{"id":"e8033e5e-cf2d-4b64-a9d0-b35a651b23a0","arxiv_id":"2412.14365","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"The iron formate perovskite has an improper modulated magnetic ground state with spins in the ab plane, perpendicular to the c-axis moments of its Ni and Co analogues.","lead":"Neutron diffraction shows that the iron formate perovskite [CH3NH3]Fe(HCOO)3 develops two incommensurate structural modulations on cooling, at 170 K and 75 K. Below 17 K its magnetic moments order in the plane perpendicular to those of the nickel and cobalt analogues, a difference attributed to magnetic anisotropy.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The improper-modulation and ab-plane-spin conclusions depend on a magnetic-space-group choice and a constrained moment refinement that are not fully settled; the 2 K data should be re-fit with Mz free and with the competing Pn'ma' model.","rationale":"The reader's conditional verdict is well aligned with my assessment. The most load-bearing weakness is the magnetic structure determination: the Pnma.1 choice and the zero-Mz constraint are underdetermined by weak, high-background data, and the final model fixes the very quantities that the 'strictly perpendicular' and 'improper' claims rest on. I see this as the primary concern because the entire contrast with the Ni compound depends on it. The DFT easy-axis mismatch is a real inconsistency but is secondary: it weakens the explanatory story but does not by itself invalidate the experimental magnetic structure if that structure is robust. I partially agree with the reader because they identified both issues in the weakest_assumption, and I agree with the verdict CONDITIONAL. The proposed re-refinement is feasible because the raw single-crystal data are deposited (ILL-DATA 5-12-339), and the alternative magnetic space groups are explicitly listed. I do not see grounds to reject the work; the structural characterization is solid, and the magnetic conclusions are plausible but need the additional analysis before being taken as settled.","tokens_in":24010,"tokens_out":2322,"duration_ms":23208,"concrete_test":"Refit the 2 K D19 single-crystal data in JANA2020 in the magnetic superspace group Pnma.1(00g)0s0 with (a) Mz free and the moment modulus unconstrained, and (b) in the alternative Pn'ma'(00g)0s0 model, using the same reflection set and weighting. Perform a Hamilton R-factor test or a log-likelihood comparison on the difference in fit quality, and report the refined Mz, |M|, and R-factors for all models. If the free-Mz refinement converges to a nonzero Mz with a statistically significant improvement, or if the Pn'ma' model fits comparably while allowing a b-axis cant, the 'strictly ab-plane' and 'improper' conclusions would need to be revised. A secondary check: recompute the DFT anisotropy with fully self-consistent non-collinear calculations (relaxing atomic positions and, for Ni, including spin-orbit coupling) and compare the predicted easy axis with the refined [1 1 0] direction.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that [CH3NH3]Fe(HCOO)3 has an improper modulated magnetic structure with Fe moments strictly in the ab plane depends on two steps in the magnetic refinement. First, the average magnetic symmetry is selected as Pnma.1 over the close competitor Pn'ma' (and other Shubnikov groups) using a neutron powder difference pattern with high incoherent hydrogen background; the decisive 1 1 1 reflection, which would detect Mz, is reported as absent but is weak and sits atop a background that the paper itself describes as substantial. Second, in the single-crystal refinement at 2 K the Mz component is 'restricted to be zero, in agreement with the magnetometry measurements' and the moment modulus is constrained to 4.0 µB, because unconstrained refinements 'convergence issues' and 'values without physical meaning' (pp. 12, 29-30). If a small Mz or alternatively the Pn'ma' model (with its allowed b-axis ferromagnetic component) were correct, the magnetic moments would not be strictly perpendicular to those of the Ni compound, and the 'improper' (non-proper-modulated) characterization would be weakened. The DFT argument is also not quantitative: the computed easy axis for Fe is [1 0 0] (Table 2), while the refined moment direction is [1 1 0], so the anisotropy-based rationalization is asserted rather than demonstrated. The structural phase sequence, in contrast, is well supported by the reported satellite evolution and refinements, so the concern is specifically about the magnetic part of the central claim.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports a combined single-crystal and powder neutron diffraction, magnetometry, and DFT+U study of the hybrid formate perovskite [CH3NH3]Fe(HCOO)3. It establishes a nuclear phase sequence: Pnma at room temperature, a modulated Pnma(00g)0s0 phase below about 170 K with q1 = 0.1662(2)c*, a second modulated phase below about 75 K with q2 = 0.1425(2)c*, and magnetic order below 17 K with propagation vector k = (0,0,0). The central magnetic claim is that the ground state has Pnma.1 magnetic superspace symmetry, with Fe moments strictly contained in the ab plane and pointing mainly along [1 1 0], with no t-dependent modulation of the moments; this is interpreted as an improper modulated magnetic structure, in contrast to the proper magnetic modulation reported for the Ni analogue. The structural phase sequence is well supported by the diffraction data and refinements. The magnetic conclusion, however, rests on a close competition between magnetic space groups and on constrained refinements whose stability is not fully documented.","tokens_in":24390,"tokens_out":4707,"duration_ms":44126,"significance":"If confirmed, the work provides a rare and valuable example of an aperiodic coordination polymer with coexisting nuclear incommensurability and commensurate magnetic order, and it offers a family comparison that could inform how magnetic anisotropy controls spin orientation in these hybrid perovskites. The paper has concrete strengths: the structural refinements are carried out at multiple temperatures with both powder and single-crystal neutron data; the crystallographic data are deposited; the raw data are made available through ILL; and the DFT calculations use experimentally determined structures with literature U values rather than parameters fitted to the target result. The magnetic discrimination between Pnma.1 and Pn'ma', however, is the load-bearing step for the 'strictly in ab plane' and 'improper modulated' claims, and that step currently depends on R-factors near 15-20% and on constraints whose effect is not quantified. The DFT anisotropy calculation also gives an easy axis that does not match the refined moment direction, so the proposed mechanism is not yet quantitatively demonstrated.","major_comments":[{"comment":"The assignment of Pnma.1 over Pn'ma' is based on Rf values of about 16% and 15%, respectively, and on the absence of the 1 1 1 reflection in a difference pattern with substantial incoherent hydrogen background. Because Pn'ma' is said to fit the 0 0 1 and 1 0 0 reflections poorly (Figure S6), while Pnma.1 is accepted only after constraining Mz to zero, the two competing models are not compared on equal footing. Please report the full agreement factors (R_F, R_wp, chi-squared, and goodness-of-fit) for all four Shubnikov models using the same reflection list, refine the Pnma.1 model with Mz free, and quote the resulting component and its uncertainty. A small nonzero Mz, or an allowed b-axis ferromagnetic component in Pn'ma', would directly modify the 'strictly in ab plane' and 'strictly antiferromagnetic' conclusions, so this discrimination needs to be quantitatively robust.","section":"Magnetic structures (NPD), pp. 23-25"},{"comment":"The single-crystal magnetic refinement fixes Mz = 0 and |M| = 4.0 micro-Bohr magnetons because, without these constraints, the refinement has convergence issues or gives values 'without physical meaning'. Yet the freely refined modulus is reported as 4.256(33) micro-Bohr magnetons, which the authors attribute to an orbital contribution. These constraints are load-bearing for the refined [1 1 0] direction, for the strictly ab-plane orientation, and for the conclusion that the three sinusoidal magnetic modes are inactive. Please provide the unconstrained (or less constrained) refinement results, the parameter correlations involving Mz, and a stability analysis (for example, refinements with Mz free from several starting values, or with the moment modulus free while damping the structure parameters). Without such tests, the absence of a c-component and the absence of t-modulation are assertions imposed by the model rather than properties established by the data.","section":"Single crystal structural and magnetic determination and refinement details, p."},{"comment":"The DFT+U calculations give [1 0 0] as the easy axis (0.00 meV per Fe) with [1 1 0] higher by 0.22 meV per Fe, whereas the refined single-crystal moment direction is described as mainly [1 1 0] and the NPD text describes the moments as 'primarily along a' with a minor b component. This mismatch means the anisotropy calculation does not quantitatively explain the observed spin direction: the claim that significant Fe anisotropy 'results in distinct spin orientations' is asserted rather than demonstrated. Please reconcile the NPD and single-crystal descriptions of the moment direction, and either refine the moment orientation with a free angle for comparison against the DFT energy surface or extend the calculations (for example, self-consistent non-collinear calculations, different U values, or inclusion of the structural modulation) to test whether the observed direction can be reproduced. The discussion should also state whether the 0.22 meV/Fe [1 1 0] energy is considered significant relative to the uncertainty of the non-self-consistent approach.","section":"Theoretical calculations, Table 2; Discussion, p. 37"}],"minor_comments":[{"comment":"Several places contain the unresolved placeholder 'Error! Bookmark not defined.' in the text and reference list; these need to be resolved before publication.","section":"Throughout"},{"comment":"The 2 K entry lists Pnma.1(00g)0s0 while the 27 K entry lists Pnma(00g)0s0; please clarify whether the 2 K entry denotes the full magnetic superspace group and why the magnetic-group label is omitted at 27 K.","section":"Table 1"},{"comment":"The terms 'proper' and 'improper' magnetic modulation are used from the abstract onward but are defined only later in the magnetic structure section; a brief definition in the introduction would improve readability.","section":"Abstract and Introduction"},{"comment":"The note that spin directions are along the simulation cell rather than along atomic positions is confusing; please specify explicitly how the SAXIS directions map onto the crystallographic axes and whether the calculations used the modulated or average structures.","section":"Table 2 caption"},{"comment":"The comparison between the Pn'ma' and Pnma.1 fits is central to the magnetic space group choice, but the figure is only in the Supplementary Information and the main text does not give quantitative agreement indices for the problematic reflections; please provide these indices in the main text.","section":"Figure S6"}],"recommendation":"major_revision","confidential_remarks":"This is a well-executed structural study with unusually complete data deposition, and the nuclear phase sequence is convincing. The magnetic conclusion is plausible but currently rests on a close space-group discrimination and on constrained refinements that should be tested with the already-collected D19 single-crystal data. The requested re-fits are feasible within the manuscript's scope, so I recommend major revision rather than rejection. Please also ensure that the formatting placeholders are cleaned up before any final version is sent to the typesetter."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The structural work here is the strong part. The authors establish two new nuclear phase transitions in [CH3NH3]Fe(HCOO)3 at 170 K and 75 K, with incommensurate wave vectors q1 = 0.1662c* and q2 = 0.1425c*, distinct from the Co and Ni analogues. The superspace refinements against single-crystal neutron data look careful, the phase sequence is convincing, and the data are deposited. That alone makes this a useful contribution to the aperiodic coordination-polymer literature.\n\nThe magnetic part is more fragile. The choice of Pnma.1 over Pn'ma' rests on R-factors of 15–20% in a difference pattern with a substantial hydrogen background, and the decisive 1 1 1 reflection is weak and reported absent rather than confidently measured. The single-crystal refinement constrains Mz to zero and the moment magnitude to 4 µB because unconstrained refinements would not converge. Those constraints are justified by the magnetometry, but they mean the strictly-ab-plane, strictly-AFM picture is not as secure as the abstract implies. Likewise, the claim that the structure is an improper modulated magnet (rather than one with a small proper modulation) is a null result: the sinusoidal magnetic modes did not improve the fit. That may well be right, but it is a negative observation.\n\nThe DFT anisotropy calculation is also not as clean a match as the text suggests. The computed easy axis is [100] (Table 2), while the refined moment direction from single-crystal data is [110]. The qualitative point—Fe has significant anisotropy, Ni does not—is plausible, but the quantitative inconsistency should be acknowledged.\n\nNone of this sinks the paper. The structural phase sequence is solid, the magnetic structure is a reasonable best model given the data, and the comparison with the Ni and Co analogues is informative. But the central 'improper modulated magnetic structure with spins perpendicular to the analogues' claim is conditionally supported, not proven. A referee should push for a re-fit of the 2 K data with Mz free and with the competing Pn'ma' model, and ask the authors to either reconcile the DFT easy axis with the refined direction or tone down the anisotropy explanation.\n\nWho is this for? Specialists in hybrid formate perovskites, superspace crystallography, and low-dimensional magnets. It deserves a serious referee; I would send it out, but expect revision.","headline":"Solid structural characterization of a new Fe formate perovskite modulated phase, but the magnetic structure and anisotropy story rest on constrained refinements that need scrutiny.","tokens_in":24931,"tokens_out":1516,"would_cite":true,"duration_ms":14459,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["75.25.-m","75.30.Gw","75.50.Ee"],"model":"deepseek-v4-flash","headline":"The iron(II) formate perovskite [CH3NH3]Fe(HCOO)3 orders its spins strictly in the ab plane along [110] with no modulation of moment size or direction, making its magnetic incommensurability improper rather than proper.","keywords":["hybrid organic-inorganic perovskite","formate perovskite","incommensurate structural modulation","magnetic superspace group","improper magnetic incommensurability","magnetic anisotropy","neutron diffraction","antiferromagnetic order"],"falsifier":"A neutron diffraction experiment on a deuterated sample, or one with enough counting time to resolve the symmetry-forbidden 1 1 1 reflection, that lets the out-of-plane moment refine without constraints: a clearly nonzero $M_z$, or a statistically better fit in the competing $Pn'ma'$ model, would falsify the claim that moments lie strictly in the $ab$ plane along $[1\\,1\\,0]$.","tokens_in":23849,"feed_emoji":"🧲","tokens_out":14794,"duration_ms":119742,"temperature":0.7,"pith_summary":"The paper asks whether the iron formate perovskite [CH3NH3]Fe(HCOO)3, an isomorph of the nickel and cobalt compounds, also carries an incommensurate magnetic ripple in its ground state. It finds that it does not: on cooling, the material develops two structural modulations, but its Fe moments lock into a strictly antiferromagnetic arrangement confined to the $ab$ plane and pointing mainly along $[1\\,1\\,0]$, with no variation in size or orientation from site to site. The incommensurability is therefore improper: the nuclear lattice is modulated while the spins are not. The authors connect this to magnetic anisotropy, arguing that Fe(II) has a strong preferred spin direction while Ni(II) is nearly isotropic, and that this difference selects the spin orientation and suppresses proper magnetic modulation.","feed_headline":"Iron-formate perovskite spins lie flat while the lattice ripples","feed_subtitle":"Unlike the nickel analogue, the iron compound's moments never modulate, so its magnetic incommensurability is improper.","key_machinery":"The load-bearing object is the magnetic superspace group $Pnma.1(00\\gamma)0s0$, an extension of the space group $Pnma$ that adds one continuous coordinate $t$ for the phase of the modulation. In this description the Fe magnetic structure splits into six modes: three constant components that set the average moment direction and three sinusoidal modes whose amplitudes would produce a proper magnetic modulation along $t$. The central mechanistic step is showing that only the constant modes are activated, so that the spins track the average structure and do not ripple with the nuclear wave. A supporting mechanism is the magnetic anisotropy landscape computed with non-collinear density functional theory including spin-orbit coupling, which shows a clear energy ordering of spin directions for Fe and essentially none for Ni.","core_discovery":"The paper claims that in [CH3NH3]Fe(HCOO)3 the Fe magnetic moments form a strictly antiferromagnetic, commensurate ($k=0$) structure described by the magnetic superspace group $Pnma.1(00\\gamma)0s0$, with moments lying in the $ab$ plane and pointing mainly along $[1\\,1\\,0]$. The three sinusoidal magnetic modes that would make the moments vary in size or orientation with the modulation coordinate do not improve the fit, so the magnetic incommensurability is improper: the nuclear lattice is modulated but the spins are not. This places the Fe compound in the same superspace-group family as the Ni analogue, yet with moments perpendicular to those of the Ni and Co compounds, where weak ferromagnetic canting and, for Ni, proper magnetic modulation occur. Density functional calculations are used to argue that Fe(II) carries strong magnetic anisotropy while Ni(II) is effectively isotropic, which the authors take to explain the different spin orientations.","pith_inferences":["Beyond the paper: the calculated easy axis for Fe is $[1\\,0\\,0]$, while the refined moments point mainly along $[1\\,1\\,0]$; a fully self-consistent calculation that includes the modulated coordinates and orbital moments would show whether anisotropy alone predicts the observed 45-degree offset.","Beyond the paper: the paper stops at magnetization and neutron data, so dielectric spectroscopy through the 17 K magnetic lock-in is an obvious next measurement; a dielectric anomaly there would indicate magnetoelectric coupling mediated by the improper magnetic structure.","Beyond the paper: by the paper's own logic, sweeping a magnetic field on the Fe compound should not reproduce the first-cycle hysteresis anomaly seen in the Ni compound, because there are no proper magnetic modes to suppress; observing such an anomaly would require revisiting the model.","Beyond the paper: the metal-site control could be tested by synthesizing Fe-doped Ni solid solutions across the full composition range and locating where the spin direction switches from c-axis to ab-plane behavior."],"forward_implications":["The iron compound becomes a clear example of improper magnetic incommensurability in a coordination polymer: the nuclear modulation is incommensurate while the magnetic order is a commensurate $k=0$ antiferromagnet.","The $Pnma.1$ symmetry forbids any net moment, which is consistent with the flat magnetization plateau and linear field dependence up to 5 T at 2 K.","Metal substitution emerges as a switch: replacing Ni by Fe changes the spin orientation by about 90 degrees and turns off the proper magnetic modulation, so the magnetic anisotropy of the B-site metal is a design parameter.","The increasing modulation onset temperatures from Ni (84 K) to Co (128 K) to Fe (170 K) suggest that less electronegative B-site metals stabilize modulated structures at higher temperatures, a trend the authors propose for tuning future frameworks."],"supporting_citations":[{"why":"Establishes the nickel analogue's proper magnetic incommensurability, the key comparison that defines what improper means here.","marker":"12"},{"why":"Reports the cobalt compound's modulated phases and hydrogen-bond frustration, the template for the transition sequence assigned to the iron compound.","marker":"11"},{"why":"Gives the cobalt magnetic structure with weak canting, the contrast that makes the strictly compensated iron structure distinctive.","marker":"13"},{"why":"Shows how B-site composition tunes ordering temperatures and modulated-phase stability in the Co/Ni solid solutions, supporting the metal-substitution argument.","marker":"15"},{"why":"Documents the magnetic behaviour of the M formate family, the source of the isotropic-versus-anisotropic comparison for Fe and Ni.","marker":"33"},{"why":"Quantifies hydrogen-bond strength in formate perovskites, used to explain the driving force for modulation.","marker":"38"},{"why":"Shows that an external field suppresses proper magnetic modes in the nickel compound, evidence that such modes are energetically unfavourable when absent.","marker":"39"},{"why":"Provides the magnetic superspace-group symmetry rules used to exclude operators and select the Pnma.1(00g)0s0 model.","marker":"36"},{"why":"Supplies the plane-wave DFT method behind the magnetic anisotropy calculations.","marker":"29"}],"fun_headline_variants":["Iron spins stay put while lattice ripples","Magnetic anisotropy freezes iron spins in place","Lattice ripples without magnetic modulation","Iron perovskite spins lie perpendicular to nickel's","Improper magnetic order: lattice only, spins static"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion depends on choosing one magnetic symmetry model (Pnma.1) over a close competitor using a few weak magnetic reflections in data with heavy hydrogen background, and on a calculated anisotropy whose easy direction is near, but not exactly, the observed spin axis.","fun_headline_variants_meta":{"raw":{"variants":["Iron spins stay put while lattice ripples","Magnetic anisotropy freezes iron spins in place","Lattice ripples without magnetic modulation","Iron perovskite spins lie perpendicular to nickel's","Improper magnetic order: lattice only, spins static"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000453,"raw_usage":{"total_tokens":2398,"prompt_tokens":1185,"completion_tokens":1213,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":801,"completion_tokens_details":{"reasoning_tokens":1145}},"tokens_in":801,"tokens_out":1213,"duration_ms":8986,"temperature":1.0,"reasoning_tokens":1145,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T12:18:29.111424+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A neutron diffraction experiment on a deuterated sample, or one with enough counting time to resolve the symmetry-forbidden 1 1 1 reflection, that lets the out-of-plane moment refine without constraints: a clearly nonzero $M_z$, or a statistically better fit in the competing $Pn'ma'$ model, would falsify the claim that moments lie strictly in the $ab$ plane along $[1\\,1\\,0]$.","supporting_citations":[],"review_version":1}