High-Speed 3D Printing Coupled with Machine Learning to Accelerate Alloy Development for Additive Manufacturing

Area of Science:

• Materials Science
• Metallurgical Engineering
• Additive Manufacturing

Background:

• Developing novel metal alloys for 3D printing is challenging due to time and resource constraints.
• Rapid screening of numerous compositions and processing parameters is crucial for alloy development.

Purpose of the Study:

• To accelerate the discovery of new alloys for 3D printing by employing high-throughput techniques.
• To explore the process-structure-properties (PSP) relationships in metastable steel-aluminum alloys.

Main Methods:

• Utilized high-throughput 3D printing to create 20 distinct steel-Al alloy combinations (0-10 wt.% Al).
• Employed in situ alloying and flexible adjustment of volumetric energy input.
• Conducted large-area crystallographic analysis, chemistry analysis, and nanoindentation for characterization.

Main Results:

• Identified significant influence of aluminum content and processing conditions on material behavior.
• Established clear process-structure-properties (PSP) relationships based on physical mechanisms.
• Validated microstructure-property relationships using explainable machine learning on limited data.

Conclusions:

• High-throughput methods enable efficient screening and development of novel 3D printable alloys.
• Understanding PSP relationships is key to optimizing alloy design for additive manufacturing.
• Machine learning can support experimental findings in materials science research.