Return to search

Cyclic Behavior of Shape Memory Alloys: Materials Characterization and Optimization

Shape memory alloys (SMAs) are unique metallic alloys which can undergo large deformations while reverting back to their undeformed shape through either the application of heat (shape memory effect) or the removal of the load (superelastic effect). A multi-scale and multi-disciplinary approach is taken to explore the use of large diameter NiTi SMAs for applications in earthquake engineering. First, a materials characterization study is performed by studying precipitate formation, grain size and orientation, thermal transformation behavior, and strength. Cyclic tensile tests on coupon specimens and full-scale large diameter bars are then used to correlate the microstructural properties to the macroscopic behavior. Further experimental studies using NiTi wire are performed in order to optimize their properties for seismic applications. The ability of mechanical training to stabilize NiTi cyclic properties, the ability of pre-straining to increase damping levels, and the influence of different types of earthquake loadings are considered. Phenomenological mechanical models are then developed based on these results. An analytical study is then used to evaluate the performance of structural systems incorporating SMAs. One type of system evaluated includes an SMA bracing system used to modify the response of a structure during a seismic event. Overall, the results of this study have shown the ability to optimize the properties of NiTi SMAs for seismic applications through material processing. The analytical results show potential for the use of SMAs in seismic applications and provide areas for continued research.

Identiferoai:union.ndltd.org:GATECH/oai:smartech.gatech.edu:1853/10497
Date05 April 2006
CreatorsMcCormick, Jason P.
PublisherGeorgia Institute of Technology
Source SetsGeorgia Tech Electronic Thesis and Dissertation Archive
Languageen_US
Detected LanguageEnglish
TypeDissertation
Format29320123 bytes, application/pdf

Page generated in 0.0022 seconds