{"id":"ce2ea5af-7948-446b-bbd8-8eefc716885b","arxiv_id":"2608.13483","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"In Mn5Si3, the M-point exchange instability is much stronger than the zone-center E2g mode that would host altermagnetism, and epitaxial strain further suppresses it.","lead":"A combined Landau, first-principles, and Monte Carlo study of Mn5Si3 finds the dominant magnetic instability at the wavevector M, not at the zone center. The result argues that the proposed altermagnetic phase of thin-film Mn5Si3 is unlikely to be stabilized by moderate epitaxial strain.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"DLM reference state fixes Mn1 moments to zero, so the Gamma-vs-M ranking is never tested against finite Mn1 fluctuations that can renormalize the Mn2 exchange kernel.","rationale":"The paper is a careful, internally consistent computational study. The Landau analysis is valuable, the DLM exchange calculations are standard, the Monte Carlo reproduces the bulk ordering scale, and the conclusion is appropriately hedged: the altermagnetic phase is disfavored in a weakly strained bulklike film, not ruled out universally. The reader identified the DLM reference state as the weakest assumption, and I agree that this is the right area. My concern sharpens that assumption: the zero Mn1 moment is not merely a detail of the reference state, it removes an entire channel of Mn2-Mn2 exchange mediated by Mn1 fluctuations. The symmetry argument in Section II B rules out bilinear coupling of the AFM2-like Mn2 mode to Mn1 order parameters, but it does not rule out the q-dependent indirect exchange that survives after integrating out paramagnetic Mn1 moments. If Mn1 moments are finite in the true paramagnet, the relative heights of lambda_max(Gamma) and lambda_max(M) could shift, and the central negative claim would need revision. The proposed test is straightforward with the same Questaal DLM machinery and would settle whether the concern is real. Because there is no current evidence that the Mn1 channel changes the ranking, rejection is not warranted; a conditional acceptance, contingent on the Mn1-moment sensitivity test, is the appropriate verdict.","tokens_in":17708,"tokens_out":5612,"duration_ms":61953,"concrete_test":"Perform constrained-moment DLM calculations with the same LDA potential, setting Mn1 local moments to 1.0, 1.5, and 2.0 muB, and also consider a spin-polarized DLM state with Mn1 included as a magnetic species. Recompute J_alpha beta(q) and the eigenvalue spectra along Gamma-M-K-Gamma. Accept the robustness claim only if lambda_max(Gamma) remains below lambda_max(M) by roughly the same margin and remains the global minimum in the q_z = 0 plane. Report the gap lambda_max(M) - lambda_max(Gamma); a reduction of more than about 30% or a sign change would show the central conclusion depends on the Mn1-moment reference state. If a finite Mn1 moment is energetically unstable in DLM, that would support the paper's zero-Mn1-moment choice.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Section III A reports that the self-consistent DLM state has m_Mn2 = 2.54 muB (LDA) or 2.85 muB (GGA) and zero moment on Mn1, and the exchange eigenvalues in Figs. 3 and 4 are for the six Mn2 sublattices only. The central claim that lambda_max(Gamma) is a global minimum in the q_z = 0 plane and far below lambda_max(M) is therefore a property of a Mn2-only exchange kernel. That property is not protected by the symmetry argument in Section II B. The statement that AFM2-like Mn2 modes exert no bilinear exchange field on Mn1 forbids a linear L_Mn2-m_Mn1 coupling, but it does not forbid an indirect Mn2-Mn2 exchange generated by integrating out fluctuating Mn1 moments. Such a term, Delta J_alpha beta(q) approximately J_12(q) chi_Mn1(q) J_21(q), respects the full paramagnetic space group and renormalizes the Mn2 kernel at every wave vector. Since Mn1 atoms carry ordered moments in the low-temperature AFM1 phase, a fluctuating Mn1 moment or susceptibility in the paramagnet is physically plausible. The sensitivity tests vary the Mn2 local moment from 2.14 to 2.85 muB, but they never vary the Mn1 moment or susceptibility. Thus the robustness claim rests on an unexplored axis of the reference-state assumption.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper examines whether the altermagnetic phase proposed for thin-film Mn5Si3, with the same (1,-1,0) Mn2 intracell pattern as the bulk AFM2 phase but at the zone center, is supported by first-principles exchange calculations. The authors construct Landau theories for both the M-star AFM2 ordering and the zone-center E2g ordering, showing that both contain the same AFM2-like pattern but differ in symmetry and order-parameter structure. They then compute the disordered-local-moment (DLM) exchange kernel J_alpha beta(q) for the paramagnetic state using TB-LMTO within LDA, scaled LDA, and GGA, and find that the leading instability is at the M point for LDA-based potentials, whereas the GGA gives a K-point instability. In all cases the largest eigenvalue at Gamma belongs to the E2g representation, but it is far below the M-point eigenvalue and is a supposed global minimum in the q_z=0 plane. Monte Carlo simulations using the LDA exchange parameters yield an orthogonal 3M ordered state with T_N about 73 K, consistent with the bulk ordering scale. Epitaxial tensile strain representative of Mn5Si3 films on Si(111) further suppresses the Gamma-point mode and reduces the M-point eigenvalue. A magnetoelastic estimate explains the experimental exchange striction and supports the single-arm AFM2 selection. The paper concludes that the high-temperature Hall-active state in Mn5Si3 films cannot be a weakly strained perturbation of bulk stoichiometric Mn5Si3.","tokens_in":17953,"tokens_out":5552,"duration_ms":49530,"significance":"If the conclusions hold, this is an important negative result for the altermagnetism field, providing a concrete counterexample to the widely discussed altermagnetic candidate Mn5Si3 and a physically transparent explanation of its bulk magnetic ordering. The paper's strengths are the combination of Landau-theory analysis, parameter-free DLM exchange calculations with multiple potentials and local-moment variations, Monte Carlo simulations that reproduce the experimental ordering temperature scale, and a magnetoelastic consistency check using the measured exchange striction. The central negative claim is falsifiable: it predicts that no Gamma-point E2g instability appears in the DLM kernel of stoichiometric Mn5Si3 under moderate strain. The main weakness is that the DLM reference state fixes Mn1 moments to zero, and the robustness of the Gamma-versus-M ranking to finite Mn1 fluctuations is not tested.","major_comments":[{"comment":"The manuscript states that interactions mediated by Mn1 can only renormalize higher-order terms in the Landau free energy, but this is not correct for the quadratic exchange kernel. Integrating out finite Mn1 fluctuations generates an effective Mn2–Mn2 exchange contribution of the form ΔJ_{αβ}(q) ~ Σ_{γδ} J_{αγ}(q) χ_{γδ}(q) J_{δβ}(q), which respects the full paramagnetic symmetry and contributes to the quadratic kernel K(q) on the same footing as the direct Mn2 exchange. The DLM calculation sets the Mn1 moment to zero self-consistently, so this term is entirely absent from the computed eigenvalues, and the sensitivity analysis varies only the Mn2 moment (2.14–2.85 μB) and strain, never the Mn1 susceptibility. Since the Mn1 sites carry ordered moments in the low-temperature AFM1 phase, finite Mn1 fluctuations in the paramagnet are physically plausible. The central claim that λ_max(Γ) is a global minimum in the q_z=0 plane is therefore not yet shown to be robust against a relevant axis of the reference-state assumption. Please provide a quantitative test, such as a constrained DLM calculation with nonzero Mn1 moments or a perturbative estimate of ΔJ using a model Mn1 susceptibility, or explicitly narrow the conclusion to the case of vanishing Mn1 moments.","section":"§II.A, §II.B, §III.A"},{"comment":"The text claims that in all three potentials the largest eigenvalue λ_max(q) has a global minimum at Γ in the whole q_z=0 plane, but Fig. 3 only displays eigenvalues along a high-symmetry path (Γ–M–K–Γ). This path samples only the boundary of the irreducible wedge of the q_z=0 plane, not the interior. No numerical scan of the full plane or analytic argument is presented. If the interior was sampled, the claim should be stated with that evidence; otherwise the strongest conclusion supported by the figure is that Γ is the minimum along the displayed path. Because the global-minimum statement is part of the argument that Γ is uniquely disfavored, this gap should be fixed either by adding a full-plane map or by softening the wording.","section":"§III.C and Fig. 3"}],"minor_comments":[{"comment":"The symbol Q_{μν} is used both for the matrix of dot products and for its individual components; please clarify the notation, e.g., by defining Q as the matrix with elements Q_{μν}.","section":"§II.B, Eq. (10)"},{"comment":"The DLM value J2 ≈ +0.055 mRy has the opposite sign from the J2 ≈ −0.16 mRy reported in Ref. [14] for the symmetry-broken reference state. This sign difference is mentioned in passing but not discussed; a short comment on its origin would help the reader.","section":"Table I and §III.B"},{"comment":"The sign of d∆λ/dε1 is given as approximately −6.0 mRy, and Eq. (19) yields γ ≈ 0.27 eV. Please check that the sign convention is consistent with the definition of ε1 in Eq. (17) and with the statement that AFM2 ordering contracts the bonds between ordered sublattices.","section":"§V"},{"comment":"The analogy with the stacked triangular Ising antiferromagnet (STAFI) is helpful, but the order-parameter space here is continuous in spin space; the phrase 'identical to that of the STAFI model' might be clarified as 'identical in its E2g angular sector'.","section":"§II.C"},{"comment":"There are a few typographical issues, such as inconsistent use of 'AFM2' versus 'AF2' and the missing accent in 'Néel' in a few places. These are cosmetic and do not affect the science.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a timely controversy and the main calculation is carefully done. My major concern is the untested dependence of the central negative claim on the DLM assumption of zero Mn1 moments; this is a load-bearing point that can be addressed with an additional calculation. The global-minimum claim also needs more evidence or qualification. If these are resolved, the paper would be a strong contribution to the altermagnetism debate."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, this is the first calculation I've seen that directly ranks the M-point and Gamma-point magnetic instabilities in Mn5Si3, and the result is clear: within the DLM reference state, the Gamma E2g mode is far below the M-star AFM2 mode and gets even weaker under the epitaxial strain relevant to Si(111) films. Second, that negative conclusion is conditional on the DLM state, and the paper never tests the one axis that could change the ranking: finite Mn1 moments.\n\nThe paper does several things well. The Landau analysis is genuinely useful. Recognizing AFM2 as a permutation-odd eigenmode at a single M-star arm, and the Gamma-point altermagnetic pattern as a special direction in the E2g order parameter space, clarifies the relationship between the two phases. The DLM calculations are careful, with two parametrizations of LDA plus a scaled-potential sensitivity check, and the exchange kernel robustly puts M above Gamma. The Monte Carlo ordering temperature (about 73 K) is in reasonable agreement with the experimental 100 K. The magnetoelastic coupling estimate is a nice consistency check: the derived gamma, combined with the calculated C66, gives an exchange striction consistent with the neutron measurement. No parameter is fitted to the conclusion.\n\nThe soft spots are real but not fatal. The DLM reference state fixes Mn1 to zero moment, and the sensitivity tests only vary the Mn2 moment. The symmetry argument forbids a bilinear coupling between the Mn2 AFM2-like mode and Mn1 order parameters, but it does not forbid Mn1-mediated Mn2-Mn2 exchange: integrating out fluctuating Mn1 moments gives a contribution of the form J_12 chi_Mn1 J_21 that respects paramagnetic symmetry and renormalizes the kernel at every q. Since Mn1 orders in the AFM1 phase, a finite Mn1 susceptibility is plausible. The paper's own statement in Section II A only mentions renormalization of higher-order terms, not the quadratic kernel. That is a gap worth probing. It is, however, a gap within a standard and credible computational approach, and the paper's language is appropriately hedged - it says 'strongly disfavor' and 'unlikely,' not 'excluded.'\n\nThe other soft spot is that the Monte Carlo simulation produces the orthogonal 3M phase, not the observed single-arm AFM2. The authors attribute this to quartic terms beyond the Heisenberg model and give a semiquantitative magnetoelastic estimate, but they do not close the gap. Again, this does not affect the central quadratic-instability ranking between M and Gamma.\n\nThe GGA result, which puts the largest eigenvalue at K, is dismissed because the GGA moment is too large. That is a judgment call, but the LDA moment is indeed much closer to the experimental ordered moments, so I find it reasonable.\n\nOverall, this is a solid, honest computational paper on a high-profile candidate. I'd send it to a serious referee. The Mn1 question is exactly what a good referee should push on, but I expect the central conclusion to survive.\n\nWho is this paper for? Condensed matter theorists and experimentalists working on altermagnetism and Mn5Si3 specifically. It provides a concrete counter-analysis to the altermagnetic interpretation and should be cited in future work.","headline":"A careful DLM-based argument that the Gamma-point altermagnetic mode in Mn5Si3 is far less competitive than the M-point AFM2 mode; the one real gap is that finite Mn1 moments are never tested.","tokens_in":18491,"tokens_out":3591,"would_cite":true,"duration_ms":30650,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["75.30.Et","75.10.Hk","71.15.Mb"],"model":"deepseek-v4-flash","headline":"This paper argues that the zone-center spin mode that would make Mn5Si3 an altermagnet is substantially weaker than the bulk M-point antiferromagnetic instability, and epitaxial strain typical of films suppresses it further, so the film's…","keywords":["altermagnetism","Mn5Si3","disordered local moment","exchange kernel","Landau theory","epitaxial strain","anomalous Hall effect","Heisenberg model"],"falsifier":"Measure the magnetic ordering wave vector of a Mn5Si3 film that shows the anomalous Hall effect in the high-temperature regime: M-point magnetic scattering persisting above 200 K would rule out the zone-center altermagnet, while a Gamma-point (0,1,-1) order would overturn the paper's case. A numerical alternative is a first-principles study that includes Mn1 moments or non-Heisenberg correlations and finds $\\lambda_{\\max}(\\Gamma) > \\lambda_{\\max}(\\mathbf{M})$.","tokens_in":17498,"feed_emoji":"🧲","tokens_out":8912,"duration_ms":68806,"temperature":0.7,"pith_summary":"This paper argues that the much-studied altermagnetic phase proposed for thin films of the metal Mn5Si3 is not a natural instability of the material's paramagnetic state. Using first-principles disordered-local-moment calculations, it shows that the strongest magnetic instability of bulk Mn5Si3 sits at the zone-boundary M point, producing the known AFM2 antiferromagnetic phase, while the zone-center mode that would give the altermagnet is substantially weaker and becomes even weaker under epitaxial strain like that in real films. The paper concludes that the high-temperature anomalous Hall effect observed in Mn5Si3 films cannot be explained as a weakly strained version of the bulk stoichiometric compound, and that some other physics must be responsible.","feed_headline":"Mn5Si3's altermagnetic order is disfavored by first-principles theory","feed_subtitle":"The zone-center spin mode is weaker than the bulk M-point order, and strain makes it worse.","key_machinery":"The load-bearing object is the Fourier-transformed exchange kernel $J_{\\alpha\\beta}(\\mathbf{q})$ on the six Mn2 sublattices, computed in the disordered-local-moment paramagnetic reference state; its largest eigenvalue $\\lambda_{\\max}(\\mathbf{q})$ locates the leading spin instability. Landau theories for the $\\mathbf{M}$ star and the zone center place the same $(1,-1,0)$ intracell pattern at a permutation-odd mode of a single $\\mathbf{M}$ arm and in the collinear $E_{2g}$ order-parameter sector at $\\Gamma$, respectively, with higher-order terms deciding the final phase. The comparison of $\\lambda_{\\max}(\\Gamma)$ with $\\lambda_{\\max}(\\mathbf{M})$, together with its response to epitaxial strain, carries the argument.","core_discovery":"The central discovery is a quantitative comparison of the two candidate magnetic instabilities of Mn5Si3 within a disordered-local-moment (DLM) description of the paramagnet. The exchange kernel $J_{\\alpha\\beta}(\\mathbf{q})$, evaluated from density functional theory in the DLM reference state, has its largest eigenvalue at the $\\mathbf{M}$ star, with an eigenvector matching the experimentally observed AFM2 pattern $(0,1,-1)$ on the three inversion-even Mn2 sublattice pairs. The same intracell pattern at the zone center belongs to the collinear branch of the $E_{2g}$ representation, so a Landau theory could produce an altermagnet from that mode; but $\\lambda_{\\max}(\\Gamma)$ is significantly below $\\lambda_{\\max}(\\mathbf{M})$ and is a global minimum within the $q_z=0$ plane. This ordering of instabilities is robust to Mn2 local moments between 2.14 and 2.54 $\\mu_B$ and to a 1% in-plane tensile strain with 0.3% $c$-axis contraction, which further suppresses the $\\Gamma$ mode. The paper concludes that the zone-center altermagnetic phase is not a close competitor in stoichiometric bulklike Mn5Si3.","pith_inferences":["Because $\\lambda_{\\max}(\\mathbf{q})$ flattens near the zone boundary under tensile strain, order-by-disorder or quartic corrections could in principle stabilize a multi-M or $\\mathbf{q}\\ne\\mathbf{M}$ state in films, but with essentially the same exchange scale; such a state still could not explain Hall signals at more than twice the bulk ordering temperature.","The negative result shifts the burden to film-specific chemistry: testing carbon-intercalated, off-stoichiometric, or Mn1-active films would directly probe whether the Hall-active state comes from a non-bulklike magnetic phase.","A straightforward experimental falsifier would be magnetic scattering on the AHE-active film: observation of M-point magnetic peaks persisting above 200 K would rule out the Gamma-altermagnet, while observation of a Gamma-point (0,1,-1) mode would overturn the paper's conclusion.","The Landau analysis shows the (0,1,-1) direction at Gamma is selected only by sixth-order terms whose sign is not fixed by symmetry, so the negative result is robust within the DLM/Heisenberg model family but could be sensitive to correlations beyond it."],"forward_implications":["Bulk AFM2 ordering is explained: the M-star eigenmode is the leading paramagnetic instability, with single-site entropy and magnetoelastic coupling able to select the single-arm AFM2 phase over the orthogonal 3M state.","The proposed altermagnetic phase of thin Mn5Si3 films is unlikely to be stabilized by moderate epitaxial strain of a stoichiometric bulklike film, since such strain reduces the leading M-point exchange eigenvalue by about 20% and suppresses the Gamma mode further.","Classical Monte Carlo simulations using the DLM exchange parameters order near 73 K into an equal-amplitude orthogonal 3M state, in reasonable agreement with the experimental $T_{N2}\\approx 100$ K given that quartic phase-selection terms lie beyond the Heisenberg model.","The high-temperature anomalous Hall and Nernst signals in films cannot be attributed to a nearby Gamma-point instability; if they are magnetic in origin, they require off-stoichiometry, intercalation, interfacial, or other non-bulklike physics.","The estimated magnetoelastic coupling is consistent with the measured orthorhombic exchange striction, identifying a concrete mechanism that favors the bulk AFM2 phase."],"supporting_citations":[{"why":"Defines the experimental AFM2 magnetic structure and its orthorhombic exchange striction, the bulk facts the Landau and magnetoelastic analysis must reproduce.","marker":"[2]"},{"why":"Reports the spontaneous anomalous Hall effect in Mn5Si3 films that motivates the altermagnetic hypothesis this paper argues against.","marker":"[3]"},{"why":"Supplies previous first-principles exchange parameters for the ordered AFM2 phase, used to show the DLM values are robust.","marker":"[14]"},{"why":"Provides the generic M-star O(3) Landau free energy whose phase diagram governs single-arm versus multi-M selection.","marker":"[20]"},{"why":"Provides the stacked-triangular antiferromagnet Landau theory used to analyze the collinear E2g sector at the zone center.","marker":"[24]"},{"why":"Gives the linear-response formalism for exchange interactions in metals on which the DLM exchange kernel is based.","marker":"[27]"},{"why":"Formulates the disordered-local-moment description of the paramagnetic state used as the reference for the exchange kernel.","marker":"[30]"},{"why":"Supplies the calculated elastic constant C66 used in the magnetoelastic-coupling estimate.","marker":"[38]"},{"why":"Underlies the Monte Carlo simulations that estimate the ordering temperature and identify the orthogonal 3M phase.","marker":"[40]"}],"fun_headline_variants":["Mn5Si3 altermagnetism loses to bulk M-point order","Theory says Mn5Si3 is no altermagnet","Bulk M-point order outcompetes Mn5Si3 altermagnetic mode","Strain kills Mn5Si3 zone-center altermagnetic instability","First principles disfavor Mn5Si3 altermagnetic phase"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion rests on the disordered-local-moment paramagnetic state with Mn2 moments of 2.14 to 2.54 mu_B and no Mn1 moment being a faithful model of the real fluctuating-moment state above the ordering temperature, so that the classical Heisenberg exchange kernel correctly ranks the competing instabilities.","fun_headline_variants_meta":{"raw":{"variants":["Mn5Si3 altermagnetism loses to bulk M-point order","Theory says Mn5Si3 is no altermagnet","Bulk M-point order outcompetes Mn5Si3 altermagnetic mode","Strain kills Mn5Si3 zone-center altermagnetic instability","First principles disfavor Mn5Si3 altermagnetic phase"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.0007,"raw_usage":{"total_tokens":3268,"prompt_tokens":1157,"completion_tokens":2111,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":773,"completion_tokens_details":{"reasoning_tokens":2033}},"tokens_in":773,"tokens_out":2111,"duration_ms":13222,"temperature":1.0,"reasoning_tokens":2033,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:02:16.633101+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the magnetic ordering wave vector of a Mn5Si3 film that shows the anomalous Hall effect in the high-temperature regime: M-point magnetic scattering persisting above 200 K would rule out the zone-center altermagnet, while a Gamma-point (0,1,-1) order would overturn the paper's case. A numerical alternative is a first-principles study that includes Mn1 moments or non-Heisenberg correlations and finds $\\lambda_{\\max}(\\Gamma) > \\lambda_{\\max}(\\mathbf{M})$.","supporting_citations":[{"cited_title":"BecauseQ 2 =L 4 −4C ×, it makes a positive contribution tov Γ","cited_arxiv_id":null,"evidence_quote":"Defines the experimental AFM2 magnetic structure and its orthorhombic exchange striction, the bulk facts the Landau and magnetoelastic analysis must reproduce."},{"cited_title":"We now consider how this ordering could emerge from a Landau theory","cited_arxiv_id":null,"evidence_quote":"Reports the spontaneous anomalous Hall effect in Mn5Si3 films that motivates the altermagnetic hypothesis this paper argues against."},{"cited_title":"Skobjin, J","cited_arxiv_id":null,"evidence_quote":"Supplies previous first-principles exchange parameters for the ordered AFM2 phase, used to show the DLM values are robust."},{"cited_title":"S¨ urgers, G","cited_arxiv_id":null,"evidence_quote":"Provides the generic M-star O(3) Landau free energy whose phase diagram governs single-arm versus multi-M selection."},{"cited_title":"Mendoza-Estrada, R","cited_arxiv_id":null,"evidence_quote":"Provides the stacked-triangular antiferromagnet Landau theory used to analyze the collinear E2g sector at the zone center."},{"cited_title":"Todoroki and S","cited_arxiv_id":null,"evidence_quote":"Gives the linear-response formalism for exchange interactions in metals on which the DLM exchange kernel is based."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the calculated elastic constant C66 used in the magnetoelastic-coupling estimate."}],"review_version":1}