| Abstract: | Laser powder bed fusion (L-PBF) additive manufacturing (AM) is an effective method of fabricating metallic structures with complex geometries, tailored material compositions and microstructures, and unique mechanical and functional properties, thus has quickly emerged as a promising technology to manufacture Shape Memory Alloy (SMA)-based components. How to identify, in an efficient and cost-effective manner, windows of processing parameter combinations resulting in defect-free parts is the first step in realizing the full application potential of L-PBF fabricated SMAs. This study assesses for the first time the printability (ability to fabricate defect-free parts) of two different types of SMAs, namely NiTiHf high temperature shape memory alloys (HT-SMAs) and NiTi SMAs with high Ni content. For NiTiHf HT-SMAs, a framework integrating an analytical thermal model with single-track experiments and uncertainty quantification (UQ) techniques is employed to map laser power, scanning speed, and hatch spacing to processing regimes leading to lack of fusion, keyholing, and balling defects. The effects of these processing parameters on the microstructure variation and phase transformation characteristics are systematically investigated. For high Ni-content NiTi SMAs, a combination of single-track experiments, specimen fabrication experiments, and machine learning classification is used for printability assessment in the presence of additional defect modes, such as warping, elevated edge/corner, delamination, etc. Another important, but less studied processing parameter, namely the laser scan strategy, on the porosity, warping deformation, phase transformation behavior, mechanical and functional properties of Ni-rich NiTi SMAs fabricated by L-PBF AM is also elucidated in detail. In addition, part-scale deformation simulations are performed by finite element analysis (FEA) for NiTi SMAs with two different Ni compositions to answer the question of why Ni-rich NiTi SMAs undergo more severe warping deformation during L-PBF AM process. The electronic version of this dissertation is accessible from https://hdl.handle.net/1969.1/198645 |
|---|