U-shaped curve against Φ, suggesting that there may be a same mechanistic cause to both.

Figure 12. (a) Yield stress of PST specimens as a function of the angle Φ at which the compression axis is inclined to the lamellar planes, from [Citation9], using data from Fujiwara et al. [Citation119] (black circles) and Nomura et al. [Citation120] (white circles). (b) Dependence of the fatigue threshold on the lamellar orientation, Φ, of the colony where the crack is propagating, from [Citation118] (reproduced with permission). Note that both the yield stress and the fatigue threshold present a U-shaped curve against Φ, suggesting that there may be a same mechanistic cause to both.

Figure 13. SEM images of similar deformation features near lamellar interfaces, interpreted as (a) ledges at lamellar interfaces [Citation123], i.e. interfacial sliding, (b) interlamellar cracking [Citation122], i.e. debonding of the lamellar interface, and (c) longitudinal slip in the γ-TiAl lamellae, near the lamellar interface [Citation124]. Above (b): AFM linescan across the feature reported as interlamellar cracking; above (c): scanning transmission electron microscopy (STEM) images of a slip step near an /γ interface. White arrows indicate the features of interest at the lamellar interfaces; black arrows are illustrative of AFM scan or cross-sectional imaging slice directions (reproduced with permission).

Figure 13. SEM images of similar deformation features near lamellar interfaces, interpreted as (a) ledges at lamellar interfaces [Citation123], i.e. interfacial sliding, (b) interlamellar cracking [Citation122], i.e. debonding of the lamellar interface, and (c) longitudinal slip in the γ-TiAl lamellae, near the lamellar interface [Citation124]. Above (b): AFM linescan across the feature reported as interlamellar cracking; above (c): scanning transmission electron microscopy (STEM) images of a slip step near an /γ interface. White arrows indicate the features of interest at the lamellar interfaces; black arrows are illustrative of AFM scan or cross-sectional imaging slice directions (reproduced with permission).

Figure 14. A selection of possible loading strategies for measuring the FCG rate as a function of , and hence determining the fatigue threshold where  m cycle−1 (growth/no-growth transition). The FCG specimen is loaded at for an extended period (as per ASTM E647 [Citation11]), followed by successive steps , and so on, until an FCG rate of  m cycle−1 is either subceeded or exceeded, depending on whether is being decreased (a, b) or increased (c, d), respectively. Constant and constant versions of each exist. Further, prior cycling at a large stress intensity range for a short period (e–h) may serve to generate a large plastic zone ahead of the crack tip within which the near-threshold crack must then grow. Finally, the large cycling may be applied between each step in , , (i–l) to remove history effects of the previous loading step by forcing the near-threshold crack to grow in the same sized large plastic zone at each step. Extended from [Citation125] (reproduced with permission).

Figure 14. A selection of possible loading strategies for measuring the FCG rate as a function of , and hence determining the fatigue threshold where  m cycle−1 (growth/no-growth transition). The FCG specimen is loaded at for an extended period (as per ASTM E647 [Citation11]), followed by successive steps , and so on, until an FCG rate of  m cycle−1 is either subceeded or exceeded, depending on whether is being decreased (a, b) or increased (c, d), respectively. Constant and constant versions of each exist. Further, prior cycling at a large stress intensity range for a short period (e–h) may serve to generate a large plastic zone ahead of the crack tip within which the near-threshold crack must then grow. Finally, the large cycling may be applied between each step in , , (i–l) to remove history effects of the previous loading step by forcing the near-threshold crack to grow in the same sized large plastic zone at each step. Extended from [Citation125] (reproduced with permission).

Figure 15. High-resolution digital image correlation strain map for a lamellar grain captured from a sample cyclically loaded (ex situ) in compression at a maximum nominal stress of 390 MPa and R=0.05. Axial (vertical here) strain accumulates inside the defined lamellar platelets which are composed by the γ-TiAl phase. From [Citation136] (reproduced with permission).

Figure 15. High-resolution digital image correlation strain map for a lamellar grain captured from a sample cyclically loaded (ex situ) in compression at a maximum nominal stress of 390 MPa and R=0.05. Axial (vertical here) strain accumulates inside the defined lamellar platelets which are composed by the γ-TiAl phase. From [Citation136] (reproduced with permission).

Figure 16. Atom probe tomography reconstruction of the /γ interface in Ti–43Al–4Nb–1Mo–0.1B–0.75C. Note the peak in C, arrowed in red, near the interface; variations in the C content at the interface may change the cohesion strength of the /γ interface. Adapted from [Citation147] (reproduced with permission).

Figure 16. Atom probe tomography reconstruction of the /γ interface in Ti–43Al–4Nb–1Mo–0.1B–0.75C. Note the peak in C, arrowed in red, near the interface; variations in the C content at the interface may change the cohesion strength of the /γ interface. Adapted from [Citation147] (reproduced with permission).

Table 2. Summary of the microstructural properties to optimise for improved HCF behaviour of fully or nearly lamellar γ-TiAl alloys as a function of lamellar orientation, Φ.