Evolution of damage and failure in an additively manufactured 316L SS structure: experimental reinvestigation of the third Sandia fracture challenge

• Published in: International journal of fracture Vol. 218; no. 1-2; pp. 63 - 84
• Main Authors: Kramer, S. L. B., Ivanoff, T. A., Madison, J. D., Lentfer, A. P.
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01-07-2019; Springer Nature B.V; Springer
• Subjects: Additive manufacturing; Austenitic stainless steels; Automotive Engineering; Characterization and Evaluation of Materials; Chemistry and Materials Science; Civil Engineering; Classical Mechanics; Computed tomography; Computer simulation; Correlation analysis; Crack initiation; Crack propagation; Deformation; Deviation; Dimpling; Ductile fracture; Evolution; Failure; Fracture surfaces; MATERIALS SCIENCE; Mechanical Engineering; Sandia Fracture Challenge 2017; Statistical tests; Steel structures; Stress concentration; Surface defects; Surface roughness; Voids; Ductile fracture; Void growth; Additive manufacturing
• ISSN: 0376-9429; 1573-2673
• DOI: 10.1007/s10704-019-00357-x

The third Sandia Fracture Challenge (SFC3) was a benchmark problem for comparing experimental and simulated ductile deformation and failure in an additively manufactured (AM) 316L stainless steel structure. One surprising observation from the SFC3 was the Challenge-geometry specimens had low variability in global load versus displacement behavior, attributed to the large stress-concentrating geometric features dominating the global behavior, rather than the AM voids that tend to significantly influence geometries with uniform cross-sections. This current study reinvestigates the damage and failure evolution of the Challenge-geometry specimens, utilizing interrupted tensile testing with micro-computed tomography (micro-CT) scans to monitor AM void and crack growth from a virgin state through complete failure. This study did not find a correlation between global load versus displacement behavior and AM void attributes, such as void volume, location, quantity, and relative size, which incidentally corroborates the observation from the SFC3. However, this study does show that the voids affect the local behavior of damage and failure. Surface defects (i.e. large voids located on the surface, far exceeding the nominal surface roughness) that were near the primary stress concentration affected the location of crack initiation in some cases, but they did not noticeably affect the global response. The fracture surfaces were a combination of classic ductile dimples and crack deviation from a more direct path favoring intersection with AM voids. Even though the AM voids promoted crack deviation, pre-test micro-CT scan statistics of the voids did not allow for conclusive predictions of preferred crack paths. This study is a first step towards investigating the importance of voids on the ductile failure of AM structures with stress concentrations.