- Email: [email protected]
In-situ scanning electron microscopy studies of small fatigue crack growth in ultrasonic consolidation bonded aluminum 2024 laminated structure
Materials Letters 112 (2013) 47–50
Contents lists available at ScienceDirect
Materials Letters journal homepage: www.elsevier.com/locate/matlet
In-situ scanning electron microscopy studies of small fatigue crack growth in ultrasonic consolidation bonded aluminum 2024 laminated structure Xiao-Hua He a, Hui-Ji Shi a,n, Yu-Duo Zhang b, Wen-Xiang Fu b, Zhi-Gang Yang b, Carey E. Wilkinson c a b c
AML, School of Aerospace, Tsinghua University, Beijing 100084, China Key Laboratory for Advanced Materials of Ministry of Education, School of Materials Science and Engineering, Tsinghua University, Beijing 100084, China The Boeing Co., USA
art ic l e i nf o
a b s t r a c t
Article history: Received 27 May 2013 Accepted 22 August 2013 Available online 30 August 2013
This work used in-situ scanning electron microscopy method to investigate the small fatigue crack behavior of the aluminum alloy 2024 laminated structure produced by ultrasonic consolidation process. The influence of local microstructure on the crack propagation of the laminates after different heat treatments (T4 and T62) has been compared and discussed. It was found that both the precipitations and the interfaces of the layers could retard crack growth and alter the growth direction, and the effect of interfaces was much stronger. Due to the higher ratio of the toughness to bond strength, the barrier effect of the interfaces on impeding the fatigue crack growth in specimens after the T4 heat treatment was more evident. & 2013 Elsevier B.V. All rights reserved.
Keywords: Aluminum alloy Fatigue In-situ SEM Interface Microstructure
1. Introduction Laminated metal structures have been utilized as structural materials in aerospace, navigation and defense industries due to their improvement in various properties, e.g. fracture toughness, impact behavior, wear and damping capacity [1–6]. As an innovative solid-state manufacturing technology, ultrasonic consolidation (UC) process has a number of advantages over the traditional techniques, e.g. internal geometry capability, reduction in residual stresses and consumed energy [7–9], and also allows embedding fibers, wirings and electronics sensors into the laminates to make smart materials [9–11]. Most of the investigations of laminated structures were focused on the fracture and impact toughness and different toughening mechanisms [2–6], but the fatigue properties were limited studied [12–14]. Besides, the researches so far on the laminates produced by the UC process were concentrated on the optimization of parameters by peel test and microstructure observation [9,15–22], but the studies on the mechanical properties were scarce and not in-depth [10,23], let alone the fatigue property. In this study, the in-situ SEM technique was used to investigate the small fatigue crack growth behavior in the UC produced Al
n
Corresponding author. Tel.: þ 86 10 62772731; fax: þ86 10 62781824. E-mail address: [email protected] (H.-J. Shi).
0167-577X/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.matlet.2013.08.093
2024 laminated structure after different heat treatments (HT). Particular emphasis is placed on examining the fatigue cracking mechanism and the effect of local microstructure on crack growth.
2. Materials and experimental details The nominal composition of Al 2024 alloy is (in wt%): 0.5Si, 0.5Fe, 3.8–4.9Cu, 0.3–0.9Mn, 1.2–1.8Mg, 0.1Cr, 0.25Zn, 0.15Ti and Al balance. Specimens were cut with gauge length parallel to the longitudinal orientation of the metal foils. The slab specimens had a dog-bone shape with a 1.4 mm by 1.5 mm gauge cross section and a U-shaped notch was prepared by WEDM. The specimens were divided into two groups and heat-treated to a T4 and a T62 temper, respectively. The surfaces of all the specimens were polished and then etched in a solution of 1% HFþ 1.5% HCl þ2.5% HNO3 þ95% H2O to reveal the microstructure. Optical microscopy (OM) and scanning electron microscopy (SEM) were performed to observe the microstructure of the laminated structure after etching. The in-situ fatigue tests were performed at room temperature (RT) in the vacuum chamber of the SEM using a specially designed servo-hydraulic testing system. A maximal stress of 300 MPa was adopted throughout the tests for all specimens. The load waveform utilized was sinusoidal with stress ratio R¼ 0.1 and frequency f ¼5 Hz.
48
X.-H. He et al. / Materials Letters 112 (2013) 47–50
3. Results and discussion Fig. 1 presents the OM and SEM observations of the microstructures on a cross-section of the laminates. No matter after T4 or T62 HT, the laminated structures contained a few small defects but no initial delaminations at the interfaces (Fig. 1b and e). The grain sizes of the specimens were not uniform, and there was no obvious difference of the grain sizes between the specimens after T4 and T62 HT. Generally, the grains near the interfaces were much smaller, as shown in Fig. 1a and d. Moreover, precipitation compounds were widely distributed in the grains, as indicated in Fig. 1c and f. The typical microstructure of Al 2024-T4 alloy laminated specimen is shown in Fig. 2a. The prepared U-shaped notch was 188 μm in depth and 485 μm in diameter. Fig. 2b depicts the initial fatigue cracks from the notch root and the precipitations after over 23,000 cycles. It was noticed that the crack did not break the precipitations but tended to change its path and bypass the small blocky
precipitation. After exceeding about 60 μm in crack length, some slip traces were observed along with the crack propagation and an obvious slip trace appeared in front of the main crack (Fig. 2c). As the crack propagated along this slip trace, a subsurface precipitation was revealed, as indicated in Fig. 2d, which explained why the slip trace appeared preferentially at that site. The interface of foils is a unique part in the laminated structure. Thus the influence of interfaces on the crack growth of the laminated specimens is of great concern. The main crack was blocked after approaching the first interface. During the following 2000 cycles, the propagation of the fatigue crack was divided into two directions, i.e. the transverse one and the longitudinal one as shown in Fig. 2e. At this stage, the longitudinal crack along interface was dominated and propagated faster. Meanwhile, the prior main crack continued to impinge the interface and its crack tip opening displacement (CTOD) visibly increased, indicating evident effect on retarding the growth of the transverse crack. This retardation by local interface delamination has also been
Fig. 1. OM observations, SEM observations and precipitations of the laminated structures after T4 HT (a)–(c) and after T62 HT (d)–(f). (The black arrows indicate the locations of interfaces.)
Fig. 2. Fatigue short crack growth in Al 2024-T4 laminated specimen. (a) 0 cycles; (b) 22,954 cycles; (c) 25,414 cycles; (d) 25,730 cycles; (e) 27,925 cycles; (f) 28,241 cycles; (g) 28,399 cycles; and (h) 28,837 cycles.
X.-H. He et al. / Materials Letters 112 (2013) 47–50
reported in the literature [14]. After a number of cycles, however, the longitudinal crack did not propagate any further. Instead, some short cracks initiated from the defect tips along the longitudinal crack as indicated in Fig. 2e. One of these cracks became the main crack and rapidly propagated. In the front region of the crack, obvious plastic deformation produced evident slip traces (Fig. 2f). Moreover, defects at the second interface and fractured precipitations due to deformation incompatibility also emerged, as indicated by the short arrows in Fig. 2f. It should be mentioned that, the slip traces occurred near the defects or pointed at the defects. Therefore, it was supposed that the slip traces indicated the crack growth path, i.e., the crack tended to connect the defects together. During the following 150 cycles, the crack propagation behavior confirmed the inference as shown in Fig. 2g. Due to the large plastic deformation and deformation incompatibility, the growth path of the crack was rugged. When arriving at the second interface, the crack behavior was similar to that at the first interface. Short cracks initiated from the defects of the other interfaces and small cracks caused by the fracture of precipitations emerged obviously (Fig. 2h). These short cracks coalesced with the main crack to induce the final rupture. The typical microstructure of Al 2024-T62 alloy laminated specimen is shown in Fig. 3a. The prepared U-shaped notch was 240 μm in depth and 430 μm in diameter. Fig. 3b depicts the initial short cracks not only from the notch root, but also from the precipitations at the first interface. The short cracks coalesced, bypassed some subsurface precipitations (small arrows in Fig. 3b) and became the main crack. Although the crack tended to propagate along the interface for a short length, the transverse crack was still dominated. During the following 1000 cycles, the main crack crossed the first interface and propagated forward stably and transgranularly. After exceeding about 135 μm in crack length, the crack changed its direction and grew along the grain boundary (GB) for a short distance. The reason for the transition was inferred to be the subsurface precipitation as indicated in Fig. 3c. The crack crossed the second and third interfaces just like the first one, i.e., the crack just grew along the interface for a short length and turned to propagate forward, subsequently.
49
The crack growth behaviors were similar when the crack went through the fourth and fifth layers, as shown in Fig. 3d and e respectively. First, a short crack initiated from the defects at the interfaces, as indicated by small arrows in Fig. 3d. Then, the crack propagated forward and tended to coalesce with the defects at the next interface. During the propagation, some defects at the GBs emerged apparently as indicated in Fig. 3d and e. However, the cracking was transgranular rather than intergranular. After the crack reached the fifth interface, the CTOD increased evidently instead of crack growth. Severe plastic deformation appeared at the crack tip, as shown in Fig. 3f. Moreover, the delamination at the sixth interface was obvious before the final fracture. According to the in-situ observations above, it could be seen that the interfaces acted as barriers to the crack growth, and the blockage effect of the interfaces of specimens after T4 HT was much stronger than that after T62 HT, which could be supported by the bigger CTOD and larger range of delaminations. In our other work, the basic mechanical properties of Al 2024 alloy used in this work were tested according to ASTM standards and the bond strength of the laminated specimens was tested by the method in the literature [20]. Some results are listed in Table 1. It should be noted that the CTOD at the maximum load was used to characterize the fracture toughness. The monolithic alloy specimens after T4 HT had the little higher tensile strength, much larger
Table 1 Mechanical properties of Al 2024 monolithic specimens and bond strength of laminated specimens. HT
Al 2024 alloy monolithic specimens
Laminated specimens
Tensile strength sUTS Elongation (MPa) (%)
Fracture toughness δa (μm)
Bond strength P (N)
36 18
10290 10560
T4 457 T62 449
21 6
a δ represents the CTOD at the maximum load applied on the specimen in fracture toughness test.
Fig. 3. Fatigue short crack growth in Al 2024-T62 laminated specimen. (a) 0 cycles; (b) 13,763 cycles; (c) 16,624 cycles; (d) 18,254 cycles; (e) 18,740 cycles; and (f) 18,903 cycles.
50
X.-H. He et al. / Materials Letters 112 (2013) 47–50
elongation and fracture toughness than those after T62 HT. However, the bond strengths of the laminated structures after T4 and T62 HT were very approximate. It was easy to understand that the higher fracture toughness meant the larger energy required for crack propagation in the materials, while the higher bond strength indicated the more energy needed for the interface delamination. Therefore, due to the higher toughness of the specimens after T4 HT, the transverse crack growth in the next layer required more energy than the longitudinal one along the interface, when the crack arrived at the interfaces. And because of this, the energy was first consumed in generating the earlier and larger range of delaminations, which led to the much more extra time for the accumulation of enough energy for the transverse crack propagation. The specimens after T62 HT exhibited the opposite behaviors. Hence, in our work, it was thought that the ratio of toughness to bond strength (i.e. n ¼ δ/P) played a critical role in the crack propagation near the interfaces. The higher n meant that the delaminations occurred earlier and more obviously, while the lower n indicated the crack was more inclined to cross the interface with no or slight deflection and propagated in the next layer. 4. Conclusion The small crack behaviors in the UC produced Al 2024 laminated structure after a T4 and a T62 heat treatment were studied by in-situ SEM technique at RT. The results demonstrated that the fatigue crack growth in this laminated structure was discontinuous, transgranular and showed great sensitivity on the microstructure. Both precipitations and interfaces acted as barriers to prevent the fatigue crack propagation and altered the crack path, and the influence of the interfaces was more obvious. The transverse crack was impeded, when arriving at the interfaces. Instead, the crack turned to propagate along the interfaces for a finite length, which resulted in the local delamination. Some short cracks caused by the fracture of precipitations and defects at the interfaces coalesced with the main crack and led to the final rupture. Moreover, the crack behavior at the interfaces, i.e. whether generated a large range of delamination firstly or just propagated forward with no or slight deflection, depended on the ratio of toughness of the materials to bond strength of the interfaces. Concerning the similar bond strength, the barrier influence of the interfaces on the cracking in specimens after T4 HT was stronger, due to the higher toughness than the specimens after T62 HT.
Acknowledgments The financial supports from Tsinghua-Boeing Joint Research Center (TBRC) and the National Natural Science Foundation of China (No. 51071094) are highly acknowledged.
References [1] Liu CY, Wang Q, Jia YZ, Jing R, Zhang B, Ma MZ, et al. Materials Science and Engineering A: Structural Materials 2012;556:1–8. [2] Cepeda-Jimenez CM, Carreno F, Ruano OA, Sarkeeva AA, Kruglov AA, Lutfullin RY. Materials Science and Engineering A: Structural Materials 2013;563: 28–35. [3] Cepeda-Jimenez CM, Garcia-Infantaa JM, Pozueo M, Ruano OA, Carreno F. Scripta Materialia 2009;61:407–10. [4] Cepeda-Jimenez CM, Pozuelo M, Garcia-Infanta JM, Ruano OA, Carreno F. Metallurgical and Materials Transactions A 2009;40A:69–79. [5] Lesuer DR, Syn CK, Sherby OD, Wadsworth J, Lewandowski JJ, Hunt WH. International Materials Reviews 1996;41:169–97. [6] Lesuer DR, Wadsworth J, Riddle RA, Syn CK, Lewandowski JJ, Hunt Jr. WH. Layered Materials for Structural Applications Symposium 1996;434:205–11. [7] White DR. Advanced Materials and Processes 2003;161:64–5. [8] Tsujino J, Ueoka T, Kashino T, Sugahara F. Ultrasonics 2000;38:67–71. [9] Yang Y, Janaki Ram GD, Stucker BE. Journal of Engineering Materials and Technology, Transactions of the ASME 2007;129:538–49. [10] Yang Y, Janaki Ram GD, Stucker BE. Journal of Materials Processing Technology 2009;209:4915–24. [11] Cheng X, Datta A, Choi H, Zhang X, Li X. Journal of Manufacturing Science and Engineering 2007;129:416–24. [12] Hassan HA, Lewandowski JJ, Abd El-Latif MH. Metallurgical and Materials Transactions A 2004;35A:45–52. [13] Hassan HA, Lewandowski JJ, Abd El-Latif MH. Metallurgical and Materials Transactions A 2004;35A:2291–303. [14] Liu HS, Zhang B, Zhang GP. Scripta Materialia 2011;65:891–4. [15] Janaki Ram GD, Robinson C, Yang Y, Stucker BE. Rapid Prototyping Journal 2007;13:226–35. [16] Obielodan JO, Ceylan A, Murr LE, Stucker BE. Rapid Prototyping Journal 2010;16:180–8. [17] Kong CY, Soar RC, Dickens PM. Materials Science and Engineering A: Structural Materials 2003;363:99–106. [18] Kong CY, Soar RC, Dickens PM. Journal of Materials Processing Technology 2004;146:181–7. [19] Janaki Ram GD, Yang Y, Stucker BE. Journal of Manufacturing Systems 2006;25:221–38. [20] Zhang C, Deceuster A, Li L. Journal of Materials Engineering and Performance 2009;18:1124–32. [21] Li D, Soar RC. Journal of Engineering Materials and Technology, Transactions of the ASME 2009;131:1–6. [22] Hopkins CD, Dapino MJ, Fernandez SA. Journal of Engineering Materials and Technology, Transactions of the ASME 2010;132:1–9. [23] Mariani E, Ghassemieh E. Acta Materialia 2010;58:2492–503.
Contents lists available at ScienceDirect
Materials Letters journal homepage: www.elsevier.com/locate/matlet
In-situ scanning electron microscopy studies of small fatigue crack growth in ultrasonic consolidation bonded aluminum 2024 laminated structure Xiao-Hua He a, Hui-Ji Shi a,n, Yu-Duo Zhang b, Wen-Xiang Fu b, Zhi-Gang Yang b, Carey E. Wilkinson c a b c
AML, School of Aerospace, Tsinghua University, Beijing 100084, China Key Laboratory for Advanced Materials of Ministry of Education, School of Materials Science and Engineering, Tsinghua University, Beijing 100084, China The Boeing Co., USA
art ic l e i nf o
a b s t r a c t
Article history: Received 27 May 2013 Accepted 22 August 2013 Available online 30 August 2013
This work used in-situ scanning electron microscopy method to investigate the small fatigue crack behavior of the aluminum alloy 2024 laminated structure produced by ultrasonic consolidation process. The influence of local microstructure on the crack propagation of the laminates after different heat treatments (T4 and T62) has been compared and discussed. It was found that both the precipitations and the interfaces of the layers could retard crack growth and alter the growth direction, and the effect of interfaces was much stronger. Due to the higher ratio of the toughness to bond strength, the barrier effect of the interfaces on impeding the fatigue crack growth in specimens after the T4 heat treatment was more evident. & 2013 Elsevier B.V. All rights reserved.
Keywords: Aluminum alloy Fatigue In-situ SEM Interface Microstructure
1. Introduction Laminated metal structures have been utilized as structural materials in aerospace, navigation and defense industries due to their improvement in various properties, e.g. fracture toughness, impact behavior, wear and damping capacity [1–6]. As an innovative solid-state manufacturing technology, ultrasonic consolidation (UC) process has a number of advantages over the traditional techniques, e.g. internal geometry capability, reduction in residual stresses and consumed energy [7–9], and also allows embedding fibers, wirings and electronics sensors into the laminates to make smart materials [9–11]. Most of the investigations of laminated structures were focused on the fracture and impact toughness and different toughening mechanisms [2–6], but the fatigue properties were limited studied [12–14]. Besides, the researches so far on the laminates produced by the UC process were concentrated on the optimization of parameters by peel test and microstructure observation [9,15–22], but the studies on the mechanical properties were scarce and not in-depth [10,23], let alone the fatigue property. In this study, the in-situ SEM technique was used to investigate the small fatigue crack growth behavior in the UC produced Al
n
Corresponding author. Tel.: þ 86 10 62772731; fax: þ86 10 62781824. E-mail address: [email protected] (H.-J. Shi).
0167-577X/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.matlet.2013.08.093
2024 laminated structure after different heat treatments (HT). Particular emphasis is placed on examining the fatigue cracking mechanism and the effect of local microstructure on crack growth.
2. Materials and experimental details The nominal composition of Al 2024 alloy is (in wt%): 0.5Si, 0.5Fe, 3.8–4.9Cu, 0.3–0.9Mn, 1.2–1.8Mg, 0.1Cr, 0.25Zn, 0.15Ti and Al balance. Specimens were cut with gauge length parallel to the longitudinal orientation of the metal foils. The slab specimens had a dog-bone shape with a 1.4 mm by 1.5 mm gauge cross section and a U-shaped notch was prepared by WEDM. The specimens were divided into two groups and heat-treated to a T4 and a T62 temper, respectively. The surfaces of all the specimens were polished and then etched in a solution of 1% HFþ 1.5% HCl þ2.5% HNO3 þ95% H2O to reveal the microstructure. Optical microscopy (OM) and scanning electron microscopy (SEM) were performed to observe the microstructure of the laminated structure after etching. The in-situ fatigue tests were performed at room temperature (RT) in the vacuum chamber of the SEM using a specially designed servo-hydraulic testing system. A maximal stress of 300 MPa was adopted throughout the tests for all specimens. The load waveform utilized was sinusoidal with stress ratio R¼ 0.1 and frequency f ¼5 Hz.
48
X.-H. He et al. / Materials Letters 112 (2013) 47–50
3. Results and discussion Fig. 1 presents the OM and SEM observations of the microstructures on a cross-section of the laminates. No matter after T4 or T62 HT, the laminated structures contained a few small defects but no initial delaminations at the interfaces (Fig. 1b and e). The grain sizes of the specimens were not uniform, and there was no obvious difference of the grain sizes between the specimens after T4 and T62 HT. Generally, the grains near the interfaces were much smaller, as shown in Fig. 1a and d. Moreover, precipitation compounds were widely distributed in the grains, as indicated in Fig. 1c and f. The typical microstructure of Al 2024-T4 alloy laminated specimen is shown in Fig. 2a. The prepared U-shaped notch was 188 μm in depth and 485 μm in diameter. Fig. 2b depicts the initial fatigue cracks from the notch root and the precipitations after over 23,000 cycles. It was noticed that the crack did not break the precipitations but tended to change its path and bypass the small blocky
precipitation. After exceeding about 60 μm in crack length, some slip traces were observed along with the crack propagation and an obvious slip trace appeared in front of the main crack (Fig. 2c). As the crack propagated along this slip trace, a subsurface precipitation was revealed, as indicated in Fig. 2d, which explained why the slip trace appeared preferentially at that site. The interface of foils is a unique part in the laminated structure. Thus the influence of interfaces on the crack growth of the laminated specimens is of great concern. The main crack was blocked after approaching the first interface. During the following 2000 cycles, the propagation of the fatigue crack was divided into two directions, i.e. the transverse one and the longitudinal one as shown in Fig. 2e. At this stage, the longitudinal crack along interface was dominated and propagated faster. Meanwhile, the prior main crack continued to impinge the interface and its crack tip opening displacement (CTOD) visibly increased, indicating evident effect on retarding the growth of the transverse crack. This retardation by local interface delamination has also been
Fig. 1. OM observations, SEM observations and precipitations of the laminated structures after T4 HT (a)–(c) and after T62 HT (d)–(f). (The black arrows indicate the locations of interfaces.)
Fig. 2. Fatigue short crack growth in Al 2024-T4 laminated specimen. (a) 0 cycles; (b) 22,954 cycles; (c) 25,414 cycles; (d) 25,730 cycles; (e) 27,925 cycles; (f) 28,241 cycles; (g) 28,399 cycles; and (h) 28,837 cycles.
X.-H. He et al. / Materials Letters 112 (2013) 47–50
reported in the literature [14]. After a number of cycles, however, the longitudinal crack did not propagate any further. Instead, some short cracks initiated from the defect tips along the longitudinal crack as indicated in Fig. 2e. One of these cracks became the main crack and rapidly propagated. In the front region of the crack, obvious plastic deformation produced evident slip traces (Fig. 2f). Moreover, defects at the second interface and fractured precipitations due to deformation incompatibility also emerged, as indicated by the short arrows in Fig. 2f. It should be mentioned that, the slip traces occurred near the defects or pointed at the defects. Therefore, it was supposed that the slip traces indicated the crack growth path, i.e., the crack tended to connect the defects together. During the following 150 cycles, the crack propagation behavior confirmed the inference as shown in Fig. 2g. Due to the large plastic deformation and deformation incompatibility, the growth path of the crack was rugged. When arriving at the second interface, the crack behavior was similar to that at the first interface. Short cracks initiated from the defects of the other interfaces and small cracks caused by the fracture of precipitations emerged obviously (Fig. 2h). These short cracks coalesced with the main crack to induce the final rupture. The typical microstructure of Al 2024-T62 alloy laminated specimen is shown in Fig. 3a. The prepared U-shaped notch was 240 μm in depth and 430 μm in diameter. Fig. 3b depicts the initial short cracks not only from the notch root, but also from the precipitations at the first interface. The short cracks coalesced, bypassed some subsurface precipitations (small arrows in Fig. 3b) and became the main crack. Although the crack tended to propagate along the interface for a short length, the transverse crack was still dominated. During the following 1000 cycles, the main crack crossed the first interface and propagated forward stably and transgranularly. After exceeding about 135 μm in crack length, the crack changed its direction and grew along the grain boundary (GB) for a short distance. The reason for the transition was inferred to be the subsurface precipitation as indicated in Fig. 3c. The crack crossed the second and third interfaces just like the first one, i.e., the crack just grew along the interface for a short length and turned to propagate forward, subsequently.
49
The crack growth behaviors were similar when the crack went through the fourth and fifth layers, as shown in Fig. 3d and e respectively. First, a short crack initiated from the defects at the interfaces, as indicated by small arrows in Fig. 3d. Then, the crack propagated forward and tended to coalesce with the defects at the next interface. During the propagation, some defects at the GBs emerged apparently as indicated in Fig. 3d and e. However, the cracking was transgranular rather than intergranular. After the crack reached the fifth interface, the CTOD increased evidently instead of crack growth. Severe plastic deformation appeared at the crack tip, as shown in Fig. 3f. Moreover, the delamination at the sixth interface was obvious before the final fracture. According to the in-situ observations above, it could be seen that the interfaces acted as barriers to the crack growth, and the blockage effect of the interfaces of specimens after T4 HT was much stronger than that after T62 HT, which could be supported by the bigger CTOD and larger range of delaminations. In our other work, the basic mechanical properties of Al 2024 alloy used in this work were tested according to ASTM standards and the bond strength of the laminated specimens was tested by the method in the literature [20]. Some results are listed in Table 1. It should be noted that the CTOD at the maximum load was used to characterize the fracture toughness. The monolithic alloy specimens after T4 HT had the little higher tensile strength, much larger
Table 1 Mechanical properties of Al 2024 monolithic specimens and bond strength of laminated specimens. HT
Al 2024 alloy monolithic specimens
Laminated specimens
Tensile strength sUTS Elongation (MPa) (%)
Fracture toughness δa (μm)
Bond strength P (N)
36 18
10290 10560
T4 457 T62 449
21 6
a δ represents the CTOD at the maximum load applied on the specimen in fracture toughness test.
Fig. 3. Fatigue short crack growth in Al 2024-T62 laminated specimen. (a) 0 cycles; (b) 13,763 cycles; (c) 16,624 cycles; (d) 18,254 cycles; (e) 18,740 cycles; and (f) 18,903 cycles.
50
X.-H. He et al. / Materials Letters 112 (2013) 47–50
elongation and fracture toughness than those after T62 HT. However, the bond strengths of the laminated structures after T4 and T62 HT were very approximate. It was easy to understand that the higher fracture toughness meant the larger energy required for crack propagation in the materials, while the higher bond strength indicated the more energy needed for the interface delamination. Therefore, due to the higher toughness of the specimens after T4 HT, the transverse crack growth in the next layer required more energy than the longitudinal one along the interface, when the crack arrived at the interfaces. And because of this, the energy was first consumed in generating the earlier and larger range of delaminations, which led to the much more extra time for the accumulation of enough energy for the transverse crack propagation. The specimens after T62 HT exhibited the opposite behaviors. Hence, in our work, it was thought that the ratio of toughness to bond strength (i.e. n ¼ δ/P) played a critical role in the crack propagation near the interfaces. The higher n meant that the delaminations occurred earlier and more obviously, while the lower n indicated the crack was more inclined to cross the interface with no or slight deflection and propagated in the next layer. 4. Conclusion The small crack behaviors in the UC produced Al 2024 laminated structure after a T4 and a T62 heat treatment were studied by in-situ SEM technique at RT. The results demonstrated that the fatigue crack growth in this laminated structure was discontinuous, transgranular and showed great sensitivity on the microstructure. Both precipitations and interfaces acted as barriers to prevent the fatigue crack propagation and altered the crack path, and the influence of the interfaces was more obvious. The transverse crack was impeded, when arriving at the interfaces. Instead, the crack turned to propagate along the interfaces for a finite length, which resulted in the local delamination. Some short cracks caused by the fracture of precipitations and defects at the interfaces coalesced with the main crack and led to the final rupture. Moreover, the crack behavior at the interfaces, i.e. whether generated a large range of delamination firstly or just propagated forward with no or slight deflection, depended on the ratio of toughness of the materials to bond strength of the interfaces. Concerning the similar bond strength, the barrier influence of the interfaces on the cracking in specimens after T4 HT was stronger, due to the higher toughness than the specimens after T62 HT.
Acknowledgments The financial supports from Tsinghua-Boeing Joint Research Center (TBRC) and the National Natural Science Foundation of China (No. 51071094) are highly acknowledged.
References [1] Liu CY, Wang Q, Jia YZ, Jing R, Zhang B, Ma MZ, et al. Materials Science and Engineering A: Structural Materials 2012;556:1–8. [2] Cepeda-Jimenez CM, Carreno F, Ruano OA, Sarkeeva AA, Kruglov AA, Lutfullin RY. Materials Science and Engineering A: Structural Materials 2013;563: 28–35. [3] Cepeda-Jimenez CM, Garcia-Infantaa JM, Pozueo M, Ruano OA, Carreno F. Scripta Materialia 2009;61:407–10. [4] Cepeda-Jimenez CM, Pozuelo M, Garcia-Infanta JM, Ruano OA, Carreno F. Metallurgical and Materials Transactions A 2009;40A:69–79. [5] Lesuer DR, Syn CK, Sherby OD, Wadsworth J, Lewandowski JJ, Hunt WH. International Materials Reviews 1996;41:169–97. [6] Lesuer DR, Wadsworth J, Riddle RA, Syn CK, Lewandowski JJ, Hunt Jr. WH. Layered Materials for Structural Applications Symposium 1996;434:205–11. [7] White DR. Advanced Materials and Processes 2003;161:64–5. [8] Tsujino J, Ueoka T, Kashino T, Sugahara F. Ultrasonics 2000;38:67–71. [9] Yang Y, Janaki Ram GD, Stucker BE. Journal of Engineering Materials and Technology, Transactions of the ASME 2007;129:538–49. [10] Yang Y, Janaki Ram GD, Stucker BE. Journal of Materials Processing Technology 2009;209:4915–24. [11] Cheng X, Datta A, Choi H, Zhang X, Li X. Journal of Manufacturing Science and Engineering 2007;129:416–24. [12] Hassan HA, Lewandowski JJ, Abd El-Latif MH. Metallurgical and Materials Transactions A 2004;35A:45–52. [13] Hassan HA, Lewandowski JJ, Abd El-Latif MH. Metallurgical and Materials Transactions A 2004;35A:2291–303. [14] Liu HS, Zhang B, Zhang GP. Scripta Materialia 2011;65:891–4. [15] Janaki Ram GD, Robinson C, Yang Y, Stucker BE. Rapid Prototyping Journal 2007;13:226–35. [16] Obielodan JO, Ceylan A, Murr LE, Stucker BE. Rapid Prototyping Journal 2010;16:180–8. [17] Kong CY, Soar RC, Dickens PM. Materials Science and Engineering A: Structural Materials 2003;363:99–106. [18] Kong CY, Soar RC, Dickens PM. Journal of Materials Processing Technology 2004;146:181–7. [19] Janaki Ram GD, Yang Y, Stucker BE. Journal of Manufacturing Systems 2006;25:221–38. [20] Zhang C, Deceuster A, Li L. Journal of Materials Engineering and Performance 2009;18:1124–32. [21] Li D, Soar RC. Journal of Engineering Materials and Technology, Transactions of the ASME 2009;131:1–6. [22] Hopkins CD, Dapino MJ, Fernandez SA. Journal of Engineering Materials and Technology, Transactions of the ASME 2010;132:1–9. [23] Mariani E, Ghassemieh E. Acta Materialia 2010;58:2492–503.