The preregistered CTRL-NIO-1 used a supercell embedding that was not magnetically compensated at the primitive level. The amended control uses the minimal Type-II embedding:
2x1x1 of the primitive fcc rock-salt cell, 4 atoms: Ni1 (0,0,0), Ni2 (1/2,1/2,0), O1 (1/2,0,0), O2 (0,1/2,1/2)
a = 4.171 A (experimental, V = 18.14 A^3 per NiO f.u.)
Seed [+2, -2] uB on the two Ni sites, net moment 0
Hubbard U on Ni 3d: 6.2 eV; kspacing 0.3 1/A; scf_thr 1e-6 Ha; scf_nmax 200.
Criterion | Bar | Amended Run | Pass |
|---|---|---|---|
Net moment | compensated | 0.002 uB/cell | yes |
Ni sublattice moments | antiparallel retained | +1.13 / -1.14 uB | yes |
Ms | < 0.1 T | 0.0004 T | yes |
SCF convergence | 1e-6 Ha within 200 | converged, step 147 | yes |
Amended control PASSES the same physical bar as the preregistered control. The route correctly reproduces the compensated antiferromagnetic ground state of NiO in the minimal Type-II embedding: sublattice moments ~1.1 uB (within the expected DFT+U range 1.0-1.3 uB), net Ms effectively zero.
The preregistered CTRL-NIO-1 failure stands as recorded and is NOT redefined. The amended run is appended to the reference dataset as a distinct control-diagnostic row with run_role=control_negative_amended, demonstrating that the original failure was attributable to the embedding geometry rather than to the route's magnetic engine.
Action: amended NiO control, 4-atom Type-II cell
Dataset row: control_negative_amended (appended, distinct from CTRL-NIO-1)