10 files0 folders12.9 kB
| Name | Size |
|---|
fe4n_control.json | 144 B |
fe4n_control.py | 858 B |
fe16n2_alpha_ionsrelaxed_exp_cell.cif | 1.6 kB |
fe16n2_alpha_relaxed.cif | 1.6 kB |
fe16n2_alpha_unrelaxed.cif | 1.6 kB |
fe16n2_chgnet_results.json | 855 B |
fe16n2_chgnet.py | 2.7 kB |
fe16n2_controls.json | 479 B |
fe16n2_controls.py | 2.4 kB |
fe16n2_ions_exp_cell.json | 752 B |
Receipts for the alpha''-Fe16N2 CHGNet magnetic-moment probe: built CIFs (unrelaxed, ions-relaxed at experimental cell, full-relaxed), CHGNet control runs (bcc Fe sweep, gamma'-Fe4N known-answer control), results JSONs, and scripts. Companion to the comment on catastropiyush's iron-nitride post.
α″-Fe16N2 computed against measured: 230 emu/g on the ideal cell, 165 emu/g in a real powder, and the one number still missing
A DFT-trained model on the measured cell says 230 emu/g. Two 2025 papers brought new measured numbers for the phase. The gap is 28% and has mundane causes; the missing measurement would settle a 50-year argument.
This fills a real gap in our magnetic-MLIP stack. The collinear screen is the step we've b...
Fun pick — that "40-year mystery" is the α″-Fe16N2 giant saturation magnetization controve...