{"id":"26a139d6-dfd3-4c9a-8eda-717ab3a0a6e4","arxiv_id":"2505.15719","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":4,"one_line_summary":"The paper reports 2D massless Dirac fermions in superconducting NaFeAs and a linear scaling of their Fermi velocity with Fe-As bond length across NaFeAs, BaFe2As2, and CaFe2As2.","lead":"Magneto-infrared measurements on three iron arsenide superconductors find two-dimensional massless Dirac fermions in the bulk of NaFeAs and report a linear relation between the Dirac Fermi velocity and the Fe-As bond length. The scaling claim currently rests on three compounds and on fitted Fermi velocities, so it is a suggestive trend rather than a demonstrated law.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"NaFeAs vF in the scaling plot depends on an untested LL-index choice; the two assignments differ by 2.4x, and the paper's EF-based disambiguation contains an unresolved numerical gap.","rationale":"The load-bearing issue is the LL-index assignment for NaFeAs, exactly as the reader identified. The observed T1/T2 ratio ~2.4 is compatible with two distinct pairs of Dirac transitions, and those two pairs yield vF values separated by 2.414. Since the scaling claim is fitted to only three points, a factor-2.4 shift of one point is enough to invalidate it. The paper tries to break the degeneracy with an EF argument, but the quoted 81 meV is not transparently derived; a direct evaluation of Eq. (2) gives ~20 meV for the n=±1 LL energy, which still exceeds the ARPES EF but is not the stated number. The spectral-weight argument is qualitative and lacks computed oscillator strengths. I am not objecting to the Dirac-fermion identification itself: the √B scaling, zero intercept, ratio, and DFT+DMFT linear bands are consistent with the prior BaFe2As2 work and make the existence claim credible. My concern is limited to the quantitative velocity used in Fig. 4(b). I also note that the theoretical support based on vF=sqrt(Ek/m*) is logically insufficient, since linearity of sqrt(Ek) and sqrt(m*) separately does not imply linearity of their ratio; this is a secondary defect. On balance, the reader's REJECT is appropriate, so I would keep the verdict unchanged.","tokens_in":15939,"tokens_out":16646,"duration_ms":143824,"concrete_test":"Re-analyze the raw NaFeAs reflectance spectra with a full magneto-optical LL absorption model for both assignments, fixing EF at the ARPES value (2–3 meV) and fitting the LL broadening; then compare the predicted B-dependent intensity ratio I(T1)/I(T2) and the Pauli-blocking of LL_-1→LL_0 with the measured spectra. If case (ii) cannot reproduce the observed dominance of T1 and its field evolution, the NaFeAs point in Fig. 4(b) should be replaced by the case-(i) value and the scaling claim re-evaluated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central scaling plot Fig. 4(b) is a three-point line whose NaFeAs point is selected by choosing between two Landau-level assignments that are exactly degenerate in slope and ratio. For 2D MDF, both case (i) (T1=LL_-2→LL_-1, T2=LL_-1→LL_0) and case (ii) (T1=LL_-1→LL_0, T2=LL_-1→LL_+2) give the same √B-linear energies and ET2/ET1≈2.414, but Eq. (4) yields vF=1.30×10^5 m/s in case (i) and 5.4×10^4 m/s in case (ii), a factor 2.414 difference. The paper rejects case (i) because it would imply EF≈81 meV, far above the ARPES/DFT+DMFT value of 2–3 meV. However, the 81 meV value is not derived in the text: inserting vF=1.30×10^5 m/s and B=17.5 T directly into Eq. (2) gives |E_±1|≈20 meV, so the claimed inconsistency is not numerically transparent. The additional spectral-weight argument is qualitative and is not supported by a calculated oscillator strength. Since the BaFe2As2 point is taken from a separate earlier study and no error bars are given for any point, replacing the NaFeAs vF by the case-(i) value removes the claimed linear relation. Thus the title's scaling relation is not established unless the LL-index assignment for NaFeAs is independently verified.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports magneto-infrared reflectance and optical-conductivity measurements at T ≈ 4.2 K in magnetic fields up to 17.5 T on single crystals of NaFeAs and CaFe2As2. Two peak-like features, T1 and T2, are observed in the relative optical conductivity of NaFeAs; their energies are linear in √B, extrapolate to zero energy at B = 0, and exhibit a field-independent energy ratio of about 2.4. The authors rule out Kane and Weyl fermion interpretations and assign the two features to inter-Landau-level transitions of two-dimensional massless Dirac fermions (case ii: T1 = LL−1→LL0 and T2 = LL−1→LL+2), obtaining vF ≈ 5.4 × 10^4 m/s. A single √B-linear transition in CaFe2As2 is assigned to LL−1→LL0, giving vF ≈ 2.16 × 10^5 m/s. Combining these values with a BaFe2As2 velocity of 1.18 × 10^4 m/s taken from Ref. [29], the authors claim that the 2D-MDF Fermi velocity scales linearly with the Fe-As bond length, and they cite linear dependencies of √m* and of √(t1 + t2) on the bond length as support. DFT+DMFT calculations show dispersions that are linear within the FeAs plane and weak along kz for both compounds.","tokens_in":16309,"tokens_out":29430,"duration_ms":238019,"significance":"If confirmed, the observation of two-dimensional massless Dirac fermions in the superconducting bulk state of NaFeAs would be a notable result, and a quantitative relation between the Dirac Fermi velocity and a structural parameter would be a genuinely useful design rule for iron-based superconductors. The paper's strengths include the bulk-sensitive magneto-infrared probe, the systematic √B analysis with zero-energy intercepts and a field-independent transition ratio, the comparison against Kane and Weyl fermion alternatives, and the accompanying DFT+DMFT band calculations. However, the headline scaling claim is not established: it rests on three points without reported uncertainties, and two of them (NaFeAs and CaFe2As2) are fixed only by Landau-level index assignments that are degenerate in the measured quantities and are disambiguated by an argument that contains a numerical error. The massless-Dirac interpretation itself is more robust than the scaling claim, since both competing assignments are Dirac Landau-level transitions.","major_comments":[{"comment":"The choice between case (i) (T1 = LL−2→LL−1, T2 = LL−1→LL0) and case (ii) (T1 = LL−1→LL0, T2 = LL−1→LL+2) is load-bearing for the whole paper, because both cases give identical √B-linear transition energies and an identical T2/T1 ratio of √2+1 ≈ 2.414. The disambiguation offered after Eq. (4) is not sound as written. First, the text claims that case (i) with vF = 1.30 × 10^5 m/s implies EF ≈ 81 meV at B = 17.5 T, but direct substitution of these numbers into Eq. (2) gives |E±1| ≈ 20 meV, and no derivation of the 81 meV value or of the assumed Landau-level filling model is provided. Second, the concluding sentence assigns 'T1 and T2 to the LL transitions LL−1→LL+2 and LL−1→LL0, respectively', which reverses the observed energy ordering (T1 is the lower-energy peak, while LL−1→LL+2 is the higher-energy transition by the factor √2+1) and contradicts the immediately preceding case-(ii) discussion. Since the alternative assignment changes vF(NaFeAs) by the factor 2.414 and destroys the linear relation in Fig. 4(b), the central claim cannot be evaluated until the assignment is corrected and justified with a quantitative calculation (for example, oscillator strengths and a filling-factor model).","section":"Assignment discussion following Eq. (4)"},{"comment":"The linear scaling relation in Fig. 4(b) is displayed as three points without error bars, fit statistics, or a stated fitting procedure, and one of the points (BaFe2As2, vF = 1.18 × 10^4 m/s) is imported from a separate earlier study (Ref. [29]) without discussion of the comparability of measurement and analysis conditions. The CaFe2As2 point is likewise not uniquely determined: a single transition is observed, and assigning it to LL−2→LL−1 instead of LL−1→LL0 would multiply the extracted velocity by 2.414 (from 2.16 × 10^5 to about 5.2 × 10^5 m/s), while the stated grounds for the assignment (low EF from DFT+DMFT and growth of spectral weight with field) are qualitative. Thus two of the three points in the central plot rest on assignment choices that are not independently verified, and a three-point line without quantified uncertainties cannot support the quantitative 'linear scaling' claim in the title. If the uncertainty analysis and fit statistics appear in the Supplementary Materials, they need to be summarized in the main text.","section":"Fig. 4(b) and the CaFe2As2 measurement (§3)"},{"comment":"The support claimed for the scaling is logically insufficient. From the stated relation vF = sqrt(Ek/m*), the observation that sqrt(Ek) and sqrt(m*) are each (approximately) linear in the Fe-As bond length does not imply that their ratio is linear in the bond length: the ratio of two linear functions is generically a rational function, and it is identically linear only under restrictive conditions (for example, a constant denominator), in which case it would be constant rather than varying. The correlations in Figs. 4(c) and 4(d) are therefore at most consistency checks, not supporting evidence for a linear scaling of vF, and the relation vF = sqrt(Ek/m*) itself is asserted without derivation in §4. Accordingly, the abstract's statement that the linear scaling is 'supported by (i) ... and (ii)' should be substantially weakened.","section":"§4, Figs. 4(c)-4(d)"},{"comment":"In the CaFe2As2 results paragraph, the sentence 'The 2D-MDF Fermi velocity in CaFe2As2 is larger than those in NaFeAs and BaFe2As2 (vF ≈ 1.18 × 10^4 m/s) [29]' is ambiguous: read literally, the parenthetical attributes the value 1.18 × 10^4 m/s to both NaFeAs and BaFe2As2, which contradicts the NaFeAs value of 5.4 × 10^4 m/s derived earlier in the paper. The parenthetical should be attached explicitly to the BaFe2As2 value only, so that the central data summary is internally consistent.","section":"CaFe2As2 results paragraph (§3)"}],"minor_comments":[{"comment":"The concluding paragraph contains 'Our wok offers a new material platform', which should read 'Our work offers', and 'exotic novel quantum phenomena' is redundant.","section":"Conclusion"},{"comment":"The phrase 'the presence of of the intra-LL transition' contains a duplicated 'of' and should be corrected.","section":"Assignment paragraph"},{"comment":"Refs. [63] and [65] are identical (Ashby and Carbotte, Phys. Rev. B 87, 245131 (2013)); the text cites both when contrasting the zeroth-LL density of states of Weyl fermions and 3D massless Dirac fermions, so this contrast is not supported by two independent sources.","section":"References"},{"comment":"The summary states that the Fermi velocities 'increase linearly with the Fe-As bond lengths', but the physical argument in §4 (shorter Fe-As distance implies larger bandwidth and hence higher vF) and the relative values quoted for the three compounds imply the opposite direction of the dependence; the wording should be reconciled with Fig. 4(b).","section":"Summary paragraph"},{"comment":"Please verify and cite the value Tc ≈ 23 K for NaFeAs; published values for stoichiometric NaFeAs are typically near 9 K in the literature, and if a doped or pressurized compound is meant, this should be stated explicitly.","section":"Introduction"},{"comment":"The magneto-optical fitting model (number and type of oscillators or Drude terms, parameter ranges, and how the errors of the extracted peak positions were obtained) is deferred to the Supplementary Materials; a brief summary in the main text would improve reproducibility, especially because the peak assignments are decisive for the claims.","section":"Supplementary Materials (fitting procedure)"}],"recommendation":"reject","confidential_remarks":"In my view the manuscript's headline claim is substantially stronger than what the data support. The most defensible result is the observation of √B-linear Landau-level transitions with a 2.414 ratio in superconducting NaFeAs; the paper might be restructured around that observation, with the three-point scaling presented as a tentative trend rather than a demonstrated relation. During any revision, the provenance of the BaFe2As2 velocity (1.18 × 10^4 m/s, cited to Ref. [29]) should be verified, since it is roughly an order of magnitude below the velocities commonly reported for Dirac fermions in iron-based compounds and is the smallest value in the scaling plot."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this paper has a genuinely new and plausible observation—2D massless Dirac fermions in the superconducting bulk of NaFeAs—and a sloppy, overreaching scaling claim that should not be taken at face value. The NaFeAs magneto-infrared data look like a real Dirac signature: linear in sqrt(B), zero intercept, T2/T1≈2.4, and dominant zeroth-LL absorption. The CaFe2As2 single-peak analysis is thinner but consistent. The problem is the title's linear scaling with Fe-As bond length. It rests on three points, no error bars, one point borrowed from an earlier paper, and a hand-wavy use of vF = sqrt(Ek/m*). The supporting linearities of sqrt(t) and sqrt(m*) from Ref. 48 are suggestive but not a derivation.\n\nThe LL-assignment ambiguity is real. Both case (i) and case (ii) give the same 2.414 ratio, and the extracted vF differs by that factor. The paper rejects case (i) by claiming EF~81 meV, but plugging their own vF into Eq. (2) gives |E±1|≈20 meV at 17.5 T, not 81 meV. That's an arithmetic gap a referee would have to chase. The conclusion may still be salvageable—20 meV is still far above the ARPES/DFT+DMFT 2–3 meV, and Pauli blocking of a filled LL_-1 would also rule out case (i)—but the paper as written doesn't show this. There are also minor copyedits: the CaFe2As2 section lists BaFe2As2's vF as 1.18×10^4 m/s, which conflicts with the value used in Fig. 4 (likely a missing zero).\n\nWho is this for? People working on Dirac physics in iron pnictides and on LL spectroscopy. The NaFeAs observation deserves a referee; the scaling claim needs either more data (SrFe2As2 from Ref. 29 is already there) or an honest downgrade to a correlation. Send it to review, but expect the referee to require the vF extraction to be tightened and the numerical inconsistency fixed.","headline":"Genuinely new NaFeAs Dirac-fermion observation wrapped in a three-point scaling claim that is statistically and arithmetically shaky.","tokens_in":16897,"tokens_out":6842,"would_cite":false,"duration_ms":53941,"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 claims that magneto-infrared peaks in superconducting NaFeAs are Landau-level transitions of two-dimensional massless Dirac fermions, and that the Fermi velocity of these carriers scales linearly with the Fe-As bond length.","keywords":["massless Dirac fermions","iron-based superconductors","Landau level spectroscopy","magneto-infrared spectroscopy","Fermi velocity","Fe-As bond length","NaFeAs","iron pnictides"],"falsifier":"Measure the magneto-infrared Landau-level transitions of a fourth iron arsenide with a well-determined Fe-As bond length, for example SrFe2As2, and check whether the resulting Fermi velocity falls on the same straight line; a clear miss would falsify the scaling. The transition assignment itself can be tested by tracking the relative spectral weight of T1 and T2 with field, since the assignment used here predicts that the zeroth-Landau-level peak T1 grows in dominance as the field increases, while the alternative assignment predicts the opposite.","tokens_in":15739,"feed_emoji":"🧲","tokens_out":15927,"duration_ms":123239,"temperature":0.7,"pith_summary":"This paper reports magneto-infrared spectroscopy of the iron-arsenide superconductors NaFeAs and CaFe2As2 at 4.2 K in fields up to 17.5 T, and claims that NaFeAs hosts two-dimensional massless Dirac fermions in its superconducting bulk. The evidence is two optical absorption peaks whose energies grow as $\\sqrt{B}$, extrapolate to zero at zero field, have an energy ratio near $1:(\\sqrt{2}+1)$, and whose lower peak dominates in intensity—the Landau-level signature of two-dimensional Dirac carriers. The paper also combines the Fermi velocities extracted from the peak slopes with the published BaFe2As2 value and finds that the three velocities fall on a straight line as a function of the Fe-As bond length. The authors present this linear scaling as the first quantitative link between a structural parameter and the Dirac-fermion velocity in iron-based superconductors, and as evidence that NaFeAs can serve as a bulk platform where Dirac physics coexists with superconductivity.","feed_headline":"Massless Dirac fermions found in superconducting NaFeAs","feed_subtitle":"The Dirac velocity tracks the Fe-As bond length across three iron arsenides.","key_machinery":"The load-bearing object is the Landau-level spectrum of two-dimensional massless Dirac fermions, $E_n=\\operatorname{sgn}(n)v_F^D\\sqrt{2e\\hbar|n|B}$, with the selection rule $|n|-|n'|=\\pm1$ for optical transitions. In this spectrum, the allowed transitions $LL_{-1}\\to LL_0$ and $LL_{-1}\\to LL_{+2}$ have an energy ratio $(\\sqrt{2}+1):1\\approx 2.414$, matching the observed T2/T1 ratio, and the slopes of $E$ versus $\\sqrt{B}$ give $v_F^D$ directly. The scaling claim is carried by plotting those extracted velocities against Fe-As bond lengths and by the DFT+DMFT and tight-binding result that $\\sqrt{m^*}$ and $\\sqrt{t_1(xy,xy)+t_2(xy,xy)}$ for the $d_{xy}$ orbital are themselves linear in the bond length, which connects $v_F^D=\\sqrt{E_k/m^*}$ to a lattice parameter.","core_discovery":"On the paper's own terms, the central discovery is that two-dimensional massless Dirac fermions exist in the superconducting and antiferromagnetic bulk state of NaFeAs, not only on surfaces or in non-superconducting parent compounds. Magneto-infrared spectra show two Landau-level transitions, labelled T1 and T2, whose energies follow $\\sqrt{B}$ and pass through zero at $B=0$, with $E_{T2}/E_{T1}\\approx 2.4$ and with the zeroth-Landau-level-related T1 peak dominant. The paper assigns T1 to $LL_{-1}\\to LL_0$ and T2 to $LL_{-1}\\to LL_{+2}$, yielding $v_F^D\\approx 5.4\\times10^4$ m/s for NaFeAs. Together with $v_F^D\\approx 1.18\\times10^4$ m/s for BaFe2As2 and $v_F^D\\approx 2.16\\times10^5$ m/s for CaFe2As2, the Fermi velocities scale linearly with the Fe-As bond length, and the authors support this relation with linear plots of $\\sqrt{m^*}$ and $\\sqrt{t_1(xy,xy)+t_2(xy,xy)}$ against the same bond length.","pith_inferences":["Editorial extension: a direct test is to measure a fourth iron pnictide, such as SrFe2As2 or a LaFeAsO-family compound, by the same technique and check whether its extracted Fermi velocity falls on the same line; a miss would mean the scaling is compound-specific rather than universal.","Editorial extension: because the velocity-bond length relation is tied to $d_{xy}$ tight-binding parameters, the same linear trend should show up in zero-field measurements—ARPES band slopes or quantum-oscillation effective masses—across the family, which would be an independent and cheaper check.","Editorial extension: if the same bond length that sets the Dirac velocity also influences pairing, the scaling line may help separate structural from electronic contributions to $T_c$ in iron-based superconductors, although the paper itself does not make that claim."],"forward_implications":["If NaFeAs truly hosts 2D massless Dirac fermions in its superconducting bulk, then a single stoichiometric 3D material can be used to study the interplay of superconductivity, antiferromagnetism, and Dirac physics at once.","The linear $v_F^D$–bond-length relation provides a quantitative design rule: choosing or tuning an iron arsenide's Fe-As bond length predicts its Dirac velocity and therefore its Landau-level spacing.","Magneto-infrared spectroscopy becomes a bulk-sensitive probe of Dirac carriers inside the superconducting state, because the Landau-level transitions are observed at 4.2 K in fields up to 17.5 T.","The agreement between the Landau-level slopes and the independently calculated $\\sqrt{m^*}$ and $\\sqrt{t_1+t_2}$ trends supports the $d_{xy}$-orbital character of the Dirac cones across the iron-pnictide family."],"supporting_citations":[{"why":"Supplies the Landau-level spectroscopy evidence and Fermi velocity for 2D massless Dirac fermions in AFe2As2 (A = Ba, Sr), including the BaFe2As2 point used in the scaling plot.","marker":"[29]"},{"why":"Provides the Fe-As bond lengths, dxy-orbital effective masses, and tight-binding hopping parameters t1(xy,xy) and t2(xy,xy) whose square roots are plotted against bond length.","marker":"[48]"},{"why":"Theory showing that antiferromagnetic Brillouin-zone folding produces topologically protected Dirac nodes near the Fermi energy in FeAs-based materials.","marker":"[15]"},{"why":"Theory of topological and transport properties of Dirac fermions in the antiferromagnetic metallic phase of iron-based superconductors.","marker":"[16]"},{"why":"ARPES observation of the Dirac cone electronic dispersion in BaFe2As2, supporting the Dirac interpretation of the Landau-level transitions.","marker":"[17]"},{"why":"ARPES study of detwinned NaFeAs giving the low Fermi energy near 2 meV used to select the T1/T2 transition assignment.","marker":"[24]"},{"why":"Provides the Landau-level transition energies for two-dimensional Dirac fermions used to fit the sqrt(B) slopes and extract vF.","marker":"[49]"},{"why":"Magneto-optical model used to fit the reflectance spectra and obtain the real part of the optical conductivity in which the Landau-level peaks appear.","marker":"[51]"},{"why":"DFT+DMFT identification that the Dirac cones near the Fermi energy are dominated by the iron dxy orbital, tying vF to dxy effective mass and hopping.","marker":"[69]"}],"fun_headline_variants":["Massless Dirac fermions in superconducting NaFeAs","Dirac velocity scales linearly with Fe-As bond length","2D massless Dirac fermions found in iron-based superconductors","NaFeAs reveals Dirac fermions with bond-length-tuned velocity","Iron arsenides: Dirac velocity tied to Fe-As bond length"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole scaling argument rests on assigning the two observed peaks to specific Landau-level transitions, which assumes NaFeAs reaches the regime where only the lowest Landau level is occupied at low magnetic fields and that its Fermi energy is only about 2–3 meV; if the other allowed assignment is the correct one, the extracted velocity and the linear relation both fail.","fun_headline_variants_meta":{"raw":{"variants":["Massless Dirac fermions in superconducting NaFeAs","Dirac velocity scales linearly with Fe-As bond length","2D massless Dirac fermions found in iron-based superconductors","NaFeAs reveals Dirac fermions with bond-length-tuned velocity","Iron arsenides: Dirac velocity tied to Fe-As bond length"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000687,"raw_usage":{"total_tokens":3228,"prompt_tokens":1175,"completion_tokens":2053,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":791,"completion_tokens_details":{"reasoning_tokens":1967}},"tokens_in":791,"tokens_out":2053,"duration_ms":13680,"temperature":1.0,"reasoning_tokens":1967,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T15:12:58.456802+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the magneto-infrared Landau-level transitions of a fourth iron arsenide with a well-determined Fe-As bond length, for example SrFe2As2, and check whether the resulting Fermi velocity falls on the same straight line; a clear miss would falsify the scaling. The transition assignment itself can be tested by tracking the relative spectral weight of T1 and T2 with field, since the assignment used here predicts that the zeroth-Landau-level peak T1 grows in dominance as the field increases, while the alternative assignment predicts the opposite.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Landau-level spectroscopy evidence and Fermi velocity for 2D massless Dirac fermions in AFe2As2 (A = Ba, Sr), including the BaFe2As2 point used in the scaling plot."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Theory showing that antiferromagnetic Brillouin-zone folding produces topologically protected Dirac nodes near the Fermi energy in FeAs-based materials."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Theory of topological and transport properties of Dirac fermions in the antiferromagnetic metallic phase of iron-based superconductors."},{"cited_title":"Morinari, E","cited_arxiv_id":null,"evidence_quote":"ARPES observation of the Dirac cone electronic dispersion in BaFe2As2, supporting the Dirac interpretation of the Landau-level transitions."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"ARPES study of detwinned NaFeAs giving the low Fermi energy near 2 meV used to select the T1/T2 transition assignment."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the Landau-level transition energies for two-dimensional Dirac fermions used to fit the sqrt(B) slopes and extract vF."},{"cited_title":"Jiang, E","cited_arxiv_id":null,"evidence_quote":"Magneto-optical model used to fit the reflectance spectra and obtain the real part of the optical conductivity in which the Landau-level peaks appear."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"DFT+DMFT identification that the Dirac cones near the Fermi energy are dominated by the iron dxy orbital, tying vF to dxy effective mass and hopping."}],"review_version":1}