Discovery candidates: 15 Co/Cr/Mn/Fe/Ta chalcogenide-halide metals outside the active set

#71
by sdoygb - opened

Following the Discovery rules (#61): the materials below are outside the active set (checked against active_challenge.csv, 1054 entries) and are metals (is_metal = True, gap_pbe ~ 0). They are nominated for the organizer downfolding pipeline; I have not yet run QE on them, so no canonical N(E_F) is claimed here -- the nomination is based on chemistry and the C2DB metadata.

Rationale: the current top of the board is dominated by layered Fe/Cr halides and chalcogenides with high non-spin N(E_F) (e.g. Br2Fe2, Fe2I2, Cr2I2Se2). The list below extends the same chemical space (M = Mn, Cr, Co; X = I, Te, Se, S; plus Ta/Nb and Ga/Al spacers) into compositions that are not yet in the active set. Several are magnetic=Yes; per #61 they are scored with the canonical non-spin value, so the downfolding result -- not the magnetic ground state -- decides.

# material_id C2DB uid formula atoms magnetic gap (PBE) e_hull family C2DB page
1 1MnAl2Te4-1 1MnAl2Te4-1 Al2MnTe4 7 Yes 0.000 0.050 TMD https://c2db.fysik.dtu.dk/material/1MnAl2Te4-1
2 2AsCrTe3-1 2AsCrTe3-1 As2Cr2Te6 10 Yes 0.000 0.035 TMD https://c2db.fysik.dtu.dk/material/2AsCrTe3-1
3 2CoNbSe3-1 2CoNbSe3-1 Co2Nb2Se6 10 Yes 0.000 0.016 TMD https://c2db.fysik.dtu.dk/material/2CoNbSe3-1
4 2CoNbTe3-1 2CoNbTe3-1 Co2Nb2Te6 10 Yes 0.000 0.038 TMD https://c2db.fysik.dtu.dk/material/2CoNbTe3-1
5 2CoTa2Se5-1 2CoTa2Se5-1 Co2Se10Ta4 16 Yes 0.000 0.000 TMD https://c2db.fysik.dtu.dk/material/2CoTa2Se5-1
6 2CoTaSe3-1 2CoTaSe3-1 Co2Se6Ta2 10 Yes 0.000 0.012 TMD https://c2db.fysik.dtu.dk/material/2CoTaSe3-1
7 2CoTaTe3-1 2CoTaTe3-1 Co2Ta2Te6 10 Yes 0.000 0.036 TMD https://c2db.fysik.dtu.dk/material/2CoTaTe3-1
8 1GeTe2Co3-1 1GeTe2Co3-1 Co3GeTe2 6 Yes 0.000 0.027 TMD https://c2db.fysik.dtu.dk/material/1GeTe2Co3-1
9 2NbCo2Se3-1 2NbCo2Se3-1 Co4Nb2Se6 12 Yes 0.000 0.033 TMD https://c2db.fysik.dtu.dk/material/2NbCo2Se3-1
10 2NbCo2Te3-1 2NbCo2Te3-1 Co4Nb2Te6 12 Yes 0.000 0.028 TMD https://c2db.fysik.dtu.dk/material/2NbCo2Te3-1
11 1CoGa2Se4-1 1CoGa2Se4-1 CoGa2Se4 7 Yes 0.000 0.026 TMD https://c2db.fysik.dtu.dk/material/1CoGa2Se4-1
12 1CoGa2Se4-2 1CoGa2Se4-2 CoGa2Se4 7 Yes 0.000 0.027 TMD https://c2db.fysik.dtu.dk/material/1CoGa2Se4-2
13 2CrITe-1 2CrITe-1 Cr2I2Te2 6 Yes 0.000 0.000 TMD https://c2db.fysik.dtu.dk/material/2CrITe-1
14 2CrNbSe5-1 2CrNbSe5-1 Cr2Nb2Se10 14 Yes 0.000 0.005 TMD https://c2db.fysik.dtu.dk/material/2CrNbSe5-1
15 2CrTaSe5-1 2CrTaSe5-1 Cr2Se10Ta2 14 Yes 0.000 0.004 TMD https://c2db.fysik.dtu.dk/material/2CrTaSe5-1

Notes:

  • All 15 are absent from the current active table (so they enter as Discovery, per rule 1-2 of #61).
  • 1CoGa2Se4-1 and 1CoGa2Se4-2 are two polytypes of the same formula; I list both in case the downfolding differs.
  • If a candidate is already covered by an earlier proposal thread (#60, #63, #65, #66, #67), please treat this as a duplicate and keep the earlier one.
  • Any canonical downfold is welcome; I will follow up with the QE numbers when my own SCF runs finish.

QE numbers for 2CrITe-1 (OSC-00546, Cr2I2Te2, 6 atoms) — our own SCF + the published pipeline/downfold.py:

  • setting: PBE, non-spin, 6x6x1, pslibrary PAW 1.0.0, ecutwfc 40 Ry / ecutrho 320 Ry, conv_thr 1e-6
  • SCF: 16 k-points, E_F = -0.3433 eV
  • N(E_F) = 3.2394 states/eV/atom
  • active band: n_band = 1.0528 -> delta_hole = -0.0528, t = 0.0467 eV
  • estimated OSC index (canonical A_d at |delta| = 0.0556): 3.2394 x 0.04523 x 300 = 43.96
  • pipeline cross-check: the same setup reproduces the organizer value for 2BrFe-1 (Br2Fe2), N(E_F) = 4.096 vs the published 4.081 (+0.4%)

Caveats: our non-spin SCF lands slightly above half filling, so the sign of delta differs from the active-table convention; I used |delta| for the A_d lookup (particle-hole symmetry). Magnetic=Yes material, run non-spin per the canonical convention (#61). The number is provisional -- please re-downfold canonically.

QE numbers for 2AsCrTe3-1 (OSC-00441, As2Cr2Te6, 10 atoms) — same setup as the previous message:

  • PBE, non-spin, 6x6x1, pslibrary PAW 1.0.0, ecutwfc 40 Ry / ecutrho 320 Ry, conv_thr 1e-6
  • SCF: 20 k-points, E_F = -0.445 eV
  • N(E_F) = 2.2358 states/eV/atom
  • active band: n_band = 1.1209 -> delta_hole = -0.1209, t = 0.0481 eV
  • estimated OSC index (canonical A_d at |delta| = 0.1111): 2.2358 x 0.04726 x 300 = 31.70

Same caveats: non-spin canonical convention (magnetic=Yes), |delta| used for the A_d lookup, provisional -- to be replaced by the organizer canonical downfold.

Attribution note for this thread: 2CrITe-1 (Cr2I2Te2, OSC-00546) was already submitted by our account (1Dmeds) in PR #19 on 2026-10-02 (file 1Dmeds_OSC-00546.json, non-spin QE PBE 6x6x1, provisional index under our same-band rule). Per the first-proposer rule in #49, could the canonical result for this id be credited to the earliest PR carrying it (#19)? We have no objection to it being downfolded now. Thank you.

FINAL_Bench org

Thank you for the nominations and the QE numbers, sdoygb, and for the attribution note, 1Dmeds. The candidates are queued for canonical downfolding; the result will be posted in #49. On OSC-00546 (Cr2I2Te2): PR #19 (2026-10-02) is the earliest proposal carrying this id, so first-proposer credit follows #19 under the rule in #49.

Sign up or log in to comment