Skip to main content
eScholarship
Open Access Publications from the University of California

UC Irvine

UC Irvine Electronic Theses and Dissertations bannerUC Irvine

Process-structure-properties relationships in laser powder bed fusion additive manufacturing

Abstract

Laser powder bed fusion (L-PBF) Additive Manufacturing (AM) has exponentially grown in the last decades and is now being used for the production of both aerospace components and biomedical components, showing great potential to become a robust technology in the near future. Because of the plethora of process parameters and variables in L-PBF, the industry has focused on optimizing the required properties (e.g. part density, yield strength, UTS) by means of brute force (i.e. conducting expensive and time-consuming experimental campaigns). As a consequence, a deep theoretical understanding of process-structure-properties relationships of LPBF-AM is still missing. In this work we investigate how these process parameters (controllable) and variables (uncontrollable), along with the physical properties of the printed alloy, determine the complex physical phenomena that dominate LPBF. Among the many findings, we demonstrate the limited validity of a physical parameter (energy density) that was previously widely used by the AM community, finding its applicability to be limited to very narrow process windows. Thanks to in-situ high-speed imaging capabilities, we offer new insight into the complex interaction between gas-entrained powder particles and the liquid metal pool. We also conduct a dedicated analysis of the solidification behavior during L-PBF AM, finding that thermal gradient (G) and solidification rate (R) are strongly dependent on AM process parameters, thus offering practical advice to control the resulting microstructure. Finally, a Design of Experiments (DoE) study is carried out to elucidate the relationships between selected process parameters and the resulting stress distribution produced in cantilever beam-shaped samples. These results suggest that - depending on sample geometry - the amount of distortion may be more closely linked to the stress gradient developed across the beam’s thickness rather than to the average magnitude of the Von Mises residual stress.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View