Fracture toughness and high-cycle fatigue behaviour of Al–Mg and Al–Mg–Sc

Abstract

Tensile, fracture toughness and high-cycle fatigue tests were carried out on Al–Mg and Al–Mg–Sc–Zr alloys. Fracture morphology and the microstructure of the alloys were observed by SEM and TEM, etc. Al–Mg–Sc–Zr alloy with fibrous fine grain structure showed higher tensile strength and fracture toughness than Al–Mg alloy, which had a larger grain size and obvious recrystallisation. In high-cycle fatigue tests, Al–Mg–Sc–Zr alloy presented a higher fatigue limit owing to its finer grains and nanosized precipitates, which increased the difficulty of fatigue crack initiation. Al–Mg alloy presented a lower fatigue limit because crack initiation caused by slip is more likely to occur. However, its rougher fracture surfaces may be beneficial to hindering the propagation of macrocracks.

Keywords

Al–Mg Al–Mg–Sc–Zr fatigue fracture morphology

Introduction

Al–Mg alloys have a good combination of strength and ductility, high fatigue resistance, good weldability, and excellent corrosion resistance [1,2]. Hence, they are widely used as lightweight structural materials in the automotive, marine, aerospace, and construction fields. The nanoscale dispersive Al₆Mn phase and the coherent Al₃Sc precipitates are very effective for pinning dislocation and can effectively refine the grains [3]. By adding a certain amount of Sc to the Al–Mg alloy during solid-solution treatment, fine coherent L1₂–Al₃Sc particles are uniformly precipitated in the supersaturated solid solution following inhomogeneous nucleation, which prevents the nucleation of precipitates at the grain boundaries. Al₃Sc precipitation can effectively fix grain and sub-grain boundaries with the dislocation motion hindered by the sub-grain boundary [4]. The tensile strength, yield strength, microhardness and thermal stability of Al–Mg–Sc alloy are significantly increased compared with Al–Mg alloy [2,5], while maintaining the same corrosion resistance as Al–Mg alloy [2], albeit the ductility is reduced [6]. In combination with Zr and Ti, the Sc doping concentration can be effectively reduced to achieve the same improved mechanical property, while adding Zr can further enhance the thermal stability of the alloy [7].

The yield strength of Al–Mg–Sc alloy is controlled by the synergistic effects of solution strengthening and precipitation strengthening. Marquis et al. found that the Orowan stress was increased with the increase of Al₃Sc precipitation particle size [8]. Meanwhile, Kim et al. observed increased Al–Zn–Mg–(Sc) alloy's tensile strength with increased Sc content [9]. Al₃Sc particles may redissolve during plastic deformation. This local dissolution can lead to cyclic softening of Al–Mg–Sc alloys with reduced particle size approaching the shear critical size for tensile deformation. Song et al. revealed that the dissolution of small Al₃(Er, Zr) particles might be caused by the diffusion of dislocation atoms along the dislocation line during dislocation shear and fatigue deformation under cyclic loading [10]. Fatigue tests on two Al–Mg–Sc alloys with different sizes of Al₃Sc precipitation revealed that specimens with smaller Al₃Sc precipitation showed cyclic softening after initial hardening, while specimens with larger Al₃Sc precipitation did not show cyclic softening. By studying the effects of Al₃Sc particle size and precipitate-free zones (PFZs) on fatigue behaviour, Watanabe et al. recognised that the fatigue softening of alloy with smaller precipitation is related to the partial redissolution of the Al₃Sc particles in the slip zone rather than owing to surface cracking. They also found that double-stage aging reduced the PFZ width in the large-grained specimens, improving the fatigue life [11]. For the 6082 aluminium alloy, adding the SiCp reﬁned the grain size, and the squeeze casting helped reduce the pore defects. Hence, small grain size and evenly distributed SiC_p were achieved with improved bonding with the Al matrix. The tensile strength, yield strength, elongation, elastic modulus and hardness were remarkably improved. Thus, adding nano-sized SiC particles has refined the microstructure with improved the mechanical properties of 6082 alloy [12,13].

For aluminium alloys with ultrafine grain, the grain size is comparable to or smaller than the typical dislocation substructure size, where the interaction of dislocations with boundaries is very important. Alloys with smaller grains have better high-cycle fatigue performance, while alloy with larger grains has better low-cycle fatigue performance. Shukla et al. studied the fatigue properties of two ultrafine grain 5024 aluminium alloys [14], and it was found that the increase in the dislocation density within the grains resulted in better cyclic stability. Meanwhile, Majchrowicz et al. also found that the high-cycle fatigue performance of the ultrafine grain 5483 alloy was significantly improved compared to the coarse grain 5483 alloy [15]. The coarse-grained 5483 alloy exhibited cyclic hardening with the increased dislocation density and a higher fatigue-limit to yield-strength ratio. Alexopoulos et al. discovered that the fatigue limit of the 2024-T3 alloy is about 40% lower than its yield stress [16]. This is owing to the combined effect of hardening and microcracking. In contrast, they found that the fatigue limit of the forged 2198-T351 alloy is only 9% lower than the yield stress, and there is no significant hardening near the yield stress.

In addition to the properties of the material itself, fatigue life can also be affected by other factors, such as sample size and shape, stress ratio, temperature, load history, load sequence effects, loading type, processing variables, environmental effects, etc. Boni et al. studied the stir friction welded 2195-T8 alloy specimens. It showed that the fatigue life of large specimens whose width is more than 13 times that of small specimens was about 40% of that of small specimens [17]. Meanwhile, the fatigue behaviour of MIG welded Al–Mg alloy was studied by Gaur et al. Their results showed that the fatigue limit decreased with the increase of the stress ratio, R. The damage is mainly initiated from the surface under a low-stress ratio, changing to defect-driven under a high-stress ratio. This change in the crack initiation site was attributed to the local cyclic plasticity caused by stress concentration [18]. Majchrowicz et al. studied the temperature effects on fatigue behaviour. They found that as the temperature increases, the Wöhler plot (S–N curve) of the superfine 5483 alloy shifts to a lower stress value. This is owing to increased surface infiltration and extrusion at high temperatures, which decreases the resistance to fatigue crack initiation [15].

In engineering applications, alternating applied loads can cause materials to suffer from fatigue problems. When the strain amplitude is mainly in the elastic range, and the stress is less than the yield limit, long fatigue life is expected with a failure cycle number of 10⁵ or more, generally called high-cycle fatigue. For notched components, their fatigue properties are also influenced by notch size [19], notch geometries and material microstructure [20]. The stress/strain concentration phenomena can dramatically reduce the fatigue life owing to the so-called ‘notch effect’ if the components contain geometric discontinuities. For ultra-high-cycle fatigue tests with a cycle number greater than 10⁷, the initiation stage accounts for most of the fatigue life [21]. Even though there are no universal criteria to define the transition between initiation and propagation. Branco et al. used an averaged total strain energy density approach based on the theory of critical distances to predict fatigue crack initiation lifetime [22]. Luo et al. presented a method based on the crack length to distinguish the crack initiation and propagation stages. The crack length, which does not affect the fatigue limit of the material, is regarded as the crack initiation stage [23].

Fracture toughness is an important material property used to assess the integrity of engineering components containing cracks [24]. Studies have shown that the plane stress fracture toughness of the ultra-ﬁne grained aluminium specimens produced through an accumulative roll bonding (ARB) process is 155% higher than that of annealed specimens after the seventh ARB cycle, accompanied by the improved tensile strength and yield strength [25]. Rahmatabadi et al. evaluated the fracture behaviour and rupture energy absorption of the Mg–Li alloys. Owing to the dependence of fracture toughness on the strength and ductility, adding lithium increased the fracture toughness compared to AZ alloys. This is owing to the compensation of the reduced strength by the significant increase in ductility [26]. For Al–Si alloys, the decreased dendrite cell size and the refinement of eutectic and intermetallic Si species benefit the fracture toughness, while the increase of the yield strength and the decrease of the plasticity caused by the refinement of primary cells and eutectics are not conducive to fracture toughness [27]. The development of high-strength weldable Al–Mg–Sc–Zr alloy is still at the transition stage to batch production. Hence, the basic performance data are not completed. For aerospace applications, the material selection and design lack guidelines. It is necessary to test and study the fatigue and fracture properties and behaviours of the Al–Mg–Sc–Zr alloys compared to the traditional Al–Mg alloy. In this paper, fracture toughness and high-cycle fatigue tests were performed on Al–Mg and Al–Mg–Sc–Zr alloys to investigate their mechanical properties and fracture characteristics under the corresponding stress conditions.

Materials and methods

The cold-rolled Al–Mg and Al–Mg–Sc–Zr alloy sheets were annealed at 320°C for 1 h. The chemical composition of both alloys is listed in Table 1. Both alloys were processed into specific specimens for tensile, fracture toughness and high-cycle fatigue tests, respectively. The setup and the instruments for testing are shown in Figure 1. The mechanical tests were carried out at room temperature and in atmospheric environments. In the tensile tests, the load rate was 0.3 kN s⁻¹. In the plane stress fracture toughness tests, thin-sheet C(T) specimens in the L–T and T–L orientations under the high constraint conditions at the crack tip were tested under uniaxial load. After fatigue pre-cracking, stress was uniformly loaded until fracture. The corresponding rectangular section specimens were used in tensile and high-cycle fatigue tests with the tensile direction parallel to the rolling direction. High-cycle fatigue tests used standard smooth specimens in the L direction and K _t = 1. Tests were performed using sinusoidal cyclic tensile load with a stress ratio of R = 0.1 at different stress levels until the specimen fractured or the cycle number reached 10⁷. The loads applied on the Al–Mg–Sc–Zr alloy are between 230 and 365 MPa, while on the Al–Mg alloy are between 160 and 300 MPa.

Table 1

Main chemical composition of alloy sheets (wt-%).

C	Sc	Zr	Mn	Mg	Al
Al–Mg	–	–	0.4	5.8–6.0	Bal
Al–Mg—Sc–Zr	0.25	0.1	0.4	5.8–6.0	Bal

After the fracture toughness and high-cycle fatigue tests, the morphology of the fracture surfaces of the corresponding specimens was studied by a field emission scanning electron microscope (SEM, SIRION200). For fracture toughness measurements, the pre-cracking zone and the final fracture zone were mainly observed; for high-cycle fatigue specimens, the fatigue crack initiation zone, crack propagation zone and final fracture zone were mainly observed.

A transmission electron microscope (TEM, TecnaiG220) was used to observe the microstructure of the two alloys and the microstructure near the fracture surface of the Al–Mg–Sc–Zr alloy after the HCF test. The fracture surface was intercepted by wire cutting and processed into specimens with a size of 10 × 10 × 0.5 mm². The sample was polished with sandpaper to a thickness of 100 μm and then processed into circular pieces with a diameter of 3 mm. TEM specimens were further thinned and perforated by double spray and were cleaned with distilled water and alcohol immediately after the perforation.

Results

Mechanical performance

Tensile tests were conducted on Al–Mg and Al–Mg–Sc–Zr alloy plates along the rolling direction. Results are shown in Figure 2. The tensile strength and yield strength of Al–Mg alloy were 361 and 172 MPa, respectively, which are lower than those of Al–Mg–Sc–Zr alloy with the corresponding values at 422 and 290 MPa, respectively. Hence, Al–Mg–Sc–Zr alloy has significantly higher tensile and yield strengths.

The fracture toughnesses of Al–Mg and Al–Mg–Sc–Zr alloy plates along the L–T and T–L orientations were also tested. Results are shown in Figure 3. For Al–Mg alloy, the fracture toughness of the T–L specimen is about 105 MPa m^1/2, and that of the L–T specimen is about 110 MPa m^1/2. For Al–Mg–Sc–Zr alloy, the fracture toughness of the T–L specimen is about 115 MPa m^1/2, and that of the L–T specimen is about 120 MPa m^1/2. Both alloys present anisotropy behaviour, with the L–T specimen having higher fracture toughness. Also, Al–Mg–Sc–Zr alloy has higher fracture toughness than Al–Mg alloy.

High-cycle fatigue tests were carried out on Al–Mg and Al–Mg–Sc–Zr alloy plates along the rolling direction. Results are shown in Figure 4. The fatigue limit of Al–Mg alloy is about 160 MPa, while Al–Mg–Sc–Zr alloy offers a much higher fatigue limit of 260 MPa. With the increase in the stress level, the corresponding number of cycles of fatigue fracture from the two alloys decreased. When the numbers of fatigue fracture cycles are similar, and both were less than 10⁵, the corresponding stress level in Al–Mg–Sc–Zr alloy is about 10 MPa to 50 MPa higher than the Al–Mg alloy. At high-stress levels, Al–Mg–Sc–Zr alloy offered a relatively small advantage in fatigue resistance, although it is always superior to Al–Mg alloy. When the stress level is about 300 MPa, the N _f of Al–Mg alloy is about 2.9 × 10⁴. For Al–Mg–Sc–Zr alloy, N _f is between 3.88 × 10⁴ and 5.29 × 10⁴. When the stress level is about 280 MPa, the N _f of Al–Mg alloy is 4.95 × 10⁴, while Al–Mg–Sc–Zr alloy is between 8.35 × 10⁴ and 1 × 10⁷. When N _f is about 1 × 10⁶, the corresponding stress level of Al–Mg alloy is about 170 MPa, and Al–Mg–Sc–Zr alloy is about 270 MPa.

Fracture morphology of fracture toughness specimens

The fracture toughness tests were performed on the Al–Mg alloy and Al–Mg–Sc–Zr alloy specimens, and the specimens were photographed after the tests, as shown in Figure 5. Both alloy plates showed oblique fractures, regardless of the orientation of the specimens. There are three fracture directions for Al–Mg alloy specimens: fracture along the notch direction, fracture at 45° to the notch direction, and fracture at 90° to the notch direction. However, Al–Mg–Sc–Zr alloy specimens have two fracture directions: the notch direction and 45° to the notch direction. When the specimens were fractured along the notch direction, the end of the fracture area showed a flat oblique section, while the specimen was fractured at 45° or 90° from the notch direction. The end of the fracture area presented a ridge-like fracture feature.

The morphological characteristics of the pre-cracked and final fracture zones of the Al–Mg–Sc–Zr alloy were observed by SEM. The images in Figure 6 were recorded at different magnifications indicated by the scale bars. Figure 6(a) shows a clear boundary between the pre-cracking and final fracture areas, with obvious differences in morphology between the two areas. In the pre-cracking stage, the fracture morphology is relatively smooth and flat. An obvious lamellar structure with typical fatigue stripes was observed in Figure 6(b) and (c). The final fracture areas, shown in the SEM images in Figure 6(a) and (d), are rough and uneven, with features of dimples, presenting morphology similar to the ductile fracture of static stretching. The dimples are randomly distributed with large and small dimples connected. In some cases, crushed coarse particles can be observed in the middle of the dimples. The height of the fracture surface varies with severe plastic deformation.

Fracture surfaces of HCF specimens

The fracture surfaces of Al–Mg and Al–Mg–Sc–Zr alloy specimens after the HCF test were observed by SEM, shown in Figures 7 and 8. The fracture surfaces of both alloys showed typical fatigue fracture morphology. The fatigue crack initiation, crack propagation, and final fracture zones were observed. The fatigue cracks sprouted from the surface of the specimens, and the macroscopic morphology of the propagation zone shows a fan-shaped dispersion from the crack initiation. In the propagation zone, the fatigue stripes can be observed perpendicular to the crack propagation direction, highly irregular crack surfaces, broken particles, and radiating cracking lines along the crack-extension direction. In the final fracture zone, typical dimple morphology is observed. The fracture surface of Al–Mg alloy is relatively rough with severe plastic deformation, which is conducive to hindering macroscopic fatigue cracking, while the fracture surface of Al–Mg–Sc–Zr alloy is relatively smooth, which is easier to develop macroscopic fatigue cracking.

TEM test results

The microstructures and the precipitated phases of Al–Mg and Al–Mg–Sc–Zr alloys before the HCF test are shown in the TEM images in Figure 9. Large grain size, flat and straight grain boundaries, and a few dislocations of the Al–Mg alloy were observed in Figure 9(a). Meanwhile, small grain size, zigzag grain boundaries, multiple sub-grains, and high density of dislocations with some precipitates were identified from the Al–Mg–Sc–Zr alloy, shown in Figure 9(b). Beam petal-like Al₃(Sc, Zr) precipitates with the size of 30–40 nm could be observed in the grains. In addition, the TEM images show that recrystallisation has occurred in the Al–Mg alloy but not in the Al–Mg–Sc–Zr alloy.

Discussion

Fracture toughness

In the fracture toughness test, with continuously applied loading on the specimen, the material gradually changes from elastic to permanent plastic deformation at the initial stage. The increase in the size of the plastic deformation zone at the crack tip increases the resistance to crack propagation. The crack-extension forces are balanced until the point of destabilisation is reached. The resistance to plastic deformation is similar to strain hardening until the specimen is fractured. Fracture toughness is normally higher if the matrix is more ductile, plastic deformation is more uniform, and there is no local shear zone.

The behaviour of a metallic material during a fracture toughness test can be described by the fracture behaviour, the strength and deformation behaviour, and the constraint effects of the geometry [28]. Crack depth, section thickness, specimen size, crack geometry and loading conﬁguration all can have strong constraint effects on the fracture toughness measurements [28]. Qian et al. studied the effect of temperature on the model parameters in local approaches (LAs) to cleavage fracture [29]. A statistical approach was adopted to correlate the fracture behaviour between a notched and a mechanically fractured specimen [30]. They investigated the constraint difference for cruciform specimens with shallow cracks, C(T) specimens, and three-point bending specimens with shallow and deep cracks [24]. Results showed the constraint is dependent on the loading and specimen geometry. C(T) specimen has the highest constraint, while the SEB specimen with a shallow crack has the lowest constraint level [24].

In this study, thin-sheet C(T) specimens with high constraints were tested. The stress intensity factor K-based resistance curves (K–R curves) have less crack size or in-plane geometry dependence. Al–Mg alloy has obvious recrystallisation with a larger grain size and a lower fracture toughness. Al–Mg–Sc–Zr alloy has no obvious recrystallisation with fibrous grain morphology and higher fracture toughness. For both alloys, the fracture toughness of L–T specimens is slightly higher than that of T–L specimens. For different specimens, the alignment of the fracture direction relative to the notch direction is different between the two alloys. For Al–Mg alloy, the fraction direction is laid between 0° and 90°, while for Al–Mg–Sc–Zr alloy, this is limited between 0° and 45°. For parallel specimens, the larger the angle, the higher the fracture toughness is.

The precipitation near the grain boundary produces stress concentration at the matrix interface with incongruent deformation, forming cavities. As the external force increases, the cavities continue to grow owing to coalescing. When the plastic deformation reaches a certain level, they are separated from the matrix and fracture forms. Inhomogeneous distribution of submicron-sized precipitates in the matrix may lead to local shear, which is not conducive to higher fracture toughness. The PFZ formed near the grain boundaries of the incoherent precipitates has lower yield stress than the grain interior. The easier dislocation movement in the PFZ can have a detrimental effect on fracture toughness.

In Al–Mg–Sc–Zr alloy, the impurity elements, such as Fe and Si, can greatly reduce the grain boundary bonding energy with them concentrated at the grain boundaries, which lead to the generation of fractures along the grains and adversely affect the fracture toughness. However, fracture toughness can be improved by uniformly and diffusely distributed coherent and semi-coherent precipitates. In the Al–Mg–Sc–Zr alloy, coherent Al₃(Sc, Zr) precipitates are present. With the increasing load, the dislocations cross a large number of coherent precipitation particles and pass through some active slip surface when the stress is higher enough to drive the dislocations moving in the matrix. Therefore, particles cut by dislocation require external stress to provide the energy needed by the particle shear, which is conducive to improving fracture toughness.

High-cycle fatigue

In the high-cycle fatigue tests, the fatigue life (N _f) of the material is formed with the initial stage of nucleation of the main crack, the propagation crack, and finally, failure. This includes the fatigue crack initiation life (N _i) and the fatigue crack propagation life (N _p). N _i includes the stages of crack initiation and microcrack propagation, while N _p includes the stages of macro crack propagation and final fracture. At a low-stress level with a negligible or weak stress concentration, good surface quality and environmental conditions, crack formation life N _i accounts for a large proportion of the total life. However, at a high-stress level with a high-stress concentration for a rough and highly defective surface under bad environmental conditions with a small amount of N _i, the N _p predominates the N _f. For a pre-cracked specimen, N _f almost equals N _p. When the fatigue life is short, the plastic strain component is dominant, and the fatigue life is determined by the ductility. The elastic strain amplitude is more significant when the fatigue life is long. The fatigue life increases with the increase of fracture strength. Prolonging either the crack initiation or propagation stage can improve fatigue strength.

Improving the tensile properties can improve the fatigue performance of many materials. Therefore, strengthening mechanisms based on a combination of solid solution, precipitation, and grain refinement have been developed. However, the tensile properties are generally affected by the overall microstructure characteristics of the material, and fatigue performance is affected by local microstructure defects, such as microstructure instability and grain coarsening. Additively manufactured metals, which have similar tensile properties to their conventional counterparts, exhibit poor fatigue performance (than conventional ones). Meneghetti et al. found that the fatigue strength of the additively manufactured specimens is lower than that of vacuum-melted bars of the same material under annealed conditions [31]. The stress level, pore size and pore-to-surface distance are the three important factors affecting fatigue behaviour and performance [32]. For aluminium alloys, it is important to improve fatigue strength. Otherwise, they cannot benefit from their high static strength. Their poor fatigue resistance is owing to the formation of microcracks and high notch sensitivity, which are susceptible to stress concentrations caused by surface roughness. Therefore, in aluminium alloys, most fatigue cracks are generated at the surface [33].

Grain boundary segregation is the main mechanism for fatigue crack initiation at near-surface grain boundaries. Local strain at the PFZ may also contribute to the crack initiation process [34]. For a smooth specimen, the dominant mechanism of fatigue crack initiation is slip nucleation driven by the local shear stress. Under the maximum shear stress, the dislocation movement at the material surface will form a small slip band. As the test continues, the number of slips will gradually increase, and the slip band will become wider and deeper, forming the so-called persistent slip band. Eventually, fatigue cracks will initiate. The persistent slip band is the cause of fatigue cracking sprouting in face-centred cubic metals. Grain boundaries are considered an insurmountable barrier to the persistent slip band, leading to stress concentration and cracking or forming a persistent slip band between adjacent grains [35]. The formation of persistent slip bands is stress dependent and only happens at low stresses and high-cycle fatigue.

Materials with low yield strength tend to produce slip bands during fatigue, followed by fatigue crack initiation at the persistent slip bands. For materials with higher yield strength, producing slip bands is more difficult. Hence, fatigue cracks are less likely to initiate. Once a fatigue crack is initiated, it generally propagates internally along the slip direction from a physically small crack to a long crack. The slip direction is mostly aligned with the direction of the maximum tangential stress. The variation in fatigue life can be explained by the changes in surface roughness and microstructure. The classical view of grain boundaries as microstructural barriers is correct when the crack length is larger than the grain size. However, at the early stage of crack development, the crack length is comparable to the grain size. Under this condition, the grain boundaries accelerate the crack growth rate and promote crack development [36]. Crack initiation can be hindered by grain refinement, solid-solution strengthening, and surface polishing, which controls the slip characteristic length, reduces the formation of persistent slip bands, and alleviates the stress concentration on a rough surface, respectively. Meanwhile, crack closure and crack front geometry can affect crack propagation.

The zones associated with crack initiation and propagation for the Al–Mg and Al–Mg–Sc–Zr alloys are relatively smooth, with finely scalloped dimples on the fatigue fracture surfaces. The crack initiation zone is roughly located at the fan shank, where the cracks initiate and microcracks propagate. The fatigue crack initiates from the surface and gradually propagates towards the inside with a small crack opening displacement. At this stage, the crack propagation is slow, and the repeated opening and closing make the two sides of the fracture squeeze each other. The initial fracture surface is smooth and flat, with some radial lines centred on the fatigue initiation zone. With further crack propagation, the fatigue crack expands along with a series of macroscopic planes of different heights.

The fatigue crack propagation zone occupies most of the fracture area. The surface is coarser than the fatigue initiation zone, and corrugated fatigue stripes are visible. The fatigue fringe centres on the fatigue source region and diffuses around the directions perpendicular to the crack propagation directions. After the crack is formed, the crack opens under tensile stress. Meanwhile, the tip is passivated and closed under compression or the release of tension. Under the cyclic stress condition, the crack tip is re-sharpened for further propagation. Owing to the high-stress concentration at the tip, subcritical extension occurs, leaving a fatigue striation. The final fracture zone is formed by the unstable expansion of the crack, which happens once the crack is larger than the critical size. The resulting fracture surface morphology is similar to that of static tensile fracture.

For the Al–Mg–Sc–Zr alloy specimens fractured in the HCF test, the fractures were intercepted and prepared for TEM analysis, shown in Figure 10. The fracture surface of Al–Mg–Sc–Zr alloy shows a high density of grain boundaries and sub-grains, diffusely distributed nanoscale Al₃(Sc, Zr) precipitates, and small grain size. Even at a certain distance from the fatigue fracture surface, deformation can still be observed. Propagation of dislocation is pinned by the interaction between the dislocations and the boundaries, the increased dislocation density inside the grains increases, and the Al₃(Sc, Zr) precipitation, which hinders the development of microcracks and is conducive to the cyclic stability.

For both alloys, standard smooth specimens were used in high-cycle fatigue tests. The fractures were located at the centre, and the cracks were initiated from the surface of the specimens. However, it was found that no obvious macroscopic cracks were observed on the surface of the specimens before the fracture occurred, which may indicate that the contribution of crack propagation in the fatigue life is small, especially at a lower stress level with high fatigue life. Since Al–Mg alloy has a larger grain size and lower yield strength, crack initiation caused by slip is more likely to occur, leading to a lower fatigue limit. However, it has good plasticity and rougher fracture surfaces, resulting in large energy consumption for macrocrack propagation. With the increase in stress level, the difference in fatigue resistance between Al–Mg alloy and Al–Mg–Sc–Zr alloy decreases owing to the increased influence of plastic deformation. Compared with Al–Mg alloy, Al–Mg–Sc–Zr alloy has a smaller grain size, higher volume fraction grain boundary, and uniformly distributed nanoscale precipitation. It has higher yield strength owing to fine grain strengthening and precipitation strengthening. Al–Mg–Sc–Zr alloy has better fatigue resistance, especially at low-stress levels. Dislocations packing near grain boundaries and pinning near precipitations were identified at the fracture surface by TEM observation. Cross slip, which is needed to avoid obstacles and cross grain boundaries, is restricted by the microstructures in the alloy. Therefore, it requires a high-stress level for microcrack initiation caused by slip accumulation, which benefits cycle stability. Therefore, Al–Mg–Sc–Zr alloy has a higher fatigue limit.

Conclusions

Al–Mg–Sc–Zr alloy has higher tensile strength and fracture toughness than Al–Mg alloy. Specimens with L–T orientation have higher fracture toughness than T–L specimens for both alloys. Fracture toughness increases with the angle between the fracture and notch directions for parallel specimens.

Al–Mg alloy with larger recrystallised grains has a lower fatigue limit owing to the easier microcrack initiation caused by slip accumulation. However, the coarser fracture surface is beneficial for prolonging the fatigue crack propagation life.

Al–Mg–Sc–Zr alloy with finer grains and fibrous structure, the higher volume fraction of grain boundaries and well-dispersed nanoscale Al₃Sc precipitation has a higher fatigue limit owing to the restriction of fatigue crack initiation caused by microstructure.

Footnotes

Disclosure statement

No potential conflict of interest was reported by the authors.

References

Roder

Wirtz

Gysler

Fatigue properties of Al–Mg alloys with and without scandium. Mater Sci Eng A. 1997;234:181–184.

Filatov

Yelagin VI Zakharov

VV.

New Al–Mg–Sc alloys. Mater Sci Eng A. 2000;280(1):97–101.

Sitdikov

Sakai

Avtokratova

Grain refinement in a commercial Al–Mg–Sc alloy under hot ECAP conditions. Mater Sci Eng A. 2007;444(1–2):18–30.

Nieh

Hsiung

Wadsworth

High strain rate superplasticity in a continuously recrystallized Al–6% Mg–0.3% Sc alloy. Acta Mater. 1998;46(8):2789–2800.

Komura

Berbon

Furukawa

High strain rate superplasticity in an Al–Mg alloy containing scandium. Scr Mater. 1998;38(12):1851–1856.

Lathabai

Lloyd P

The effect of scandium on the microstructure, mechanical properties and weldability of a cast Al–Mg alloy. Acta Mater. 2002;50(17):4275–4292.

Lee

Utsunomiya

Akamatsu

Influence of scandium and zirconium on grain stability and superplastic ductilities in ultrafine-grained Al–Mg alloys. Acta Mater. 2002;50(3):553–564.

Marquis

Seidman

Dunand

DC.

Effect of Mg addition on the creep and yield behavior of an Al–Sc alloy. Acta Mater. 2003;51(16):4751–4760.

Kim

Yeom

Effect of scandium content on the hot extrusion of Al–Zn–Mg–(Sc) alloy. J Mater Process Technol. 2007;187:635–639.

10.

Song

Huang

Deformation-induced dissolution and growth of precipitates in an Al–Mg–Er alloy during high-cycle fatigue. Acta Mater. 2014;81:409–419.

11.

Watanabe

Monzen

Tazaki

Effects of Al₃Sc particle size and precipitate-free zones on fatigue behavior and dislocation structure of an aged Al–Mg–Sc alloy. Int J Fatigue. 2008;30(4):635–641.

12.

Jiang

Zhu

Enhanced mechanical properties of 6082 aluminum alloy via SiC addition combined with squeeze casting. J Mater Sci Technol. 2021;88:119–131.

13.

Zhu

Jiang

Microstructure and mechanical properties of SiCnp/Al6082 aluminum matrix composites prepared by squeeze casting combined with stir casting. J Mater Process Technol. 2020;283:116699.

14.

Shukla

Komarasamy

Mishra

RS.

Grain size dependence of fatigue properties of friction stir processed ultrafine-grained Al-5024 alloy. Int J Fatigue. 2018;109:1–9.

15.

Majchrowicz

Pakieła

Giżyński

High-cycle fatigue strength of ultrafine-grained 5483 Al–Mg alloy at low and elevated temperature in comparison to conventional coarse-grained Al alloys. Int J Fatigue. 2018;106:81–91.

16.

Alexopoulos

Migklis

Stylianos

Fatigue behavior of the aeronautical Al–Li (2198) aluminum alloy under constant amplitude loading. Int J Fatigue. 2013;56:95–105.

17.

Boni

Lanciotti

Polese

“Size effect” in the fatigue behavior of friction stir welded plates. Int J Fatigue. 2015;80:238–245.

18.

Gaur

Enoki

Okada

A study on fatigue behavior of MIG-welded Al–Mg alloy with different filler-wire materials under mean stress. Int J Fatigue. 2018;107:119–129.

19.

Shen

Zeng

Investigating the effect of notch size on critical distance and fatigue limit by coupling the theory of critical distance and finite fracture mechanics. Theor Appl Fract Mech. 2022;122:103566.

20.

Yang

Zhang

Notch size influence on fatigue limit of steels pertinent to grain size. Int J Fatigue. 2022;156:106642.

21.

Marines-Garcia

Paris

Tada

Fatigue crack growth from small to long cracks in VHCF with surface initiations. Int J Fatigue. 2007;29(9–11):2072–2078.

22.

Branco

Costa

Berto

Fatigue crack initiation behaviour of notched 34CrNiMo6 steel bars under proportional bending-torsion loading. Int J Fatigue. 2020;130:105268.

23.

Luo

Yao

Susmel

Prediction of fatigue damage region with the use of the notch critical plane approach for crack initiation and propagation. Int J Fatigue. 2020;135:105533.

24.

Qian

Cao

Niffenegger

comparison of constraint analyses with global and local approaches under uniaxial and biaxial loadings. Eur J Mech-A/Solids. 2018;69:135–146.

25.

Rahmatabadi

Hashemi

Mohammadi

Experimental evaluation of the plane stress fracture toughness for ultra-fine grained aluminum specimens prepared by accumulative roll bonding process. Mater Sci Eng A. 2017;708:301–310.

26.

Rahmatabadi

Pahlavani

Bayati

Evaluation of fracture toughness and rupture energy absorption capacity of as-rolled LZ71 and LZ91 Mg alloy sheet. Mater Res Express. 2018;6(3):036517.

27.

Di Giovanni

de Menezes

JTO

Cerri

Influence of microstructure and porosity on the fracture toughness of Al–Si–Mg alloy. J Mater Res Technol. 2020;9(2):1286–1295.

28.

Zhu

Joyce

JA.

Review of fracture toughness (G, K, J, CTOD, CTOA) testing and standardization. Eng Fract Mech. 2012;85:1–46.

29.

Qian

Lei

Niffenegger

On the temperature independence of statistical model parameters for cleavage fracture in ferritic steels. Philos Mag. 2018;98(11):959–1004.

30.

Qian

Lei W

Peng

Statistical assessment of notch toughness against cleavage fracture of ferritic steels. Fatigue Fract Eng Mater Struct. 2018;41(5):1120–1131.

31.

Meneghetti

Rigon

Gennari

An analysis of defects influence on axial fatigue strength of maraging steel specimens produced by additive manufacturing. Int J Fatigue. 2019;118:54–64.

32.

Tang

The effect of porosity defects on the mid-cycle fatigue behavior of directed energy deposited Ti–6Al–4V. Theor Appl Fract Mech. 2022;119:103322.

33.

Benedetti

Fontanari

Bandini

High-and very high-cycle plain fatigue resistance of shot peened high-strength aluminum alloys: The role of surface morphology. Int J Fatigue. 2015;70:451–462.

34.

Nanninga

White

Mills

Effect of specimen orientation and extrusion welds on the fatigue life of an AA6063 alloy. Int J Fatigue. 2010;32(2):238–246.

35.

Qiu

Zhu

Intergranular crack during fatigue in Al–Mg–Si aluminum alloy thin extrusions. Int J Fatigue. 2017;100:105–112.

36.

Le V

Morel

Bellett

Simulation of the Kitagawa–Takahashi diagram using a probabilistic approach for cast Al–Si alloys under different multiaxial loads. Int J Fatigue. 2016;93:109–121.

Fracture toughness and high-cycle fatigue behaviour of Al–Mg and Al–Mg–Sc–Zr alloys

Abstract

Keywords

Introduction

Materials and methods

Results

Mechanical performance

Fracture morphology of fracture toughness specimens

Fracture surfaces of HCF specimens

TEM test results

Discussion

Fracture toughness

High-cycle fatigue

Conclusions

Footnotes

References