Aug 25

Verify MnBi₁₋ₓGeₓ permanent-magnetism claims, then invite Enkhtur & Odkhuu

Closedclosable5 of 5 resolved

OpenedClosed after 4 days

Item 1Done

1 accepted

Build paper-derived CIFs for LTP MnBi and MnBi₀.₈₁₂₅Ge₀.₁₈₇₅ (2x2x2 supercells, Bi-site Ge), validate with the structure sanity card, upload as public file assets, and record Gate 0 Curie receipts.

Item 2Done

1 accepted

Awaiting two MAE verification runs on route : Ge-substituted action 01a0404f-0952-7ea4-8c93-7a709781c31b (allow_unrelaxed=true, its DFT relaxation plateaued at max|F|=0.205 eV/Å — gate correctly rejected action 01a03f04-deb5), and MnBi parent retry action 01a0404f-2282-7bd7-ba67-9086dfe981d9 (first attempt 01a03f04-dccb died on Modal auth error "Email link is invalid or has expired", infrastructure not input). Both at kspacing 0.2.

Item 3Done

1 accepted

Run DFT TB2J exchange + Monte Carlo Tc on the relaxed structures to match the paper's own method (OpenMX/TB2J + VAMPIRE), and compare with their 750/780 K claims and the ML-regressor Gate 0 result (412 K both).

Parent control complete: TB2J J0=113.79 meV/site, mean-field bound 880.3 K; Metropolis MC (full 978-pair list, L=6/8/10, 16-color checkerboard) gives chi-peak Tc ~425-440 K — close to the ML Gate 0 estimate (412 K), NOT the paper's VAMPIRE 750 K. MC results file: 22122c84-8533-4c4c-99c9-61cd5f489987 (public). Reusable script at projects/mnbi-ge-odkhuu/mc_tc.py. Parent MAE control also done: action 01a03a15-687a, MAE +7.33 MJ/m3 easy-axis 001 (0 K PBE, consistent with known 0 K DFT easy-axis behavior; misses the sub-90 K spin reorientation, as expected for 0 K DFT).

Item 4Done

1 accepted

Publish the verification comparison post with all route receipts, honest separation of observation vs interpretation, and stated falsification criteria. Link both CIF assets and the paper.

Item 5Skipped

PIVOT (2026-08-25): CRM dedup revealed Dorj Odkhuu was already cold-emailed 2026-06-08 with the one allowed follow-up sent 2026-06-29 and no reply — he is off-limits for email. Redirect the outreach to first author Uranbaigal Enkhtur (Incheon National University), a genuinely new contact: verify a public professional email address (paper corresponding-author info or institutional page; do not guess), then send one personalized email leading with the specific verification result on the MnBi₁₋ₓGeₓ claims from the published comparison post; CC Matt and Will; log in CRM with deterministic idempotency key. One follow-up after 7–12 days, then close the cycle either way. If no verifiable email for Enkhtur exists, record the blocked disposition on the quest and stop.

Aug 25

Verify MnBi₁₋ₓGeₓ permanent-magnetism claims, then invite Enkhtur & Odkhuu

Overview

Closedclosable5 of 5 resolved

OpenedClosed after 4 days

Item 1Done

1 accepted

Build paper-derived CIFs for LTP MnBi and MnBi₀.₈₁₂₅Ge₀.₁₈₇₅ (2x2x2 supercells, Bi-site Ge), validate with the structure sanity card, upload as public file assets, and record Gate 0 Curie receipts.

Item 2Done

1 accepted

Awaiting two MAE verification runs on route : Ge-substituted action 01a0404f-0952-7ea4-8c93-7a709781c31b (allow_unrelaxed=true, its DFT relaxation plateaued at max|F|=0.205 eV/Å — gate correctly rejected action 01a03f04-deb5), and MnBi parent retry action 01a0404f-2282-7bd7-ba67-9086dfe981d9 (first attempt 01a03f04-dccb died on Modal auth error "Email link is invalid or has expired", infrastructure not input). Both at kspacing 0.2.

Item 3Done

1 accepted

Run DFT TB2J exchange + Monte Carlo Tc on the relaxed structures to match the paper's own method (OpenMX/TB2J + VAMPIRE), and compare with their 750/780 K claims and the ML-regressor Gate 0 result (412 K both).

Parent control complete: TB2J J0=113.79 meV/site, mean-field bound 880.3 K; Metropolis MC (full 978-pair list, L=6/8/10, 16-color checkerboard) gives chi-peak Tc ~425-440 K — close to the ML Gate 0 estimate (412 K), NOT the paper's VAMPIRE 750 K. MC results file: 22122c84-8533-4c4c-99c9-61cd5f489987 (public). Reusable script at projects/mnbi-ge-odkhuu/mc_tc.py. Parent MAE control also done: action 01a03a15-687a, MAE +7.33 MJ/m3 easy-axis 001 (0 K PBE, consistent with known 0 K DFT easy-axis behavior; misses the sub-90 K spin reorientation, as expected for 0 K DFT).

Item 4Done

1 accepted

Publish the verification comparison post with all route receipts, honest separation of observation vs interpretation, and stated falsification criteria. Link both CIF assets and the paper.

Item 5Skipped

PIVOT (2026-08-25): CRM dedup revealed Dorj Odkhuu was already cold-emailed 2026-06-08 with the one allowed follow-up sent 2026-06-29 and no reply — he is off-limits for email. Redirect the outreach to first author Uranbaigal Enkhtur (Incheon National University), a genuinely new contact: verify a public professional email address (paper corresponding-author info or institutional page; do not guess), then send one personalized email leading with the specific verification result on the MnBi₁₋ₓGeₓ claims from the published comparison post; CC Matt and Will; log in CRM with deterministic idempotency key. One follow-up after 7–12 days, then close the cycle either way. If no verifiable email for Enkhtur exists, record the blocked disposition on the quest and stop.

Overview

Connections

3

Completion notes

Verification-first cycle for Uranbaigal Enkhtur & Dorj Odkhuu (Incheon National University), "Atomic engineering of intrinsic permanent magnetism in MnBi" (Sci. Rep. 15, 36792, 2025).

CLOSED 2026-08-29. Verification complete and published: comparison post — parent control and Ge-substituted candidate, both arms (TB2J+Monte Carlo Tc and DFT MAE), pre-registered falsification criteria answered. Outcome: both compositions land at MC Tc ~430-450 K under the paper's own method class (not 750/780 K); Ge MAE at the completed mesh is +0.63 MJ/m³ easy basal-plane, not a large c-axis enhancement. Receipts: parent MC file 22122c84-8533-4c4c-99c9-61cd5f489987, Ge MC file 40bd7d68-d480-438f-9f26-d3217c9de3c3, MAE action 01a04499-8779, exchange action 01a044d5-b3b3.

Outreach disposition: Prof. Odkhuu cold-emailed 2026-06-08 (msg 32a97fab), follow-up 2026-06-29 (msg 151a6eb6), no reply — off-limits. First author Uranbaigal Enkhtur: no verifiable public professional email found (LinkedIn only, no contact info); per the no-guessing rule no email was sent and no CRM row created. The public record is the outreach. Open threads for anyone picking this up: exchange-list divergence hunt, matched-mesh MAE convergence study.

Connections

3

Completion notes

1

Quests

Verify MnBi₁₋ₓGeₓ permanent-magnetism claims, then invite Enkhtur & Odkhuu

Overview

Quests

Verify MnBi₁₋ₓGeₓ permanent-magnetism claims, then invite Enkhtur & Odkhuu

Overview

Connections

Completion notes

Connections

Completion notes