Tuning molten pool dynamics and keyhole formation during laser powder bed fusion through laser beam shaping
Keyhole porosity remains a critical barrier to broader adoption of laser powder bed fusion (PBF-LB) AM. The short lifespan and rapid dynamics of the molten pool pose significant challenges in probing pore formation mechanisms. To address these challenges, a high-fidelity thermo-fluid model was implemented that integrates laser-material interactions, vapor recoil, multiple reflections, and Marangoni effects. Simulations reveal keyhole pore formation follows a cycle of localized overheating, wall instability, and collapse, with subsequent bubble evolution—entrapment, coalescence, or surface escape—dictated by molten pool flow dynamics. By employing this model as a computational testbed, we evaluate a beam-shaping strategy to mitigate porosity and compare four distinct energy distributions: Gaussian, flat-top, longitudinal elliptical, and transverse elliptical (TE). The TE profile, with its major axis perpendicular to the scan direction, uniquely mitigates localized energy concentration and reduces detrimental multi‑reflections, thereby stabilizing the keyhole and reducing susceptibility to porosity under the studied conditions. Crucially, the results suggest pore susceptibility can be reduced with a modest decrease in volumetric melt rate, while the predicted solidification conditions may favor equiaxed-grain formation. This simulation-derived beam optimization framework provides fundamental qualitative insights and guidance for reducing susceptibility to keyhole pores in high-integrity, microstructurally tailored materials produced by PBF-LB.
Authors
- Xiaowu Jiang (ORCID: https://orcid.org/0000-0002-9951-9114)
- Xingjun Liu (ORCID: https://orcid.org/0000-0002-3015-8525)
- Rongpei Shi (ORCID: https://orcid.org/0000-0002-5007-4249)
- Shuai Wang (ORCID: https://orcid.org/0009-0001-6162-4269)
- Xu Tang (ORCID: https://orcid.org/0000-0001-9437-2662)
- Zhifu Yao (ORCID: https://orcid.org/0000-0003-1554-5276)
- Jun Wei (ORCID: https://orcid.org/0000-0002-6726-1972)
Institutions
- Harbin Institute of Technology (CN)
- China Iron and Steel Research Institute Group (CN)
- Shenzhen Technology University (CN)
- Zhejiang University (CN)
Publication Details
- Journal
- npj Computational Materials
- Published
- 2026-11-14
- DOI
- https://doi.org/10.1038/s41524-026-02394-2
- Primary Topic
- Additive Manufacturing Materials and Processes
- Type
- article
- Field-Weighted Citation Impact
- 0.00