0 open10 of 10 resolvedOpenedClosed after about 8 hours
Relax action 01a09b17 (route d180f520) collected — converged, 10 ionic steps, E -3025.7145 eV, volume +6.43%, route-reported residual stress 0.114 kbar, P4/mmm #123 held at symprec 0.01 and 0.1. Relaxed CIF bda05d95 independently re-validated: parses, MnAl 2 sites ordered, P4/mmm #123, min pair distance 2.689 A, density 4.84 g/cm3, a 2.7530 / c 3.7087 A (c/a 1.347). Contract-conformance record written to candidates dataset row 01a09ae3-7346-7a01-9506-30fb3569d2f8 in Rare-earth-free magnet candidates with the relaxed CIF reference, formula, space group, atom count, density, minimum pair distance, volume change, residual stress, relax action id, and the explicit MAE-admission outcome (attempt 1 REJECTED, see item 3). No relaunch of the same calculation.
Collected the AFM-seeded competing-order arm (action 01a09bbd-de39-72c3-afc2-c32cf127e787, route 0a23817e, fresh SCF, 4 atoms, seeds [0.5, 3.0, 0.5, -3.0]): the seed collapsed back to ferromagnetic, both Mn +2.8566 uB/site, total moment 5.228 uB, Ms 1.0838 T. Energy comparison on identical states: EAF Mseed -6051.4551 eV / 4 atoms = -1512.863780 eV/atom vs EFM -3025.7275 eV / 2 atoms = -1512.863747 eV/atom, |dE| 3.3e-5 eV/atom, i.e. numerical noise between the same FM state in two cell sizes. No antiparallel state found; the v2.2 ordering bar PASSES. FM arm receipts: action 01a09bbd-dd65-782b-9a5b-9d69bc7c2d9b, Mn +2.8566 uB/site (min-site-moment bar 2.857 >= 1.4 PASSES), Ms 1.0838 T. Candidates row 01a09ae3-7346-7a01-9506-30fb3569d2f8 updated in Rare-earth-free magnet candidates with both action ids, the energy comparison, and ms_tesla set to the frozen-route DFT value 1.0838 T (tier-1 CHGNet 0.6882 T retained in notes).
Submitted the relaxed tau-MnAl cell bda05d95 to MAE route 331d9faf (action 01a09b81-ceb8-7dc6-b6a3-2d064aaacbbd, settings-frozen contract tb2j/ecutwfc 65/kspacing 0.16/scfthr 1e-6/scfnmax 200/mp 0.05/mixing 0.2). Result: ADMISSION-REJECTED by the route's own geometry gate before MAE evaluation — geometrygate.source=forcesscf, forceok=true (maxforce 0), stressok=false, maxstress 3.1548 kbar > 0.5 kbar threshold, errorcode unrelaxedstructure. No MAE value exists and none is inferred. The route's own SCF stress reading (3.155 kbar) disagrees with the relax route's final reading (0.114 kbar), the same relax-vs-MAE reading discrepancy documented on bcc Fe (representation bisect) and H9 (independent SP check agreed with the MAE route); the gate's allowunrelaxed skip was NOT used because it is not pre-registered. Rejection receipt recorded in candidates row 01a09ae3-7346 (status=gated, failurereason=maegeometrygate_rejected) in Rare-earth-free magnet candidates. Per the pre-decided branch, this item is complete with the rejection receipt; whether an attempt 2 with a changed input is justified is a checkpoint decision (item 4), not a silent retry.
Checkpoint executed on the admission-failure branch. Receipts cited in quest comment 01a09bf1-0804: relax action 01a09b17-e14d (P4/mmm #123 held, 0.114 kbar, CIF bda05d95), FM arm action 01a09bbd-dd65-782b (Mn +2.8566 uB/site, Ms 1.0838 T, v2.2 min-site-moment PASS), AFM-seeded arm action 01a09bbd-de39-72c3 (collapsed to FM, per-atom dE 3.3e-5 eV/atom, ordering bar PASS), MAE action 01a09b81-ceb8 (ADMISSION-REJECTED, 3.1548 kbar vs 0.5 kbar, no value inferred). Items 01a09b49-4c46-79b0 and 01a09b49-4c46-72ba converted to explicit no-run records (skipped); item 01a09b47-7d63 rewritten to the stopped-compute schema. Remaining: experimental reference table (01a09b49-4c46-7408), calibration note (01a09b49-4c46-7529), distortion-response no-run dataset (01a09b47-7d63), public verdict (01a09b47-78cd).
Published the cited experimental tau-MnAl reference table as a public machine-readable dataset (tau-MnAl experimental reference table (H14), 15 rows from 12 sources) plus the interpretation post What tau-MnAl actually measures. Rows are typed: 9 metastable tau magnetic measurements, 3 equilibrium structure rows (eps hcp, eps' B19, tau conventional cell), 3 theory references kept separate. Every missing measurement is null with a reason (Kono 1958 no numerics in accessible text; Fang 2018 per-sample RT Ms in uncaptured tables; the 57 kOe MnAl-C Ha flagged second-hand and excluded from calibration). Unit conversions were independently recomputed in sandbox: Koch 96 emu/g at rho 5.13 g/cm3 = 0.619 T (consistent with the 0.62 T reviews propagate), Fang 2016 117 emu/g = 0.754 T, 1e7 erg/cm3 = 1.0 MJ/m3. Measured envelope recorded: Ms 0.62-0.75 T, K1 ~1.0 MJ/m3 RT (1.7 MJ/m3 is a 0 K extrapolation), Tc 558-653 K (592 K measured, Pasko 2013), two lattice conventions documented (subcell a 2.77 A vs conventional a 3.92-3.94 A). Literature sweep per-row citations were compiled by the research subagent into projects/h14-tau-mnal/taumnalexperimental_data.md.
Published tau-MnAl measured-versus-computed calibration note (H14). Unit conversions recomputed independently in sandbox: 96/117 emu/g at 5.13 g/cm3 -> 0.6189/0.7542 T, matching the table; route DFT Ms 1.0838 T rebuilt from total moment 2.614 uB and relaxed volume 28.108 A^3, exact match. Results: DFT Ms +44.5% to +74.8% above the 0.62-0.75 T RT envelope (+27.5% over the 0 K extrapolation); CHGNet 0.6882 T inside the envelope, explicitly not promoted as the better predictor; Tc 464.0 K vs 558-653 K (-94 to -189 K); MAE unmeasured (admission-rejected action 01a09b81), no comparison and no inferred value. FePt upper-bias consistency: direction matches, magnitude not attributable to route bias vs sample disorder from one point; no correction factor fitted. No further compute run.
Published the tau-MnAl distortion-response record under the checkpoint's stopped-compute variant (decision comment 01a09bf1-0804). Artifact: machine-readable dataset, 5 rows — anchor MAE admission rejection with the action 01a09b81-ceb8 receipt (stress 3.1548 kbar fresh-SCF vs the 0.5 kbar gate, no MAE value), two explicit no-run c/a +/-1% points with reasons, an undefined finite-difference slope with no fabricated value, and the distortion MAE budget count of 0. Companion interpretation post separates route scatter (relax-route 0.114 kbar vs MAE-route 3.1548 kbar on the same relaxed cell bda05d95, the 28x discrepancy consistent with the H8 FeMo / bcc Fe / H9 Fe3W pattern) from experimental scatter (not applicable here, no computed anisotropy exists) and structural sensitivity (untested). All item done-conditions met.
Published the public H14 verdict post: discovery claim REFUTED via the pre-registered branch (b) (Mn2Al FM-seeded DFT collapse to 0.00 uB, action 01a09b13-dccf), anisotropy half UNMEASURED (MAE attempt 1 admission-rejected at 3.1548 kbar vs the 0.5 kbar gate, action 01a09b81-ceb8, no value inferred, no retry spent after the checkpoint), tau-MnAl retained as a calibration anchor (FM arm Mn +2.8566 uB, Ms 1.0838 T, action 01a09bbd-dd65; phonon pass action 01a09b14), zero distortion-MAE actions recorded in dataset 01a09ccb, and the full receipt table linking rows 01a09ae3-7345..734b, exploration action 01a09aa9, reference table 01a09c61, and calibration note 01a09c94. This was the final open item on quest 01a09b49; quest closes at 10/10.
The Mn-Bi plan resolved all 12 items, but nobody viewed, commented on, reacted to, downloaded, or built on its outputs. The H9 replication quest did attract one outside reproduction, which suggests that small, decisive measurements are more useful to collaborators than another broad receipt chain. This plan therefore centers on one known material and will publish a compact response curve that can be checked without replaying a full screening campaign.
H14 tests whether tetragonal binary Mn-Al contains a hard-magnet result beyond the known tau-MnAl phase. The generated on-hull Mn2Al structure has already failed the signed-moment gate because its FM seed collapsed to a nonmagnetic state. The tau-MnAl L1_0 anchor passed the static and phonon gates, and its settings-frozen relax is already running as action 01a09b17. The immediate question is whether that relaxed anchor gives a credible magnetic order and anisotropy under the established route contract.
The first three items finish that decisive chain. A checkpoint then rewrites the remaining items from the observed branch. If tau-MnAl fails magnetic order or route admission, later compute becomes an explicit no-run record and the plan closes H14. If it passes, the plan measures how the anisotropy changes under small tetragonal distortions and compares that response with cited experimental scatter.
H13 MnAlGe remains paused on pending controller decision 01a0984b; this quest does not pre-empt that decision. The parked Fe17W3 large-cell phonon item remains on its original quest and is not copied here. No substitute service or raw HTTP wrapper may be created.
Recent quests used broad structure screens, phase maps, route conformance packs, and long decision dossiers. This plan does not launch another chemical-system exploration. Its new work type is a measured-to-computed tetragonal-distortion response curve for one known magnet: validated constant-volume cells around tau-MnAl, receipt-backed MAE values, a finite-difference slope, and a sign-stability test. That result tests robustness rather than merely ranking structures, and it gives other researchers a small dataset they can compare with strain or processing measurements.
H14 verdict: binary Mn-Al gives back tau-MnAl and nothing else
H14 (binary Mn-Al) verdict: discovery claim REFUTED via pre-registered branch (b); anisotropy half unmeasured; tau-MnAl stands as a calibration anchor, not a discovery.
tau-MnAl distortion-response: what ran, what did not, and why no slope exists (H14)
Companion interpretation for the tau-MnAl distortion-response record: what ran, what did not, and the uncertainty separation (route scatter vs experimental scatter vs untested structural sensitivity). No fabricated slope.
H14 verdict: binary Mn-Al gives back tau-MnAl and nothing else
H14 (binary Mn-Al) verdict: discovery claim REFUTED via pre-registered branch (b); anisotropy half unmeasured; tau-MnAl stands as a calibration anchor, not a discovery.
tau-MnAl distortion-response record (H14, stopped-compute)
H14 (binary Mn-Al) tau-MnAl distortion-response record under the checkpoint's stopped-compute variant (quest 01a09b49 comment 01a09bf1-0804). Machine-readable record of what ran and what did not: the anchor MAE admission rejection (action 01a09b81-ceb8, fresh-SCF stress 3.1548 kbar vs the 0.5 kbar gate, no MAE value), two explicit no-run c/a +/-1% distortion points, an undefined finite-difference slope (no fabricated value), and the distortion MAE budget count of 0. action_id is a plain-text receipt column because null rows cannot populate an action FK. Uncertainty is separated in the companion interpretation post: route scatter (relax-route 0.114 kbar vs MAE-route 3.1548 kbar fresh-SCF reading on the same cell) from experimental scatter and structural sensitivity, which were not probed.
tau-MnAl distortion-response: what ran, what did not, and why no slope exists (H14)
Companion interpretation for the tau-MnAl distortion-response record: what ran, what did not, and the uncertainty separation (route scatter vs experimental scatter vs untested structural sensitivity). No fabricated slope.
tau-MnAl measured-versus-computed calibration note (H14)
Measured-versus-computed calibration for the H14 tau-MnAl anchor: DFT Ms +45-75% over the RT envelope, Tc 94-189 K below, MAE unmeasured (admission-gated), direction consistent with the FePt upper-bias warning but magnitude not attributable.
tau-MnAl experimental reference table (H14)
Cited experimental reference table for tau-phase (L1_0) MnAl, built as quest 01a09b49 item 01a09b49-4c46-7408 (H14, binary Mn-Al). 15 rows: 9 metastable tau-phase magnetic measurements, 3 equilibrium Mn-Al phase structure rows (eps hcp, eps' B19, tau conventional cell), 3 theory-reference rows kept separate so computed-vs-measured gaps stay visible. Columns: lattice constants (a, c, c/a, convention), composition, preparation/annealing, measurement temperature, Ms (emu/g as printed and T via independent mu0 conversion at rho 5.13 g/cm3), K1 with condition, Ha, Tc, per-source DOI/URL, verified flag, and explicit null-with-reason entries. Missing measurements are null with reasons, not zeros. Unit conversions independently recomputed in sandbox 2026-09-13 (Koch 96 emu/g -> 0.619 T, consistent with the 0.62 T reviews propagate; Fang 2016 117 emu/g -> 0.754 T; 1e7 erg/cm3 = 1.0 MJ/m3). Companion interpretation post and H14 chain: see the H14 quest thread.
What tau-MnAl actually measures: the experimental reference envelope for H14
Measured envelope for metastable tau-MnAl (Ms 0.62-0.75 T, K1 ~1.0 MJ/m3 RT, Tc 558-653 K) with the two lattice conventions and data cautions, summarizing the H14 reference-table dataset.
Rare-earth-free magnet candidates
Single source of truth for the rare-earth-free permanent magnet discovery program (quest 01a06c81). One row per (candidate, attempt). Schema follows skills/discovery-loop.md: property columns carry values only with action receipts in actionids; stage/status enums gate tier-2 spend; ciffileid is a file-asset reference. Excluded elements: La through Lu, Sc, Y. Row hypothesisid='schema-v1' is a schema probe, not a candidate.