# Notch Corrosion Fatigue Behavior of Ti-6Al-4V

^{1}

^{2}

## Abstract

**:**

_{t}values and the role of different inert or corrosive environments. This alloy is widely used in naval-structures and aero-engine communities and the outcomes of the work will have direct relevance to industrial service operations. Axial fatigue tests (R = 0.1; 2 × 10

^{5}cycles; f = 10 Hz) were carried out on smooth and high notched (K

_{tmax}= 18.65) flat specimens in laboratory air, paraffin oil, laboratory air + beeswax coating, recirculated 3.5% NaCl solution. The step loading procedure was used to perform the fatigue tests and the surface replica method and crack propagation gages were used to check crack nucleation and propagation until failure. Log-Log plots of σ

_{max}vs. K

_{t}showed a bilinear behavior and enabled the demonstration of the presence of a threshold stress intensity factor (K

_{t}= 8–9), after which the environment has no effect on the fatigue damage for all the tested environments.

## 1. Introduction

^{9}load cycles. The material mechanical properties were: UTS = 968 MPa and YS = 930 MPa. The tests were carried out at 60 and 20 kHz. After around 200,000 load cycles, the fatigue limit of the Titanium alloy is approximately constant regardless of the test frequency. In [3], Bellows et al. used a step loading procedure to carry out axial fatigue tests at 10

^{7}cycles on smooth cylindrical Ti-6Al-4V samples. The specimens were formed from fan blade forgings with their axis parallel to the longest direction in the forging. The material UTS and YS were 978 and 930 MPa, respectively. R-ratios equal to −1, 0.1, 0.5 and 0.8 were investigated and the test frequency was 60 Hz. Room temperature endurance limits and constant-life Haigh (modified-Goodman) diagrams for smooth Ti-6Al-4V specimens generated by both the step method and conventional method (using S-N curves) were analyzed and statistically compared. The conclusion is that step testing yields results that are within the statistical limits of conventional S-N curve results and therefore is a valid method for generating endurance limits and therefore Haigh diagrams for Ti-6Al-4V specimens.

_{t}= 2.0, 2.7 and 4.1) cylindrical samples were tested under axial fatigue with R-ratios = −1, 0.1, 0.5, 0.65 and 0.8. The samples were all stress relieved after the machining operations. The fatigue limit at 10

^{6}cycles was estimated using the step loading method [3] at a test frequency of 50 Hz. Finite element solutions were generated to provide stress distributions for the notched gage sections. The stress distributions were used in the search for a critical distance over which the quantities of mean stress, stress range, or elastic strain energy may contribute to the fatigue process and can be correlated to similar quantities from smooth, unnotched specimens.

^{6}cycles. The material mechanical properties in the longitudinal direction were: UTS = 978 MPa and YS = 930 MPa. The samples were all stress relieved after the machining operations. Stress concentration factors of 1.97, 2.72, 2.85, 2.86, 4.07 were obtained and R-ratios = −1, 0.1, 0.5, 0.65 and 0.8 were investigated. A step loading procedure with a stress level increment of 5% of the previous step was implemented for generating points on a Haigh diagram. The experimental data were used in combination with finite element solutions for all specimen geometries to determine a “critical distance” parameter, determined from the stress distribution surrounding the notch in combination with fatigue limit stress data from unnotched specimens, that allows fatigue resistance evaluation.

^{5}cycles for R = −1 and 0.1, respectively, are also indicated. The data collected for R= −1 are in good agreement with the literature references mentioned in previous reports [2]. The results also show that after around 100,000 load cycles the fatigue limit of the Titanium alloy is approximately constant regardless of the number of cycles. In Bellows et al. [3], the fatigue limits were also approximately constant with the number of cycles for a loading ratio R = 0.1.

**Figure 2.**S-N data for Ti-6Al-4V alloy plotted in terms of σ

_{max}vs. the number of cycles to fail for various constant R-ratios [11].

_{2}) which tends to passivate the action of the external environment [16]. However, the influence of a mechanical action or abrasive can remove this surface layer, generating a direct interaction between the Titanium alloy and the external environment, leading to the appearance of relevant phenomena of stress corrosion cracking in an aqueous environment [17]. This interaction is also generated by the presence of surface discontinuities, such as cracks, damage and notches, which break the continuity of the passivating layer of oxide, generating in this case also the susceptibility to stress corrosion cracking in water [13,14].

_{t}= 1.18) Ti-6Al-4V specimens immersed in solution at various concentrations of methanol, however, have shown that, in the presence of dynamic loads applied (R = 0.1), the effect of corrosion appears to be significant, even for very high amounts of water in solution [18]. There is also an obvious correlation between the concentration of methanol and the breaking stress. This marked sensitivity of Titanium alloys exposed to mixtures of water and methanol can give rise to safety problems in the aeronautical field, since the injection of such mixtures in the compression stage of the turbine engine is used to retrieve the performance under conditions low density outside air [19]. These problems can be extended to other structural elements made of Titanium alloy, whereas the majority of aircraft components is constantly fatigue loaded, for the heavy and repeated dynamic loads to which aircraft are typically subjected. The few experimental results related to the study of Titanium alloys in a solution of water, NaCl solution and methanol environment, presented in [10,15,16,18], and related to different geometries of the specimens, do not allow to quantify, at the project level, the actual margin of safety for the design of components in Ti-6Al-4V.

_{t}and the role of inert or corrosive environment. The Titanium alloy was tested in several environmental conditions in order to study the fatigue crack initiation/growth mechanisms under: (1) recirculated 3.5% NaCl solution; (2) inert environments like laboratory air, paraffin oil, laboratory air + beeswax coating. The samples were machined with a sharp notch having a stress concentration factor K

_{t}ranging from 1.16 to 18.65. Such tests would allow to decouple the environmental and stress effects and understand the corrosion fatigue mechanism in terms of chemical and mechanical driving forces. More than 50 flat smooth and V-notched specimens (K

_{t}= 1.16–18.65) made of Ti-6Al-4V (ASTM B265-99 [20]; STOA treatment; UTS = 990 MPa; YS = 945 MPa) were tested. The test time for each sample was about 1 week for a total testing time of 70 weeks (very long time experiments).

## 2. Experimental Techniques

#### 2.1. Material and Samples Geometry

^{5}cycles, f = 10 Hz were carried out on Ti-6Al-4V smooth and notched flat specimens. The fatigue tests were carried out on mild notched and notched specimens in different inert or aggressive test environments: laboratory air (relative humidity 30%; T = 22 °C), paraffin oil, laboratory air + beeswax coating and recirculated 3.5% NaCl solution.

**Figure 3.**Sketch of the specimens’ orientation with respect to the parent Ti-6Al-4V plate rolling direction.

_{t}= 1.16–18.65). The values of the numerical K

_{t}for each notch tip radius ρ is reported in Table 1. The notches were obtained by milling at very low cutting speed to limit the residual stresses, then by electrical discharge machining (EDM) to obtain the precise values of the different notch root radii ρ. All the samples were finally stress relieved at 700 °C in a vacuum for 1 h.

Types | Values of the numerical
K_{t} for each notch tip radius ρ | ||||||||
---|---|---|---|---|---|---|---|---|---|

ρ (mm) | 30 | 30 | 2.5 | 1.5 | 0.45 | 0.26 | 0.06 | 0.05 | 0.025 |

K_{t} | 1.16 | 1.18 | 2.55 | 3.10 | 5.17 | 6.63 | 13.34 | 14.34 | 18.65 |

#### 2.2. Experimental Equipment and Procedures

- (1)
- A reference step number of cycles equal to 200,000 was considered;
- (2)
- Only two specimens were needed for each experimental condition. The second specimen was the confirmation one and was tested at the fatigue limit found with the first step loaded specimen;
- (3)
- Reference Haigh diagrams extrapolated from the literature were used to define the initial test loads. Each initial test load was defined so that the applied maximum stress was calculated by lowering by a certain percentage the limiting stress condition from the diagram;
- (4)
- The loading condition of each block was defined by increasing by 5%–10% the maximum load of the previous one;
- (5)
- Every 10,000–20,000 load cycles of each loading block replicas were taken at the notch tips on both the sample faces until the nucleation occurrence has been noticed. After the nucleation the replicas were taken more frequently. The procedure was carried on until the complete failure of the sample;
- (6)

**Figure 5.**(

**a**) Picture of a specimen mounted on the gripping devices; and (

**b**) detail of the closed cell mounted on a specimen during the tests.

**Figure 6.**(

**a**) Sketch of an acetate strip for surface replication placed over the notch root area; and (

**b**) picture of the acetate strip on the specimen.

**Figure 7.**(

**a**) Surface replica; and (

**b**) microscope observation of the sample in the same area (mirror image).

**Figure 8.**(

**a**) Sketch of the electrical circuit including the CPG; and (

**b**) picture showing the circuit assembled with the indication of each component.

**Figure 9.**(

**a**) Micrograph (20×) of the crack propagation gage (about 30 strands are broken); and (

**b**) detail (50×) of the crack propagation gage at the notch root area: the fatigue crack crosses only the first 4 strands of the crack propagation gage. Each strand rupture is related to the increment in crack propagation (with constant distance between the strands).

## 3. Results and Discussion

_{t}. With K

_{t}ranging from 1 to 15, Frost and Dugdale [7,21] found that a critical threshold stress concentration factor, approximately equal to three, is present for mild steel. After such value of K

_{t}the complete fracture line and the non-propagating cracks line bifurcate, giving evidence of a lower effect of the stress concentration factor on the fatigue resistance after such threshold. The critical stress concentration factors in different environments, given by the intersection between the complete fracture lines and the initiation curves, in this papers were estimated for the Titanium alloy.

_{t}diagram at a constant life of 200,000 cycles for the samples tested in air, Paraffin oil and 3.5% NaCl solution are shown respectively.

_{max}= 627K

_{t}

^{−0.964}MPa; 1 ≤ K

_{t}≤ 8 ailure

_{max}= 84 MPa; K

_{t}> 8 Failure

_{max}= 670K

_{t}

^{−0.930}MPa; 1 ≤ K

_{t}≤8 Failure, Lanning et al. [22]

_{max}= 592K

_{t}

^{−0.938}MPa Nucleation

_{max}= 544K

_{t}

^{−0.801}MPa; 1 ≤ K

_{t}≤ 9 Failure

_{max}= 94 MPa; K

_{t}> 9 Failure

_{max}= 529K

_{t}

^{−0.84}MPa Nucleation

_{max}= 443 × K

_{t}

^{−0.832}MPa; 1 ≤ K

_{t}≤ 9 Failure

_{max}= 76 MPa; K

_{t}> 9 Failure

_{max}= 433 × K

_{t}

^{−0.833}MPa; K

_{t}≤ 9 Nucleation

_{t}= 1 (Sadananda et al. [11], Peters et al. [23]) as shown in Figure 11 and Figure 12.

**Figure 11.**Log-limiting maximum stress (R = 0.1, 10 Hz) at initiation and failure vs. Log-K

_{t}at a constant life of 200,000 cycles for the samples tested in air.

**Figure 12.**Log-limiting maximum stress (R = 0.1, 10 Hz) at initiation and failure vs. Log-K

_{t}at a constant life of 200,000 cycles for the samples tested in Paraffin oil.

**Figure 13.**Log-limiting maximum stress (R = 0.1, 10 Hz) at initiation and failure vs. Log-K

_{t}at a constant life of 200,000 cycles for the samples tested in NaCl 3.5 wt%.

_{t}= 8–9 after which σ

_{max}remains constant, regardless of K

_{t}. After the threshold K

_{t}there is a region of non-propagating cracks between crack initiation and crack failure, in accordance with the literature for other materials at lower K

_{t}[24]. Referring to [7], in which Frost and Dugdale fatigue tested flat mild steel smooth and notched specimens, a relatively low threshold K

_{t}equal to three was found. This Titanium alloy shows a different behavior both in inert, air and in corrosive environments. This means that, regardless of K

_{t}, if K

_{t}is higher that 8–9, σ

_{max}remains constant for a notched component and the environment does not have a great influence on the fatigue resistance. This seems to be a relative important result in case of the aeronautical Ti-6Al-4V components that work in aggressive environments like a 3.5% NaCl solution. The non-propagating cracks curve and the almost triangular area after bifurcation of the curve after the K

_{t}= 3, clearly visible in Figure 10 and obtained for a mild steel, is equal to 8–9 in Figure 11, Figure 12 and Figure 13 for Ti-6Al-4V. Furthermore, crack nucleation and complete failure curves are very close for Ti-6Al-4V. The Titanium alloy exerts a much higher crack propagation rate [24] with respect to mild steel, for all the K

_{t}and test environment considered in the present paper. This means that after nucleation the crack rapidly propagates until complete fracture of the specimens. Moreover, crack growth rates in 3.5% NaCl solution proved to be higher than they were in air at all the K

_{t}investigated [24].

_{t}.

_{t}in 3.5 wt% NaCl solution (Figure 14). Between the two points (1 and 2) of the fracture surfaces there is no difference of morphology. As K

_{t}increases the fracture surface becomes smoother and less corrugated with no dimples. The explanation is that lower K

_{t}means higher applied loads in order to reach the stress for crack nucleation at the notch tip; on the other hand, because of the steep stress gradient, higher K

_{t}means lower applied loads in order to reach the stress for crack nucleation at the notch tip.

_{t}is surely strongly dependent on a lot of parameters among which the most important might be cyclic frequency, microstructure and the notch depth. For very shallow notches the threshold K

_{t}shifts towards lower values of K

_{t}and the non-propagating domain starts at very low values of K

_{t}. After increasing the notch value beyond a critical value the non-propagating domain disappears completely. With shallow notches the plastic zone at the root of the notch is much larger with respect to deep notches. The plastic enclave that arises at the notch root of shallow notches would embed the nucleated cracks in this plastic shelter stopping their propagation. With deep notches nucleated cracks fast overcome the small plastic zone and become macro cracks [21]. The effect of the stress gradient is also shown in Figure 15.

_{t}and high K

_{t}need the same stress σ* for crack nucleation but different applied loads to have such stress at the notch tip. Furthermore, low K

_{t}means mild stress gradient and a high applied load needed to reach the nucleation stress σ* at the notch tip; on the other hand high K

_{t}means steep stress gradient and a low applied load needed to reach the same nucleation stress σ* at the notch tip. Moreover, the steep stress gradient at high K

_{t}reduces the mean stress at the notch tip (Figure 15). The aggressive environment is another driving force for crack nucleation and, together with the stress gradient and maximum stress, has a big influence on the nucleation process of cracks. The driving force for crack propagation is the applied load: thus the applied stress far from the notch tip (crack tip after nucleation). Such stress is much higher in case of low K

_{t}, thus reducing the fatigue resistance with respect to higher K

_{t}specimens. An example of the different stress gradients at low and high K

_{t}, having the same maximum stress σ* at the notch tip, is shown in Figure 15.

_{t}because of the higher crack growth rate in NaCl solution and of the higher applied loads, needed to reach the nucleation stress at the notch tip, together with the stress gradient effect (higher mean loads at the tip of the notch as shown in Figure 15).

## 4. Conclusions

- (1)
- Tests in lab air and paraffin oil gave approximately the same maximum stresses at nucleation and complete fracture, while a 20% reduction occurred for the tests in 3.5% NaCl solution, for all the values of K
_{t}. - (2)
- The microstructure consisted in primary alpha and beta-transformed grains due to the performed heat treatment. The tensile results are compatible with such microstructure;
- (3)
- The fatigue limit at 200.000 load cycles was strongly affected by the experimented notches, even if the fatigue life reached a steady state after K
_{t}= 8–9. This means that if K_{t}is higher than 8–9, σ_{max}does not vary for a notched component and the environment does not influence its fatigue resistance. This seems to be a relative important results in case of the aeronautical Ti-6Al-4V components that work in aggressive 3.5% NaCl solution environments. - (4)
- From the analysis of the fracture surfaces, no significant difference was observed for air or NaCl solution tested samples. This is probably related with the short propagation time required by the fatigue cracks, not enough to make a general corrosion process start.

## Acknowledgements

## Nomenclature

K_{t} | stress concentration factor |

R | load ratio |

N_{f} | number of cycles to failure in the final loading block |

σ_{FL, max} | interpolated fatigue limit (maximum stress over the fatigue cycle) at 2 × 10 ^{5} load cycles (MPa) |

σ_{prior, max} | maximum applied stress on minimum cross section in the loading block prior to that of failure (MPa) |

σ_{final, max} | maximum applied stress on minimum cross section in the failure block (MPa) |

σ_{limit, max} | limiting maximum fatigue stress at failure on minimum cross section (MPa) |

E | Modulus of elasticity (MPa) |

T | Temperature (°C) |

YS | Yield strength (MPa) |

UTS | Ultimate tensile strength (MPa) |

## Conflicts of Interest

## References

- Leyens, C.; Peters, M. Titanium and Titanium Alloys: Fundamentals and Applications; Wiley-VCH: Weinheim, Germany, 2005. [Google Scholar]
- Morrissey, R.J.; Nicholas, T. Fatigue strength of Ti-6Al-4V at very long lives. Int. J. Fatigue
**2005**, 27, 1608–1612. [Google Scholar] [CrossRef] - Bellows, R.S.; Muju, S.; Nicholas, T. Validation of the step test method for generating Haigh diagrams for Ti-6Al-4V. Int. J. Fatigue
**1999**, 21, 687–697. [Google Scholar] [CrossRef] - Lanning, D.B.; Nicholas, T.; Haritos, G.K. On the use of critical distance for the prediction of the high cycle fatigue limit stress in notched Ti-6Al-4V. Int. J. Fatigue
**2005**, 27, 45–57. [Google Scholar] [CrossRef] - Lanning, D.B.; Nicholas, T.; Palazotto, A. The effect of notch geometry on critical distance high cycle fatigue predictions. Int. J. Fatigue
**2005**, 27, 1623–1627. [Google Scholar] [CrossRef] - Haritos, G.K.; Nicholas, T.; Lanning, D.B. Notch size effects in HCF behavior of Ti-6Al-4V. Int. J. Fatigue
**1999**, 21, 643–652. [Google Scholar] - Frost, N.E.; Marsh, K.J.; Pook, L.P. Metal Fatigue; Clerendon Press: Oxford, UK, 1974. [Google Scholar]
- Morrissey, R.J.; McDowell, D.L.; Nicholas, T. Frequency and stress ratio effects in high cycle fatigue of Ti-6Al-4V. Int. J. Fatigue
**1999**, 21, 679–685. [Google Scholar] [CrossRef] - Lu, H.; Gao, K.W.; Qiao, L.J.; Wang, Y.B.; Chu, W.Y. Stress Corrosion Cracking Caused by Passive Film-Induced Tensile Stress. In Proceedings of the CORROSION/2000, Orlando, FL, 26–31 March 2000; pp. 1112–1118.
- Baragetti, S.; Foglia, C.; Gerosa, R. Fatigue crack nucleation and growth mechanisms for Ti-6Al-4V in different environments. Key Eng. Mater.
**2013**, 525–526, 505–508. [Google Scholar] [CrossRef] - Sadananda, K.; Sarkar, S.; Kujawskj, D.; Vasudevan, A.K. A two-parameter analysis of S-N fatigue life using Δσ and σ
_{max}. Int. J. Fatigue**2009**, 31, 1648–1659. [Google Scholar] - Aladjem, A. Anodic oxidation of titanium and its alloys. J. Mater. Sci.
**1973**, 8, 688–704. [Google Scholar] [CrossRef] - Pilchak, A.L.; Young, A.H.; Williams, J.C. Stress corrosion cracking facet crystallography of Ti-8Al-1Mo-1V. Corros. Sci.
**2010**, 52, 3287–3296. [Google Scholar] [CrossRef] - Brown, B.F. Stress Corrosion Cracking in High Strength Steels and in Titanium and Aluminum Alloys; Naval Research Lab: Washington, DC, 1972. [Google Scholar]
- Johnston, R.L.; Johnson, R.E.; Ecord, G.M.; Castner, W.L. Analysis of Failures Occurred in Pressurized Aerospace Methanol Tanks. NASA Technical Note, D-3868; NASA: Washington, DC, USA, 1967. [Google Scholar]
- Chen, C.M.; Kirkpatrick, H.B.; Gegel, H.L. Stress Corrosion Cracking of Titanium Alloys in Methanolic and Other Media; Air Force Materials Laboratory, Wright-Patterson Air Force Base: Dayton, OH, USA, 1972. [Google Scholar]
- Trasatti, S.P.; Sivieri, E. Corrosion behavior of titanium in non-aqueous solvents. Mater. Chem. Phys.
**2005**, 92, 475–479. [Google Scholar] [CrossRef] - Baragetti, S. Corrosion fatigue behavior of Ti-6Al-4V in methanol environment. Surf. Interface Anal.
**2013**, 45, 1654–1658. [Google Scholar] - Soares, C. Gas Turbines: A Handbook of Air, Land and Sea Applications; Elsevier: Amsterdam, the Netherland, 2008. [Google Scholar]
- ASTM B 265-99. Standard Specification for Titanium and Titanium Alloy Strip, Sheet, and Plate; ASTM (American Society for Testing and Materials): West Conshohocken, PA, USA, 2013.
- Milella, P.P. Fatigue and Corrosion in Metals; Springer: Berlin, Germany, 2013. [Google Scholar]
- Lanning, D.B.; Nicholas, T.; Palazotto, A. HCF notch predictions based on weakest-link failure models. Int. J. Fatigue
**2003**, 25, 835–841. [Google Scholar] - Peters, J.O.; Boyce, B.L.; Chen, X.; Mcnaney, J.M.; Hutchinson, J.W.; Ritchie, R.O. On the application of the Kitagawa–Takahashi diagram to foreign-object damage and high-cycle fatigue. Key Eng. Mater.
**2002**, 69, 1425–1446. [Google Scholar] - Baragetti, S.; Cavalleri, S.; Tordini, F. Fatigue behavior of notched Ti-6Al-4V in air and corrosive environment. Procedia Eng.
**2011**, 10, 2435–2440. [Google Scholar] [CrossRef] - Sposito, G. Advances in Potential Drop Techniques for Non-Destructive Testing. Ph.D. Thesis, Imperial College, London, UK, 2009. [Google Scholar]
- Baragetti, S.; Gerosa, R. Notch effect and fatigue performance of Ti-6Al-4V sheets in saline environment. Mater. Perform. Charact.
**2014**, 3, 1–9. [Google Scholar]

© 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).

## Share and Cite

**MDPI and ACS Style**

Baragetti, S.
Notch Corrosion Fatigue Behavior of Ti-6Al-4V. *Materials* **2014**, *7*, 4349-4366.
https://doi.org/10.3390/ma7064349

**AMA Style**

Baragetti S.
Notch Corrosion Fatigue Behavior of Ti-6Al-4V. *Materials*. 2014; 7(6):4349-4366.
https://doi.org/10.3390/ma7064349

**Chicago/Turabian Style**

Baragetti, Sergio.
2014. "Notch Corrosion Fatigue Behavior of Ti-6Al-4V" *Materials* 7, no. 6: 4349-4366.
https://doi.org/10.3390/ma7064349