Electrical machines account for a significant fraction of the world’s energy consumption, hence there is a pressing need for the next generation of such devices to have improved efficiency. Commonly used soft magnetic materials (SMM) for rotating electrical machines and transformers cores include Fe-Si based and Fe-Ni alloys, which have their own set of limitations in terms of properties and ease of manufacturing; in all cases these alloys currently lack a good balance of functional and structural properties.

Accelerated materials development utilizing a combinatorial approach of high throughput experiments and AI/machine learning are vital in the search for promising new SMM alloy compositions with a good combination of both functional and structural properties.

Several techniques, including high throughput spark plasma sintering (SPS) and additive manufacturing (AM), were used to quickly create material libraries of Fe-Co-Ni, Ni-Co and Fe-Si-Al. These libraries were subsequently rapidly characterized and multiple properties were evaluated.

Through our accelerated methodology, multiple properties of many compositional regions of the relevant ternary phase diagram could be rapidly determined. The optimum material compositions for a specific property set can be identified. Further materials development can be built upon these results by machine learning. Investigation into alloy systems such as Fe-Ni-Si, Fe-Si and Fe-Si-Al, using this methodology, are ongoing.