No abstract

No abstract

It is well accepted that the steady-state fatigue crack growth rates of long cracks depend uniquely on AK for a fixed load ratio R and test environment. Anomalous growth behavior of short cracks in either inert or deleterious environments that has been reported [1][2][3][4][5][6][7], however, calls into question the validity of using only long-crack results in evaluating the service life of a structural component and argues for the need for considering the effects of crack size. Crack-size effects have been extensively studied from the perspectives of mechanical, metallurgical and chemical principles. Microstructurally and mechanically short cracks [1][2][3] are associated with the influences of fine-scale microstructure, the limitation of continuum mechanics (or LEFM-linear elastic fracture mechanics), or crack closure. Chemically short cracks [4][5][6][7], on the other hand, normally extend to longer lengths and are attributed to the differences in local crack-tip chemistry (e.g., pH, [Oz], potential) between the long and short cracks and also, perhaps, from the bulk solution. The objective of this study is to explore the effect of crack size in various environments and to identify the key variables that affect crack growth behavior. The crack size investigated herein was chemically short but microstructurally and mechanically long.A 1.6 mm thick 2024-T3 (bare) aluminum alloy was used and its chemical composition (in wt%) is as follows: 4.24 Cu, 1.26 Mg, 0.65 Mn, 0.15 Fe, 0.08 Zn, 0.06 Si, 0.031 Ti, <0.01 Cr and balance A1. The elongation, 0.2% offset yield strength and tensile strength are 17.0%, 355 MPa (51 ksi) and 480 MPa (70 ksi), respectively. The specimens were cut in the L-T orientation as dog-bone shaped, center-pin-loaded, single-edge-cracked specimens with a 31.75 mm wide and 76.2 mm long (final dimension) mid-section. A 1.6 mm long electro-discharge machined (EDM) starter notch was introduced at the center of one edge of each specimen. The specimens were precracked by fatigue in air to a crack length of 3.42 mm (including the notch). Extra material (2.92 mm wide) was included on the notched side, and was removed by EDM to produce the final specimen, with a symmetrical test section and an approximaely 0.5 mm long edge crack. The following K calibration equation was used:P is the applied tensile load; B is the specimen thickness; W is the specimen width and a is the crack length. Equation (2) is an empirical formula obtained by Tada [8] and (3) is the correction accounting for the bending effect due to rotation at the loading pins [9]. A PC-controlled closed-loop electrohydraulic testing machine was used for the crack growth tests. Crack lengths were monitored by an a.c. potential system [10]. The specimens were loaded under pseudo-constant AK condition by automatically shedding the load range after each crack length determination (about every 10 ~m). The load ratio, frequency and temperature for all tests were R=0.1, f=10 Hz and T = 295 K, respectively. Crack growth rates at two AK levels (5 and 10 ...

An investigation of the interactions between mechanical and environmental variables on the shortfatigue-crack growth rate (FCGR) for a 2024-T3 aluminum alloy in 0.5M NaCl solution was carried out. Fatigue-crack growth tests were performed under a constant stress-intensity-factor range (⌬K ) control using single-edge-cracked tension specimens. The relationship between FCGR and crack length (0.5 to 15 mm) was determined at a cyclic frequency of 10 Hz over six ⌬K levels (4, 5, 6, 7, 8, and 10 MPaΊm), two load ratios (R) (0.1 and 0.5), and three dissolved oxygen concentrations (0, 7, and 30 ppm). Tests in gaseous environments (namely, high-purity oxygen) were also conducted for comparison. Short-crack effects were observed, with the FCGR in the short-crack regime accelerated by as much as a factor of 2. The observed crack-size effects tend to appear only at the lower loading levels (⌬K Ͻ 10 MPaΊm and R ϭ 0.1) and are more pronounced at higher oxygen levels. Fractographic examinations suggested that hydrogen embrittlement is responsible for the environmental enhancement of the FCGR for both short and long cracks in this material/environment system. A transport model was developed to estimate the crack-tip oxygen concentration and to examine its correlation to changes in the FCGR with crack length. The model correctly accounted for the decrease in short-crack effect on the FCGR with crack length under a given mechanical condition at each oxygen level, but did not explain the disappearance of short-crack effects at ⌬K Ն 10 MPaΊm.

No abstract