Microscale Phase Transformation in Shape Memory Alloy Actuators

Shape Memory Alloy;NiTi;Nitinol;Shape Memory Effect;Actuator;Phase Transformation;Martensite;Austenite;in-situ;SEM;Digital Image Correlation;Materials Science and Engineering;Mechanical Engineering;Engineering;Mechanical Engineering

Gong, YueMankame, Nilesh ;

University of Michigan

关键词: Shape Memory Alloy;    NiTi;    Nitinol;    Shape Memory Effect;    Actuator;    Phase Transformation;    Martensite;    Austenite;    in-situ;    SEM;    Digital Image Correlation;    Materials Science and Engineering;    Mechanical Engineering;    Engineering;    Mechanical Engineering;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/137027/gongyue_1.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Shape memory alloys (SMAs) are a class of metallic alloys with the ability to recover large mechanical deformations with little residual strain, and the ability to revert to a previously defined shape or size when deformed and then heated past a set transformation temperature. The prior ability is termed superelasticity (SE) or pseudoelasticity, and the latter is termed the shape memory effect (SME). The most widely used shape memory alloy is Nickel-Titanium, also known as Nitinol, which is a nearly equiatomic mixture of nickel and titanium that undergoes a phase transformation between a B2 cubic austenite and a B19’ martensite in order to give rise to these unique properties. Although Nitinol is utilized in a variety of actuation applications, there is much that is unknown about the microscale characteristics of the transformation that underlies the shape memory effect, particularly under cyclic actuation. At the macroscopic length scale, global measures like stress-strain relationships are used to improve prediction capabilities. These studies provide valuable information, but are averaged measures that cannot capture the transformation heterogeneity inherent to these materials, nor the dependence of this heterogeneity on the underlying microstructure. As such, an opportunity to control material performance through targeted processing of the microstructure is lost. In this thesis, the transformation characteristics of fine-grained and coarse-grained shape memory NiTi microwires (grain diameters of 20-40nm and 10 μm, respectively) are investigated using a custom experimental methodology combining deformation tracking with scanning electron microscopy in order to map the microscale strains associated with phase transformation during shape memory actuation. Regardless of the grain diameter, a strongly heterogeneous strain distribution arose with martensite detwinning upon application of load, and remained consistent with actuation cycling. Sub-grain regions that accommodated high amounts of strain in the detwinned state early in the cycling process were more likely to accommodate high amounts of strain in subsequent cycles, and showed a faster accumulation of residual strain in subsequent cycles as well, indicating a link between sub-grain residual strain and strain at detwinned state that is established early and can negatively impact actuator performance. Additionally, correlation coefficients between individual data points tended to increase with thermo-mechanical cycling, indicating that there is a settling in process where the similarity induced by detwinning and plasticity becomes stronger as cycling progresses. In the coarse-grained specimens, neither grain diameter nor grain orientation affected the transformation characteristics of the grain, supporting the hypothesis that it is rather the interactions with the surrounding microstructure that determine transformation characteristics at a point. The standard deviation of the strain increased with the average grain diameter - this strain heterogeneity was inversely related to the ratio between the length scale of the microstructure and the field of view, but was not related to the actuation stroke.