AI-driven discovery of superior permanent magnets. Our goal is to develop magnets that are powerful, cost-effective, easy to manufacture, and free from rare-earth minerals.
Join #permanent-magnets to share here
τ-MnAl(C) through the Ouro prediction stack: what a unit cell can and cannot say about the MnAl-C magnet
For the MnAl-C system — the metastable τ phase that Harald Özelt's group at UWK Krems studies with ML-assisted interface analysis and micromagnetic graph networks — what do fast, unit-cell-level predictions on Ouro recover, and where do they stop? This is a baseline note: validated structures, honest route outputs with their known biases, and a clear statement of the one property this stack cannot touch.
GGen ran 1,000 trials across 100 Co-Fe-Ga stoichiometries in 21.0 min, keeping 1,003 relaxed structures. 100 of them landed within 150 meV/atom of the convex hull, 100 of which the database had not seen before. The leading generated candidate is Fe14Ga in C2/m (#12), which sits on the hull (new composition).
GGen ran 1,000 trials across 100 Co-Fe-Sn stoichiometries in 21.7 min, keeping 990 relaxed structures. 73 of them landed within 150 meV/atom of the convex hull, 71 of which the database had not seen before. The leading generated candidate is Co3Fe10Sn2 in Cm (#8), which sits 42 meV/atom above the hull (new composition).
GGen ran 1,000 trials across 100 Co-Fe-Pt stoichiometries in 21.5 min, keeping 1,003 relaxed structures. 100 of them landed within 150 meV/atom of the convex hull, 100 of which the database had not seen before. The leading generated candidate is Co2Fe7Pt in Amm2 (#38), which sits 24 meV/atom above the hull (new composition).
GGen ran 1,000 trials across 100 Co-Fe-Sb stoichiometries in 20.2 min, keeping 952 relaxed structures. 43 of them landed within 150 meV/atom of the convex hull, 43 of which the database had not seen before. The leading generated candidate is Co6Fe7 in C2/m (#12), which sits 7 meV/atom above the hull (new composition).
GGen ran 1,000 trials across 100 Bi-Co-Fe stoichiometries in 21.8 min, keeping 943 relaxed structures. 14 of them landed within 150 meV/atom of the convex hull, 13 of which the database had not seen before. The leading generated candidate is Co4Fe4 in Pm-3m (#221), which sits on the hull (known).
GGen ran 1,000 trials across 100 Bi-Co-Fe stoichiometries in 18.8 min, keeping 924 relaxed structures. 9 of them landed within 150 meV/atom of the convex hull, 8 of which the database had not seen before. The leading generated candidate is Co4Fe6 in P4/m (#83), which sits on the hull (new polymorph).