{"id":"1192c161-a674-4875-b36c-98d3fa355cd2","arxiv_id":"1908.08701","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Field-induced suppression of the spin-Seebeck effect in LiCuVO4 is attributed to growing spin-nematic correlations that bind magnons into spin-2 pairs and block spin-1 interfacial exchange.","lead":"This paper reports that the spin-Seebeck effect in the frustrated quantum magnet LiCuVO4 is suppressed by an applied magnetic field even though the magnetization stays linear in field. The authors explain this as evidence for spin-nematic correlations that bind magnons into pairs, which cannot inject spin current at the interface.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Figure 3b aligns theory and data by normalizing B with different saturation fields (93 T vs 43 T); when plotted on a common physical-field axis the theoretical peak is near 9 T while the measured suppression at T=4 K turns over near 4-5 T, so the central quantitative comparison may be an artifact…","rationale":"The reader's conditional verdict is appropriate, but the single most load-bearing weakness is more specific than the reader's stated weakest assumption. The paper's only direct quantitative comparison between theory and experiment, Fig. 3(b), plots theory against B/Bs with Bs=93 T and data against B/Bs with Bs=43 T. This separate rescaling makes the curves overlap at B/Bs~0.1 even though the corresponding physical fields differ by a factor of about two: the theoretical peak is near 9 T, while the measured S at T=4 K shows negative slope above 5 T and a peak near 4-5 T. The authors' defense that the difference is not essential because Ebind is B-linear is not sufficient: B-linearity makes the gap a fixed function of physical B, but the peak position in Eq. (2) is set by the competition between the growing spectral weight (related to m(B/Bs)) and the growing gap. That competition can depend on Bs, so the separate normalizations may be hiding a real quantitative discrepancy. This is a concrete, addressable issue. The experimental observation itself is new and the magnon-pair mechanism is plausible; the problem is the strength of the quantitative support, not the viability of the physical picture. A replot on a common physical-field axis, or a refit with a consistent Bs inside the quoted experimental range, would settle the question. Secondary concerns including the neglect of interchain coupling and magnon-pair spin current are noted by the reader and would likely soften rather than eliminate the suppression, so they are less decisive. The conditional verdict remains unchanged.","tokens_in":13066,"tokens_out":9268,"duration_ms":102284,"concrete_test":"Generate Js(B) from Eq. (2) using exactly the stated parameters (J1/J2=-1, J2=50 K, Bs=93 T) and overlay the measured S(B) at T=4 K from Fig. 2(c) on a single physical-field axis in Tesla, without dividing either curve by its own Bs. Then repeat the calculation with J2 rescaled so that Bs=43 T while preserving the B-linear form of Ebind(B). If the theoretical peak remains near 9 T and does not shift toward ~4 T, the authors' claim that the Bs mismatch is inessential fails, and the agreement in Fig. 3(b) is an artifact of separate normalizations.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In the 'Comparison between experimental and theoretical results', the quantitative support for the central claim rests on Fig. 3(b), where the theoretical spin current Js is plotted against B/Bs with Bs=93 T (J2=50 K, J1/J2=-1) while the measured S is plotted against B/Bs with Bs~43 T. The two curves are therefore not compared at the same physical magnetic field: the theoretical peak at |B|~9 T corresponds to B/Bs~0.1, and the measured S at T=4 K also turns over near B/Bs~0.1, but that is |B|~4-5 T in the experiment. The authors argue the Bs difference is not essential because Ebind is B-linear. However, B-linearity only fixes the gap as a function of B; the prefactor and spectral weight in Eq. (2) involve the magnetization m(B/Bs), whose growth competes with the gap to set the peak position. Thus the peak field can depend on Bs even when Ebind(B) does not. The main text offers no replot on a common Tesla axis, so the claim that the B dependence of S is quantitatively captured by magnon-pair formation is not yet established.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports spin-Seebeck effect (SSE) measurements on LiCuVO4/Pt junctions. The authors observe a field-induced suppression of the transverse thermopower at low temperatures even though the magnetization is B-linear, and interpret it as evidence for the growth of spin-nematic (magnon-pair) correlations that bind single magnons into spin-2 pairs, thereby suppressing interfacial spin-1 exchange. They support this interpretation with a theoretical calculation of the injected spin current from a spin-nematic Tomonaga-Luttinger liquid with a B-linear magnon-pair binding energy, and claim good agreement with the B and T dependence of the data.","tokens_in":13296,"tokens_out":6663,"duration_ms":66100,"significance":"If the interpretation holds, the work would establish SSE as a probe of spin-nematic correlations in frustrated quantum magnets, a potentially valuable advance. The experimental data are clean and show a robust, reproducible suppression that is difficult to explain by simple Zeeman physics. The theoretical calculation is based on a transparent formula and shows that a single-magnon gap set by the magnon-pair binding energy can produce a broad peak in the spin current. However, the quantitative comparison in Fig. 3(b) is weakened by the use of different saturation fields for theory and experiment, and the temperature comparison in Fig. 4 is not calibrated against the measured magnetization. These issues are central to the claimed agreement.","major_comments":[{"comment":"The central quantitative comparison is made after normalizing B by different saturation fields: Bs = 93 T for the theory and Bs ≈ 43 T for the experiment. The theoretical spin current peaks near |B| ≈ 9 T (B/Bs ≈ 0.1) while the measured S at T = 4 K turns over near |B| ≈ 4–5 T (also B/Bs ≈ 0.1). The authors argue that the B-linearity of Ebind makes the Bs difference non-essential, but this argument is incomplete: in Eq. (2), the single-magnon spectral weight is also controlled by the magnetization m(B/Bs), and the peak position is set by the competition between the growth of m and the growth of the gap. Since m(B/Bs) changes with Bs at fixed physical B, the peak field can depend on Bs even if Ebind(B) is linear. The manuscript provides no replot on a common physical-field axis, so the claimed quantitative agreement is not established. Please show the calculation for the experimental Bs (or for a range of Bs values) on the same Tesla axis as the data and discuss the resulting peak position.","section":"Comparison between experimental and theoretical results, Fig. 3(b)"},{"comment":"The temperature comparison in Fig. 4 maps experimental B values to theoretical magnetization m values, but the manuscript does not use the measured M(B) curve from Fig. 2(c) to make this mapping. The theoretical m(B) depends on the same J1/J2 and Bs parameters that are already mismatched between theory and experiment. Without a calibrated mapping, the agreement in peak position and peak shift shown in Fig. 4 is qualitative. The authors should use the measured M(B) curve, or at least explicitly show the B-to-m relation used, to place the theoretical curves at the experimental fields.","section":"Fig. 4 and 'Comparison between experimental and theoretical results'"},{"comment":"The theoretical treatment assumes that the spin dynamics of LiCuVO4 in the measured B-T window is described by a purely 1D spin-nematic TLL, ignoring interchain coupling and coexisting spiral/SDW correlations. The authors acknowledge this and argue that integrating the other effects will yield a more quantitative result. However, the central claim of a 'well reproduced' B and T dependence rests on this simplified model, and no estimate is given for how the neglected interchain coupling J' (of order a few K) or the 3D spiral correlations might shift the peak field or suppress the signal. A quantitative error estimate on the predicted Js would help assess whether the agreement is meaningful.","section":"Spin-nematic nature of LiCuVO4 and 'Comparison between experimental and theoretical results'"}],"minor_comments":[{"comment":"Reference [19] is incomplete; it lacks the journal name, volume, and page numbers.","section":"References"},{"comment":"The sentence 'spin-Seebeck coefficients exhibit positive sign along with the same B dependences as those of M' should read 'positive sign and the same B dependence as that of M'.","section":"Introduction and experimental results"},{"comment":"The caption of Fig. 1(a) would be clearer if it explicitly stated which panel is the purely 1D phase diagram and which is the quasi-1D one; currently this is only discernible from the figure itself.","section":"Fig. 1 caption"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a timely topic and the experimental observation is interesting. The main risk is the overstatement of quantitative agreement between theory and experiment, particularly the different Bs used in Fig. 3(b). I would encourage the editor to require the authors to address the saturation-field issue before publication, either by replotting on a common Tesla axis or by fitting the theory with the experimental Bs."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The genuinely new thing here is the experimental observation: the spin-Seebeck signal in LiCuVO4 is suppressed by a field above about 2 T while the magnetization is strictly B-linear. That is a real, reproducible effect, and it is not something the existing SSE literature on ferromagnets or paramagnets would predict. Attaching it to magnon-pair binding is a fresh and reasonable hypothesis. To their credit, the theory uses an independent Ebind from the spin-nematic TLL literature rather than fitting it to the SSE data; the B and T dependences are captured qualitatively, including the field-driven shift of the peak in S(T). That is honest, and the paper openly states the main simplifying assumption: a purely 1D spin-nematic TLL with weak interchain coupling ignored, plus a coexistence of spiral/SDW correlations above 3 K. The logic is coherent and the authors do not oversell it as more than a consistency check.\n\nWhere it gets soft is the quantitative anchoring. The comparison in Fig. 3(b) is normalized and uses different saturation fields (93 T for theory, 43 T for data). The stress-test note argues this makes the common-field comparison meaningless. I do not entirely buy that specific objection: in the low-field linear regime, the peak position in this model is set by the ratio of the gap slope to temperature, not by the magnetization prefactor, so a different Bs would not move the theoretical peak much. The real problem is the opposite: the theory peak sits near 9 T while the data turn over near 4-5 T. That means the effective gap slope needed to suppress the SSE at 4 K is about twice what the calculation produces. The paper does not address this mismatch directly; it simply asserts that the Bs difference is not essential. A careful reader will want to see the same curves on a common Tesla axis, and ideally with parameters that reproduce the experimental saturation field. Also, the model ignores any magnon-pair contribution to the spin current, and the raw data are shown without error bars. These are common simplifications, but they mean the central claim rests on a qualitative consistency check rather than a quantitative derivation.\n\nWho is this for? Experimentalists working on spin Seebeck and quantum magnets, and theorists interested in multipolar correlations. It is a good candidate for a serious referee: the experiment is new and clean, and the interpretation, while not airtight, is well-motivated and falsifiable. I would send it to peer review, asking for a common-field replot and a direct discussion of the gap-slope discrepancy.","headline":"A real effect with a plausible but under-quantified magnon-pair interpretation; referee, but hold the authors to a common-field comparison.","tokens_in":13884,"tokens_out":5954,"would_cite":true,"duration_ms":63179,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["75.10.Pq","75.40.Gb","72.25.Ba","85.75.-d"],"model":"deepseek-v4-flash","headline":"In LiCuVO4, an applied magnetic field suppresses the spin-Seebeck effect despite a linear magnetization, because growing spin-nematic correlation binds single magnons into spin-2 pairs that cannot inject spin into the metal contact.","keywords":["spin-Seebeck effect","spin-nematic correlation","magnon pairs","LiCuVO4","frustrated spin-1/2 chain","Tomonaga-Luttinger liquid","spin current injection","quantum magnet"],"falsifier":"Measure the low-energy spin excitations of LiCuVO4 by inelastic neutron scattering at $B=9$ T and $T=4$ K: the claim predicts a single-magnon gap of about 3 K, equal to the magnon-pair binding energy, opening above about 2 T. If the spectrum instead shows a gapless continuum of single magnons in the same field and temperature range where the spin-Seebeck signal is suppressed, the central claim is falsified.","tokens_in":12837,"feed_emoji":"🧲","tokens_out":15167,"duration_ms":136549,"temperature":0.7,"pith_summary":"This paper reports that the spin-Seebeck voltage in a Pt/LiCuVO4 junction is suppressed by an applied magnetic field above about 2 T, even though the magnetization keeps growing linearly with field. The authors attribute the suppression to a field-induced growth of spin-nematic correlation: single magnons bind into spin-2 magnon pairs, opening an energy gap in the single-magnon spectrum and shutting off the spin-1 exchange that injects spin current into the platinum. A microscopic calculation of the interfacial spin current, in which single magnons carry a gap equal to the magnon-pair binding energy, reproduces the field and temperature dependence of the measured signal. The result matters because it turns the spin-Seebeck effect into a tool for detecting spin-nematic states and for identifying the spin quantum number of low-energy excitations.","feed_headline":"Field-induced magnon pairs quench a spin current in LiCuVO4","feed_subtitle":"In a frustrated spin chain, a magnetic field binds magnons into spin-2 pairs that cannot inject spin into platinum.","key_machinery":"The paper's working engine is the spin-nematic (magnon-pair) correlation of a frustrated spin-$\\frac12$ chain: a quadrupolar correlation in which magnons pair up into spin-2 objects, so that single magnons open an energy gap equal to the binding energy $E_{\\rm bind}$ while pairs remain gapless. The load-bearing formula is the linear-response expression for the interfacial spin current $\\tilde{J}_s$, which weights the single-magnon dynamical susceptibility $\\chi^{-+}_{\\rm mag}$ by a thermal kernel and shows that a gap $E_{\\rm bind}$ suppresses the spin-1 current injected into the metal contact. The calculation sets $E_{\\rm bind}$ from the $J_1$-$J_2$ chain with $J_1/J_2=-1$ and $J_2=50$ K, finding $E_{\\rm bind}$ to grow linearly with $B$ in the low-field regime, which converts a $B$-linear magnetization into a gapped single-magnon response at the interface.","core_discovery":"In a LiCuVO4/Pt junction, the transverse thermopower S associated with the spin-Seebeck effect initially grows with magnetic field but, above about 2 T, deviates below the $B$-linear trend and even acquires a negative slope beyond about 5 T at $T=4$ K, while the isothermal magnetization remains linear in $B$ over the same range. The paper's claim is that this suppression is a crossover from single-magnon to magnon-pair (spin-nematic) correlation: the field increases the binding energy $E_{\\rm bind}$ of spin-2 magnon pairs, opening a gap in the single-magnon spectrum, and because the interfacial exchange with conduction electrons transfers spin-1, the pair-carrying channel cannot inject spin current into the platinum. A calculation of the interfacial spin current from a spin-nematic Tomonaga-Luttinger description, with the single-magnon susceptibility gapped by $E_{\\rm bind}$, reproduces both the field dependence of $S$ and the broad peak in its temperature profile that shifts upward with increasing field. The claimed conclusion is that spin-Seebeck measurements probe spin-1 magnetic excitations selectively and can detect spin-nematic order in quantum magnets.","pith_inferences":["If the mechanism is correct, the same suppression should show up in other quasi-1D $J_1$-$J_2$ chain compounds, with the onset field scaling with $J_2$; comparing the spin-Seebeck field profiles of such compounds would test the generality of the magnon-pair picture.","A direct check would be to measure the field-dependent single-magnon gap spectroscopically (for instance, by inelastic neutron scattering or Raman scattering) and to verify that a gap close to $E_{\\rm bind}$ opens above roughly 2 T and tracks the spin-Seebeck suppression.","Since the authors ignore the magnon-pair contribution to the injected spin current, a residual, weakly field-dependent spin-Seebeck signal at high fields would bound the importance of higher-order interfacial processes that transfer spin-2.","The normalized comparison could be turned into a parameter-free test by extracting $E_{\\rm bind}(B)$ directly from the measured suppression and comparing it with the theoretical curve, without fixing $J_1$ and $J_2$ in advance."],"forward_implications":["The spin-Seebeck effect can serve as a probe of spin-nematic correlation: a field-induced suppression of the spin-Seebeck signal in a material whose magnetization remains $B$-linear is a sign that single magnons are being bound into spin-2 pairs.","The interfacial spin transfer in the spin-Seebeck effect is dominated by spin-1 exchange, so spin-2 magnon pairs do not inject spin current into the metal contact; this makes the effect a selective detector of the spin quantum number of magnetic excitations.","The broad peak in the temperature dependence of the spin-Seebeck signal, and its shift to higher temperatures as the field grows, arise from the competition between the magnon-pair gap, which suppresses single magnons at low temperature, and the increasing angular momentum per single magnon, which enhances the signal at higher temperature.","Because $E_{\\rm bind}$ grows linearly with $B$ in the low-field regime of the spin-nematic Tomonaga-Luttinger description, the field scale for the spin-Seebeck suppression is set by the exchange couplings of the chain, not by the Zeeman energy alone."],"supporting_citations":[{"why":"Supplies the linear-response theory that links the spin-Seebeck voltage to interfacial spin current produced by spin-1 exchange.","marker":"[3]"},{"why":"Previous one-dimensional spin-current study whose method and normalization the present spin-Seebeck analysis builds on.","marker":"[16]"},{"why":"Provides the nonequilibrium transport kernel used in deriving the spin-current formula (2).","marker":"[68]"},{"why":"Establishes the spin-nematic Tomonaga-Luttinger phase in the $J_1$-$J_2$ chain that the paper assumes for LiCuVO4.","marker":"[32]"},{"why":"Supplies the calculation of the magnon-pair binding energy $E_{\\rm bind}(B)$ used to set the single-magnon gap.","marker":"[36]"},{"why":"Reports field suppression of the spin-Seebeck effect in ferrimagnets and paramagnets, the Zeeman-gap interpretation the paper argues against.","marker":"[8]"},{"why":"Determines the exchange couplings $J_2$ and $J_1/J_2$ of LiCuVO4 from neutron scattering, fixing the model parameters.","marker":"[42]"},{"why":"Provides the $B$-$T$ phase diagram of LiCuVO4 used to place the spin-Seebeck suppression above the ordering temperatures.","marker":"[45]"}],"fun_headline_variants":["Field-induced magnon pairs suppress spin-Seebeck effect","Spin-nematic correlation quenches spin injection in LiCuVO4","Magnon pairs trap spin-1, stifling spin-Seebeck current","LiCuVO4: field binds magnons into pair-blocked spin current","Spin-2 magnon pairs gate spin-Seebeck response"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The measured suppression is produced solely by a field-induced energy gap in LiCuVO4's single-spin-flip excitations, equal to the magnon-pair binding energy computed for an isolated one-dimensional $J_1$-$J_2$ chain, with weak interchain coupling, coexisting spiral or spin-density-wave correlations, and any magnon-pair-driven spin current all making negligible contributions.","fun_headline_variants_meta":{"raw":{"variants":["Field-induced magnon pairs suppress spin-Seebeck effect","Spin-nematic correlation quenches spin injection in LiCuVO4","Magnon pairs trap spin-1, stifling spin-Seebeck current","LiCuVO4: field binds magnons into pair-blocked spin current","Spin-2 magnon pairs gate spin-Seebeck response"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000197,"raw_usage":{"total_tokens":1376,"prompt_tokens":965,"completion_tokens":411,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":581,"completion_tokens_details":{"reasoning_tokens":318}},"tokens_in":581,"tokens_out":411,"duration_ms":4093,"temperature":1.0,"reasoning_tokens":318,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:31:12.208689+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the low-energy spin excitations of LiCuVO4 by inelastic neutron scattering at $B=9$ T and $T=4$ K: the claim predicts a single-magnon gap of about 3 K, equal to the magnon-pair binding energy, opening above about 2 T. If the spectrum instead shows a gapless continuum of single magnons in the same field and temperature range where the spin-Seebeck signal is suppressed, the central claim is falsified.","supporting_citations":[{"cited_title":"Adachi, J","cited_arxiv_id":null,"evidence_quote":"Supplies the linear-response theory that links the spin-Seebeck voltage to interfacial spin current produced by spin-1 exchange."},{"cited_title":"Hirobe, M","cited_arxiv_id":null,"evidence_quote":"Previous one-dimensional spin-current study whose method and normalization the present spin-Seebeck analysis builds on."},{"cited_title":"Jauho, N","cited_arxiv_id":null,"evidence_quote":"Provides the nonequilibrium transport kernel used in deriving the spin-current formula (2)."},{"cited_title":"Hikihara, L","cited_arxiv_id":null,"evidence_quote":"Establishes the spin-nematic Tomonaga-Luttinger phase in the $J_1$-$J_2$ chain that the paper assumes for LiCuVO4."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the calculation of the magnon-pair binding energy $E_{\\rm bind}(B)$ used to set the single-magnon gap."},{"cited_title":"Kikkawa, K","cited_arxiv_id":null,"evidence_quote":"Reports field suppression of the spin-Seebeck effect in ferrimagnets and paramagnets, the Zeeman-gap interpretation the paper argues against."},{"cited_title":"Enderle, C","cited_arxiv_id":null,"evidence_quote":"Determines the exchange couplings $J_2$ and $J_1/J_2$ of LiCuVO4 from neutron scattering, fixing the model parameters."},{"cited_title":"B¨ uttgen, P","cited_arxiv_id":null,"evidence_quote":"Provides the $B$-$T$ phase diagram of LiCuVO4 used to place the spin-Seebeck suppression above the ordering temperatures."}],"review_version":1}