{"id":"875b4000-b949-4c33-8243-eb278f6b6dd0","arxiv_id":"2608.02925","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"A symmetry-first high-throughput DFT+U screen of the MC2D database predicts 24 exfoliable two-dimensional altermagnets, 20 of which are new candidates.","lead":"This paper screens 2,710 two-dimensional materials from an existing database and predicts that 24 of them, 20 for the first time, can be exfoliated as altermagnets, a magnetic phase useful for spintronics. A reader interested in new magnetic materials for future electronics gets a concrete candidate list and an automated screening recipe.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Ground-state labels are not established because the energy comparison only includes altermagnetic and ferromagnetic configurations; non-altermagnetic AFM and noncollinear orders are never computed.","rationale":"The reader's weakest assumption correctly identifies the configuration-space restriction as the load-bearing step. The central claim requires the searched space to contain the true magnetic ground state, but the workflow explicitly scans only altermagnetic configurations plus FM, so non-AM AFM and noncollinear orders are absent. This concern lands as a real limitation, and it is correctable with additional calculations on a subset of candidates. I do not base the verdict on the four negative stability entries alone, because the table caption and the text can be made consistent by reading the column as E_NAM - E_AM; the notation is ambiguous but the numerical story is plausible. The arithmetic inconsistency in the candidate counts (42 -> 36 -> 3 exclusions -> 24, versus Methods' 42 self-consistent U values) is real and should be fixed, but it is secondary to the missing competitor search. The paper contributes a reusable symmetry-first workflow and a candidate list, and the self-consistent U pipeline is a genuine methodological step, so conditional acceptance with a request for the additional magnetic-order checks and a corrected count is the appropriate outcome.","tokens_in":17665,"tokens_out":15058,"duration_ms":133529,"concrete_test":"For a representative subset (e.g., CuH8N2(OF2)2, SrCo2Te3(ClO4)2, Fe2Si2SbO9, CoBrO, and Fe2SeTe), enumerate all collinear magnetic orderings compatible with the primitive magnetic cell (extending to a 2x1 or 2x2 supercell where necessary) plus a noncollinear calculation such as a constrained spin spiral or full vector relaxation, using the same PBE+U settings, pseudopotentials, and unrelaxed structures as in Table I at the self-consistent Hubbard U. If any non-AM AFM or noncollinear state is lower than the reported AM state by more than 1 meV/atom, the '24 refined ground-state altermagnets' claim is overstated and the count should be revised; if the AM state remains lowest or within 1 meV/atom in every tested case, the ground-state label for those compounds is supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that 24 monolayers are refined ground-state altermagnets. The workflow (Section I.A and Figure 2, AMPWScanWorkChain) computes DFT+U energies for the symmetry-generated altermagnetic configurations plus one ferromagnetic configuration, and defines E_GS as the lowest energy among all configurations considered for the same U. No non-altermagnetic AFM order and no noncollinear or incommensurate state is calculated. Therefore the |E_AM - E_GS| < 1 meV/atom criterion is evaluated against a configuration set whose only non-AM member is FM, and the Table I 'energetic stability' column cannot establish that the AM state is the lowest non-AM magnetic state. If any conventional AFM or spiral order lies lower than the AM state by more than 1 meV/atom at the self-consistent U, the ground-state label for that candidate is wrong. This is a gap between the searched configuration space and the claimed ground state, not merely a DFT accuracy issue. The manuscript's own count is also hard to audit: Section I.B reports 42 candidates, self-consistent U for 36, three exclusions, and then 24 final materials, while Methods states U values were obtained for all 42; the missing 9 are unexplained. The sign convention of the stability column (AM - NAM) is ambiguous, so the four negative entries should not be used as the primary basis for the verdict.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a high-throughput computational search for two-dimensional altermagnets in the MC2D database (2710 monolayers). The screening is symmetry-first: candidates are filtered by chemical composition, compensated magnetization, and the absence of connecting symmetries (I, T, C2z, Mz) between opposite-spin sublattices, yielding 79 easily exfoliable candidates. For these, the authors run DFT+U calculations over a grid of Hubbard U values (0–8 eV) for all symmetry-generated altermagnetic configurations plus one ferromagnetic configuration, and apply quantitative criteria on magnetization compensation, spin splitting magnitude, zero splitting at Gamma, and energetic proximity to the lowest-energy calculated state (|E_AM - E_GS| < 1 meV/atom). This stage produces 42 altermagnetic candidates. A refinement stage determines self-consistent Hubbard U values from DFPT, re-evaluates the magnetic states, and applies structural relaxation, resulting in a final set of 24 'refined ground-state altermagnets', including 4 previously reported and 20 new. The paper highlights Fe2SeTe, Fe2Si2SbO9, and CoBrO with large spin splittings, and provides an automated AiiDA workflow and a database for future use.","tokens_in":17860,"tokens_out":6069,"duration_ms":50842,"significance":"If the ground-state labels are correct, this work substantially expands the known family of 2D altermagnets and introduces a useful methodological template: symmetry analysis before expensive magnetic ground-state searches. The use of self-consistently determined Hubbard U from DFPT, the openness of the workflow, and the direct computation of momentum-dependent spin splittings are concrete strengths. The paper is likely to be influential for the 2D magnetism and spintronics communities. However, the central claim that the 24 materials are ground-state altermagnets is currently weakened by an incomplete search over magnetic configurations and by several unresolved numerical inconsistencies in the manuscript. These issues are fixable but must be addressed before publication.","major_comments":[{"comment":"The ground-state criterion is evaluated against a restricted set of magnetic configurations. The AMPWScanWorkChain computes DFT+U energies only for the symmetry-generated altermagnetic configurations and one ferromagnetic configuration, and E_GS is defined as the lowest-energy configuration among those considered for the same U. Non-altermagnetic antiferromagnetic orders (e.g., conventional stripe or zigzag AFM) and noncollinear or incommensurate states are never computed. Therefore the abstract's claim of 'reliably capture[ing] magnetic ground states' and the Table I label 'refined ground-state altermagnets' are not established unless the authors either extend the energy competition to include explicit non-altermagnetic AFM configurations for the 24 finalists, or rephrase the claims as 'lowest-energy among the calculated collinear AM and FM states.' This is a load-bearing issue because the numerical highlight of the paper (24 robust ground-state altermagnets) depends on it.","section":"Section I.A and Figure 2"},{"comment":"The material counts are internally inconsistent. Section I.B states that self-consistent U values are obtained for 36 of the 42 candidates, that three materials are excluded during refinement (Co2NO6, Co2TaTe2, CoH2SeO4), and that 24 materials remain. However, 36 minus 3 is 33, not 24; the fate of the other 9 materials is never explained. The Methods section compounds the inconsistency by saying the automated Hubbard workflow converged for 24 materials and the remaining 18 were done manually, implying all 42 received U values, while later listing several materials (TlCr4BiO14, FePSe3, Cu2Fe4S7, etc.) for which U could not be completed. The paper must reconcile these numbers (42, 36, 24, plus the individual exclusions) so that the final count is auditable.","section":"Section I.B and Methods"},{"comment":"The sign convention of the 'energetic stability' column is contradictory. The caption defines ΔE_min_AM-NAM as the energy difference between the altermagnetic ground state and the lowest-energy non-altermagnetic magnetic state. Under this definition, a negative value means the altermagnetic state lies lower in energy and is therefore the ground state. However, the text in Section I.B states that the negative values for CuH8N2(OF2)2, SrCo2Te3(ClO4)2, V3(H3O5)2, and Al2CuCl8 'indicate that the altermagnetic state for these materials is not the lowest-energy state.' This is backwards under the stated definition. The authors should either flip the sign in the table/definition or correct the textual interpretation; the classification of these four materials as refined ground-state altermagnets should be re-examined once the convention is fixed.","section":"Table I and Section I.B"},{"comment":"The transition from 'three materials are excluded' to '24 materials are confirmed' leaves 9 unaccounted systems. Even if the missing 9 are removed because they no longer satisfy the altermagnetic criteria at their self-consistent U, the manuscript should state this explicitly and list them. As written, the paper cannot be independently audited from the information given in the main text.","section":"Section I.B"}],"minor_comments":[{"comment":"The caption reports ΔE in meV/magnetic atom, while the screening criterion in Section I.A uses meV/atom; please clarify whether these are the same normalization and how the number of magnetic atoms is counted for each compound.","section":"Table I caption"},{"comment":"In the criterion list, the expression '|E AM−E GS|<1 meV/atom' and the phrase 'the last criterion i.e.,|E AM−E GS|<1 meV/atom' should be reformatted with proper spacing for readability.","section":"Section I.A"},{"comment":"The starting magnetization values are denoted 'm_i = 0.1, 0.5' but the units are not given; please indicate whether these are μB per magnetic site or normalized values.","section":"Figure 2"},{"comment":"The claim that 24/2710 is a 'significantly larger fraction' than prior searches is not quantified; adding the comparison with the 2/194 result of Haddadi et al. would make this statement more precise.","section":"Abstract and Summary"},{"comment":"The Methods section describes both a 'balanced protocol' for the AiiDA Hubbard workflow and a manual DFPT iteration with a convergence threshold of 0.1 eV; please state whether the two approaches give consistent U values for the systems where both were applied.","section":"Methods"}],"recommendation":"major_revision","confidential_remarks":"The paper is a worthwhile computational contribution with a strong methodological core: the symmetry-first approach, the use of self-consistent DFPT Hubbard U, and the open workflow are all positive features. The main unresolved problems are the incomplete configuration space used to define ground states, the numerical inconsistencies in the candidate counts, and the sign error in the stability column. These are fixable within the scope of a revision, but they affect the central claim and must be addressed before the paper can be accepted. I would not reject the manuscript, but I would not accept it in its current form."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [Colleague],\n\nTwo things you should know about this paper. The symmetry-first screening is a real contribution, and the candidate list is probably worth keeping on your desk. But the central claim that all 24 materials are 'refined ground-state altermagnets' is not established by the evidence as presented.\n\nWhat is genuinely good: the workflow inverts the usual pipeline. Instead of relaxing every magnetic ground state first, they use symmetry to generate candidate altermagnetic states before any DFT, which lets them cover larger unit cells than earlier screens. The Hubbard U is determined by DFPT, not fitted to the target property, so the circularity burden is low. They find 20 new candidate formulas, and two of them are genuinely notable: metallic Fe2Si2SbO9 with ~294 meV splitting near EF, and insulating CoBrO with 330 meV splitting. Prior work is cited fairly.\n\nThe soft spots are real. Most important: the energy comparison that defines the 'ground state' only includes altermagnetic configurations plus one ferromagnetic configuration. No conventional AFM order, no noncollinear or incommensurate state is computed. So |E_AM - E_GS| < 1 meV/atom shows the AM state beats the FM state, not that it is the true magnetic ground state. That is a gap between the searched configuration space and the claimed label. For a few top candidates, checking other collinear and possibly noncollinear orders would tighten the claim a lot.\n\nThe internal counts also do not add up. Results say self-consistent U was obtained for 36 of 42, three more are excluded, leaving 24, but 36-3=33. Methods implies all 42 got U (24 automated, 18 manual). Some of this may be presentation, but as written the reader cannot audit the funnel. And the 'energetic stability' column: four materials have negative ΔE_AM-NAM, and the text says they are not the lowest-energy state, yet they are still in the 'ground-state altermagnet' list. That label overstates a threshold. Being within 1 meV/atom of the ground state is a useful screening criterion, but it is not the ground state.\n\nMinor items: Fe2SeTe is a substitutionally disordered alloy modeled as an ordered crystal; the authors give a reasonable justification but experiments may not honor the ordering. The code and data are promised, not yet released, so reproducibility is currently unverifiable.\n\nWho is this for? Experimental groups looking for 2D spintronic candidates and people building high-throughput magnetic workflows. The list is a useful shortlist even after you discount the 'ground-state' wording. I would send it to peer review, with a clear request for revision: fix the counts, relabel or reclassify the four negative-stability entries, add noncollinear checks for the top candidates, and release the workflow and database with versioned identifiers.\n\nMy recommendation: engage with it, but don't cite the 24 as established ground states until the configuration space is widened and the numbers are reconciled.","headline":"Useful symmetry-first candidate list, but 'ground-state' labels outrun the computed configuration space and the funnel counts don't add up.","tokens_in":18493,"tokens_out":4618,"would_cite":true,"duration_ms":40111,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A symmetry-first computational screen of 2710 exfoliable monolayers finds 24 two-dimensional altermagnets, 20 of them new predictions.","keywords":["altermagnetism","two-dimensional materials","high-throughput screening","Hubbard U","density functional theory","spin splitting","magnetic ground states","exfoliable monolayers"],"falsifier":"Pick one candidate, say CoBrO or Fe2Si2SbO9, and determine its magnetic order experimentally in an exfoliated monolayer using neutron or resonant x-ray diffraction to check collinearity and compensation, and its band structure using spin-resolved photoemission to check for momentum-dependent splitting; alternatively, compute total energies allowing noncollinear or spin-spiral states and check whether any falls below the reported altermagnetic configuration. A noncollinear ground state or a lack of the predicted spin-split bands would falsify the altermagnet label for that material.","tokens_in":1916,"feed_emoji":"🧲","tokens_out":2288,"duration_ms":81348,"temperature":0.7,"pith_summary":"The paper claims that a symmetry-first computational screen of 2710 exfoliable two-dimensional crystals can find altermagnets before any magnetic ground-state energy is calculated, and that this approach, combined with density-functional theory plus self-consistently determined Hubbard U corrections, identifies 24 monolayers with altermagnetic ground states, 20 of them new predictions. Altermagnets are collinear antiferromagnets whose spin splitting varies with momentum, combining advantages of ferromagnets and antiferromagnets for spintronics without stray fields. The significance is practical: the 24 candidates are all easily exfoliable from known bulk crystals, and several show large spin splittings, up to 487 meV, making them immediate targets for device experiments. The paper also claims its automated workflow makes the discovery reproducible and extendable to other material families.","feed_headline":"Symmetry-first search finds 24 2D altermagnets, 20 new","feed_subtitle":"Verified with self-consistent Hubbard U, spin splittings reach 487 meV in exfoliable monolayers.","key_machinery":"The load-bearing object is the symmetry-guided screening protocol, specifically the step that generates all spin configurations consistent with altermagnetic symmetry and rejects those connected by inversion, translation, the out-of-plane twofold rotation C2z, or the horizontal mirror Mz. An altermagnet is a collinear compensated magnet whose opposite-spin sublattices are linked by a rotation, so spin-up and spin-down bands split in momentum space while remaining degenerate at certain points. The protocol combines this symmetry analysis with density-functional theory plus Hubbard U scans over a range of U values and two starting magnetization moments, then applies quantitative criteria for compensation, splitting size, and energetic proximity. In the refinement stage, U values are recomputed self-consistently with density-functional perturbation theory and structures are relaxed, which removes 18 of the 42 initial candidates and yields the final 24. This sequence is what allows the search to be broad, covering all 2710 monolayers, yet still land on ground-state altermagnets rather than merely symmetry-permitted ones.","core_discovery":"The central claim is that 24 monolayers are refined ground-state altermagnets: they have compensated magnetization (net moment below 0.01 μB per cell), altermagnetic spin splitting (maximum splitting above 5 meV, splitting at the Γ point below 5 meV), and an altermagnetic configuration within 1 meV per atom of the lowest magnetic state, all evaluated with Hubbard U parameters determined self-consistently from first principles. Four of these, RuF4, VF4, MnMoTeO6, and Fe2SeTe, were previously reported; twenty are new. The screening inverts the usual order, applying symmetry analysis to all 2710 monolayers before any energy calculation, so that no symmetry-permitted altermagnet is excluded because its unit cell is too large to scan. After the high-throughput stage identifies 42 candidates for at least one U value, refinement with self-consistent U and structural relaxation leaves the final 24. Prominent examples are metallic Fe2Si2SbO9 with a 294 meV splitting near the Fermi level and insulating CoBrO with a 330 meV splitting.","pith_inferences":["The screening tests only collinear spin configurations, so any candidate whose true magnetic ground state is noncollinear or incommensurate would not actually be a ground-state altermagnet; extending the workflow to noncollinear order is a direct test.","Because the refinement uses a Hubbard-corrected density functional with self-consistent U, switching to hybrid functionals or adding spin-orbit coupling could shift energy differences and splitting magnitudes, though the symmetry classification itself should persist.","The transition-metal filter is conservative; compounds containing other magnetic elements, such as AgF2, may also be two-dimensional altermagnets, so the 24 candidates should be viewed as a lower bound.","If Fe2SeTe behaves as a doped FeSe analogue, it may bridge altermagnetism and superconductivity in one material, a possibility the paper raises but does not establish."],"forward_implications":["Twenty new exfoliable monolayers become candidate two-dimensional altermagnets, considerably expanding the known set relative to earlier screens of the same database.","The metal Fe2Si2SbO9 offers a 294 meV spin splitting near the Fermi level, a usable scale for spin-transport devices, while insulating CoBrO combines a 1.31 eV gap with a 330 meV splitting for spin-filtering.","The symmetry-first workflow can be applied to other crystal databases without the small-unit-cell constraint that limited earlier searches.","Because the candidates are easily exfoliable, they are ready for van der Waals heterostructure experiments where gating, proximity effects, and moiré engineering could tune altermagnetic order.","The energetic-stability criterion admits a few materials, such as CuH8N2(OF2)2 and SrCo2Te3(ClO4)2, whose altermagnetic state is not the lowest-energy state, so their classification as ground-state altermagnets rests on the 1 meV per atom tolerance."],"supporting_citations":[{"why":"Supplies the database of experimentally derived exfoliable two-dimensional crystals that the screening starts from.","marker":"[56]"},{"why":"Expands that database to the 2710 monolayers used in this search.","marker":"[57]"},{"why":"Provides the symmetry-analysis code adapted to generate altermagnetic spin configurations and enforce two-dimensional symmetry constraints.","marker":"[70]"},{"why":"Prior high-throughput screen whose small-unit-cell limitation motivates the symmetry-first approach.","marker":"[53]"},{"why":"Density-functional perturbation theory method for computing Hubbard U parameters self-consistently.","marker":"[61]"},{"why":"Implements the density-functional perturbation theory Hubbard U calculation used in the refinement stage.","marker":"[62]"},{"why":"Defines the moderate protocol for k-point sampling and convergence settings in the high-throughput calculations.","marker":"[72]"},{"why":"Automated workflow for first-principles Hubbard parameters used to obtain self-consistent U values.","marker":"[73]"}],"fun_headline_variants":["2710 monolayers screened, 24 altermagnets, 20 new","Symmetry-first screen nets 20 new 2D altermagnets","Hubbard-corrected DFT finds 24 stable 2D altermagnets","20 novel 2D altermagnets from 2710-material screen","2D altermagnet hunt finds 20 new monolayers"],"cache_read_input_tokens":20480,"weakest_assumption_plain":"The list of 24 ground-state altermagnets stands on the assumption that the only magnetic configurations that matter are collinear and that density-functional theory with a Hubbard correction and self-consistently determined U ranks their energies correctly; if a candidate's true magnetic ground state is noncollinear, incommensurate, or otherwise absent from the calculated set, that candidate is not actually a ground-state altermagnet.","fun_headline_variants_meta":{"raw":{"variants":["2710 monolayers screened, 24 altermagnets, 20 new","Symmetry-first screen nets 20 new 2D altermagnets","Hubbard-corrected DFT finds 24 stable 2D altermagnets","20 novel 2D altermagnets from 2710-material screen","2D altermagnet hunt finds 20 new monolayers"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001087,"raw_usage":{"total_tokens":4601,"prompt_tokens":1063,"completion_tokens":3538,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":679,"completion_tokens_details":{"reasoning_tokens":3436}},"tokens_in":679,"tokens_out":3538,"duration_ms":25362,"temperature":1.0,"reasoning_tokens":3436,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T14:55:12.842388+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Pick one candidate, say CoBrO or Fe2Si2SbO9, and determine its magnetic order experimentally in an exfoliated monolayer using neutron or resonant x-ray diffraction to check collinearity and compensation, and its band structure using spin-resolved photoemission to check for momentum-dependent splitting; alternatively, compute total energies allowing noncollinear or spin-spiral states and check whether any falls below the reported altermagnetic configuration. A noncollinear ground state or a lack of the predicted spin-split bands would falsify the altermagnet label for that material.","supporting_citations":[{"cited_title":"Mounet, M","cited_arxiv_id":null,"evidence_quote":"Supplies the database of experimentally derived exfoliable two-dimensional crystals that the screening starts from."},{"cited_title":"Campi, N","cited_arxiv_id":null,"evidence_quote":"Expands that database to the 2710 monolayers used in this search."},{"cited_title":"Smolyanyuk, L","cited_arxiv_id":null,"evidence_quote":"Provides the symmetry-analysis code adapted to generate altermagnetic spin configurations and enforce two-dimensional symmetry constraints."},{"cited_title":"Timrov, N","cited_arxiv_id":null,"evidence_quote":"Density-functional perturbation theory method for computing Hubbard U parameters self-consistently."},{"cited_title":"Timrov, N","cited_arxiv_id":null,"evidence_quote":"Implements the density-functional perturbation theory Hubbard U calculation used in the refinement stage."},{"cited_title":"de Miranda Nascimento, F","cited_arxiv_id":null,"evidence_quote":"Defines the moderate protocol for k-point sampling and convergence settings in the high-throughput calculations."},{"cited_title":"Bastonero, C","cited_arxiv_id":null,"evidence_quote":"Automated workflow for first-principles Hubbard parameters used to obtain self-consistent U values."}],"review_version":1}