Development of aluminum oxide slurries for additive manufacturing by Bayesian optimization

Abstract

Additive manufacturing by vat photopolymerization (VPP) enables the flexible production of ceramic components. The process requires ceramic slurries consisting of a photosensitive binder system and ceramic powder. To prevent defects during debinding and sintering, the highest possible content of ceramic particles is desired. At the same time, a certain viscosity must not be exceeded to ensure the processability in the VPP process. This conflict of objectives requires a precise adjustment of the large amount of slurry constituents. Hence, an experimental slurry development and optimization is very expensive and time-consuming. Therefore, Bayesian optimization, an artificial intelligence (AI) approach, was used to enhance an experimental optimization of the slurry composition. Using this approach, it was possible to achieve in less than 40 optimization steps an aluminum oxide (Al2O3) slurry suitable for VPP with a content of 65 vol.% ceramic powder, the highest currently known fraction for Al2O3 in VPP slurries.

Keywords

• Additive manufacturing
• Vat photopolymerization
• Artificial Intelligence
• Bayesian optimization
• Slurry development