¡¡
A Study of the Role of Grain-Boundary Engineering
in Promoting High-Cycle Fatigue Resistance

Fatigue Failure:
Failure analysis and prevention are important to all of the engineering disciplines, especially for the aerospace industry. Aircraft accidents are remembered by the public because of the unusually high loss of life and broad extent of damage.

High-cycle fatigue (HCF), proved as the severest cause of turbine engine failures in aircraft, has resulted in the loss of aircraft and human life. Due to the high vibration frequencies, microcracks can initiate from defects, which are often associated with microstructural damage caused by fretting or foreign object impact, subsequently propagate to failure in short time periods, possibly within a single flight segment. Therefore, HCF-critical materials in engine components must be used below the fatigue-crack initiation or growth threshold such that crack propagation rate is lower than 109 m/cycle. There is an extensive database on the high-cycle fatigue threshold behavior of the materials used in turbine engines, however, the question as to whether microstructures of these materials can be optimized to promote HCF resistance has not been addressed.

Microstructurally designed materials are playing an increasingly important role in material processing and manufacturing because such materials are important not only as components with superior mechanical performance but also in hush-environment applications. For example, Ni- or Ti-base alloys are extensively applied in aviation and power generation because they have reliable and excellent mechanical properties and chemical inertness even at high temperatures. Material properties are invariably related to their microstructure. A thorough understanding of these structure-property relations gains particular importance in most of materials processing, wherein the desired properties are determined by the level of microstructure control.

We aim to understand and optimize the microstructural dependence of both fracture toughness and high cycle fatigue life for Ni- and Ti base superalloys used in aircraft.

Fatigue Crack Propagation Bahavior:

Large-crack Propagation Behavior:

(a) Ambient Temperatures

The variation in the fatigue-crack propagation rates, da/dN, of large (~8-20 mm), through-thickness cracks in the as-received, grain-boundary engineered and grain-coarsened microstructures in ME3 at 25º§³ (with R = 0.1) are shown in Figure 1 as a function of the stress-intensity range. The most striking feature of these results is the marked influence of grain size; values of the ¦¤KTH fatigue thresholds, increase linearly from 5.6 MPa/m for a 1.3 µm grain size (as-received fine-grain alloy) to 11.5 MPa/m for a 17 µm grain size (grain-coarsened alloy). However, when compared at comparable grain sizes (~13-15 µm), it is clear that there is little independent effect of grain-boundary engineering on fatigue-crack propagation and threshold behavior at ambient temperatures. ¦¤KTH thresholds are comparable (within 5%) for the as-received coarse-grain (with fN ~ 0.28) and grain-boundary engineered (with fN ~ 0.42) structures, despite a ~50% increase in the special grain-boundary fraction; furthermore, they are ~17% higher in the grain-coarsened microstructure (where fN ~ 0.30). The absence of an effect of grain-boundary engineering at 25º§³ is consistent with the observed mechanisms of crack extension at this temperature.


Figure 1.  Fatigue-crack propagation rates, da/dN, at 25º§³ for large (~8-20 mm) cracks in the nickel-base superalloy ME3, as a function of the stress-intensity range, ¦¤K, for the as-received, grain-boundary engineered and grain-coarsened microstructures. Note the large effect of grain size on near-threshold behavior.

SEM fractography of the fatigue surfaces in the as-received and grain-boundary engineered microstructures, shown in Figure 2(a,b) at near-threshold levels and in Figure 2(c,d) at higher growth-rate behavior, reveal a predominantly transgranular cracking mode. Typical of many superalloys with low stacking-fault energy, such transgranular crack growth was highly planar and often crystallographic in character, with the facets on the fracture surfaces consistent with slip along {111} planes.

Figure 2.  SEM fractography of ambient temperature fatigue-crack propagation in ME3 at near-threshold stress intensities, for (a,c) as-received and (b,d) grain-boundary engineered microstructures.

(b) Elevated Temperatures

To evaluate the effect of grain-boundary engineering at elevated temperatures, which are more representative of the practical applications for this alloy, large-crack fatigue-crack growth-rate data for the as-received and the grain-boundary engineered microstructures of similar average grain sizes (13-15 µm) were compared at 700 and 800º§³ with growth-rate data at 25º§³. Results, shown in Figure 3, indicate that crack-growth rates in both microstructures are typically faster by one to two orders of magnitude at 700-800º§³ than at ambient temperature. More importantly, although there is no difference in behavior above ~10-7 m/cycle, there is a definitive, albeit small, increase in crack-growth resistance in the grain-boundary engineered microstructures at near-threshold levels. Specifically, with the increase in fraction of special boundaries, near-threshold growth rates (at a specific ¦¤K) are some 5 to 10 times lower, and ¦¤KTH thresholds are ~10% higher at 700º§³ and over 20% higher at 800º§³.


(a)
  
(b)
Figure 3.  Fatigue-crack propagation behavior at elevated temperatures for large (~8-20 mm) cracks in ME3, as a function of the stress-intensity range, ¦¤K, for in the as-received and grain-boundary engineered microstructures. Shown are results at (a) 700º§³ and (b) 800º§³, as compared to behavior at 25º§³.

This beneficial effect of grain-boundary engineering at elevated temperatures can also be traced to the mechanisms of cyclic crack extension. Quantitatively, the proportion of intergranular fracture could be best determined from area fractions on lower magnification SEM fractographs. As shown in Figure 4, near-threshold fatigue-crack growth in ME3 is characterized by an increasing proportion of intergranular cracking with increase in temperature. Measurements on the as-received material at a  ¦¤K ~ 10 MPa/m revealed an area fraction of intergranular facets of ~40% at 700º§³ and as high as ~75% at 800º§³. Most importantly, however, after increasing the fraction of special boundaries by grain-boundary engineering, there was a definite reduction in the relative proportion of intergranular crack growth, specifically by some 20 to 25%.

Figure 4.  SEM fractography of near-threshold fatigue-crack growth at elevated temperatures in ME3, showing a comparison between (a,c) as-received and (b,d) grain-boundary engineered microstructures (a,b) at 700º§³ and (c,d) at 800º§³. Note the lower proportion of intergranular fracture in the grain-boundary engineered microstructures.