In brief

A novel maraging steel customised for laser additive manufacturing through machine learning surpassed traditional strength and elongation metrics without post-processing heat treatments, paving the way for more energy-efficient manufacturing.

© Shutterstock

Green steel cuts carbon emissions

1 Jul 2024

Aided by machine learning, researchers develop a novel high-performance steel optimised for more eco-friendly 3D printing.

When you print a document, inkjet printers deploy tiny dots of ink to create text and images. Laser additive manufacturing (LAM) works on a similar principle, but in three dimensions. LAM builds objects from the bottom up, printing one precise metal layer at a time according to a digital blueprint.

There’s a downside to the metal powders traditionally used in LAM, however; they're not very eco-friendly, according to Chaolin Tan, a Senior Scientist at A*STAR’s Singapore Institute of Manufacturing Technology (SIMTech).

“LAM-processed commercial metallic materials usually need extensive post-heat treatments (PHT) to achieve good mechanical performance, which releases carbon dioxide and consumes a lot of time and energy,” said Tan.

To address this, Tan and collaborators from the City University of Hong Kong, China, and Pennsylvania State University, US, searched for more sustainable LAM building blocks. They aimed to harness the intrinsic heat treatment (IHT) effect found in LAM printing processes, which naturally strengthens the materials used as they are deposited.

“IHT happens as LAM subjects materials to rapid heating and cooling cycles, creating short temperature spikes that can effectively harden the material as it is deposited,” Tan added.

The challenge lay in finding the right composition of alloys that relies solely on LAM's intrinsic thermal cycles to produce strong end products, eliminating the need for energy-intensive PHT. The team turned to machine learning (ML) for help, using it to analyse vast amounts of data and quickly identify materials with the most promising properties.

Their research led them to develop a novel maraging steel formulation of iron, nickel, titanium and aluminium (Fe-Ni-Ti-Al) using a Random Forest ML model known for its accuracy. This new steel formulation successfully formed hard precipitates during the printing process itself.

When tested, the deposited material achieved an impressive tensile strength of 1538 MPa and a uniform elongation of 8.1 percent, outperforming LAM-printed traditional high-strength steels which usually need PHT to achieve similar parameters.

Tan noted that artificial intelligence techniques like ML are helping to facilitate 4D printing, which integrates what would normally be time-dependent processes like PHT reshaping into the 3D geometry of LAM, cutting manufacturing time significantly. “This not only makes LAM processes more efficient, but also reduces their energy use and carbon footprint,” Tan said.

Tan added that the metallurgical industry is a major emitter of greenhouse gases, with a large portion attributed to heat treatment. “Our approach to sustainable alloy design could significantly reduce the industry’s greenhouse gas emissions, helping it become more environmentally friendly,” said Tan.

The A*STAR-affiliated researchers contributing to this research are from the Singapore Institute of Manufacturing Technology (SIMTech).

Want to stay up to date with breakthroughs from A*STAR? Follow us on Twitter and LinkedIn!


Tan, C., Li, Q., Yao, X., Chen, L., Su, J., et al. Machine learning customized novel material for energy-efficient 4D printing. Advanced Science 10 (10), 2206607 (2023). | article

About the Researchers

Chaolin Tan is a Senior Scientist and Doctoral Supervisor at SIMTech. He obtained his PhD from South China University of Technology. He was a Visiting Scholar (in 2018) and Honorary Research Fellow at the University of Birmingham (in 2019); and an Associate Professor at Guangdong University of Technology (in 2019). He is a Fellow of the International Association of Advanced Materials (FIAAM) and listed in the World’s Top 2% Scientist Rankings in 2022 and 2023. He has 10 years of research experience in Laser Additive Manufacturing (LAM), with research interests in materials innovation, machine learning and heterostructured materials around LAM and energy-field assisted additive manufacturing. He is on the Editorial Board of the International Journal of Machine Tools and Manufacture, and Youth Editor of the International Journal of Extreme Manufacturing; Journal of Materials Science & Technology; Materials Research Letters; and Rare Metals. He has also contributed to more than 70 journal papers and two books, including 36 Science Citation Index (SCI) papers as lead author, and a few Essential Science Indicators (ESI) Highly Cited and Hot Papers.
Youxiang Chew has over nine years of experience in Laser Directed Energy Deposition (LDED) technology, focusing on processes, residual stress modelling and the integration of machine learning to optimise laser deposition toolpaths. Chew’s other research interests include developing high entropy alloys and bulk spatially patterned heterostructured multimaterials parts using LDED. Beyond the LDED process, Chew is now looking into new process innovation by combining different advanced manufacturing processes for multimaterial processing. The group is also exploring hybrid DED processes and new friction stir processing for repair and remanufacturing, aiming to streamline the manufacturing of high value components.

This article was made for A*STAR Research by Wildtype Media Group