The production of high-performance titanium alloys for applications in the aerospace, medical technology and defense industries has traditionally involved a great deal of effort. Despite advances in metal 3D printing, the search for optimal manufacturing parameters requires extensive testing. A research team from the Johns Hopkins Applied Physics Laboratory (APL) and the Whiting School of Engineering has now used AI-supported models to identify new manufacturing approaches for the Laser Powder Bed Fusion (L-PBF) process in order to increase the manufacturing speed and improve the material properties at the same time.
The research focuses on the titanium alloy Ti-6Al-4V, which is characterized by high strength and low weight. By using machine learning, the scientists were able to analyze manufacturing conditions that were previously considered ineffective and determine new process parameters that enable a denser and more resistant material structure.
“The nation faces an urgent need to accelerate manufacturing to meet the demands of current and future conflicts,” said Morgan Trexler, program manager for Science of Extreme and Multifunctional Materials in APL’s Research and Exploratory Development Mission Area. “At APL, we are advancing research in laser-based additive manufacturing to rapidly develop mission-ready materials, ensuring that production keeps pace with evolving operational challenges.”
Traditional methods for process optimization are based on time-consuming tests. In contrast, the research team used AI models that efficiently identify suitable process parameters through Bayesian optimization. This method makes it possible to simulate thousands of configuration options and validate only the most promising ones in real tests. This showed that certain combinations of laser power, scanning speed and track spacing increase the material strength without losing ductility.
“For years, we assumed that certain processing parameters were ‘off-limits’ for all materials because they would result in poor-quality end product,” said Brendan Croom, a senior materials scientist at APL. “But by using AI to explore the full range of possibilities, we discovered new processing regions that allow for faster printing while maintaining — or even improving — material strength and ductility, the ability to stretch or deform without breaking. Now, engineers can select the optimal processing settings based on their specific needs.”
The findings could have far-reaching implications for various sectors of industry. In aerospace and shipbuilding in particular, more efficient production of highly resilient titanium components can lead to cost reductions and improved performance characteristics. At the same time, the method could be extended to other alloys and additive manufacturing techniques to enable further material innovations. The team is already working on integrating AI-powered simulations to predict material behavior in extreme environments. This could help to qualify new materials more quickly and optimize them for industrial use.
Future developments could also include real-time monitoring during the printing process in order to make adjustments directly during production.
Steve Storck describes a vision of the future in which state-of-the-art metal-based additive manufacturing works as seamlessly as 3D printing at home: “We envision a paradigm shift where future additive manufacturing systems can adjust as they print, ensuring perfect quality without the need for extensive post-processing and that parts can be born qualified.”
The research results could therefore fundamentally change the 3D printing of high-performance materials and open up new potential in additive manufacturing
Subscribe to our Newsletter
3DPresso is a weekly newsletter that links to the most exciting global stories from the 3D printing and additive manufacturing industry.
