{"id":"95a8c26a-66a5-4681-a94b-07a7db2f5c86","arxiv_id":"2607.10877","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.5,"correctness_risk":"low","formal_verification":"none","parameter_count":4,"one_line_summary":"Cu NMR spectra up to 41.9 T confirm that the 1/8 and 2/15 plateaus of SrCu2(BO3)2 are Wigner crystals of pinwheel spin-2 bound states, incompatible with extended-triplet crystals.","lead":"High-field copper NMR shows that the 1/8 and 2/15 magnetization plateaus in SrCu2(BO3)2 are crystals of pinwheel-shaped spin-2 bound triplet pairs, not ordinary triplets. This experimentally establishes Wigner crystals of bound states as a distinct mechanism for magnetization plateaus.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"No significant objection identified","rationale":"The reader correctly isolates the transferability of the hyperfine set and the T2 interpretation of missing lines as the softest points. Both are real but secondary: the qualitative mismatch (non-physical B/C hierarchy plus absence of the most-positive Hi predicted by extended triplets) survives even if A is floated by the observed 12 % magnetostriction and even if the wiped-out intensity is reassigned. No internal inconsistency or hidden assumption that would reverse the pinwheel-versus-triplet conclusion is present. The verdict therefore remains ACCEPT with high confidence.","tokens_in":25151,"tokens_out":429,"duration_ms":5510,"concrete_test":"Re-fit the 1/8 and 2/15 Hi distributions while allowing A to vary by the full 12 % discrepancy reported between 41.9 T and the low-field Acc(q=0); if the pinwheel structures still yield |B| < |C| and the extended-triplet structures still require |B| > |C| with a missing high-frequency line, the class discrimination is robust.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim holds under scrutiny. The 1/3-plateau calibration of A (and the sum-rule Acc(q=0)) is performed on an uncontroversial three-site stripe structure that all methods agree on; the resulting |A| matches independent Cu2+ values, and |C|,|B| remain small as expected. When these couplings are transferred to the dilute plateaus, pinwheel histograms reproduce the observed Hi < 0 lines with physical B,C while extended-triplet candidates force non-physical |B| > |C| and predict an isolated most-positive Hi that is absent from the data. Residual issues (field-dependent transferred terms, T2 wipe-out of near-zero lines, neglected DM/3D) affect quantitative details of B,C and line intensities but do not invert the class-level discrimination between pinwheels and extended triplets.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript reports 63,65Cu NMR spectra of SrCu2(BO3)2 up to 41.9 T in the 1/8, 2/15 and 1/3 magnetization plateaus, combined with high-accuracy iPEPS spin structures. From the uncontroversial three-site stripe structure of the 1/3 plateau the authors extract the on-site hyperfine constant A ≈ −20.64 T/μB (and small transferred C, B) via Eq. (2). Transferring this A and fitting only B and C, they show that the measured local-field distributions in the 1/8 and 2/15 plateaus match the histograms of pinwheel-shaped spin-2 bound states and are incompatible with crystals of extended triplets (which force non-physical |B| ≫ |C| and predict an isolated most-positive Hi that is absent experimentally). The work thereby supplies the first real-space experimental confirmation that the low-magnetization plateaus are Wigner crystals of bound triplet pairs rather than ordinary triplon crystals.","tokens_in":25419,"tokens_out":1082,"duration_ms":11692,"significance":"If the interpretation holds, the paper establishes a new fundamental paradigm for magnetization plateaus: crystallization of composite spin-2 bound states rather than elementary triplons or semiclassical collinear configurations. The experimental discrimination rests on a clean calibration of the dominant hyperfine coupling from an uncontroversial plateau and on a direct, falsifiable comparison of local-field histograms; residual free parameters (B, C) cannot invert the class-level distinction. The result is of immediate interest to the frustrated-magnetism and flat-band communities and is supported by state-of-the-art iPEPS data that improve on earlier variational energies.","major_comments":[{"comment":"Sec. IV and Fig. 4: the decisive discrimination between pinwheels and extended triplets relies on the assumption that the on-site A fixed at 41.9 T remains field-independent down to 27–29 T, while only the transferred B and C are allowed to vary. The authors themselves note a 12 % discrepancy between Acc(q=0) at 41.9 T and the low-field value, which they attribute to magnetostriction of the transferred terms. A short quantitative estimate of how large a field-induced change in A would have to be to make an extended-triplet structure produce physical B, C (or to spoil the pinwheel fit) would strengthen the claim that the class-level conclusion is robust.","section":null},{"comment":"Sec. IV, Fig. 4(b,c) and SM Table I: several predicted lines near Hi ≈ 0 (and one weakly positive line) are missing and are ascribed to short-T2 wipe-out. While this is plausible for nearly unpolarized spins, the argument is currently qualitative. A brief estimate of the expected T2 contrast, or a statement that the missing intensity cannot be reassigned to a different spin structure without destroying the fit of the observed Hi < 0 lines, would close this residual loophole.","section":null}],"minor_comments":[{"comment":"Fig. 2 and SM Fig. S2: the numerical spin-polarization values are given, but a compact table of the grouped/averaged polarizations actually used in the least-squares fits of Fig. 4 would make the comparison fully reproducible without re-digitizing the histograms.","section":null},{"comment":"Eq. (2) and the subsequent fit: the relative signs and magnitudes of B and C obtained for the optimal rhomboid structure (B = −5.07 T/μB, C = 0.21 T/μB) differ markedly from the 1/3-plateau values; a one-sentence physical comment on why the transferred couplings can change so strongly between dense and dilute plateaus would help the reader.","section":null},{"comment":"SM Sec. VI (3D stacking): the ABAB stacking argument for the 1/3 plateau is convincing; a short remark on why analogous inter-plane splittings are not resolved in the dilute 1/8 and 2/15 spectra would complete the discussion.","section":null},{"comment":"Typographical: “MA TERIAL”, “RESUL TS”, “POLARIZA TION” etc. contain spurious spaces; “sextuplets” is used consistently but “sextuplet” would be more standard.","section":null}],"recommendation":"minor_revision","confidential_remarks":"The central experimental discrimination is solid and the paper is already close to publishable. The two major points I raise are refinements rather than obstacles; either can be addressed with a short paragraph or an additional SM estimate. I see no reason to delay publication beyond a minor revision."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This paper finally gives a clean spectroscopic answer to a long-standing question in SrCu2(BO3)2. Using full 63,65Cu spectra up to 41.9 T, they fix the on-site hyperfine A from the uncontroversial three-site 1/3 stripe (A ≈ −20.6 T/μB, physically sensible and consistent with other Cu2+ oxides), then show that the local-field distributions at 27.6 T and 29 T match the iPEPS pinwheel spin-2 histograms and are incompatible with extended-triplet crystals (the latter force non-physical |B| > |C| and predict an isolated most-positive Hi that is simply missing).\n\nWhat is new is not the pinwheel idea itself (Corboz & Mila 2014) but the decisive on-site discrimination. Earlier Cu and 11B work mapped the plateaus without being able to tell the two classes of textures apart. The 1/3 calibration plus the improved high-frequency signal quality make the comparison quantitative enough to reject one class. The iPEPS structures are treated as competing hypotheses; experiment selects among them. Histograms, sum-rule Acc(q=0), and the ABAB stacking analysis for the 1/3 line splittings all look carefully done. Citations are appropriate and the math/data chain is transparent.\n\nSoft spots are real but secondary. A is transferred from 41.9 T dense packing to 27–29 T dilute packing; transferred B and C are fitted (and show some field dependence already noted by the authors). Near-zero and weakly positive lines are wiped out and attributed to short T2; that is plausible but not proven. DM and 3D couplings are neglected for the dilute phases. None of these invert the class-level conclusion: pinwheels fit with physical parameters, triplets do not. No public code/data is a minor practical annoyance, not a soundness issue.\n\nThis is for people who work on magnetization plateaus, Shastry-Sutherland physics, or flat-band composites. It deserves a serious referee and should be engaged with; I would cite the spectroscopic discrimination.","headline":"Solid experimental discrimination: Cu NMR calibrated on the 1/3 plateau selects pinwheel spin-2 crystals over extended triplets for the 1/8 and 2/15 plateaus.","tokens_in":26058,"tokens_out":543,"would_cite":true,"duration_ms":6564,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["75.10.Jm","75.25.-j","76.60.-k","75.30.Kz"],"model":"grok-4.5","headline":"NMR spectra of SrCu2(BO3)2 show that its low-field magnetization plateaus are Wigner crystals of pinwheel-shaped spin-2 bound triplet pairs, not ordinary triplets.","keywords":["magnetization plateaus","Shastry-Sutherland","spin-2 bound states","Wigner crystal","63,65Cu NMR","iPEPS","SrCu2(BO3)2","triplon pairs"],"falsifier":"A copper NMR spectrum in the 1/8 or 2/15 plateau that recovers the full intensity of the near-zero and weakly positive local-field lines and shows the isolated most-positive field expected for extended triplets, or a re-fit that forces non-physical transferred couplings larger than the on-site term.","tokens_in":26067,"feed_emoji":"🧲","tokens_out":642,"duration_ms":7099,"temperature":0.7,"pith_summary":"For years the low-magnetization plateaus of the quantum antiferromagnet SrCu2(BO3)2 could not be explained by the usual pictures of ordered up-down spins or crystals of elementary triplets. Tensor-network calculations had suggested instead that the 1/8 and 2/15 plateaus are lattices of composite spin-2 objects, each a bound pair of triplets that spreads over eight dimers in a pinwheel pattern. High-field copper NMR now maps the local spin polarizations directly. Hyperfine couplings fixed from the well-understood 1/3 plateau turn the measured local-field histograms into a fingerprint that matches the pinwheel polarizations and rules out the extended-triplet alternatives. The result establishes a new, experimentally verified mechanism for magnetization plateaus: first form a bound multiparticle state, then let those composites crystallize.","feed_headline":"NMR confirms pinwheel spin-2 crystals in quantum magnet plateaus","feed_subtitle":"Low-field plateaus of SrCu2(BO3)2 are lattices of bound triplet pairs, not ordinary triplets","key_machinery":"Pinwheel-shaped spin-2 bound triplet pairs: composite objects delocalized over eight dimers whose Wigner-crystal arrangements produce the distinctive intermediate positive and weakly negative spin polarizations seen in the NMR histograms.","core_discovery":"Using 63,65Cu NMR spectra up to 41.9 T and hyperfine couplings constrained by the non-controversial 1/3 plateau, the local-field distributions of the 1/8 and 2/15 plateaus match iPEPS spin polarizations of pinwheel-shaped spin-2 bound states and are incompatible with crystals of extended triplets.","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["NMR spectra match pinwheel spin-2 bound states in low plateaus","Pinwheel spin-2 Wigner crystals confirmed for 1/8 and 2/15 plateaus","Bound triplet pairs, not triplets, form crystals in SrCu2(BO3)2 plateaus","Hyperfine data rule out extended triplets in 1/8 and 2/15 plateaus","Local fields confirm spin-2 pinwheels stabilize magnetization plateaus"],"cache_read_input_tokens":16512,"weakest_assumption_plain":"The on-site hyperfine constant fixed in the dense 1/3 plateau at 41.9 T, together with only two adjustable transferred couplings, still correctly maps local fields to spins in the dilute 1/8 and 2/15 plateaus at lower field, and the missing near-zero lines are only a relaxation effect.","fun_headline_variants_meta":{"raw":{"variants":["NMR spectra match pinwheel spin-2 bound states in low plateaus","Pinwheel spin-2 Wigner crystals confirmed for 1/8 and 2/15 plateaus","Bound triplet pairs, not triplets, form crystals in SrCu2(BO3)2 plateaus","Hyperfine data rule out extended triplets in 1/8 and 2/15 plateaus","Local fields confirm spin-2 pinwheels stabilize magnetization plateaus"]},"model":"grok-4.5","effort":"low","cost_usd":0.00554,"raw_usage":{"total_tokens":1456,"prompt_tokens":751,"num_sources_used":0,"completion_tokens":115,"cost_in_usd_ticks":55400000,"prompt_tokens_details":{"text_tokens":751,"audio_tokens":0,"image_tokens":0,"cached_tokens":128},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":590,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":751,"tokens_out":115,"duration_ms":5929,"temperature":1.0,"reasoning_tokens":590,"cache_read_input_tokens":128,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-14T08:35:44.541336+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"A copper NMR spectrum in the 1/8 or 2/15 plateau that recovers the full intensity of the near-zero and weakly positive local-field lines and shows the isolated most-positive field expected for extended triplets, or a re-fit that forces non-physical transferred couplings larger than the on-site term.","supporting_citations":[],"review_version":1}