{"id":"b2cedcac-121f-4a29-88bf-c82e2b569106","arxiv_id":"2507.17238","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Polarized infrared transmission and Faraday rotation in FePS3 reveal field-split magnons, circular dichroism, and phonon Faraday rotation, evidence of spin-phonon coupling.","lead":"This paper measures how infrared light passes through a thin magnetic crystal, FePS3, in a magnetic field and at low temperatures, and shows that a magnetic excitation and several lattice vibrations respond to the field in a way that reveals their coupling. The measurements give physicists a new way to see how spins and lattice vibrations interact in two-dimensional magnets, which could matter for future spintronic devices.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Appendix D's claim that birefringence cancels from the field-reversed Faraday angle is questionable: in a birefringent slab the B-odd rotation acquires a δ-dependent prefactor, so unquantified δ(ω) near 129 cm^-1 can distort σ_xy and the reduced-dichroism claim.","rationale":"The paper presents a rich, credible dataset: polarized infrared magneto-transmission, full-protocol Faraday rotation, temperature- and field-dependent studies, DFT phonon calculations, and Raman spectroscopy. The qualitative observations—field-split 122 cm^-1 magnon, phonon Faraday rotation, and in-plane phonon anisotropy below TN—are well-supported and consistent with prior literature. The central novelty is quantitative: the first reconstruction of circular optical conductivities of the split magnon and the conclusion that the upper branch near 129 cm^-1 has reduced dichroism due to magnon-phonon hybridization. That conclusion hinges on the extraction of σ_xy from Faraday angles via Eq. D29. The reader's weakest_assumption correctly identifies the unquantified birefringence and the effective orthorhombic approximation. My stress-test goes further: the Appendix D cancellation argument itself is internally problematic. The eigenvalues of the in-plane dielectric tensor are even in ε_xy, but the measured Faraday angle in a birefringent medium depends on the eigenvectors as well, so δ enters the magnitude of the B-odd signal and does not simply cancel under field reversal. Given the 26 µm thickness and the strongly polarization-selective phonons near the upper branch, this distortion could be material and would directly affect the reduced-dichroism ratio. This does not invalidate the qualitative observations, but it makes the headline quantitative claim conditional on a test the manuscript does not provide. I therefore keep the reader's CONDITIONAL verdict; the concern is addressable by the forward-modeling check described above.","tokens_in":27180,"tokens_out":11878,"duration_ms":127196,"concrete_test":"Compute δ(ω) = (ε_xx − ε_yy)/2 at 5 K and 0 T from the published σ_1(ω) (VDF or Drude-Lorentz) for Pol 1 and Pol 2 in the 110–140 cm^-1 range. Then forward-model the full 2×2 slab transfer matrix (including Fresnel coefficients and Fabry-Pérot interference) with in-plane dielectric tensor [[ε̄+δ, i g],[−i g, ε̄−δ]], using the paper's extracted g = ε_xy(ω), for B = ±7 T, and predict the Faraday angle spectra. Compare with the measured Faraday angles in Fig. 9. If the predicted B-odd angle deviates from the measured one by more than the quoted uncertainty (≈2 mrad) at either the 116 cm^-1 or 129 cm^-1 branch, Eq. D29 is invalid and the reduced upper-branch dichroism is not established. Alternatively, report the ratio δ(129 cm^-1)/|ε_xy(129 cm^-1)|; if it is not ≪1, the extracted σ_xy is birefringence-contaminated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The quantitative central claim—reconstruction of circular optical conductivities and the reduced dichroism of the 129 cm^-1 upper branch—rests on Eq. D29, which converts the measured Faraday angle into ε_xy via n_± = n0 ± ε_xy/(2n0). This formula assumes circular eigenmodes (δ=0). For δ≠0, the eigenvalues of ε_circ (D10) give n^2_± = ε̄ ± sqrt(δ^2 − ε_xy^2) (D18), an expression even in ε_xy. The paper argues that birefringence cancels in the field-reversal analysis because ε_xy^2 is symmetric under B→−B. That conclusion is not correct: the measured Faraday angle is not just the eigenvalue phase difference; the eigenmodes are elliptical, and their orientation depends on the sign of ε_xy. A Jones calculation for a slab with in-plane tensor [[ε̄+δ, i g],[−i g, ε̄−δ]] gives, to leading order in g, θ_F ≈ −g sin(2kdn0√(δ^2+g^2))/(2√(δ^2+g^2)), which reduces to θ_F ≈ −g kd n0 only if δ=0. Thus δ does not cancel; it multiplies the B-odd signal by a δ- and thickness-dependent prefactor. The sample is 26 µm thick, and near the upper branch δ(ω) is plausibly large because the 128.5 and 133.1 cm^-1 modes are strongly polarization-selective (Table I). The expansion in Eq. D31 is also mis-stated: it requires δ≫ε_xy, not δ≪ε̄. Without quantifying δ(ω)/|ε_xy(ω)| at 116 and 129 cm^-1, the reconstructed σ_xy—and hence the headline reduced-dichroism claim—is not established. This is an internal issue in the analysis chain, not a disagreement with outside consensus.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports polarization-resolved infrared magneto-transmission and Faraday rotation measurements on a 26 µm FePS3 crystal between 5 and 150 K and up to ±7 T. Below TN ≈ 118 K, the phonon spectrum develops a strong anisotropy between the two principal in-plane polarizations, whereas a mode at 122 cm⁻¹ remains polarization-independent, hardens on cooling with an order-parameter-like temperature dependence (Eq. 2), and splits linearly with applied magnetic field with a gyromagnetic ratio near the free-electron value, identifying it as a magnetic excitation. From absolute transmission and a full-protocol Faraday angle measurement at ±7 T, the authors reconstruct the circular optical conductivities σ± and report a pronounced dichroism of the field-split branches, a reduced dichroic response of the upper branch near 129 cm⁻¹ that they attribute to hybridization with a nearby infrared phonon, and Faraday rotation of several phonon modes (108, 133.1, 165.6 cm⁻¹) interpreted as evidence of spin–phonon coupling. The paper also provides DFT phonon frequencies, Raman data, a MnPS3 comparison, and a public data repository.","tokens_in":27586,"tokens_out":27239,"duration_ms":264103,"significance":"If the quantitative reconstruction were shown to be sound, the paper would provide the first quantitative mapping of the circular optical conductivities of the field-split magnon branches in FePS3 and direct magneto-optical evidence for spin–phonon coupling in a 2D antiferromagnet. The qualitative phenomenology is solid: the field splitting of the 122 cm⁻¹ mode is compared against the free-electron gyromagnetic ratio without adjustment (Fig. 8), the order-parameter-like temperature dependence is fitted consistently for both polarizations, the full-protocol Faraday measurements document a ~2 mrad noise floor with explicit transmission masking, and the polarized dataset is openly archived. The DFT phonon catalog, the MnPS3 control, and the authors' candor about the limitations of the Lorentz fits, the indicative symmetry assignments, and the Appendix C multiplet model are all strengths.","major_comments":[{"comment":"The analysis of the birefringent case in Appendix D is not valid, and the stress-test concern about the cancellation argument lands. Equations D17–D18 give eigenmode refractive indices n²± = ε̄ ± sqrt(δ² − ε_xy²), which are even in ε_xy; the text then argues that the birefringent contribution cancels in the field-reversal analysis because ε_xy² is symmetric under B → −B. This conclusion does not follow, because the measured Faraday angle is not the eigenvalue phase difference: for δ ≠ 0 the propagating eigenmodes are elliptical, and a Jones-matrix treatment of the 26 µm slab gives a B-odd rotation proportional to ε_xy × sin(2kdn0 sqrt(δ² + ε_xy²))/sqrt(δ² + ε_xy²) (transparent limit), which reduces to the δ = 0 result of Eq. D29 only when δ = 0. Field reversal cancels B-even contributions; it does not remove the δ- and thickness-dependent prefactor multiplying ε_xy. In addition, the expansion in Eq. D31 is a power series in ε_xy/δ and therefore requires δ ≫ |ε_xy|, not the stated condition δ ≪ ε̄; conversely, the regime δ ≪ |ε_xy|, which is the one the authors need, is not treated. These are load-bearing for the headline result because δ(ω) is resonantly enhanced near 129 cm⁻¹, where the 128.5 and 133.1 cm⁻¹ modes are strongly polarization-selective in the zero-field transmission (Table I). The reconstructed σ_xy, and with it the reduced-dichroism claim, is therefore not established until the authors either perform an explicit Jones propagation using the independently determined σ_xx(ω) and σ_yy(ω) or quantify δ(ω)/|ε_xy(ω)| at 116 and 129 cm⁻¹.","section":"Appendix D, Eqs. D17–D31 and Eq. D29"},{"comment":"The reduced dichroism of the upper branch (σ+ ≈ σ−/3 at 129 cm⁻¹) is the central quantitative result, but its robustness is not demonstrated. First, the two independent determinations of σ_xy (Pol 1 and Pol 2) disagree locally precisely around 129 cm⁻¹, as acknowledged in Sec. IVB and visible in Fig. 18; averaging the two removes the discrepancy rather than resolving it. The argument that a frequency-local discrepancy cannot be systematic is not compelling, because birefringence-related distortions are themselves resonant and frequency-local near the anisotropic phonons at 128.5 and 133.1 cm⁻¹. The error budget for the σ+/σ− ratio should include the spread between the two polarizations and the sensitivity to the averaging choice. Second, the upper branch sits on the shoulder of the strong 128.5 cm⁻¹ phonon (Pol 1), which contributes an unpolarized background (σ_xx + σ_yy)/2 to both σ+ and σ−; a competing explanation of the reduced ratio is simply that this phonon background fills in σ+, and the paper does not subtract it with an explicit multi-oscillator fit. Establishing the hybridization interpretation therefore requires either a quantitative coupled-mode model or a demonstration that the reduced dichroism persists after phonon baseline subtraction.","section":"Sec. IVB, Figs. 12 and 18"}],"minor_comments":[{"comment":"The identification of the 122 cm⁻¹ (15 meV) and 320 cm⁻¹ (40 meV) features with magnon excitations is justified by comparison with neutron-scattering energies measured at finite wave vectors, while optical spectroscopy probes q ≈ 0; the text acknowledges this tension but should state explicitly how the zone-center status of these excitations is established (e.g., magnetic-supercell folding, as in Refs. [43, 45, 46]) before describing the reconstruction as a mapping of the magnon circular conductivity.","section":"Sec. IIIC, last two paragraphs"},{"comment":"The sentence 'the lower mode is completely dichroic since the spectral weight of σ− in that excitation is the same magnitude as that of σ1 at zero field while σ+ shows a clear peak' is internally confusing, since a peak in σ+ at 116 cm⁻¹ would contradict complete dichroism; please specify which channel carries the 116 cm⁻¹ peak and which is suppressed.","section":"Sec. IVB, paragraph after Fig. 12"},{"comment":"Please state the sign convention that connects σ+ and σ− to the spin-sublattice and to the sign of the applied field; as written, the assignment of the upper and lower branches to specific circular polarizations is implicit and cannot be checked against Fig. 9.","section":"Eq. (4) and Fig. 12"},{"comment":"The relation between the quoted gyromagnetic ratio γ ≈ 0.94 cm⁻¹/T and the plotted splitting should be clarified: the text quotes branch positions near 116 and 129 cm⁻¹ at 7 T, implying a full splitting of ≈13 cm⁻¹ = 2 × 0.94 × 7, so if γ denotes the per-branch Zeeman shift, the 'expected linear dependence' in Fig. 8 should be drawn with slope 2γ for the full splitting.","section":"Fig. 8 and Sec. IIIB"},{"comment":"The 320 cm⁻¹ feature is described as a dip; please specify how the baseline was defined for this feature and whether the 1% transmission mask affects this spectral region at 7 T.","section":"Fig. 10"},{"comment":"Please correct 'the the optical phonons' (Sec. IID), the inconsistent 'Drude-Lorenz'/'Drude-Lorentz' spelling, 'Sibolometer' (Sec. IIB), and the missing space in 'The fitting yieldsa=' (Fig. 6 caption).","section":"various"}],"recommendation":"major_revision","confidential_remarks":"I recommend major revision rather than rejection because the load-bearing problem is fixable within the scope of the paper: the authors already possess σ_xx(ω) and σ_yy(ω) from the two-polarization transmission analysis and the sample thickness, so a Jones-matrix propagation of the measured Faraday geometry is a well-defined, bounded task that would either validate or correct Eq. D29 and the σ+/σ− ratio. The qualitative observations (magnon Faraday rotation, phonon Faraday rotation, anisotropic phonon spectrum) are solid and would survive even if the quantitative dichroism claim needed softening. I would also ask the editor to consider whether the novelty statement is sufficiently sharp relative to the recent FePS3 literature the paper cites (Refs. [39], [46], [48], [50]), which already reports related avoided crossings and chiral-phonon effects; the paper should position its 'first quantitative circular conductivity' claim explicitly against those works."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a look: this is the first polarized IR magneto-transmission and Faraday-rotation study of FePS3 that I know of. The raw observations are credible and new: the 122 cm^-1 magnon splits linearly with field, several phonons show B-odd Faraday rotation at the 10–50 mrad level, and the low-temperature phonon spectrum becomes strongly anisotropic. The authors also did a careful full-protocol measurement rather than the fast four-point scheme, which matters for a monoclinic sample. The DFT phonon table and the MnPS3 control strengthen the paper. Credit where it's due.\n\nThe problem is the extraction of σ_xy and the claim that the upper branch at ~129 cm^-1 shows reduced dichroism. Appendix D's derivation assumes δ=0 when it converts Faraday angle to ε_xy (Eq. D29). For δ≠0, the eigenmodes are elliptical and the measured rotation is not simply (ωd/2c) Re(n_+ − n_−). A Jones calculation gives θ_F ≈ −g sin(2kdn0√(δ²+g²))/(2√(δ²+g²)), so δ does not cancel in the field-reversal analysis; it appears as a multiplicative prefactor that is frequency-dependent and potentially large near 129 cm^-1, where two strongly polarization-selective phonons sit. The expansion in Eq. D31 is also mis-stated: the branch used requires δ≫|ε_xy|, not δ≪ε̄. So the reconstructed σ_xy—and the headline reduced-dichroism claim—is not established as presented. This is an internal analysis-chain issue, not a disagreement with outside consensus. The raw Faraday rotation spectra may show the effect, but the paper doesn't provide the Jones-matrix analysis that would validate it.\n\nMinor: the 320 cm^-1 feature is weak; the symmetry assignments are explicitly approximate (fine, but not to be leaned on); the data repository reference [31] has no accessible link.\n\nBottom line: this is a serious experimental paper that deserves review, but the referee should insist on a quantitative treatment of birefringence, either by measuring δ(ω) and using a full Jones transfer-matrix fit, or by showing the reduced dichroism survives in the raw B-reversed Faraday spectra. The other observations—magnon splitting, phonon Faraday rotation, anisotropy—are likely robust and should survive revision.","headline":"Worth a look: the raw Faraday-rotation dataset is new and likely robust, but the central quantitative claim about reduced dichroism at 129 cm^-1 does not survive scrutiny of the birefringence analysis.","tokens_in":28188,"tokens_out":4084,"would_cite":false,"duration_ms":44492,"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":"Polarized infrared light maps the circular dichroism of the field-split 122 cm-1 magnon in FePS3 and shows lattice phonons acquiring magnetic-field-dependent optical activity.","keywords":["FePS3","van der Waals antiferromagnet","spin-phonon coupling","Faraday rotation","circular optical conductivity","magnon","infrared magneto-spectroscopy","2D magnets"],"falsifier":"Measure the zero-field linear birefringence $\\delta = (\\varepsilon_{xx}-\\varepsilon_{yy})/2$ in the same 100-150 cm-1 range and compare it with $\\varepsilon_{xy}$ obtained from field reversal; if $\\delta$ is comparable to or larger than $\\varepsilon_{xy}$, the cancellation no longer isolates the magneto-optical term. A thickness-dependence or in-plane crystal-rotation check of the reconstructed $\\sigma_{xy}$ would also show whether residual cross-polarization or monoclinic terms contaminate the result.","tokens_in":26945,"feed_emoji":"🧲","tokens_out":8333,"duration_ms":78606,"temperature":0.7,"pith_summary":"This paper seeks to establish that the 122 cm-1 (15 meV) excitation of the two-dimensional van der Waals antiferromagnet FePS3 is a magnon whose field-split branches have measurable circular dichroism, and that this dichroism is modified where the magnon meets lattice vibrations. The authors combine polarized infrared transmission with Faraday rotation measurements up to 7 T to reconstruct the circular optical conductivities $\\sigma_+$ and $\\sigma_-$. They find the lower branch is almost fully dichroic while the upper branch near 129 cm-1 retains about one-third of the opposite circular weight, which they attribute to hybridization with nearby infrared phonons; several phonons also show Faraday rotation, implying lattice vibrations acquire magnetic-field-dependent optical activity. If right, this provides a quantitative look at spin-phonon coupling in a 2D antiferromagnet and a route to probing magnetic order with polarization-resolved infrared light.","feed_headline":"Magnon split by field shows phonon-modified dichroism","feed_subtitle":"Faraday rotation exposes circular dichroism of both split magnon branches and optical activity of phonons.","key_machinery":"The central objects are the circular optical conductivities $\\sigma_\\pm(\\omega) = (\\sigma_{xx}+\\sigma_{yy})/2 \\pm i\\sigma_{xy}$, which the paper obtains by combining absolute polarized transmission with Faraday rotation angle measurements using a full polarizer-rotation protocol at $\\pm 7$ T. The reconstruction assumes an effective orthorhombic in-plane dielectric tensor with a single antisymmetric off-diagonal component $\\sigma_{xy}$, and uses field reversal so that the linear birefringence term $\\delta = (\\varepsilon_{xx}-\\varepsilon_{yy})/2$ cancels to leading order; the off-diagonal term is then extracted as $\\varepsilon_{xy}(\\omega) = (2c/\\omega d)\\, n_0(\\omega)\\Theta(\\omega)$. This is what converts measured rotation angles into the dichroism of the split magnon branches and of the phonon modes.","core_discovery":"The central claim is that the 122 cm-1 mode of FePS3 is a genuine magnetic excitation: it is polarization-independent, hardens on cooling with an order-parameter-like temperature dependence around $T_N \\approx 118$ K, and splits linearly with magnetic field at a gyromagnetic ratio near $0.94$ cm$^{-1}$/T. Combining absolute polarized transmission with Faraday rotation measured by a full polarizer-rotation protocol at $\\pm 7$ T, the authors reconstruct $\\sigma_\\pm(\\omega) = (\\sigma_{xx}+\\sigma_{yy})/2 \\pm i\\sigma_{xy}$ and show that the two field-split branches carry opposite circular dichroism. The lower branch is essentially fully dichroic, whereas the upper branch near 129 cm-1 shows a reduced dichroic response, with $\\sigma_+$ approximately one-third of $\\sigma_-$; this anomaly is attributed to hybridization with nearby infrared-active phonons at 128 and 133 cm-1. Phonon modes at 108, 133.1, and 165.6 cm-1 also exhibit Faraday rotation of up to roughly 50 mrad at 7 T, providing evidence that lattice vibrations acquire magnetic-field-dependent optical activity through spin-phonon coupling.","pith_inferences":["Beyond the paper, the same full-protocol Faraday technique should be transferable to monolayer FePS3, where the 122 cm-1 mode persists; measuring its circular dichroism would test whether spin-phonon coupling survives the two-dimensional limit.","Beyond the paper, the upper-branch hybridization hypothesis predicts an avoided crossing between the 129 cm-1 magnon and the 133 cm-1 phonon as a function of magnetic field, and resolving that crossing would yield a direct magnon-phonon coupling strength.","Beyond the paper, the authors' tentative crystal-field assignment leaves open whether the 122 cm-1 transition arises from the 6.4 meV or 17.9 meV spin-orbit level; circular-dichroism measurements as a function of field direction relative to the c-axis could distinguish the two."],"forward_implications":["The 122 cm-1 mode is a magnetic excitation: its frequency hardens on cooling along an order-parameter curve with $T_N \\approx 118$ K and its field splitting is linear at about 0.94 cm-1/T up to 7 T.","The lower split branch is essentially fully dichroic, with the $\\sigma_-$ spectral weight matching the zero-field mode, so the two branches can be assigned to opposite circular polarizations of the two spin sublattices.","The upper branch's reduced dichroism, with $\\sigma_+$ about one-third of $\\sigma_-$, is a signature of hybridization with nearby infrared phonons, providing direct evidence of spin-phonon coupling in the collective excitation spectrum.","Phonon modes at 108, 133.1, and 165.6 cm-1 exhibit Faraday rotation up to roughly 50 mrad at 7 T, showing that lattice vibrations acquire magnetic-field-dependent optical activity.","Polarization-resolved infrared magneto-spectroscopy can quantify spin-lattice coupling and circular dichroism in low-dimensional antiferromagnets."],"supporting_citations":[{"why":"Neutron scattering study that identified the 15 meV and 40 meV magnetic excitations of FePS3 at finite momentum, anchoring the optical assignment of the 122 cm-1 mode as a magnon.","marker":"[14]"},{"why":"Magneto-Raman study showing the 122 cm-1 mode splits linearly with magnetic field, which the paper reproduces and extends to infrared transmission.","marker":"[43]"},{"why":"Report of magnon polarons and an avoided crossing between the lower magnon branch and the 108 cm-1 phonon, used to explain the field dependence and the anomalous upper-branch response.","marker":"[46]"},{"why":"High-field study of high-angular-momentum excitations in FePS3 that documents avoided crossings; the paper invokes it for the 108 cm-1 coupling and as a model for the 129 cm-1 upper branch.","marker":"[48]"},{"why":"Provides the magneto-optical Kramers-Kronig procedure and fast-protocol Faraday measurements that the paper adapts to a full-protocol anisotropic analysis.","marker":"[19]"},{"why":"The variable dielectric function method used to extract the optical conductivity from transmission data without imposing full Drude-Lorentz line shapes.","marker":"[32]"},{"why":"Early Raman scattering work that assigned the 122 cm-1 feature in FePS3 and FePSe3, providing the historical identification the paper refines as a magnetic excitation.","marker":"[36]"},{"why":"Theoretical and experimental work on chiral phonons from magnon-phonon coupling in zigzag antiferromagnets, used as a qualitative framework for phonon Faraday rotation.","marker":"[50]"}],"fun_headline_variants":["FePS3 magnon splits in field, upper branch feels phonon","Faraday rotation exposes phonon-modified magnon dichroism","Spin-phonon coupling makes lattice modes optically active in FePS3","Field-split magnon branches show opposite circular dichroism"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the in-plane optical response is effectively orthorhombic with a single antisymmetric off-diagonal component and that linear birefringence cancels under field reversal, so that if monoclinic off-diagonal terms or birefringence are not negligible, the reconstructed circular conductivities and the reported dichroism would be distorted.","fun_headline_variants_meta":{"raw":{"variants":["FePS3 magnon splits in field, upper branch feels phonon","Faraday rotation exposes phonon-modified magnon dichroism","Spin-phonon coupling makes lattice modes optically active in FePS3","Field-split magnon branches show opposite circular dichroism"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000315,"raw_usage":{"total_tokens":1857,"prompt_tokens":1091,"completion_tokens":766,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":707,"completion_tokens_details":{"reasoning_tokens":691}},"tokens_in":707,"tokens_out":766,"duration_ms":11273,"temperature":1.0,"reasoning_tokens":691,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T14:54:54.901189+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the zero-field linear birefringence $\\delta = (\\varepsilon_{xx}-\\varepsilon_{yy})/2$ in the same 100-150 cm-1 range and compare it with $\\varepsilon_{xy}$ obtained from field reversal; if $\\delta$ is comparable to or larger than $\\varepsilon_{xy}$, the cancellation no longer isolates the magneto-optical term. A thickness-dependence or in-plane crystal-rotation check of the reconstructed $\\sigma_{xy}$ would also show whether residual cross-polarization or monoclinic terms contaminate the result.","supporting_citations":[{"cited_title":"Lançon, H","cited_arxiv_id":null,"evidence_quote":"Neutron scattering study that identified the 15 meV and 40 meV magnetic excitations of FePS3 at finite momentum, anchoring the optical assignment of the 122 cm-1 mode as a magnon."},{"cited_title":"McCreary, J","cited_arxiv_id":null,"evidence_quote":"Magneto-Raman study showing the 122 cm-1 mode splits linearly with magnetic field, which the paper reproduces and extends to infrared transmission."},{"cited_title":"Vaclavkova, M","cited_arxiv_id":null,"evidence_quote":"Report of magnon polarons and an avoided crossing between the lower magnon branch and the 108 cm-1 phonon, used to explain the field dependence and the anomalous upper-branch response."},{"cited_title":"Wyzula, I","cited_arxiv_id":null,"evidence_quote":"High-field study of high-angular-momentum excitations in FePS3 that documents avoided crossings; the paper invokes it for the 108 cm-1 coupling and as a model for the 129 cm-1 upper branch."},{"cited_title":"Levallois, I","cited_arxiv_id":null,"evidence_quote":"Provides the magneto-optical Kramers-Kronig procedure and fast-protocol Faraday measurements that the paper adapts to a full-protocol anisotropic analysis."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The variable dielectric function method used to extract the optical conductivity from transmission data without imposing full Drude-Lorentz line shapes."},{"cited_title":"Scagliotti, M","cited_arxiv_id":null,"evidence_quote":"Early Raman scattering work that assigned the 122 cm-1 feature in FePS3 and FePSe3, providing the historical identification the paper refines as a magnetic excitation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Theoretical and experimental work on chiral phonons from magnon-phonon coupling in zigzag antiferromagnets, used as a qualitative framework for phonon Faraday rotation."}],"review_version":1}