{"id":"9ad96e90-fb3b-412f-8460-4067f1baf70e","arxiv_id":"2507.21403","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"In La2CuO4, the thermal Hall conductivity is below 1e-4 W/mK^2 between 2 and 20 K in fields up to 10 T, contradicting earlier reports of a large signal.","lead":"A high-resolution thermal conductivity study of the parent cuprate La2CuO4 finds no detectable thermal Hall signal, setting an upper bound 10 to 30 times smaller than previously reported signals. The result challenges the chiral phonon interpretation of the thermal Hall effect in the cuprate pseudogap if confirmed.","discovery_kind":"replication","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Unquantified hysteresis residual in the in-out subtraction could shift κxy/T by more than the claimed 1e-4 W/mK2 bound.","rationale":"The paper is a careful, high-resolution search, and the longitudinal channel is validated by agreement with prior κxx data. However, the transverse channel's power to set a 10−4 W/mK² bound rests entirely on the in-out subtraction. The material is known to exhibit weak ferromagnetism and metamagnetic hysteresis in this field range; the authors demonstrate hysteresis in ΔTxx but do not propagate it through the subtraction. This is the most direct route by which a spurious signal could survive or a real signal be cancelled. Other concerns (single sample, 10 T vs 15 T, lack of data deposits) are secondary: the field difference is too small to explain the factor-of-30 discrepancy, and the single-sample issue is acknowledged. Thus I agree with the reader's identification, though I would phrase the assumption as 'reproducibility between up and down sweeps' rather than 'antisymmetry.' The test I propose is quantitative and can be done with data already in hand; it should settle whether the hysteresis contaminates the bound. If the residual αR(H) is below 1×10−4 across the entire field range, the conditional verdict can be accepted; if not, the bound must be raised or the subtraction method revised.","tokens_in":8281,"tokens_out":12439,"duration_ms":151717,"concrete_test":"Quantify the residual in κxy/T units: from the raw up/down field sweeps, compute R(H) = ½[ΔTxx_up(H) − ΔTxx_down(−H)], convert to κxy/T using the same calibration as the transverse channel, and estimate α by a controlled misalignment procedure (e.g., measuring the apparent transverse gradient with the heat current deliberately tilted, or by comparing the two transverse thermometers' responses to a known longitudinal gradient). If |αR(H)| exceeds 1×10−4 W/mK² at any H in 0–10 T, the claimed bound is not conservative and the verdict should be revised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central null result depends on the in-out subtraction in the field-sweep protocol: δTxy(H) = ½[ΔTxy(H) − ΔTxy*(−H)]. This cancels longitudinal contamination αδTxx only if δTxx is the same in the two sweeps. The paper reports 'a small hysteresis in the longitudinal κxx channel' (Sec. III C 2, Fig. 4), with loops in the field interval ±(4–7) T, so δTxx depends on sweep direction. In that interval, the contamination does not cancel; the residual ½α[δTxx_up(H) − δTxx_down(−H)] leaks into the extracted κxy/T. The leakage ratio α (transverse–longitudinal crosstalk from thermometer misalignment or differential magnetoresistance) is never measured, and the residual is never quoted in κxy/T units. The claimed bound is 1×10−4 W/mK², and with the stated conversion 30 µK ↔ 4×10−5 W/mK², a residual of only ~75 µK in the transverse channel saturates that bound. The authors' own statement that 'a naive antisymmetrization in ΔTxy could result in misleading thermal Hall signals' acknowledges the hazard, yet the FS results are presented without a quantitative residual-subtraction analysis, so the null result is not yet proven robust against this known systematic.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports high-resolution measurements of the longitudinal (κxx) and transverse (κxy) thermal conductivity in the parent cuprate La2CuO4, using both field-sweep (FS) and temperature-sweep (TS) protocols, for temperatures 2–20 K and magnetic fields up to 10 T. The authors find no resolvable thermal Hall signal and set a conservative upper bound |κxy/T| < 1×10^-4 W m^-1 K^-2, which is 10–30 times smaller than the signal reported by Grissonnanche et al. (2020) in the same material. The longitudinal conductivity κxx/T agrees well with previous studies, and weak hysteretic features near ±(4–7) T are observed and attributed to the metamagnetic transition of the weakly canted antiferromagnetic order.","tokens_in":8519,"tokens_out":8005,"duration_ms":91794,"significance":"If correct, this null result is significant for the ongoing debate on the thermal Hall effect in cuprate parent compounds and the chiral-phonon scenario. The paper's strengths include a detailed description of the FS protocol, explicit treatment of relaxation spikes, in-situ thermometer calibration with magnetoresistance correction, and the use of in-out symmetrization. The authors also provide a conservative upper bound and acknowledge the hazard of naive antisymmetrization. However, the central claim is not yet fully supported because key systematic uncertainties—specifically the hysteresis residual in the in-out subtraction and the statistical basis of the upper bound—are not quantified.","major_comments":[{"comment":"The in-out subtraction δTxy(H) = ½[ΔTxy(H) − ΔTxy*(−H)] removes longitudinal contamination αδTxx only if δTxx is exactly symmetric under field reversal. The hysteresis in ΔTxx presented in Fig. 4(a) violates this condition in the field interval ±(4–7) T, leaving a residual ½α[δTxx_up(H) − δTxx_down(−H)] in the extracted signal. The paper neither measures α nor translates the observed hysteresis into an upper bound on this residual in κxy/T units. Given the stated conversion 30 µK ↔ 4×10^-5 W m^-1 K^-2, a residual of only ~75 µK in the transverse channel would saturate the claimed bound. The authors' own remark that 'a naive antisymmetrization in ΔTxy could result in a misleading thermal Hall signal' (Sec. III C2) shows that this is a known hazard, yet no quantitative residual-subtraction analysis is provided. The null result is therefore not yet demonstrated to be robust against this systematic.","section":"IIA, III C2, Fig. 4"},{"comment":"The conservative upper bound is justified by the statement that averaging the curves 'by eye' suggests a weak signal of about 0.1×10^-3 W m^-1 K^-2, but that this conclusion is 'seen to be invalid' when the curves are displaced (Sec. III C3). This is not a quantitative analysis. For a null result that claims to improve on previous bounds by an order of magnitude, the bound should be derived from a statistical analysis (e.g., mean and standard deviation of κxy/T over the measured field and temperature range, and a test for a field-odd component). Such an analysis would also establish whether the apparent positive offsets at 5.5 and 10.6 K are statistically significant.","section":"III C3, Fig. 3"},{"comment":"The temperature-sweep data are offered as confirmation of the FS null result, but the TS protocol does not apply the in-out cancellation and is therefore subject to the same longitudinal contamination and hysteresis effects, with no correction. Because the TS data are also substantially noisier (as acknowledged in Sec. IIIA), they cannot provide independent confirmation. The paper should either apply an equivalent quantitative cancellation analysis to the TS data, or explicitly exclude them from the upper-bound claim.","section":"IIB, Fig. 2b"}],"minor_comments":[{"comment":"Sec. IIA states a resolution of ±100 µK at 2.3 K, while Sec. IIIA quotes δTxy ≈ 30 µK; please reconcile these numbers and state the resolution consistently.","section":"IIA, IIIA"},{"comment":"The statement that the resolution bounds for ΔTxy are independent of T from 2 to 10 K appears to conflict with the later statement that the He-3 insert stability degrades at higher temperatures; please clarify the temperature dependence of the resolution.","section":"IIIA"},{"comment":"There are several typographical errors, e.g., 'nomincal' for 'nominal' in Sec. IIIB and 'thermal hall' for 'thermal Hall' in the abstract; these should be corrected.","section":"IIIB"},{"comment":"The caption of Fig. 4(a) is confusing: it says solid circles are the down sweep from 10 T to −10 T and hollow ones the up sweep, but the text in Sec. III C2 says 'Open and solid red circles correspond to up-field and down-field scans, respectively.' Please make the caption and text consistent.","section":"IIIC2, Fig. 4"},{"comment":"The abstract and Sec. IV quote the upper bound as 1×10^-4 W m^-1 K^-2, while Sec. IIIA says the resolution threshold is 10^-5 to 10^-4 W m^-1 K^-2; please state a single, precise number with the conditions under which it applies.","section":"Abstract, IIIA, IV"},{"comment":"No raw data or data repository is mentioned; given the importance of the null result, the authors should make the raw data for Figs. 2–4 available.","section":"Data availability"}],"recommendation":"major_revision","confidential_remarks":"This is a carefully conducted experimental study on a controversial topic. The main weakness is that the central null result depends on the in-out subtraction, but the quantitative impact of the observed hysteresis is not assessed, and the upper bound is justified by an 'eyeball' analysis rather than statistics. These issues are addressable with additional analysis of the existing data, so I recommend major revision rather than rejection. The TS data should not be used as supporting evidence without the same cancellation analysis."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Let me give you the bottom line: this is a careful experimental null result that directly contradicts the widely cited Nature Physics claim of a large thermal Hall effect in La2CuO4. If it survives scrutiny, it is a significant result. The field-sweep protocol gives an upper bound ten to thirty times smaller than the earlier signal, and the longitudinal conductivity checks out against previous data, which makes the sample credible.\n\nThe experiment is well executed. The authors wait out the transient spikes after field steps, calibrate their thermometers in situ, correct for magnetoresistance, and report a ±0.1 mK resolution at 2.3 K. I also believe they are honest: they explicitly point out that naive antisymmetrization of ΔTxy can produce misleading signals. That is exactly the right concern.\n\nThe soft spot is that they do not close that concern. The in-out subtraction assumes δTxx(H) = δTxx*(-H). But the paper reports hysteresis in the longitudinal channel in the 4–7 T range, which means the assumption is not strictly true. The leakage ratio α from longitudinal to transverse is never quantified, and the resulting residual in κxy/T units is never estimated. At their stated conversion, 30 µK corresponds to 4e-5 W/mK2, so even a small residual could sit at the edge of the claimed bound. They warn about the hazard, but then present the FS results without a quantitative residual analysis. That leaves the central null not fully proven.\n\nMinor points: the data are not deposited, only one sample is used, the field goes to 10 T rather than the 15 T of the earlier work, and the 'by eye' averaging is informal. These are manageable.\n\nThis paper should definitely go to referees. The question is whether the hysteresis contamination can be quantified or bounded. I would want the authors to deposit raw data and to show that the in-out residual is small compared to the 1e-4 bound. My own verdict would be conditional on that. It is a serious piece of work that deserves a serious referee.","headline":"A careful null result that challenges a high-profile thermal Hall claim, but the unquantified hysteresis residual in the in-out subtraction keeps the bound from being fully closed.","tokens_in":9079,"tokens_out":5457,"would_cite":true,"duration_ms":60670,"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":"This paper reports that high-resolution field-sweep measurements resolve no thermal Hall signal in La2CuO4 between 2 and 20 K in fields up to 10 T, setting an upper bound 10 to 30 times smaller than previously reported signals.","keywords":["thermal Hall effect","thermal conductivity","La2CuO4","cuprates","field-sweep protocol","chiral phonons","hysteresis","metamagnetic transition"],"falsifier":"A direct measurement of the hysteresis-loop asymmetry: if the difference between up-sweep and down-sweep branches of deltaTxx, evaluated at the same |H|, exceeds the claimed transverse resolution (about 30 microK at 2.3 K), then the in-out antisymmetrization is not exactly cancelling longitudinal contamination, and the null result could be questioned. Alternatively, repeating the measurement on the same crystals at H = 15 T and finding no signal would directly contradict the 15-T report, while finding a signal would reveal a steep field threshold.","tokens_in":8065,"feed_emoji":"🧲","tokens_out":4849,"duration_ms":49577,"temperature":0.7,"pith_summary":"The paper attempts to establish that the thermal Hall effect in the parent cuprate La2CuO4, previously reported to be large and attributed to chiral phonons, is not intrinsic: using a high-resolution field-sweep protocol, the authors resolve no transverse thermal gradient in the range 2 to 20 K and 0 to 10 T. The claimed conservative upper bound, |kappa_xy/T| < 1e-4 $Wm^{-1}$$K^{-2}$, is 10 to 30 times smaller than the signal reported by Grissonnanche et al. (2020). The longitudinal conductivity kappa_xx/T agrees with previous studies in magnitude and temperature dependence, including a weak hysteresis tied to the metamagnetic transition of canted spins. If correct, the result challenges the chiral-phonon interpretation of the cuprate thermal Hall effect and points to sample differences or extrinsic artifacts as the source of the earlier signal.","feed_headline":"No thermal Hall signal in La2CuO4 at a 10-30x tighter bound","feed_subtitle":"Field-sweep measurements put the upper bound 10 to 30 times below the previously reported chiral-phonon signal.","key_machinery":"The field-swept (FS) protocol is the load-bearing experimental mechanism: instead of cooling in a fixed field, H is stepped in increments of 0.1 to 1 T with T rigorously regulated, waiting 3 to 8 minutes after each step so that magnetocaloric and eddy-current temperature spikes (1 to 6 mK, relaxing in 1 to 10 ms) die out before data are acquired. The in-out antisymmetrization, deltaTxy(H) = 1/2 (DeltaTxy(H) - DeltaT*xy(-H)), cancels the longitudinal contamination alpha deltaTxx under the assumption deltaTxx(H) = deltaT*xx(-H). The authors also perform in-situ thermometer calibration under field, with heat current off, to correct for the magnetoresistance of the thermometers.","core_discovery":"The central claim is a null result with a quantified bound: in as-grown La2CuO4 crystals with Neel temperature near 255 K, neither field-sweep nor temperature-sweep protocols resolve a thermal Hall conductivity, and the conservative upper bound on |kappa_xy/T| is 1e-4 $Wm^{-1}$$K^{-2}$ across 2 to 20 K in H up to 10 T. The authors emphasize that their resolution is 10 to 30 times smaller than the magnitude reported in Ref. [8], where kappa_xy/T was approximately 3e-3 W/$mK^{2}$ near 7 K in 15 T, and that their resolution is independent of T from 2 to 10 K, incompatible with the exponential decay claimed previously. They attribute the high resolution to the field-sweep protocol, in which the bath temperature is actively regulated, post-step relaxation spikes (magnetocaloric and eddy-current) are waited out, and the in-out antisymmetrization cancels longitudinal contamination. The paper also documents an unexplained weak anomaly near 7 K in kappa_xx/T and a field-induced up-turn in kappa_xx at low T, while noting that the absence of a resolvable Hall signal is intrinsic and not an experimental artifact.","pith_inferences":["If the null result holds generally, the reported thermal Hall signals in other cuprate parent and pseudogap materials may also contain a large extrinsic component, such as resonant skew scattering by defects, rather than an intrinsic chiral-phonon response.","The hysteresis asymmetry the authors note could be tested directly: measuring the up- and down-sweep branches of deltaTxx at fixed T and checking whether the difference is antisymmetric under H -> -H would quantify the residual systematic error in any in-out subtraction.","The unexplained 7-K anomaly in kappa_xx/T may be a signature of a low-energy magnetic or structural mode; a targeted specific-heat or neutron-scattering measurement near 7 K could test whether it is intrinsic.","A direct extension would be to repeat the measurement at 15 T and below 2 K: if a signal still fails to appear, the upper bound would directly confront the conditions of the original report."],"forward_implications":["The previously reported large, exponentially decaying thermal Hall signal in undoped La2CuO4 is not reproduced at fields up to 10 T, so chiral-phonon explanations based on that signal lose their main empirical support for this material.","Sample stoichiometry becomes a plausible controlling variable: if the effect is extrinsic, its magnitude should vary with preparation, and the paper names deoxygenation as one possible difference.","The 10 to 30 times tighter upper bound sets a new benchmark that future thermal Hall experiments on cuprate parent compounds must beat or explain.","The documented hysteretic contamination in the longitudinal channel implies that naive antisymmetrization of transverse gradients can produce spurious Hall-like signals, so the field-sweep protocol with careful relaxation control should become standard for small-signal searches.","The linear suppression of kappa_xx with H and the low-T up-turn above 7.5 T are new field-dependence features that invite theoretical explanation."],"supporting_citations":[{"why":"Reports the large thermal Hall signal in La2CuO4 that this paper fails to reproduce; it is the benchmark for the upper bound.","marker":"[8]"},{"why":"Supplies the field-swept protocol and in-out antisymmetrization technique adopted for high-resolution measurements.","marker":"[13]"},{"why":"Continues the field-swept protocol reference used for the in-out cancellation and data acquisition.","marker":"[14]"},{"why":"Previous longitudinal thermal conductivity data in La2CuO4 used to validate the magnitude and temperature dependence of kappa_xx.","marker":"[16]"},{"why":"Provides the antisymmetric-exchange and spin-flop explanation used to interpret the weak ferromagnetic hysteresis in the longitudinal channel.","marker":"[17]"},{"why":"Neutron-scattering study of the transition to weak ferromagnetic order, supporting the metamagnetic interpretation of the hysteresis.","marker":"[19]"},{"why":"High-field magnetism study of La2CuO4 giving transition-field values that bracket the hysteresis loops observed here.","marker":"[20]"},{"why":"Recent neutron-scattering work on spin canting and lattice symmetry, cited as support for the field-induced metamagnetic transition.","marker":"[22]"}],"fun_headline_variants":["La2CuO4: thermal Hall signal absent, bound 30x tighter than prior","No thermal Hall in La2CuO4; new bound is 30x lower","La2CuO4 thermal Hall ruled out down to 30x smaller signal","No detectable thermal Hall in La2CuO4, 30x tighter limit","Thermal Hall null in La2CuO4 with 30x higher sensitivity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The in-out subtraction assumes the longitudinal contamination in the transverse channel is exactly antisymmetric under field reversal; the paper's own data show weak hysteresis in the longitudinal channel, and if that hysteresis is not perfectly antisymmetric, the subtraction could cancel a real signal or create a spurious one.","fun_headline_variants_meta":{"raw":{"variants":["La2CuO4: thermal Hall signal absent, bound 30x tighter than prior","No thermal Hall in La2CuO4; new bound is 30x lower","La2CuO4 thermal Hall ruled out down to 30x smaller signal","No detectable thermal Hall in La2CuO4, 30x tighter limit","Thermal Hall null in La2CuO4 with 30x higher sensitivity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000674,"raw_usage":{"total_tokens":3097,"prompt_tokens":1001,"completion_tokens":2096,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":617,"completion_tokens_details":{"reasoning_tokens":1987}},"tokens_in":617,"tokens_out":2096,"duration_ms":17464,"temperature":1.0,"reasoning_tokens":1987,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T12:46:39.533968+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct measurement of the hysteresis-loop asymmetry: if the difference between up-sweep and down-sweep branches of deltaTxx, evaluated at the same |H|, exceeds the claimed transverse resolution (about 30 microK at 2.3 K), then the in-out antisymmetrization is not exactly cancelling longitudinal contamination, and the null result could be questioned. Alternatively, repeating the measurement on the same crystals at H = 15 T and finding no signal would directly contradict the 15-T report, while finding a signal would reveal a steep field threshold.","supporting_citations":[{"cited_title":"Czajka, T","cited_arxiv_id":null,"evidence_quote":"Reports the large thermal Hall signal in La2CuO4 that this paper fails to reproduce; it is the benchmark for the upper bound."},{"cited_title":"Sun, J.-Y","cited_arxiv_id":null,"evidence_quote":"Supplies the field-swept protocol and in-out antisymmetrization technique adopted for high-resolution measurements."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Continues the field-swept protocol reference used for the in-out cancellation and data acquisition."},{"cited_title":"Hirschberger, Quasiparticle excitations with Berry curvation in insulating magnets and Weyl semimetals , Phd thesis, Princeton University (2017)","cited_arxiv_id":null,"evidence_quote":"Previous longitudinal thermal conductivity data in La2CuO4 used to validate the magnitude and temperature dependence of kappa_xx."},{"cited_title":"Czajka, Exotic thermal transport in a Kitaev magnet , Phd thesis, Princeton University (2022)","cited_arxiv_id":null,"evidence_quote":"Provides the antisymmetric-exchange and spin-flop explanation used to interpret the weak ferromagnetic hysteresis in the longitudinal channel."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Neutron-scattering study of the transition to weak ferromagnetic order, supporting the metamagnetic interpretation of the hysteresis."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"High-field magnetism study of La2CuO4 giving transition-field values that bracket the hysteresis loops observed here."},{"cited_title":"Kastner, R","cited_arxiv_id":null,"evidence_quote":"Recent neutron-scattering work on spin canting and lattice symmetry, cited as support for the field-induced metamagnetic transition."}],"review_version":1}