{"id":"b4c6be13-9a3d-4ee4-bde9-7883c814d27f","arxiv_id":"2607.17688","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"In Fe1-x(Mn/Co)xRh, Mn hole doping collapses spin polarization (P crosses zero near x=0.5) and suppresses the Curie temperature by ~450 K through near-cancelling exchange, while Co electron doping preserves |P|≈0.75 and TC>800 K: d-band filling relative to the majority-spin pseudogap is the control","lead":"Quantum simulations of FeRh alloys show that swapping iron for manganese removes electrons and collapses the magnetic stability — the Curie temperature drops by about 450 K — while swapping in cobalt adds electrons and keeps the magnetism strong. The design rule this suggests — the Fermi level's position relative to a spin-split pseudogap controls magnetic fate — matters for tunable magnetocaloric and spintronic materials.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"FM-reference Jij may be invalid in Mn-rich regime because only G-type AFM-II was tested and PBE likely overstabilizes FM; the paper's own caveat admits other orderings may compete.","rationale":"The reader's weakest_assumption identified exactly this issue: the FM reference is validated only against G-type AFM-II, and the paper's caveat admits other orderings may compete in the Mn-rich regime. My stress-test pass confirms this is the single most load-bearing concern because it directly undermines the premise that the FM state is well separated and therefore the exchange-topology analysis is physically grounded. The concern is concrete: the computed FM stability at x=0.8 (ΔE≈0.35 eV/atom) conflicts with the known AFM order of MnRh, and the PBE functional's FM bias (ref [30]) makes this discrepancy plausible. I agree with the reader's conditional verdict: the paper is internally consistent and the Co-series results appear robust, but the Mn-rich conclusions hinge on an untested assumption. A decisive computational test—checking other AFM orderings—would determine whether the central claim survives. No change to the reader's CONDITIONAL verdict is needed because the concern is already reflected in that verdict; if the test fails, the verdict would need to move to REJECT for the Mn-rich part of the claim. I therefore set verdict_should_be to UNCHANGED.","tokens_in":18757,"tokens_out":3858,"duration_ms":45658,"concrete_test":"Using the same SPR-KKR-CPA/PBE setup, compute total energies for additional collinear AFM orderings (A-type, C-type, and at least one planar spin-spiral) at Mn compositions x=0.5 and x=0.8, optimizing the lattice parameter for each ordering. If any of these configurations has E < E_FM, then the FM reference is not the ground state in the Mn-rich regime, and the exchange-topology conclusions drawn from FM-reference Jij are not valid for those compositions. This would directly test whether the paper's own third caveat is realized in practice.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that d-band filling relative to the majority-spin pseudogap controls magnetic stability—rests on the FM state being the appropriate magnetic reference for the Liechtenstein analysis and on the FM state remaining well separated from AFM configurations across the full composition range (Abstract; Sec. 3, Fig. 1b). However, the FM reference is validated only against a single competing ordering, G-type AFM-II (Sec. 3, Fig. 1b). At Mn-rich compositions, the paper's own third caveat (Sec. 4.2) concedes that 'magnetic configurations beyond the collinear FM state—including, but not restricted to, the G-type AFM-II ordering—may become energetically competitive in the Mn-rich regime.' This is not a mere technicality: the end-member MnRh is antiferromagnetic, yet Table 1 reports ΔE ≈ 0.35 eV/atom favoring FM at x=0.8, a result that is suspiciously large given that PBE is known to overstabilize the FM state in FeRh (ref [30]). If any other AFM ordering (e.g., A-type, C-type, or a spin spiral) lies below FM at high Mn content, then the FM-reference Jij, the η_Mn<0 signature, and the computed TC suppression are all affected; the 'itinerant magnetic softness' would be an artifact of using an unstable reference state rather than a genuine finite-temperature instability of the equilibrium phase. Since the abstract explicitly asserts the FM state is 'well separated' and the title claims broad control by d-band filling, this unresolved possibility is the most load-bearing concern.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript uses SPR-KKR-CPA calculations with the PBE functional to study B2-ordered Fe1-x(Mn/Co)xRh alloys (x = 0.1–0.8), reporting spin-resolved DOS, magnetic moments, spin polarization, Liechtenstein exchange parameters, MFA Curie temperatures, and FM–G-type-AFM-II energy differences. The central claim is that d-band filling relative to the majority-spin pseudogap is the primary control parameter of magnetic stability. Mn substitution (hole doping) is said to move the Fermi level out of the pseudogap, collapse the spin polarization (P crosses zero near x ≈ 0.5), introduce negative Mn-centred exchange competition (η_Mn < 0), and suppress the MFA Curie temperature from 834 K to 382 K, even though the FM state remains higher-lying than the G-type AFM-II state by up to ~0.35 eV/atom. Co substitution (electron doping) is said to keep EF pinned in the pseudogap, preserving |P| ≈ 0.75, η_Co ≳ 0.37, and TC > 800 K. The authors introduce the concept of ‘itinerant magnetic softness’ to describe the Mn-driven near-cancellation of competing exchange interactions.","tokens_in":19066,"tokens_out":7443,"duration_ms":75004,"significance":"If the results hold, the paper would provide a simple, band-filling-based design rule for magnetic ordering temperatures in a technologically important FeRh alloy family, with falsifiable predictions (e.g., a ~450 K TC suppression in Mn-substituted films and preserved high TC in Co-substituted films). The systematic composition series, the use of CPA plus Liechtenstein exchange analysis, the site-resolved decomposition of the competition parameter, and the generally self-consistent tabulated data are strengths. The paper is also honest in listing caveats, and one of these caveats is directly load-bearing: the FM reference is validated only against a single AFM ordering. Because the FM reference underlies the Jij, η, and TC results, the central claim is not yet established at high Mn content.","major_comments":[{"comment":"The FM reference for the Liechtenstein analysis is validated only against the G-type AFM-II configuration. The paper's own third caveat in Sec. 4.2 states that other magnetic configurations ‘may become energetically competitive in the Mn-rich regime’. This is not a remote possibility: the end-member MnRh is antiferromagnetic, yet Table 1 reports ΔE = 0.346 eV/atom favouring FM at x = 0.8, which is hard to reconcile with the known ground state and with the acknowledged PBE tendency to overstabilize FM (ref. [30]). If an A-type, C-type, or spin-spiral state lies below FM at high Mn content, then the FM-reference Jij, the η_Mn < 0 signature, and the computed TC cascade are all evaluated about an unstable reference, and the abstract's statement that the FM state is ‘well separated’ would be incorrect. I request explicit total-energy comparisons against at least A-type, C-type, and one simple","section":"Sec. 3, Fig. 1b; Sec. 4.2, third caveat"},{"comment":"For the Mn series, η_Mn < 0 and the net Mn-centred exchange nearly vanishes. In the MFA, TC is obtained from the largest eigenvalue λ_max of the Heisenberg exchange matrix. When the exchange competition is strong, λ_max may correspond to an AFM or canted eigenmode, not to the ferromagnetic mode the paper identifies as the Curie temperature. The authors do not report the eigenvector associated with λ_max, nor do they verify that the classical ground state of the fitted Heisenberg model is FM. If the leading instability is non-FM, then the plotted ‘TC’ is not the ordering temperature of the FM phase, and the ‘itinerant magnetic softness’ narrative would need to be reformulated. Please report the eigenvector(s) at the instability, and if needed, compute the FM-mode TC separately and reconcile the Heisenberg ground state with the DFT ΔE of the G-type AFM-II and other orderings.","section":"Sec. 3.4, Eq. (4), Table 3"},{"comment":"The separation of d-band-filling effects from magneto-volume effects is inferred from the different lattice-parameter evolution of the two series (Mn: ~0.1% change, TC drops; Co: ~0.9% change, TC stable). However, no fixed-lattice calculation is reported. Since a change in volume also shifts EF and modifies the Jij, the observed trends do not uniquely identify chemical filling as the causal factor. The claim that d-band filling is ‘primary’ rather than magneto-volume would be substantially strengthened by computing the two series at a common lattice constant (e.g., the parent FeRh value) or by an explicit decomposition of the TC change into volume and substitutional contributions. As written, the magneto-volume argument is suggestive but not conclusive.","section":"Sec. 3.2, Tables 1–2"}],"minor_comments":[{"comment":"The minority-channel DOS N↓ is listed with a negative sign in Tables 1 and 2, but Eq. (1) is written with N↓(EF) as a positive magnitude. The reported P values correspond to using |N↓|. Please state this convention explicitly near Eq. (1) to avoid ambiguity.","section":"Eq. (1), Tables 1–2"},{"comment":"The definition of η in Eq. (5) sums over all i,j, but the text then evaluates it separately for Mn- or Fe-centred environments. Please specify precisely which Jij pairs enter each restricted sum, and state the real-space cutoff or number of coordination shells used in the sums.","section":"Eq. (5), Sec. 3.4"},{"comment":"The LKAG integral in Eq. (3) contains the Fermi–Dirac function with β = 1/kBT, but the smearing temperature used in the numerical integration is not stated. Please report this value and any convergence tests with respect to it and the k-mesh.","section":"Eq. (3)"},{"comment":"The schematic expression J(R) ~ cos(2kF R + φ)/R^3 and the associated single-kF language are illustrative only for a multiband d-electron system. Please mark this explicitly as a schematic picture to avoid implying that a single Fermi-surface spanning vector was extracted from the DFT calculations.","section":"Sec. 4.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is in scope for the journal and the computational methodology is appropriate. The core issue is the FM-reference validation: the own-caveat in Sec. 4.2, combined with the known AFM ground state of MnRh and the suspiciously large FM favouring ΔE at x = 0.8, makes it essential to test additional orderings before the central claim can be accepted. The MFA eigenmode issue should also be addressed, since it concerns the physical meaning of the reported TC. A fixed-lattice control calculation would solidify the causal claim. With these additions, the paper could become a valuable contribution; as it stands, the central claim is defensible only conditionally."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Candid take: this is a solid, internally consistent first-principles paper whose central contrast—Mn suppresses the FM Curie temperature while Co preserves it—holds up within the model. The reader's conditional verdict is fair.\n\nWhat's new: the full composition map (x=0.1–0.8) for both series, site-resolved LKAG exchange interactions, the eta = sum J / sum |J| decomposition separating net cancellation from overall weakening, and the spin-polarization sign reversal near x=0.5. The paper is also honest: it states its own bounding caveats, including PBE's known FM bias and the transverse-only nature of the Liechtenstein Jij.\n\nSoft spots, in proportion. The most load-bearing one is the FM reference. Only the G-type AFM-II ordering was tested as a competitor; the end-member MnRh is antiferromagnetic, and the paper itself concedes in Sec. 4.2 that other non-collinear or AFM configurations may become competitive in the Mn-rich regime. If one of those lies below FM, then the Jij and eta describe a metastable state, and 'magnetic stability' is really 'stability of the FM phase within the model.' That is more than a technicality, but it doesn't destroy the compositional trend: both series are treated in the same FM reference, and the qualitative difference between hole and electron doping would likely survive. I would not call it fatal, but it should be fixed or explicitly scoped.\n\nSecond, the explanation has a tautological edge: TC is computed from the largest eigenvalue of the J matrix, and eta is a decomposition of the same J, so 'exchange competition suppresses TC' is partly restating the definition. The authors do provide some independent evidence—the Fe-centred total exchange magnitude weakens ~21% across the Mn series while staying flat in Co—so the softness is not purely a label. Still, 'itinerant magnetic softness' is a description, not a mechanism.\n\nMinor: no input files, and the Fermi-Dirac smearing and real-space cutoff for the Jij sums are not stated.\n\nWho it's for: researchers working on FeRh-based magnetocaloric or spintronic alloys, and anyone using CPA+LKAG to rationalize substitution trends. It deserves a serious referee, not a desk reject. I'd ask the referee to require one or two additional AFM orderings at high Mn, a fixed-volume control, and ideally deposition of input files. With those in place, the paper would be a useful contribution.","headline":"Solid full-composition computational map of Mn/Co in FeRh; the FM-reference caveat is real but not fatal, and the composition trend survives.","tokens_in":19747,"tokens_out":2745,"would_cite":false,"duration_ms":29284,"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":"Mn hole doping collapses FeRh's Curie temperature by about 450 K, while Co electron doping keeps it above 800 K.","keywords":["FeRh","d-band filling","itinerant magnetism","Curie temperature suppression","exchange interactions","coherent potential approximation","spin polarization","pseudogap"],"falsifier":"A total-energy search over several antiferromagnetic and non-collinear orderings for Fe1-xMnxRh at high x (say 0.6–0.8); finding any configuration below the ferromagnetic state would invalidate the FM-reference exchange couplings and the predicted ~450 K Curie-temperature drop. Experimentally, measuring the Curie temperature and spin polarization of Mn-substituted FeRh films across x would test the predicted collapse and sign reversal near x≈0.5.","tokens_in":18456,"feed_emoji":"🧲","tokens_out":4963,"duration_ms":42166,"temperature":0.7,"pith_summary":"This paper argues that the magnetic stability of B2-ordered FeRh alloys is set by where the Fermi level sits relative to a pseudogap in the majority-spin band, not by volume or by distance to an antiferromagnetic state. Replacing Fe with Mn removes electrons, dragging the Fermi level out of that gap; spin polarization collapses and flips sign near x≈0.5, Mn-centred exchange turns antiferromagnetic, and the Curie temperature falls from about 834 K to 382 K. Replacing Fe with Co adds electrons, pins the Fermi level in the gap, and leaves the ferromagnet strong, with Curie temperature above 800 K and spin polarization near 0.75. The paper's central distinction is that a thermally fragile ferromagnet can still be energetically well separated from a competing antiferromagnetic ordering; the instability is 'itinerant magnetic softness' from near-cancellation of exchange interactions rather than from AFM–FM energy proximity.","feed_headline":"Mn doping drops FeRh's Curie temperature by ~450 K","feed_subtitle":"The magnetic fate of Fe1-x(Mn/Co)xRh hinges on where the Fermi level sits relative to a spin-channel pseudogap.","key_machinery":"The load-bearing object is the majority-spin pseudogap of B2 FeRh and the Fermi level's position within it; the paper tracks how substitution shifts the Fermi level across the spin-resolved density of states. The quantitative machinery is the Liechtenstein exchange formula, which maps the itinerant electronic structure onto effective Heisenberg couplings Jij, and the competition parameter η = ΣJij/Σ|Jij|, which separates ferromagnetic from antiferromagnetic pathways. A decomposition of η into net (ΣJij) and total-magnitude (Σ|Jij|) parts distinguishes genuine weakening of the ferromagnetic backbone from near-cancellation of competing couplings.","core_discovery":"The central claim is that d-band filling relative to the majority-spin pseudogap is the primary control parameter for magnetic stability in Fe1-x(Mn/Co)xRh. Mn hole doping moves the Fermi level out of the pseudogap, raising the majority-spin DOS at the Fermi level and lowering the minority-spin DOS, so spin polarization P falls through zero near x≈0.5 and reverses sign; Mn-centred exchange couplings develop antiferromagnetic components, quantified by a negative ηMn, and the mean-field Curie temperature drops by about 450 K even though the ferromagnetic state remains lower in energy than the G-type AFM-II state by up to about 0.35 eV/atom. Co electron doping leaves the Fermi level pinned in t","pith_inferences":["The same Fermi-level control parameter may allow graded design of magnetocaloric or spintronic materials, where the Curie temperature is deliberately placed near an operating temperature; the paper hints at this but does not explore device consequences.","A direct experimental test would compare the predicted TC(x) curve for Mn substitution against magnetometry on thin films; deviations at high x would signal that orderings beyond the collinear ferromagnetic state intervene.","The rigid-band picture suggests that other hole-doping substituents such as Cr or V on the Fe site should mimic Mn's softening, and other electron donors such as Ni should mimic Co's hardening—an extension the paper proposes but does not compute.","Because the analysis is performed in the ferromagnetic reference, the negative ηMn predicts that Mn moments in the FM host prefer antiferromagnetic alignment; measurements sensitive to short-range Mn–Mn correlations could test this locally."],"forward_implications":["Tuning the valence-electron count of Fe-sublattice substituents can position the Curie temperature of FeRh alloys over a ~450 K range while the ferromagnetic state stays lower in energy than the G-type antiferromagnetic state.","A ferromagnet can be thermally fragile even when its ground state is well separated from a competing antiferromagnet, because the ordering temperature tracks the small net exchange field rather than the large total exchange scale.","Co substitution acts as a magnetic hardener: high spin polarization (|P|≈0.75) and ferromagnetic exchange persist across x=0.1–0.8, keeping the Curie temperature above 800 K even as the total moment falls by about 20%.","The same d-band-filling mechanism should operate in other itinerant magnets whose Fermi level sits near a one-spin-channel pseudogap, making hole doping soften and electron doping harden ferromagnetism."],"fun_headline_variants":["d-band filling is the master dial for FeRh magnetism","Mn doping softens FeRh magnets, Co hardens them","Fermi level vs pseudogap decides FeRh's magnetic fate"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The analysis assumes the ferromagnetic state is the correct reference for extracting exchange couplings, but it verifies that assumption against only one competing magnetic order (the G-type antiferromagnetic state); the paper concedes other orderings could become competitive at high Mn content, and if any sits lower in energy, the computed exchange parameters, the negative Mn signature, and the Curie-temperature suppression would all be affected.","fun_headline_variants_meta":{"raw":{"variants":["d-band filling is the master dial for FeRh magnetism","Mn doping softens FeRh magnets, Co hardens them","Fermi level vs pseudogap decides FeRh's magnetic fate"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000775,"raw_usage":{"total_tokens":3330,"prompt_tokens":872,"completion_tokens":2458,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":616,"completion_tokens_details":{"reasoning_tokens":2401}},"tokens_in":616,"tokens_out":2458,"duration_ms":16041,"temperature":1.0,"reasoning_tokens":2401,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T17:17:24.771205+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A total-energy search over several antiferromagnetic and non-collinear orderings for Fe1-xMnxRh at high x (say 0.6–0.8); finding any configuration below the ferromagnetic state would invalidate the FM-reference exchange couplings and the predicted ~450 K Curie-temperature drop. Experimentally, measuring the Curie temperature and spin polarization of Mn-substituted FeRh films across x would test the predicted collapse and sign reversal near x≈0.5.","supporting_citations":[],"review_version":1}