收缩率
多孔性
材料科学
融合
微观结构
复合材料
冶金
选择性激光熔化
语言学
哲学
作者
William Frieden Templeton,Shawn Hinnebusch,Seth Strayer,Albert C. To,Petrus Christiaan Pistorius,Sneha Prabha Narra
出处
期刊:Acta Materialia
[Elsevier]
日期:2023-12-29
卷期号:266: 119632-119632
被引量:12
标识
DOI:10.1016/j.actamat.2023.119632
摘要
This work focuses on the occurrence of shrinkage porosity and its manifestation in Alloy 718 during laser-powder bed fusion (PBF-LB) processing. Shrinkage porosity is a common defect in metal castings that degrades part performance. In traditional metal castings, the Niyama criterion is a reliable heuristic for predicting the formation of shrinkage porosity. However, the Niyama criterion's applicability in the PBF-LB process remains unexplored, and there is no known and evaluated heuristic to predict shrinkage porosity in the PBF-LB manufactured parts. This work employs microstructure characterization and an analytical heat transfer model to develop a mechanistic explanation for the formation of shrinkage porosity in the PBF-LB process. The results show that the Niyama criterion could not effectively predict the occurrence of shrinkage porosity. Further, the formation of shrinkage porosity is primarily driven by secondary dendrite arm growth in the solidifying microstructure, where the transition to cellular growth at high cooling rates during solidification mitigates porosity by removing locations for pore formation. This means a heuristic based on solidification cooling rate could reliably predict the occurrence of shrinkage porosity. For practical use, shrinkage porosity process maps are presented for use in process design and control to directly aid shrinkage pore mitigation in process planning. The process-shrinkage porosity relationship results also indicate that trends in PBF-LB manufacturing toward higher deposition temperatures and higher throughput are likely to aggravate the conditions for shrinkage pore formation and further elevate the importance of the mitigation strategies presented in this work.
科研通智能强力驱动
Strongly Powered by AbleSci AI