Suppression of Void Nucleation in High-Purity Aluminum via Dynamic Recrystallization
- PDF / 11,435,519 Bytes
- 13 Pages / 593.972 x 792 pts Page_size
- 8 Downloads / 161 Views
I.
INTRODUCTION
BEFORE 1950, the question of how and where voids nucleate during ductile rupture was hotly debated. Several researchers proposed that voids nucleated at the head of blocked slip bands by a cleavage-like mechanism,[1,2] while others hypothesized that voids formed by dislocation reactions.[3–5] The invention of the scanning electron microscope enabled the discovery by Tipper and others[6,7] that voids nucleate at second-phase particles. Based on this discovery, Cottrell hypothesized that ‘‘if such particles were not present, the specimen would pull apart entirely by the inward growth of the external neck, giving nearly 100 pct reduction in area.’’[5] Several critical experiments on high-purity face-centered-cubic (FCC) metals, primarily aluminum (Al), strongly supported this conclusion.[7–12] For example, using aluminum, Chin et al.[10] observed that the dimple density on the fracture surface decreased with increasing sample purity. In the extreme case, zone-refined (approximately 99.999 pct Al) aluminum failed by necking to a chisel point rather than by cavitation.[10] Similar trends were reported in high-purity lead (Pb),[9] though the resolution of the characterization techniques used in both studies was likely insufficient to detect voids smaller than 10 lm. These results led to the conclusion
PHILIP J. NOELL and BRAD L. BOYCE are with the Sandia National Laboratories, P.O. Box 5800, Albuquerque, NM 87185-0889. Contact e-mail: [email protected] RYAN B. SILLS is with the Sandia National Laboratories, P.O. Box 969, Livermore, CA 94551-9035. Manuscript submitted April 5, 2019.
METALLURGICAL AND MATERIALS TRANSACTIONS A
that generalized void nucleation does not occur in bulk, particle-free FCC metals that deform by slip[13–15] except, perhaps, when supersaturated hydrogen creates micropores during processing.[16] In contrast, Boyce et al.[17] observed that high-purity tantalum (Ta), a body-centered-cubic (BCC) metal, failed in a ductile manner by void nucleation, growth, and coalescence. No second-phase particles or inclusions were observed in this material. Subsequent analysis showed that voids in Ta nucleated at deformation-induced dislocation boundaries, i.e., dislocation cell walls and cell block boundaries.[18] Similar grain subdivision into groups of dislocation cells, separated from each other by lamellar cell block boundaries, occurs in all wavy glide materials, including Al, during plastic deformation.[19–25] In general, dislocation cell walls are primarily formed of statistically stored dislocations[26] and are low-angle (< 1 deg) boundaries; cell block boundaries contain many geometrically necessary dislocations[26] and generally accommodate misorientations greater than 2 deg.[21,22] The observation that voids nucleate at these boundaries in Ta raises an important question: why is void nucleation at dislocation boundaries suppressed in Al and, by extension, other FCC metals? It is possible that the relative ease of slip in FCC metals compared to BCC metals suppresses void nucleation at di
Data Loading...
