Leveraging the digital thread for physics-based prediction of microstructure heterogeneity in additively manufactured parts

Research output: Contribution to journalArticlepeer-review

4 Scopus citations

Abstract

A major limitation of additive manufacturing (AM) processes is that local conditions of material deposition frequently lead to unintentional heterogeneities in microstructure and properties within a single component, despite nominally uniform process conditions. Up to now, there has been no way to a priori determine the distribution of these heterogeneities, requiring expensive trial-and-error approaches to fabrication, testing, and characterization. Here, a physics-based framework for creating a digital representation of the laser powder bed fusion (PBF) process is proposed to predict the variation in solidification behavior that leads to heterogeneous microstructures in an as-built part. By leveraging in situ process data stored in the part's digital thread, the scan path and process parameters were input into a heat transfer model which predicted solidification data at the melt pool scale. A two-step unsupervised clustering algorithm was used to first cluster the local solidification conditions (12.5 µm3 voxels) and then to cluster the regional behavior on the scale of multiple scan passes and print layers (250 µm3 super-voxels). This process was used to identify regions with similar solidification characteristics for multiple locations in a Stainless Steel 316-L component. The corresponding as-built part was sectioned and characterized using electron backscatter diffraction (EBSD). Quantitative analysis of the pole figures confirmed that the predicted regions of heterogeneity in the solidification conditions corresponded with differences in the observed microstructure. This work shows a viable path for estimating the microstructural heterogeneity for additively manufactured parts to either limit microstructural variation throughout a part or to enable functionality-based variation of the microstructure.

Original languageEnglish
Article number103861
JournalAdditive Manufacturing
Volume78
DOIs
StatePublished - Sep 25 2023

Funding

The work related to the development of digital tools associated with Peregrine and the characterization of materials was funded by the Department of Energy Advanced Materials & Manufacturing Technologies Office and utilized resources at the ORNL Manufacturing Demonstration Facility . The development of the clustering algorithm was supported by the Advanced Materials and Manufacturing Technologies ( AMMT ) program sponsored by the Department of Energy Office of Nuclear Energy . This research used resources of the Compute and Data Environment for Science ( CADES ) at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725 . Additionally, the authors thank Selda Nayir for insightful technical discussion on microstructure analysis and Sarah Graham for metallographic specimen preparation.

Keywords

  • Additive manufacturing
  • Digital thread
  • Heterogeneity
  • Microstructure
  • Solidification
  • Stainless steel 316 L

Fingerprint

Dive into the research topics of 'Leveraging the digital thread for physics-based prediction of microstructure heterogeneity in additively manufactured parts'. Together they form a unique fingerprint.

Cite this