Experimentation and Model Development of High Temperature Shape Memory Alloy Actuation Cycling Under Tension, Compression, and Bending

Abstract

Shape Memory Alloys (SMAs) are a unique class of intermetallic alloys that can cyclically sustain large deformations and recover a designed geometry through a solid-to-solid, martensitic phase transformation. The multifunctional behavior of SMAs, being both structural and active, along with a favorable actuation energy density, make SMA actuators practical for volume, mass, and mechanical improvements to a system. For aerospace engineering applications, High Temperature SMAs (HTSMAs) are specialized to operate in the extreme environmental conditions necessary for safe operation in a higher temperature range. The main limitations to HTSMAs, however, are the unpredictable cyclic response and stability, large scale manufacturing inconsistency, and lack of commercially available design tools to accurately capture the macroscopic response. This study will address these limitations by characterizing cyclic evolution, using a single extrusion of high temperature SMA material for manufacturing and property consistency, and by developing and validating a finite element model to predict cyclic actuation response in bending loading conditions typical in SMA actuator components. Results show experiments for Ni50.3TiHf20 High Temperature SMA (HTSMA) including: differential scanning calorimetry, thermomechanical actuation cycling in tension and compression, thermomechanical actuation cycling under pure bending, and C-Ring bending. Since there is significant tension - compression asymmetry in SMA phase transformation, a full-field strain response is quantified for bending cases using Digital Image Correlation (DIC). The four point bending results contain a neutral axis shift due to the asymmetry, and during actuation cycling, the neutral axis continues to shift as a consequence of remnant, unrecovered plastic strains. C-ring tests show the martensitic phase transformation initiate in tension followed by compression during forward phase transformation. Lastly, two phenomenological, macroscopic SMA constitutive models that address cyclic behavior and anisotropy in SMA actuation are addressed. The constitutive models are used in a finite element software, Abaqus, with a user-defined material subroutine (UMAT) to address model improvements in light of the experimental results. Needed updates to the UMAT are discussed that will improve accuracy and prediction of the SMA actuation response in bending.