AI-supported optimization of 3D-printed titanium alloys

A research team from the Korea Advanced Institute of Science and Technology (KAIST), in collaboration with Pohang University of Science and Technology (POSTECH), has developed a method to optimize the mechanical properties of 3D-printed Ti-6Al-4V. Using artificial intelligence (AI), the scientists were able to overcome the challenge of simultaneously achieving high strength and high ductility in this widely used titanium alloy. The research results were published in the journal Nature Communications.

Ti-6Al-4V is a commonly used titanium alloy with high strength and good biocompatibility, which is used in aerospace and medical technology, among others. Additive manufacturing using laser powder bed fusion (LPBF) enables the precise production of complex components from this alloy. However, previous manufacturing processes often led to an increase in strength at the expense of ductility and vice versa. The conventional approach to optimizing material properties is based on adjusting pressure parameters and heat treatment processes, which is time-consuming and resource-intensive due to the large number of possible combinations with experimental and simulation-based methods.

The AI-supported active learning framework developed by KAIST and POSTECH enables a targeted selection of process parameters and heat treatment settings. The model analyzes mechanical properties such as tensile strength and elongation at break, taking into account uncertainties in the predictions. The proposed parameters are tested experimentally in iterative learning loops, the results are fed into the model and further optimized. After just five iterations, a combination of compression and post-treatment parameters was identified that achieves a tensile strength of 1190 MPa and an elongation at break of 16.5%.

Professor Seungchul Lee commented, “In this study, by optimizing the 3D printing process parameters and heat treatment conditions, we were able to develop a high-strength, high-ductility Ti-6Al-4V alloy with minimal experimentation trials. Compared to previous studies, we produced an alloy with a similar ultimate tensile strength but higher total elongation, as well as that with a similar elongation but greater ultimate tensile strength.” He added, “Furthermore, if our approach is applied not only to mechanical properties but also to other properties such as thermal conductivity and thermal expansion, we anticipate that it will enable efficient exploration of 3D printing process parameters and heat treatment conditions.”

The research was funded by the Nano & Material Technology Development Program and the Leading Research Center Program of the National Research Foundation of Korea. The results show the potential of AI-supported process design in additive manufacturing and lay the foundation for more efficient material development in the industry.