{"id":"005a1319-97f8-44a2-9cf8-db5632b49cf7","arxiv_id":"1908.09994","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Landau level optical transitions in TaAs show sqrt(B) scaling at low energy and B-linear scaling above 17 meV, indicating a crossover from Dirac to parabolic bands.","lead":"This paper measured how the optical response of the Weyl semimetal TaAs changes in a magnetic field. The results show Landau level transitions with two distinct scaling laws, revealing a crossover from linear Dirac bands to parabolic free-electron bands at about 17 meV.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The B-linear IR peak claim rests on peak identifications in R(B)/R(0) ratios without error bars, and the 1/sqrt(B) intensity scaling is a visual collapse without quantitative residuals; the central evidence is suggestive but not yet quantitative.","rationale":"The reader's weakest assumption and my load-bearing concern converge on the same point: the Kramers-Kronig extrapolation scheme is unvalidated and sits exactly where the THz peak is measured. I agree that the paper is not fatally flawed: the observed magnetic-field dependence of R and sigma is real data, the sqrt(B) behavior of the dominant THz peak is visually plausible, and the theoretical reference is external and appropriate. However, the central quantitative claims—the scaling collapse and the linear-B IR peak energies—lack error bars, quantitative fitting, and robustness checks. The most load-bearing single concern is the KKA extrapolation boundary at 2-4 T, because the THz sigma(omega) below ~5 meV (including the 2 T peak at ~4 meV) is precisely where the extrapolation changes, so this choice could directly distort the peak positions and intensities that are the evidence for the scaling. The concrete test I propose (re-analyzing with alternative extrapolations and comparing fitted exponents and collapse residuals) would settle whether the scaling is robust. The verdict remains CONDITIONAL, since the requested quantitative checks and error analysis are prerequisites for fully supporting the strongest claim.","tokens_in":6244,"tokens_out":1761,"duration_ms":15032,"concrete_test":"Re-analyze the raw R(omega,B) spectra (Figs. 1 and 3a) with two independent low-energy extrapolations: (i) Hagen-Rubens for all B, and (ii) constant for all B, in addition to the published piecewise rule; for each extrapolation, extract the THz sigma(omega) peak position and amplitude for B = 1, 2, 4, 6 T and compute the best-fit sqrt(B) exponent and the 1/sqrt(B) scaling collapse residual (peak amplitude times sqrt(B), and chi-squared of the collapsed curves). If the fitted exponents and collapse residuals change substantially between extrapolations (e.g., exponent shifts by more than 0.1, or residuals exceed the spread of the data), the claimed scaling is not robust to the KKA extrapolation choice.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim has two quantitative pillars: sqrt(B) THz peak energy with 1/sqrt(B) intensity scaling, and B-linear IR peak energies. The THz pillar is mainly supported by fitted a=2.65 meV/T^1/2 and a visual collapse in Fig. 2c; the paper reports no numerical comparison between the scaled curves, no residual scatter, and no error analysis on peak positions or amplitudes. The IR pillar is weaker: peaks are read from R(omega,B)/R(omega,0) ratios (changes below ~2%) with no uncertainty on peak picking, and Fig. 3b shows points with no error bars; the claim that the IR peaks converge to ~17 meV at B=0 relies on extrapolation of a few points, and the 'linear-B' guide lines are guides to the eye. A more critical concern is that the Kramers-Kronig extrapolation changes at the 2-4 T boundary: for B<=2 T a Hagen-Rubens function and for B>=4 T a constant low-energy extrapolation. Since the THz sigma(omega) below ~5 meV is largely determined by this choice, and the broad peak at 2 T at ~4 meV sits exactly in the regime where the extrapolation changes, the reported peak energies and amplitudes, especially at intermediate fields, could be systematically biased rather than intrinsic. The theoretical prediction being tested (Ashby-Carbotte) generates both energy and intensity scaling shapes, so a clean quantitative test should compare single-scaling collapse residuals and fit peak positions with uncertainties; absence of these leaves the strongest claim not fully supported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports magneto-optical reflectivity and optical conductivity measurements of the type-I Weyl semimetal TaAs at 10 K in magnetic fields up to 6 T, covering the THz and IR ranges (3 meV to 54 meV, with B=0 data to 30 eV for Kramers-Kronig analysis). The central claim is that the THz optical conductivity develops a gap-like peak whose energy grows as sqrt(B) with intensity scaling as 1/sqrt(B), consistent with the Ashby–Carbotte theory of Landau-level transitions in linear (Dirac) bands, while IR reflectivity peaks show a linear-B energy dependence, interpreted as transitions in parabolic (free-electron-like) bands. The authors conclude that the two behaviors signal a crossover from Dirac to free-electron-like dispersion at about 17 meV.","tokens_in":6680,"tokens_out":2880,"duration_ms":32213,"significance":"If the evidence is accepted, the paper would provide one of the first experimental demonstrations of the 1/sqrt(B) intensity scaling predicted for Landau-level optical absorption in Weyl semimetals, and a rare observation of a field-dependent crossover in the effective band dispersion. The measurement strategy, combining THz and IR reflectivity with Kramers-Kronig analysis over a very wide range, is appropriate and the comparison with Ashby–Carbotte theory is explicit and falsifiable. However, the quantitative support for the key scaling laws is incomplete, and the central claim is therefore not yet fully established.","major_comments":[{"comment":"The Kramers-Kronig extrapolation changes from a Hagen-Rubens function for B <= 2 T to a constant for B >= 4 T. The THz sigma(omega) below about 5 meV is largely determined by this choice, and the peak assigned at 2 T sits at about 4 meV, i.e., exactly in the regime where the extrapolation scheme changes. This could systematically bias the peak positions and amplitudes in Fig. 2a. The authors should validate the extrapolation, for example by comparing the low-energy conductivity with an independent measurement or by reporting the sensitivity of the extracted peak energy and intensity to alternative low-energy extrapolations.","section":"Section 2 and Fig. 2a"},{"comment":"No uncertainties are reported for the peak positions or for the fitted exponent a = 2.65 meV/T^1/2 in Fig. 2b. The IR peaks in Fig. 3a are extracted from reflectivity ratios with changes below 2%, and Fig. 3b shows points with no error bars and only guide lines for the B-linear dependence. The paper should provide error estimates propagated from spectral noise and from the extrapolation ambiguity, fit the IR slopes quantitatively, and test whether the data are statistically consistent with B-linear versus other functional forms.","section":"Fig. 2b, Fig. 3b"},{"comment":"The scaling collapse in Fig. 2c is judged visually. Because the horizontal axis is normalized by sqrt(B) using the very peak positions that establish the sqrt(B) law, the energy-axis collapse is partly a restatement of the fitted scaling. A quantitative residual analysis of the collapsed curves is needed, and the 1/sqrt(B) intensity scaling should be tested independently of the peak-energy normalization. This is a correctness-risk concern, not a claim that the theory is circular: the theory is taken from the external reference [11].","section":"Fig. 2c"},{"comment":"The IR peaks are identified as down arrows in Fig. 3a, but the paper gives no criterion for peak selection or uncertainty in locating them. With spectral changes below 2%, the assignment would be considerably strengthened by fitting the line shapes, by reproducing the peaks in sigma(omega) rather than only R(omega)/R(0), and by checking stability against the 0 T reference spectrum.","section":"Fig. 3a and Section 3"}],"minor_comments":[{"comment":"The caption contains an incomplete sentence and typos: \"spectrm\" should be \"spectrum\" and the phrase \"with the Ha\" is cut off.","section":"Fig. 2 caption"},{"comment":"The description \"appropriate extrapolations\" is vague; the authors should specify the exact extrapolation functions used at low and high energy (beyond 30 eV) in the Kramers-Kronig analysis.","section":"Section 2"},{"comment":"The term \"W2 points\" is introduced without definition; it should be defined or accompanied by a reference to the earlier band-structure discussion in Ref. [8].","section":"Section 3"},{"comment":"The phrase \"could be scaled not only in the energy scale by sqrt(B) but also in the intensity by 1/sqrt(B)\" is grammatically awkward; \"the intensity was scaled\" would be clearer.","section":"Abstract"},{"comment":"Reference [14] is cited as an arXiv preprint (arXiv:1503.02630); if a published version exists, it should be cited instead.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The manuscript reports an interesting experiment with a clear theoretical anchor, but the current version does not yet provide the quantitative support that its strong claims require. The most pressing issue is the unvalidated change in the Kramers-Kronig extrapolation at 2-4 T, which sits exactly where the key THz peak is located. The other major comments concern missing uncertainties and the visual-only scaling collapse. These are fixable within the manuscript's scope with additional analysis of existing data, so I recommend major revision rather than rejection. The novelty claim (first experimental observation of the 1/sqrt(B) intensity scaling) should be checked carefully against the cited NbAs work [19], which may already contain related evidence."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper’s real news is the first experimental report of the predicted 1/sqrt(B) intensity scaling for inter-Landau-level absorption in a Weyl semimetal, and it is the same material in which the authors see both sqrt(B) and B-linear regimes. That is a nice, clean result to put on the record. The THz data, the comparison to Ashby and Carbotte, and the identification of the 17 meV crossover to the saddle point are all coherent. The self-citation for that energy is not a problem; the earlier paper did the band-structure work.\n\nThe soft spots are in the quantitative backing. The Kramers-Kronig extrapolation changes at the 2–4 T boundary: Hagen-Rubens below, constant above. The 2 T peak at ~4 meV sits right in that window, so the intermediate-field peak positions and intensities could be systematically biased. The paper gives no error bars on the peak energies or on the fitted a = 2.65 meV/T^1/2, and the scaling collapse in Fig. 2c is judged by eye. The IR peaks are read from R(B)/R(0) ratios with changes below 2% and no uncertainty estimate, and the B-linear lines are guides to the eye. The extrapolation of those IR peaks to ~17 meV rests on a few points. These are addressable, and none of them clearly kills the core claim—the sqrt(B) scaling is consistent with prior Dirac-material work, and the intensity scaling is a new observation that fits theory. But as it stands the strongest claim is suggestive, not demonstrated.\n\nThe reader’s circularity concern is fair but minor: using the fitted peak positions to normalize the horizontal axis of the collapse makes the energy collapse partly a restatement of the fit. The intensity collapse is less affected. A clean test would show scaled traces with residuals and a stated procedure for peak picking.\n\nWho gets value from this: experimentalists working on Weyl and Dirac semimetal magneto-optics, and theorists who want a data point for the Ashby-Carbotte scaling. It deserves a serious referee—conditional acceptance with a request for error analysis, a validation of the extrapolation scheme, and quantitative collapse metrics. I would probably cite it as the first report of the intensity scaling, with a cautious qualifier pending the revision.","headline":"A plausible, useful magneto-optical study of TaAs with a genuinely new scaling claim, but the quantitative support is thinner than the abstract suggests—worth refereeing with a request for error bars and a check on the Kramers-Kronig extrapolation.","tokens_in":7145,"tokens_out":1344,"would_cite":true,"duration_ms":14978,"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":"Magnetic-field optical spectra reveal a Dirac-to-parabolic band crossover in TaAs at about 17 meV.","keywords":["Weyl semimetal","TaAs","magneto-optics","optical conductivity","Landau levels","terahertz","infrared spectroscopy","band dispersion"],"falsifier":"Measure the low-energy optical conductivity of TaAs directly, for example by time-domain THz transmission, at 10 K in fields from 0 to 6 T, and compare the peak position and intensity with the Kramers-Kronig-reconstructed spectra; if the apparent peak near 5 meV at 4 T does not move as $\\sqrt{B}$ with amplitude proportional to $1/\\sqrt{B}$, the central claim fails. Equally decisive would be a band-structure calculation of the field-dependent inter-Landau-level absorption: if no transition with the reported scaling appears near the quoted energies, the interpretation is wrong.","tokens_in":6007,"feed_emoji":"🧲","tokens_out":9943,"duration_ms":97799,"temperature":0.7,"pith_summary":"Using magnetic-field-dependent reflectivity of the type-I Weyl semimetal TaAs, a material whose low-energy electrons mimic massless Weyl fermions, the paper reconstructs the optical conductivity at 10 K and reads the band shape from how Landau-level absorption peaks move with field. In the THz range, a broad absorption peak shifts as $\\sqrt{B}$ while its amplitude falls as $1/\\sqrt{B}$, the predicted signature of optical transitions in linearly dispersing Weyl bands; the spectra collapse onto a single curve under that scaling. In the infrared range above about 17 meV, reflectivity peaks move linearly in $B$, the signature of parabolic free-electron-like bands. The paper concludes that TaAs's band dispersion crosses over from Dirac-like near the Fermi level to free-electron-like near the saddle points, with the crossover around 17 meV, and reports the intensity scaling as an experimental first for Weyl semimetals.","feed_headline":"TaAs bands switch from Dirac to parabolic at 17 meV","feed_subtitle":"THz peaks scale with √B, infrared peaks with B, exposing a Weyl-band crossover.","key_machinery":"The engine of the argument is optical absorption between magnetic-field-induced Landau levels, detected as reflectivity changes and converted to optical conductivity by Kramers-Kronig analysis. In a linear band, inter-Landau-level transition energies scale as $\\sqrt{B}$ and the absorption strength as $1/\\sqrt{B}$; in a parabolic band, the transition energies scale as $B$. The paper uses those two scaling laws as a dispersion meter: the field dependence of each peak labels which kind of band produces it, and the crossover energy labels where the Weyl band ceases to be linear.","core_discovery":"The central claim is that in TaAs the conduction band changes dispersion character with energy: near the Fermi level it is linear, so the inter-Landau-level optical peak in the THz region moves as $\\hbar\\omega_{\\mathrm{peak}} = a\\sqrt{B}$ with $a = 2.65\\,\\mathrm{meV}/\\mathrm{T}^{1/2}$ and its spectral weight shrinks as $1/\\sqrt{B}$; above roughly 17 meV the bands are parabolic, so the infrared Landau-level peaks move linearly in $B$ and extrapolate to about 17 meV at zero field. That extrapolation point is identified with the saddle points between the Weyl points, and the crossover between the two field dependencies is offered as evidence that the linearly dispersing Weyl bands become free-electron-like away from the Weyl points. The simultaneous collapse of energy and intensity scaling is the paper's main experimental result.","pith_inferences":["Extended to higher fields, the same measurement could track the crossover continuously and yield a dispersion curve rather than a single crossover energy.","A low-energy measurement that avoids Kramers-Kronig reconstruction, such as time-domain THz transmission, could independently test the $\\sqrt{B}$ peak shift and $1/\\sqrt{B}$ intensity collapse.","If the same scaling analysis were applied to other type-I Weyl semimetals, each would be expected to show its own crossover energy set by its saddle-point position, making the method a general probe of band shape."],"forward_implications":["The conduction band of TaAs changes from linear Dirac to parabolic around 17 meV, so magneto-optical Landau-level spectroscopy can map band dispersion directly.","The $\\sqrt{B}$ energy scaling paired with the $1/\\sqrt{B}$ intensity scaling gives a concrete experimental fingerprint for identifying Weyl-band Landau levels in other materials.","The suppression of the Drude spectral weight with field ties the optical response to the giant magnetoresistance of TaAs.","Extrapolating the infrared peaks to zero field places the saddle-point energy near 17 meV, connecting the optical measurements to the underlying band structure."],"supporting_citations":[{"why":"Supplies the theoretical prediction that inter-Landau-level absorption peaks in Weyl bands scale as $\\sqrt{B}$ in energy and $1/\\sqrt{B}$ in intensity.","marker":"[11]"},{"why":"Provides the previous optical-conductivity study that identified TaAs as a type-I Weyl semimetal and located the saddle-point energy used for comparison.","marker":"[8]"},{"why":"Gives the Dirac-band result $\\sigma(\\omega) \\propto \\omega$ used to interpret the zero-field low-energy spectrum as Weyl-band absorption.","marker":"[15]"},{"why":"Reports similar $\\sqrt{B}$ scaling of Landau-level peaks in the Dirac semimetal ZrTe5, serving as a comparison.","marker":"[17]"},{"why":"Reports similar $\\sqrt{B}$ scaling behavior in the Dirac semimetal Cd3As2, supporting the interpretation.","marker":"[18]"},{"why":"Reports similar magneto-optical Landau-level behavior in the Weyl semimetal NbAs, a direct material analog.","marker":"[19]"},{"why":"Provides the Kramers-Kronig analysis procedure used to obtain optical conductivity from reflectivity.","marker":"[12]"}],"fun_headline_variants":["TaAs magneto-optics: √B vs B scaling reveals band crossover at 17 meV","TaAs Landau levels: THz gap scales as √B, IR as B, crossover at 17 meV","Weyl semimetal TaAs: THz √B, IR B scaling, band crossover at 17 meV","TaAs THz gap opens as √B, IR peaks as B: Dirac to free-electron at 17 meV","Magneto-optics of TaAs: √B and B scalings mark band crossover at 17 meV"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central claim rests on the assumption that the reflectivity below the measured spectral range is correctly reconstructed by the chosen piecewise extrapolation rules in the Kramers-Kronig transform; if that reconstruction is off, the peak positions and intensities that produce the reported $\\sqrt{B}$ and $B$ scalings could shift.","fun_headline_variants_meta":{"raw":{"variants":["TaAs magneto-optics: √B vs B scaling reveals band crossover at 17 meV","TaAs Landau levels: THz gap scales as √B, IR as B, crossover at 17 meV","Weyl semimetal TaAs: THz √B, IR B scaling, band crossover at 17 meV","TaAs THz gap opens as √B, IR peaks as B: Dirac to free-electron at 17 meV","Magneto-optics of TaAs: √B and B scalings mark band crossover at 17 meV"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001125,"raw_usage":{"total_tokens":4682,"prompt_tokens":954,"completion_tokens":3728,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":570,"completion_tokens_details":{"reasoning_tokens":3585}},"tokens_in":570,"tokens_out":3728,"duration_ms":24653,"temperature":1.0,"reasoning_tokens":3585,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:55:00.126565+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the low-energy optical conductivity of TaAs directly, for example by time-domain THz transmission, at 10 K in fields from 0 to 6 T, and compare the peak position and intensity with the Kramers-Kronig-reconstructed spectra; if the apparent peak near 5 meV at 4 T does not move as $\\sqrt{B}$ with amplitude proportional to $1/\\sqrt{B}$, the central claim fails. Equally decisive would be a band-structure calculation of the field-dependent inter-Landau-level absorption: if no transition with the reported scaling appears near the quoted energies, the interpretation is wrong.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the theoretical prediction that inter-Landau-level absorption peaks in Weyl bands scale as $\\sqrt{B}$ in energy and $1/\\sqrt{B}$ in intensity."},{"cited_title":"Kimura, H","cited_arxiv_id":null,"evidence_quote":"Provides the previous optical-conductivity study that identified TaAs as a type-I Weyl semimetal and located the saddle-point energy used for comparison."},{"cited_title":"Hosur, S","cited_arxiv_id":null,"evidence_quote":"Gives the Dirac-band result $\\sigma(\\omega) \\propto \\omega$ used to interpret the zero-field low-energy spectrum as Weyl-band absorption."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports similar $\\sqrt{B}$ scaling of Landau-level peaks in the Dirac semimetal ZrTe5, serving as a comparison."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports similar $\\sqrt{B}$ scaling behavior in the Dirac semimetal Cd3As2, supporting the interpretation."},{"cited_title":"Y uan, Z","cited_arxiv_id":null,"evidence_quote":"Reports similar magneto-optical Landau-level behavior in the Weyl semimetal NbAs, a direct material analog."},{"cited_title":"Kimura and H","cited_arxiv_id":null,"evidence_quote":"Provides the Kramers-Kronig analysis procedure used to obtain optical conductivity from reflectivity."}],"review_version":1}