{"id":"1ac35187-ebf7-4bcb-9b79-d8eed56f4e7c","arxiv_id":"2508.12759","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Pulsed-field 31P NMR detects a double-peak spectral change above 29 T along [001] in Pb(TiO)Cu4(PO4)4, consistent with the theoretically predicted phase III magnetic structure.","lead":"Pulsed-field 31P NMR reveals a new magnetic transition near 29 tesla in the square cupola antiferromagnet Pb(TiO)Cu4(PO4)4, seen as a double-peak spectral change along [001]. The data support a spin reorientation into the predicted phase III, helping explain an unsolved dielectric anomaly in this multiferroic material.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Low-field hyperfine calibration in §III B fails to reproduce the measured Bz component; the 25%-shifted phase-III comparison in Fig. 6(c) is therefore not a clean confirmation.","rationale":"The reader's conditional verdict is well calibrated, but the strongest specific weakness is not simply the external possibility that A_i are wrong; it is an internal inconsistency in the stated derivation. Under the same assumptions, the published A_i predict Bz≈31 mT while the measured value is 48 mT. This is not a negligible detail: the H∥[001] high-field analysis compares only two branches of Bint, and the comparison is completed by shifting the entire calculated spectrum by 25%, which is comparable to the calibration error. Re-solving the equations shows a self-consistent solution exists with somewhat different A_i, so the issue is fixable; until it is redone, Fig. 6(c) should be treated as an illustrative comparison rather than as definitive evidence for phase III. The double-peak NMR observation above 29 T remains a solid experimental result, so a conditional verdict is appropriate and no change from the reader's verdict is needed.","tokens_in":11745,"tokens_out":8570,"duration_ms":84393,"concrete_test":"Re-fit A1..A4 and My (and, if needed, the off-diagonal hyperfine components) to the full measured Bhf=(13,-10,48) mT at 7 T using the equations in §III B, without imposing My=0 or A3=A4, and with Ap fixed at 635 mT/µB. Then use the best-fit couplings to recompute the phase-III Bint curves in Fig. 6(a) and the spectrum at 32.2 T without the ad hoc 25% shift. If the double-peak fit requires a different shift or fails, the phase-III identification is not supported by the present analysis.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing step is the low-field calibration of the P-site hyperfine couplings used to simulate the high-field internal fields. The stated constants A1=185, A2=136, A3=A4=157 mT/µB, together with Mx=0.48, Mz=0.64 and My≈0, do not satisfy the same equations from which they are said to come. With these numbers, Bx=(A1−A3)Mx=13.4 mT and By=(A2−A4)Mx=−10.1 mT match the reported hyperfine field, but Bz=(A1−A2−A3+A4)Mz=(A1−A2)Mz=31.4 mT, against the reported 48 mT. Re-solving all three equations plus Ap=635 with A3=A4 yields a different set (A1≈197, A2≈122, A3=A4≈158 mT/µB, My≈0.02µB) that does reproduce Bz. The paper never reports this discrepancy and later rescales all phase-III Bint values by 25% to obtain the match in Fig. 6(c). Because the H∥[001] phase assignment is based on comparing only two Bint branches, a calibration that already fails one measured component by roughly 35% is a serious unaddressed source of error; the agreement at 32.2 T cannot be regarded as a clean confirmation of phase III until the hyperfine tensor is re-derived consistently.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports 31P NMR measurements in flat-top pulsed magnetic fields up to 32.2 T on the square cupola antiferromagnet Pb(TiO)Cu4(PO4)4, extending the field range accessible to NMR studies of this material. For H along the [100] and [001] directions, the authors observe field-induced changes in the NMR spectra and identify a double-peak structure emerging above about 29 T for H || [001], which they attribute to a magnetic transition into a phase previously denoted phase III. They calibrate site-dependent 31P hyperfine couplings from the low-field antiferromagnetic state, simulate the internal fields at the P sites using cluster mean-field and classical mean-field models, and compare the simulated spectra with the high-field NMR data. The paper proposes that phase III is stabilized above 29 T and discusses the possible role of magnetostriction in suppressing the chiral rotation of the cupola clusters.","tokens_in":11992,"tokens_out":3353,"duration_ms":34752,"significance":"If the phase assignment is correct, the paper provides the first microscopic NMR evidence for the high-field phase III in this magnetoelectric square cupola compound, connecting a previously unexplained dielectric anomaly at 26 T with a magnetic transition. The experimental achievement is substantial: high-resolution NMR spectra in flat-top pulsed fields up to 32.2 T, with a consistent frequency-sweep technique, are not routine. The paper also ships a quantitative framework linking hyperfine couplings, dipole fields, and mean-field magnetic structures. However, the central phase III claim rests on a low-field calibration of hyperfine constants that is internally inconsistent with the measured internal-field components, and on a 25% uniform rescaling of the simulated internal fields. Until these points are addressed, the comparison in Fig. 6(c) cannot be regarded as a clean confirmation of the phase III structure.","major_comments":[{"comment":"The hyperfine constants A1 = 185, A2 = 136, and A3 = A4 = 157 mT/µB quoted in Sec. III B do not reproduce the reported hyperfine field component Bz = 48 mT. Using the same equations and the stated Mx = 0.48 µB, Mz = 0.64 µB, and My = 0, one obtains Bx = 13.4 mT and By = -10.1 mT (consistent), but Bz = (A1 - A2 - A3 + A4)Mz = (A1 - A2)Mz = 31.4 mT, which is about 35% smaller than the measured 48 mT. This internal inconsistency in the calibration is load-bearing because these A_i are used for all subsequent simulations of the high-field internal fields. The authors should re-derive the A_i by solving the full set of equations, including the small My term, and should report the resulting values and the residuals for all three components.","section":"Sec. III B"},{"comment":"The comparison between the simulated and experimental 31P-NMR spectrum at 32.2 T is made after uniformly shifting the whole simulated spectrum by 25% to 'compensate the anisotropy of hyperfine coupling constants.' This is a large ad hoc rescaling: the measured anisotropy of Ap between the [100] and [001] directions is about 12%, roughly half of the applied shift. The paper does not provide a quantitative derivation of the 25% correction from the angle-dependent hyperfine coupling. Since the phase III assignment rests on the agreement of only two internal-field branches after this rescaling, the authors should either justify the shift with a concrete model or show the comparison without it, so that the level of agreement is transparent.","section":"Sec. IV, Fig. 6(c)"},{"comment":"The classical mean-field calculation that produces the phase III magnetic structure uses model parameters that were 'slightly tuned to better reproduce the magnetization curve' (footnote [26]), specifically D = 0.8, θ = 100°, and φ = 2°. The paper does not state which values of these parameters are used for the other phases, nor how sensitive the predicted Bint values are to these adjustments. Because the identification of phase III and the resulting internal-field branches depend on this tuned model, the authors should provide a small parameter-sweep or robustness test showing that the double-peak assignment is not an artifact of the tuning.","section":"Sec. IV, footnote [26]"}],"minor_comments":[{"comment":"The manuscript contains numerous typographical errors, including 'antiferromganetic' (Introduction), 'Metoropolitan' (affiliation), 'solides' (Ref. [1]), 'expacsion' (Ref. [5]), 'qupola' (Ref. [12]), 'fo' (Ref. [5]), 'dn' (Ref. [19]), 'Tha' (footnote [26]), and 'intneisty' (Fig. 3 caption). These should be corrected.","section":"Global"},{"comment":"The caption of Fig. 6(c) states that the whole spectrum was uniformly shifted by 25% but does not specify the direction or the absolute value of the shift. Please state whether the shift is in Bint or frequency and in which direction, and give the magnitude in mT.","section":"Sec. IV, Fig. 6(c)"},{"comment":"The statement that the coupling constants for the [100] and [001] directions are 662 mT/µB and 564 mT/µB, while the values after dipole subtraction are 674 and 557 mT/µB, could be made clearer: it is not immediately obvious why the dipole subtraction increases the [100] value and decreases the [001] value. A brief parenthetical explanation would help.","section":"Sec. III A"},{"comment":"The local coordinate definitions for M1 in Fig. 3(c) (x, y, z parallel to [010], [‾100], and [001], respectively) are mentioned in the text but are not reproduced in the figure caption. Adding them to the caption would make the equations easier to follow.","section":"Sec. III B"}],"recommendation":"major_revision","confidential_remarks":"The calibration inconsistency in Sec. III B is verifiable from the equations given in the paper and is not merely a matter of presentation. The 25% uniform rescaling in Fig. 6(c) should also be addressed head-on. If the authors can re-derive consistent hyperfine constants and justify or remove the rescaling, the paper could be publishable. The experimental data on the double-peak structure above 29 T are convincing; the issue is the interpretation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, the pulsed-field 31P NMR data above 29 T are genuinely new, and the double-peak splitting at Bc2 is the first microscopic evidence of the transition that previously showed up only as a dielectric anomaly near 26 T. That part is solid and worth publishing. Second, the assignment to phase III is not as clean as the paper claims. The low-field hyperfine calibration in Section III B is internally inconsistent: the reported A1 = 185, A2 = 136, A3 = A4 = 157 mT/µB, together with Mx = 0.48 µB, Mz = 0.64 µB and My ≈ 0, give Bz = (A1 - A2) Mz ≈ 31 mT, not the measured 48 mT after dipole subtraction. Re-solving the same equations with Ap = 635 mT/µB and A3 = A4 yields A1 ≈ 197, A2 ≈ 122, A3 = A4 ≈ 158 mT/µB, and My ≈ 0.02 µB. That My is small, but it shifts Bx - By by roughly Ap·My ≈ 13 mT, so neglecting it is not harmless. The paper never reports this discrepancy, and later rescales the computed phase-III internal fields by 25% to make the comparison work. The agreement at 32.2 T in Fig. 6(c) is therefore a plausibility argument, not a confirmation. The experimental method is state of the art, the spectra are shown in full, and the H // [100] analysis using cluster mean-field theory is reasonable. The observation of successive transitions is robust. But the phase assignment is model-dependent: the mean-field parameters are tuned to the magnetization curve, and the alternative scenario of a field-induced reduction of the chiral angle is acknowledged but not tested. A high-field neutron or x-ray diffraction experiment would settle it. This paper deserves a serious referee, not a desk reject. The new data warrant publication, but the hyperfine tensor needs to be re-derived consistently and the 25% rescaling must be justified or removed before the phase III identification becomes convincing. I would send it to review with a request for major revision.","headline":"Genuinely new pulsed-field NMR data, but the phase III assignment rests on an internally inconsistent hyperfine calibration.","tokens_in":12618,"tokens_out":4486,"would_cite":true,"duration_ms":42064,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims that 31P NMR in pulsed fields up to 32.2 T reveals a successive magnetic transition in Pb(TiO)Cu4(PO4)4 above 29 T along [001], and that the spectrum at 32.2 T matches the internal fields expected for the theoretically…","keywords":["31P NMR","pulsed high magnetic fields","square cupola antiferromagnet","magnetic quadrupole order","field-induced phase III","cluster mean-field theory","metamagnetic transition","hyperfine coupling constants"],"falsifier":"A direct neutron or x-ray diffraction measurement of the magnetic structure above 29 T along [001] would settle whether the spin reorientation predicted for phase III actually occurs; the proposed phase III structure predicts two distinct internal fields at phosphorus sites, so an NMR measurement that fully accounts for the anisotropic hyperfine coupling could also test whether the 25% correction applied to the simulated spectrum hides a different magnetic structure.","tokens_in":11489,"feed_emoji":"🧲","tokens_out":4067,"duration_ms":35710,"temperature":0.7,"pith_summary":"In the square cupola antiferromagnet Pb(TiO)Cu4(PO4)4, the paper uses 31P NMR in pulsed magnetic fields up to 32.2 T to track how the ordered copper moments rearrange as the field grows. It finds a new spectral splitting above 29 T applied along [001] that signals a successive magnetic transition, and it shows that the measured spectrum at 32.2 T matches the internal fields expected for the theoretically predicted phase III. If correct, this identifies the previously unresolved high-field phase and its magnetic structure.","feed_headline":"NMR in pulsed fields reveals new magnetic phase above 29 tesla","feed_subtitle":"Phosphorus lines split again at 29 T in Pb(TiO)Cu4(PO4)4, evidence for the predicted phase III spin structure.","key_machinery":"The analysis rests on 31P NMR spectroscopy under flat-top pulsed magnetic fields, which keeps the field constant long enough to record frequency-swept spectra up to 32.2 T. The internal field at each of the eight phosphorus sites is computed as the sum of hyperfine fields from four neighboring Cu2+ moments, using site-dependent coupling constants A1 = 185, A2 = 136, and A3 = A4 = 157 mT/µB extracted from the low-field antiferromagnetic state, plus the direct dipole fields. These simulated internal fields are compared with measured peak positions for fields along [100] and [001], and the match above 29 T is used to identify phase III.","core_discovery":"The paper establishes that above 29 T, applied along the c-axis [001], the 31P NMR spectrum of Pb(TiO)Cu4(PO4)4 changes from a broad single shape into a double-peak structure, evidencing successive magnetic transitions in a field range inaccessible to superconducting magnets. Simulating the internal fields at the phosphorus sites from the candidate magnetic structures, the authors find that the observed double-peak spectrum at 32.2 T is consistent with the phase III spin configuration obtained from classical mean-field theory, leading to the proposal that phase III is stabilized above 29 T. This assignment resolves the dielectric anomaly at 26 T that had no clear magnetization counterpart, and it provides a microscopic picture of the field-induced spin reorientation.","pith_inferences":["The 25% uniform correction applied to the simulated internal field suggests the anisotropic part of the hyperfine coupling is not negligible; a full tensor determination, for example from measurements at intermediate orientations, could replace this correction and test the phase III assignment more strictly.","If magnetostriction suppresses the chiral rotation angle of the cupola clusters in high fields, as the authors speculate, then high-field x-ray diffraction might detect a lattice deformation accompanying the 29 T transition.","The same pulsed-field NMR approach could be applied to the isostructural Ba and Sr compounds, where the phase III window may shift and provide a systematic test of the mean-field prediction."],"forward_implications":["Above 29 T along [001], the double-peak NMR spectrum is a marker of a successive magnetic transition, so future magnetization and electric polarization measurements can be re-examined in that field range.","The proposed phase III spin reorientation accounts for the 26 T dielectric anomaly that had no clear magnetization signature, linking a purely electric response to a magnetic transition.","The site-dependent hyperfine coupling constants derived here (A1 = 185, A2 = 136, A3 = A4 = 157 mT/µB) allow the local spin arrangement to be inferred from NMR in both low-field and field-induced states.","The combination of flat-top pulsed fields with fast frequency-skipping NMR makes microscopic magnetic-structure studies possible well beyond the reach of superconducting magnets, up to at least 32 T."],"supporting_citations":[{"why":"Provides the low-field magnetic quadrupolar structure and neutron diffraction result used to fix the copper moment components in the hyperfine analysis.","marker":"[9]"},{"why":"Supplies the spin model and the magnetization and electric polarization data to which the cluster mean-field calculation is tuned.","marker":"[16]"},{"why":"Gives the critical fields Ba = 14.8 T and Bc1 = 12.3 T used to scale the theory and to identify the field range of the observed transitions.","marker":"[17]"},{"why":"Introduces the cluster mean-field theory and the phase Y and phase III labels used throughout the analysis.","marker":"[18]"},{"why":"Extends the cluster multipole model to comparison compounds and gives the parameter set used for the theoretical calculations.","marker":"[19]"},{"why":"Reports the 26 T dielectric anomaly that the proposed phase III assignment is invoked to explain.","marker":"[20]"},{"why":"Describes the fast frequency-skipping NMR spectrometer used to record transient Fourier-transform spectra within a single field pulse.","marker":"[21]"},{"why":"Describes the flat-top pulsed-field generation with feedback control that makes constant-field NMR measurements possible.","marker":"[25]"}],"fun_headline_variants":["Pulsed-field NMR catches new magnetic phase above 29 T","Double-peak NMR spectra mark new phase above 29 T","High-field NMR exposes hidden phase in Pb(TiO)Cu4(PO4)4","29 T NMR double-peak signals successive spin transitions"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The identification of phase III relies on hyperfine coupling constants derived assuming A3 = A4, taking the copper moment components Mx = 0.48 µB and Mz = 0.64 µB from neutron diffraction, and neglecting the small My component and off-diagonal anisotropic hyperfine terms; if these site-dependent couplings or the neglected anisotropy are wrong, the simulated internal fields and the phase assignment would change.","fun_headline_variants_meta":{"raw":{"variants":["Pulsed-field NMR catches new magnetic phase above 29 T","Double-peak NMR spectra mark new phase above 29 T","High-field NMR exposes hidden phase in Pb(TiO)Cu4(PO4)4","29 T NMR double-peak signals successive spin transitions"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001061,"raw_usage":{"total_tokens":4402,"prompt_tokens":851,"completion_tokens":3551,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":467,"completion_tokens_details":{"reasoning_tokens":3476}},"tokens_in":467,"tokens_out":3551,"duration_ms":27879,"temperature":1.0,"reasoning_tokens":3476,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T17:18:31.921596+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct neutron or x-ray diffraction measurement of the magnetic structure above 29 T along [001] would settle whether the spin reorientation predicted for phase III actually occurs; the proposed phase III structure predicts two distinct internal fields at phosphorus sites, so an NMR measurement that fully accounts for the anisotropic hyperfine coupling could also test whether the 25% correction applied to the simulated spectrum hides a different magnetic structure.","supporting_citations":[{"cited_title":"The modification in the spectral shape above Bc1 6 0 2 4 6 8 5 10 15 20 25 30 35 NMR intensity offset by� 0H (arb","cited_arxiv_id":null,"evidence_quote":"Gives the critical fields Ba = 14.8 T and Bc1 = 12.3 T used to scale the theory and to identify the field range of the observed transitions."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports the 26 T dielectric anomaly that the proposed phase III assignment is invoked to explain."}],"review_version":2}