{"id":"84bdd7f6-608c-40b1-848e-78289dbb47bb","arxiv_id":"2509.05819","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A single pass through a high-mobility 2D electron gas rotates microwave polarization by 82 degrees, approaching the ideal 90-degree Faraday limit.","lead":"Physicists rotated the polarization of microwaves by 82 degrees in a single pass through an ultra-clean 2D electron gas, close to the theoretical 90-degree limit. The result shows that the ordinary Hall effect in a nearly loss-free electron layer could underpin compact microwave isolators and circulators.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Appendix A Eq. (A9) has an algebraic error in the gain calibration: the fitted amplitude a is α/√(1+α), not α√(1+α), so the inferred Faraday angle may be systematically biased.","rationale":"The reader correctly identified the calibration of θF as the weakest point. I agree with that assessment, but I can make it more concrete: the calibration equations in Appendix A contain a specific algebraic error. Eq. (A9) misstates the prefactor a by a factor (1+α), which would directly bias the gain ratio and hence the inferred angle. This is exactly the kind of technical detail that should be checked before taking the 82° value at face value. I do not find an independent fatal flaw; the Drude model fits are plausible, the collisionless-regime reasoning is sound, and the Verdet comparison is conservative. However, the algebraic inconsistency and the lack of quantified uncertainty leave the central claim conditional. The reader's verdict of CONDITIONAL remains appropriate, so I recommend UNCHANGED.","tokens_in":8371,"tokens_out":27966,"duration_ms":290645,"concrete_test":"Recompute the Faraday angle using the corrected gain relation a=(g⊥/g∥)γα/√(1+α) or, better, determine g∥ and g⊥ from an independent VNA calibration at the reference plane and recompute θF(B). If the peak angle at 10.2 GHz differs by more than 5° from 82°, the near-total rotation claim is not robust to the calibration error.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central 82° claim rests on extracting θF from t⊥/t∥ after dividing by the complex gains g∥, g⊥. In Appendix A, the authors use Eq. (A7) z = a B B0/(B² + B0²) to fit the field dependence, and Eq. (A8) (μ̃B0)² = 1+α. Substituting the Drude tensor (A4, A5) into the transmission model (A1–A3) yields S⊥/S∥ = γα μ̃B/[1+α+(μ̃B)²]. Using (A8), this becomes z = (g⊥/g∥) γα B B0/[√(1+α)(B² + B0²)], so the correct prefactor is a = (g⊥/g∥) γα/√(1+α). Eq. (A9) instead gives a = (g⊥/g∥) α√(1+α) γ — an extra factor of (1+α). At 10.2 GHz, α ≈ 0.5 (from B*/Bc ≈ 1.23), so this is a ~50% error in the relation between the fitted amplitude and the gain ratio. If Eq. (A9) was used to infer g⊥/g∥, the extracted tanθ is overestimated by ~1.5, shifting the reported 82° downward by several degrees. The paper provides no error bars, and the theoretical curves in Fig. 3 are amplitude-scaled, so this systematic bias is not checked.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports microwave transmission measurements through a high-mobility GaAs 2DEG placed in a circular waveguide behind a conducting iris. From field-dependent parallel and perpendicular transmission coefficients, the authors extract a Faraday rotation angle θF = 82° (1.43 rad) at 10.2 GHz and B ≈ 30 mT on a single pass, corresponding to V = 9.5×10^8 rad T⁻¹ m⁻¹. The rotation is interpreted with a classical Drude conductivity tensor in the collisionless limit (ωτ ≫ 1) with reactive electromagnetic coupling, modeled by tanθF = S⊥/S∥ = α μ̃ B / (B² − B0²). The model is fitted to data at three frequencies, and the frequency dependence of α, μ̃, and B0 is presented.","tokens_in":8815,"tokens_out":6672,"duration_ms":75698,"significance":"If the 82° result holds, it is a striking demonstration: the reported Verdet constant exceeds previously studied 2D materials and atomic systems by a large margin, and the collisionless reactive-coupling mechanism would be relevant for nonreciprocal devices. The paper's strengths are the direct measurement of S∥ and S⊥, the compact analytic Drude description, and the explicit parameter-extraction procedure. However, the central quantitative claim depends on a calibration step in Appendix A that contains an algebraic error, and the paper provides no uncertainty analysis for the extracted angle. These issues are load-bearing and must be resolved before the claim can be accepted.","major_comments":[{"comment":"The prefactor in the fit z = a B B0/(B² + B0²) is algebraically incorrect. Substituting (A4)–(A5) into (A1)–(A3) gives tanθ = S⊥/S∥ = γα μ̃ B / [1 + α + (μ̃B)²]. Using (A8), (μ̃B0)² = 1 + α, this becomes tanθ = γα B / [μ̃ (B² + B0²)] = γα B B0 / [√(1+α)(B² + B0²)]. Hence a = (g⊥/g∥) γα / √(1+α), not (g⊥/g∥) γα √(1+α) as written in Eq. (A9). The published relation is too large by a factor (1+α), which is ≈1.5 at 10.2 GHz. If Eq. (A9) was used to determine g⊥/g∥ from the fitted amplitude a, the inferred S⊥/S∥ and therefore θF are overestimated. Please correct Eq. (A9) and re-evaluate the reported 82° value and the quoted Verdet constant.","section":"Appendix A, Eq. (A9)"},{"comment":"The extraction of θF from the measured scattering parameters rests on the even/odd mixing removal and on the complex gains g∥ and g⊥, but no uncertainties or calibration-error propagation are given. Near B*, |S∥| is small, so θF is highly sensitive to any residual offset or gain error in either channel. The theoretical curves in Fig. 3 are amplitude-scaled to match the data, so a systematic calibration bias would not be visible in the fits. Please provide error bars on |S∥|, |S⊥|, and θF, and a sensitivity analysis of θF to plausible misalignment, gain-ratio, and orthomode-transducer non-idealities.","section":"Sec. III.A and Appendix A"},{"comment":"The dashed theoretical curves in Fig. 3 use per-frequency fitted α and B0, and the caption states that the theoretical amplitude is scaled to match the data. The model therefore has at least two complex free parameters per frequency plus an overall scale, so the 'excellent agreement' is not a parameter-free validation of the underlying mechanism. In particular, the divergence of the peak as Im B0 → 0 is inferred from fits to the same data. Please state explicitly which quantities are predicted from independent inputs (e.g., dc mobility, geometry) and which are fitted, and show a comparison using only independently determined parameters.","section":"Sec. III.A, Fig. 3"}],"minor_comments":[{"comment":"The caption says the theoretical amplitudes are scaled to match experiment, but the scaling procedure is not described in the text. Please specify how the scale was determined and whether it is a free parameter in the fits.","section":"Fig. 3 caption"},{"comment":"The figures would benefit from error bars; currently no measurement uncertainty is shown for the transmission coefficients or for the extracted Faraday angle.","section":"Figs. 2, 5, 6"},{"comment":"In the text following Eq. (7), the branch of the square root in (μ̃B0)² = 1+α is not specified. Since α and μ̃ are complex, please define the chosen branch to avoid ambiguity in the plotted B0.","section":"Sec. III.B"},{"comment":"The iris diameter is given as 9 mm, but the sample is 1 cm × 1 cm; a sentence clarifying the relative alignment and the electrical contact between the metal plate and the 2DEG would help reproducibility.","section":"Sec. II"}],"recommendation":"major_revision","confidential_remarks":"The algebraic error in Eq. (A9) is concrete and easily corrected. The key question is whether the 82° value survives a corrected gain calibration and a proper uncertainty analysis; if it does, the paper would be a strong experimental result. The use of the same group's earlier linear-transport ansatz [8] for the interpretation is reasonable but should be stated more clearly as a model assumption rather than a derivation from first principles."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the paper reports something worth knowing—near-total single-pass Faraday rotation in a classical 2DEG, not the quantum regime—and the reactive-coupling/iris story is coherent. But the quantitative headline (82°, Verdet record) is softer than the prose suggests. The amplitude-scaled fits and missing error bars are one issue; more concretely, I checked the appendix algebra and Eq. (A9) has an extra factor. With z = (g⊥/g∥) S⊥/S∥ and the Drude tensor (A4, A5), you get S⊥/S∥ = γα μ̃B/[1+α+(μ̃B)^2] = γα B B0/[√(1+α)(B^2+B0^2)], so the prefactor a should be (g⊥/g∥) γα/√(1+α), not (g⊥/g∥) γα√(1+α). At α≈0.5 this is a ~50% error in a, so if A9 was used to infer g⊥/g∥, the extracted tanθ is inflated by about 1.5. That shifts the 82° down by several degrees—not enough to destroy the qualitative claim, but enough to matter when you compare Verdet constants.\n\nWhat's genuinely good: the experiment is clean in conception, the iris engineering is thoughtful, and the Drude fits capture the peak shape and detuning B* > Bc. The collisionless-limit interpretation (imaginary μ̃⁻¹, reactive α) is physically reasonable, and the paper is honest about the standing-wave ripple. The Verdet constant uses the 50 nm quantum well thickness, which is a convention you can quibble with but not an error.\n\nThe soft spots, in order: (1) the appendix calibration error; (2) theoretical curves scaled to match data—this means the fits don't test the absolute amplitude, which is where the Verdet value comes from; (3) no error bars on any quoted number. These are fixable. The paper deserves a serious referee, but I would not cite the quantitative claim until the authors correct A9 and give a calibration-uncertainty budget.","headline":"Near-total Faraday rotation looks plausible, but Appendix A has a calibration algebra error that biases the headline 82° and there are no error bars.","tokens_in":9304,"tokens_out":2578,"would_cite":false,"duration_ms":26983,"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":"A classical Hall effect in a high-mobility 2DEG rotates microwave polarization by 82 degrees on a single pass, approaching the ideal 90-degree limit.","keywords":["Faraday rotation","Hall effect","two-dimensional electron gas","microwave transmission","cyclotron resonance","Verdet constant","non-reciprocal devices","Drude conductivity"],"falsifier":"Measure the transmitted polarization directly at the waveguide output with a rotating analyzer at f = 10.2 GHz and B ≈ B* ≈ 0.13 T, without relying on the S-parameter gain calibration of Appendix A; the major axis of the transmitted polarization would have to lie at 82° from the incident axis.","tokens_in":8316,"feed_emoji":"🌀","tokens_out":11335,"duration_ms":117157,"temperature":0.7,"pith_summary":"The paper sets out to show that the ordinary, classical Hall effect—not a quantum effect—can rotate the polarization of a microwave beam by almost the maximum possible 90 degrees in a single pass through a two-dimensional electron gas. In a high-mobility GaAs quantum well at 4 K, the authors measure 82 degrees of rotation at 10.2 GHz with a small magnetic field of 30 mT, corresponding to a Verdet constant of about 9.5×10^8 rad T^-1 m^-1, roughly an order of magnitude larger than earlier reported values in graphene and other 2D materials. The key regime is ωτ≫1, where the electron gas behaves reactively rather than resistively, combined with a deliberately weak, capacitive coupling provided by a metal iris in the waveguide. A classical Drude model with a single complex pole reproduces the field dependence of the rotation at three frequencies. If correct, the result overturns the common assumption that dissipation in a conducting 2D layer prevents large Faraday rotation, and it points to a simple route to non-reciprocal microwave devices.","feed_headline":"2D electron gas rotates microwaves 82 degrees in one pass","feed_subtitle":"The 50-nanometer GaAs layer beats benchmark Verdet constants by an order of magnitude.","key_machinery":"The central object is the complex pole B0 in tan(θF) = α μ̃ B/(B² − B0²), obtained from the Drude conductivity tensor via tan(θF)=γZσyx/(K+Zσxx). Here α=(Z/K)σ̃0 is the dimensionless 2DEG-waveguide coupling, μ̃=μ/(1−iωτ) the inertia-delayed mobility, and B0=i(1+α)^{1/2}/μ̃. The iris is the control: its sub-cutoff aperture makes Z/K predominantly imaginary (capacitive), while in the ωτ≫1 limit μ̃^-1 is also almost purely imaginary; the two combine so that B0 has a small imaginary part. The tangent then peaks sharply at B*=|B0|, and its maximum grows as Im B0 → 0, pushing θF toward 90°. A corollary is |B0|=|1+α|^{1/2}Bc: the iris sets how far the peak sits from cyclotron resonance.","core_discovery":"The paper claims that near-total Faraday rotation can be produced by the classical Hall effect alone when the 2DEG is collisionless (ωτ≫1, here ωτ>17) and weakly, reactively coupled to the field. In this regime, the measured parallel and perpendicular transmissions through a 50 nm GaAs quantum well give tan θF = S⊥/S∥, peaking at B* ≈ 0.13 T at 10.2 GHz, detuned from cyclotron resonance; at the peak the phase delay vanishes, leaving a linear polarization rotated by 82°, with Verdet constant 9.5×10^8 rad T^-1 m^-1. The field dependence is reproduced by the Drude conductivity via tan θF = α μ̃ B/(B² − B0²), whose complex pole B0 has small imaginary part in this regime, allowing θF to approach","pith_inferences":["If the mechanism is as general as the paper hints, the same iris-plus-collisionless-2DEG recipe should work in graphene or topological surface states and at terahertz or optical frequencies whenever ωτ≫1; Verdet comparisons would then need to be renormalized by layer thickness.","The linear-polarization-at-peak property suggests a concrete isolator layout: a 90° Hall rotator between two angled polarizers should block reverse transmission; a testable extension is measuring isolation ratio in such a two-port configuration.","The detuning relation B* = √(1+α) Bc gives a direct experimental handle: varying the iris diameter should move the rotation peak and change its sharpness in a predictable way, a check that requires no new physics beyond the paper's model."],"forward_implications":["A classical, non-quantized Hall response can deliver essentially total Faraday rotation on a single pass, so the usual assumption that 2DEG dissipation caps the rotation angle does not apply in the ωτ≫1 regime.","The reported Verdet constant exceeds the paper's cited graphene THz value by roughly an order of magnitude, making the geometry attractive for compact rotators.","At the rotation peak the transmitted polarization is linear, which is the input state a Faraday isolator or circulator needs; non-reciprocal devices based on the classical Hall effect become plausible.","The peak field B* is set by √(1+α) times the cyclotron field, so the iris geometry directly tunes where and how sharply the near-90° rotation occurs.","The Drude model with the complex pole B0 quantitatively reproduces |tan θF| versus B at 9.20, 10.20, and 11.20 GHz, giving a predictive account of the frequency dependence."],"supporting_citations":[{"why":"Supplies the linear transport ansatz tan(θF)=γZσyx/(K+Zσxx) and the waveguide/iris measurement geometry this analysis builds on.","marker":"[8]"},{"why":"Earlier demonstration of microwave Faraday rotation in magnetically biased graphene by the classical Hall effect, the direct precedent this work extends.","marker":"[14]"},{"why":"Field-effect tuning of microwave Faraday rotation and isolation in graphene, establishing the classical-Hall rotator context at microwaves.","marker":"[15]"},{"why":"Provides the monolayer-graphene THz Verdet constant used as the main benchmark for the reported enhancement.","marker":"[7]"},{"why":"Provides a state-of-the-art slow-light atomic vapour Verdet constant in the microwave domain, another baseline for comparison.","marker":"[19]"}],"fun_headline_variants":["82° Faraday rotation from Hall effect alone","2DEG rotates microwaves 82° in one pass","Record Verdet constant from 2D electron gas","Near-total Faraday rotation in single 2DEG pass","Classical Hall effect spins microwaves 82°"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The load-bearing premise is the Appendix A calibration: the measured parallel and perpendicular transmissions are assumed, after even/odd symmetrization, to be proportional to the true S∥ and S⊥ with known complex gains, and because S∥ is small near the peak, any residual calibration offset can shift the reported 82° and the inferred Verdet constant.","fun_headline_variants_meta":{"raw":{"variants":["82° Faraday rotation from Hall effect alone","2DEG rotates microwaves 82° in one pass","Record Verdet constant from 2D electron gas","Near-total Faraday rotation in single 2DEG pass","Classical Hall effect spins microwaves 82°"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000259,"raw_usage":{"total_tokens":1459,"prompt_tokens":815,"completion_tokens":644,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":559,"completion_tokens_details":{"reasoning_tokens":566}},"tokens_in":559,"tokens_out":644,"duration_ms":7656,"temperature":1.0,"reasoning_tokens":566,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T04:58:08.833829+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the transmitted polarization directly at the waveguide output with a rotating analyzer at f = 10.2 GHz and B ≈ B* ≈ 0.13 T, without relying on the S-parameter gain calibration of Appendix A; the major axis of the transmitted polarization would have to lie at 82° from the incident axis.","supporting_citations":[{"cited_title":"Suresh, E","cited_arxiv_id":null,"evidence_quote":"Supplies the linear transport ansatz tan(θF)=γZσyx/(K+Zσxx) and the waveguide/iris measurement geometry this analysis builds on."},{"cited_title":"Sounas, H","cited_arxiv_id":null,"evidence_quote":"Earlier demonstration of microwave Faraday rotation in magnetically biased graphene by the classical Hall effect, the direct precedent this work extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Field-effect tuning of microwave Faraday rotation and isolation in graphene, establishing the classical-Hall rotator context at microwaves."},{"cited_title":"Siddons, N","cited_arxiv_id":null,"evidence_quote":"Provides a state-of-the-art slow-light atomic vapour Verdet constant in the microwave domain, another baseline for comparison."}],"review_version":1}