{"id":"e68bd103-26c0-46eb-975f-47d8f4e9a8b2","arxiv_id":"2501.19242","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Gypsum thin films support hyperbolic, canalized, and elliptical shear phonon polaritons, with a topological transition between regimes and group velocities down to 0.0005c.","lead":"Researchers imaged phonon polaritons, light and lattice vibration hybrids, in thin flakes of the mineral gypsum and watched their propagation shape change from open hyperbolas to closed ellipses as the frequency changed. This is the first time these shear polariton states, including a canalization regime, have been seen in an exfoliable thin crystal, pointing to slow-light and nanophotonics applications.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim rests on a 1983 bulk permittivity tensor being valid for thin exfoliated flakes; the paper's own nano-FTIR shows ~3 cm^-1 TO shifts, so the canalization frequency and regime labels may be mis-assigned.","rationale":"The reader's weakest-assumption analysis and this pass converge on the same point: the scientific claim of hyperbolic-to-elliptical shear polariton transition is interpreted through a bulk permittivity tensor from 1983, and the paper never validates that tensor for the specific exfoliated flakes. The concern is not merely hypothetical—the manuscript's own nano-FTIR data show a ~3 cm^-1 shift of the in-plane TO features, and SI Figure S7 reports flake-to-flake variation. Because the Reststrahlen bands are only 12-35 cm^-1 wide and the experimental images are spaced by 5-10 cm^-1, a small shift can change which regime is observed at a given nominal frequency. That makes the permittivity transfer the most load-bearing assumption in the central argument. I do not see a more fundamental internal flaw: the tensor comes from an independent 1983 reflectance fit rather than from fitting the polariton images, the near-field fringes are directly observed, and the TMM and COMSOL results agree qualitatively with the data. The unproved analytical IFC formula in the Methods is a genuine gap, but it is not load-bearing because the TMM IFCs in Figure 4G-J already support the same topological transition. The group-velocity values are also model-dependent, but they are secondary to the transition claim. The concrete check described above would settle whether the bulk tensor is adequate; until then, the correct assessment remains a qualified acceptance with the permittivity assumption as the principal condition.","tokens_in":16181,"tokens_out":7672,"duration_ms":74111,"concrete_test":"Extract the thin-film permittivity of the 75-nm and 150-nm gypsum flakes by fitting Lorentz oscillators to the polarized FTIR transmittance and nano-FTIR amplitude/phase spectra, then recompute the TMM IFCs at 1115, 1125, 1135, 1140, and 1150 cm^-1. If the best-fit TO frequencies differ from Aronson's 1110.2/1138.2 cm^-1 by more than ~3 cm^-1, or if the IFC at 1140 cm^-1 is no longer flat/canalized, the topological-transition assignment and the slow-light numbers need to be revised. A cheaper sensitivity check is to perturb the bulk oscillator frequencies by the observed -3 to +5 cm^-1 shifts and test whether the canalization window moves by more than the claimed ~5 cm^-1 range.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Every theoretical panel in the paper—the Reststrahlen-band classification, TMM dispersions, IFCs, and COMSOL simulations—uses the Lorentz-oscillator permittivity of bulk gypsum from Aronson et al. (1983), with no independent measurement of the thin-film permittivity (Section 2, Figure 1D, SI Tables S1/S2). This is load-bearing because the experimental claim is not just that fringes exist, but that specific frequencies correspond to hyperbolic, canalized, and elliptical shear polaritons, and those labels are read off the sign changes of the bulk-tensor components. The manuscript itself gives reason to worry: the s-SNOM nano-FTIR line scans (Figure 2B,C) place the in-plane TO resonances at about 1105 and 1135 cm^-1, whereas Aronson's tensor has them at 1110.2 and 1138.2 cm^-1. SI Figure S7 additionally states that the 1135 cm^-1 TO is 'slightly shifted to higher frequencies' in the 150-nm flake relative to the 75-nm flake. A 3-5 cm^-1 shift is comparable to the widths of the narrow Reststrahlen bands used to define the regimes (e.g., 1110-1122, 1122-1138, 1138-1172 cm^-1), so the canalization window, and hence the claimed topological transition, could move by more than the frequency spacing of the experimental images. If the thin-film oscillator strengths or damping also differ, the TMM dispersion fits and the extracted group velocities (down to 0.0005c) are affected. The qualitative observation of fringes is credible, but the quantitative assignment of the canalization transition and slow-light numbers depends on an unvalidated transfer of bulk optical constants.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports the first observation of elliptical and canalized shear phonon polaritons in thin films of gypsum, an exfoliable monoclinic crystal. Using s-SNOM nano-imaging and nano-FTIR spectroscopy on 75-nm and 150-nm flakes, the authors visualize a frequency-driven transition from hyperbolic shear polaritons through a canalization regime to elliptical shear polaritons. The experimental dispersion, extracted by fitting real-space fringes, is compared with transfer-matrix-method (TMM) calculations and full-wave COMSOL simulations, both based on the bulk dielectric tensor of gypsum from Aronson et al. (1983). The manuscript also reports unusually low group velocities, down to 0.0005c, and attributes the shear character to non-zero off-diagonal permittivity components in the monoclinic plane. The central claim is the topological transition of shear polaritons in a thin-film exfoliable material, supported by real-space images and theoretical IFCs.","tokens_in":16558,"tokens_out":3307,"duration_ms":31740,"significance":"If the central claim holds, this work extends shear polariton physics from bulk non-van der Waals crystals to exfoliable thin films, which is significant for nanophotonic integration and for exploring slow-light and non-Hermitian phenomena. The study is strengthened by the combination of real-space nano-imaging, nano-FTIR spectroscopy, TMM dispersion calculations, and COMSOL simulations, and by using an external bulk permittivity tensor rather than fitting the polariton data itself. The observation of canalized and elliptical shear polaritons in a natural material is potentially impactful. However, the quantitative regime assignments and the slow-light figures depend on assumptions that are not fully validated in the manuscript, as detailed below.","major_comments":[{"comment":"The central regime assignment—hyperbolic, canalized, elliptical—relies entirely on the bulk permittivity tensor of Aronson et al. (1983), yet the paper's own nano-FTIR data place the in-plane TO resonances at approximately 1105 and 1135 cm−1, whereas the Aronson tensor has them at 1110.2 and 1138.2 cm−1. SI Figure S7 further notes a shift of the 1135 cm−1 TO to higher frequencies in the 150-nm flake. Because the Reststrahlen bands used to define the regimes are narrow (e.g., 1110–1122, 1122–1138, 1138–1172 cm−1), shifts of 3–5 cm−1 are comparable to the band widths and can move the canalization frequency and the topological transition by more than the experimental frequency step. The authors should provide an independent determination of the thin-film permittivity, or at minimum test the sensitivity of the TMM/COMSOL predictions and regime labels to the observed TO shifts and to realistic variations of oscillator strengths and dampings.","section":"§2, Figure 2B/C, SI Figure S7, Tables S1/S2"},{"comment":"The analytical IFC formula used to produce Figure 4K is presented without proof, with the text stating that 'a detailed proof of the former result will be given elsewhere.' Since Figure 4K is a key theoretical corroboration of the claimed topological transition, this is load-bearing. The authors should either provide a complete derivation in the paper or supplementary material, cite a published derivation if it exists, or explicitly validate the formula against the TMM IFCs over the full frequency and wavevector range shown. Without this, the analytical IFC panel cannot be considered independent support for the central claim.","section":"Methods, 'Analytical approximations to the dispersion of shear polaritons in thin films'"},{"comment":"The group velocities down to 0.0005c and lifetimes of 0.6–2 ps are extracted by fitting the experimental dispersion with a power law y = a x^b and then differentiating. The manuscript reports no uncertainties on the fit parameters and no goodness-of-fit metrics, and the experimental dispersion points in Figure 2G,H carry sizable error bars. Given that the slow-light claim is a highlighted result, the authors should report confidence intervals on the group velocity and lifetime, and demonstrate that the power-law form is appropriate over the fitted range rather than an arbitrary smoothing of the data.","section":"§2, Figure 2G,H and SI Figure S5"},{"comment":"The frequency-dependent rotation angle γ(ω) used to diagonalize the real part of the permittivity is derived in the lossless limit, as the authors acknowledge in the text ('the derivation of γ(ω) considers a lossless scenario and it may be valid only when the losses are small'). The experimental frequencies are close to TO phonons where losses are large, so the rotated-frame classification of hyperbolic and elliptical regimes may be quantitatively unreliable. The authors should quantify the error in γ(ω) and in the regime boundaries when the full complex tensor is used without neglecting losses, or justify more rigorously that the lossless rotation remains a valid diagnostic for the IFC topology in this frequency range.","section":"§2, Eq. (1) and SI Figure S2"}],"minor_comments":[{"comment":"The main text refers to 'Figure 4C-F' for the real-part field simulations, while the caption lists panels '(C-E)' for four frequencies (1120, 1130, 1140, 1150 cm−1). Please correct the panel numbering for consistency.","section":"Figure 4 caption and main text"},{"comment":"The power-law fit y = a x^b is described without specifying which quantity is x and which is y (frequency versus wavevector) and over what range. Please define these variables explicitly in the caption or methods.","section":"SI Figure S5"},{"comment":"The paper frequently references prior work on hyperbolic shear metasurfaces [29] for the rotation-angle framework, but it would help the reader to state explicitly which formulas are taken from [29] and which are new to this work, particularly Eq. (1) and the analytical IFC expression.","section":"Introduction, Ref. [29]"},{"comment":"The abstract claims 'the first observation of elliptical shear and canalized shear phonon polaritons in gypsum thin films'; the word 'first' is strong given that the paper itself notes the narrowness of the frequency bands and the reliance on an external bulk permittivity. Please consider softening the wording or providing a more detailed contextual comparison in the introduction.","section":"Abstract"}],"recommendation":"major_revision","confidential_remarks":"The manuscript presents a visually compelling and potentially important observation, but the quantitative regime assignment and the slow-light figures hinge on the bulk permittivity tensor and on an unproved analytical formula. The issues are fixable within the scope of a revision, so I recommend major revision rather than rejection. The main risk is that the thin-film permittivity may differ from the bulk values by enough to shift the canalization frequency; the authors should address this explicitly with sensitivity analyses or independent thin-film characterization."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is worth taking seriously. It reports the first elliptical and canalized shear phonon polaritons, and the first shear polaritons in an exfoliable thin film, with real-space s-SNOM images that show the hyperbolic-to-elliptical transition through a canalization regime. The core observation is convincing: the fringes at 1115 and 1125 cm^-1 are hyperbolic, the parallel fringes near 1135–1140 cm^-1 look canalized, and the 1150 cm^-1 images are elliptical. The TMM and COMSOL calculations are consistent with this picture, and the permittivity inputs come from an independent 1983 bulk reflectance fit, not from fitting the near-field data, so the central prediction is not circular. The Asymmetric propagation and rotation of optical axes are nice signatures of shear behavior, and the confinement factors are respectable.\n\nThe soft spots are real but not fatal. First, the analytical IFC formula in Methods is explicitly unproved — \"detailed proof will be given elsewhere\" — yet it underlies Figure 4K, a key summary figure. That should be fixed or removed. Second, the stress-test's concern about permittivity transfer is legitimate: the paper's own nano-FTIR places the in-plane TOs at ~1105 and ~1135 cm^-1, a few wave numbers below the 1110.2 and 1138.2 cm^-1 in Aronson's tensor. Given the narrow Reststrahlen bands, a 3–5 cm^-1 shift can move the canalization onset and blur the regime labels at the 5 cm^-1 spacing of the experimental images. This does not overturn the qualitative transition, but it does weaken the precision of the \"topological transition\" and the extracted group velocities. The slow-light numbers come from a power-law fit with no error propagation, so the 0.0005c figure should be treated as indicative, not exact. Finally, no raw data or code are shipped, which makes independent verification harder than it should be.\n\nOverall: the central claim is plausible and likely correct in its qualitative form. The paper deserves a serious referee, but it needs revision to prove or cite the analytical formula, quantify the TO shifts, and provide error bars on the group velocities. I would not desk-reject it.","headline":"A credible first visualization of elliptical and canalized shear polaritons in thin gypsum, with a real but manageable soft spot: the bulk permittivity used to assign regimes is shifted by a few cm^-1 from the film's own TO resonances.","tokens_in":17206,"tokens_out":3575,"would_cite":true,"duration_ms":33157,"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":"Thin films of gypsum are shown to carry shear phonon polaritons that pass through a canalization regime as frequency rises.","keywords":["shear phonon polaritons","gypsum","monoclinic crystals","near-field nano-imaging","phonon polaritons","topological transition","canalization","slow light"],"falsifier":"Measure the permittivity of an exfoliated gypsum flake independently, for example by infrared ellipsometry, and recompute the isofrequency contours; if the sign pattern of $\\Re\\{\\varepsilon_{mm}\\}$, $\\Re\\{\\varepsilon_{nn}\\}$, and $\\Re\\{\\varepsilon_{zz}\\}$ or the off-diagonal component differs from the bulk tensor, the assigned hyperbolic, canalized, and elliptical frequency bands would shift or disappear. A simpler check is near-field imaging at the nominal canalization frequency: the two parallel fringes should appear only there, and a closed elliptical contour at that frequency would rule out the reported transition.","tokens_in":15952,"feed_emoji":"🔬","tokens_out":8027,"duration_ms":67456,"temperature":0.7,"pith_summary":"This paper reports the first observation of shear phonon polaritons in thin films of an exfoliable crystal, gypsum (calcium sulfate dihydrate). Using near-field nano-imaging and spectroscopy, it follows the polariton propagation as the frequency rises and finds a topological transition from hyperbolic shear propagation to elliptical shear propagation, with a shear canalization regime in between. The same transition appears in transfer-matrix dispersions, analytical isofrequency contours, and full-wave simulations. The paper also measures unusually low group velocities, down to 0.0005c, which makes gypsum a candidate platform for slow-light and nanophotonic devices.","feed_headline":"Shear polaritons in gypsum films change shape with frequency","feed_subtitle":"Near-field images trace a hyperbolic-to-elliptical transition, with light slowed to 0.0005c.","key_machinery":"The central object is the monoclinic dielectric permittivity tensor of gypsum, whose non-zero off-diagonal component $\\varepsilon_{xy}(\\omega)$ cannot be removed at all frequencies by a single rotation. The paper rotates the monoclinic plane by the frequency-dependent angle $\\gamma(\\omega) = 0.5 \\tan^{-1}(2\\,\\Re\\{\\varepsilon_{xy}\\}/(\\Re\\{\\varepsilon_{xx}\\}-\\Re\\{\\varepsilon_{yy}\\}))$, producing a coordinate frame where the real part of the off-diagonal term vanishes. The signs of $\\Re\\{\\varepsilon_{mm}\\}$, $\\Re\\{\\varepsilon_{nn}\\}$, and $\\Re\\{\\varepsilon_{zz}\\}$ in that frame define the hyperbolic type I, hyperbolic type II, and elliptical regimes, while the canalized regime sits at the crossing $\\Re\\{\\varepsilon_{mm}\\} \\approx 0$; transfer-matrix methods and analytical isofrequency contours turn this tensor into the predicted polariton dispersions that the near-field images are compared against.","core_discovery":"The central claim is that gypsum thin films support shear phonon polaritons in three distinct propagation regimes within a narrow mid-infrared window around 1100–1200 cm$^{-1}$: hyperbolic shear, canalized shear, and elliptical shear. These regimes are identified by the signs of the real parts of the permittivity components in a frequency-dispersive coordinate frame, and the paper assigns them to Reststrahlen bands built from two in-plane sulphate stretching phonons at about 1110 and 1138 cm$^{-1}$. The canalization occurs where the relevant permittivity component crosses zero, and the experimental images show asymmetric flattened wavefronts that the paper interprets as shear canalization, a form not reported before. Along the a axis the fitted polariton dispersion gives group velocities as low as 0.0005c with lifetimes around 0.6–2 ps. The paper concludes that gypsum is the first exfoliable material shown to host these shear phenomena in thin-film form.","pith_inferences":["Editorial inference: because all regime assignments use the bulk permittivity tensor, the exact transition frequencies for nanoscale flakes should be checked with an independent thickness-dependent measurement; small shifts would move the canalization band without changing the overall picture.","Editorial inference: the paper notes that oscillator angle and losses control the shear asymmetry, so deliberately varying losses (for example by temperature or by coupling to metal antennas) could act as a tuning knob for the propagation direction.","Editorial inference: the canalized shear regime, with energy flux along one direction for all allowed wavevectors, could be tested as a directional coupler by placing a second gypsum flake or a plasmonic antenna in the path of the canalized fringes."],"forward_implications":["Gypsum becomes the first exfoliable crystal platform demonstrated to support shear phonon polaritons in thin films.","The hyperbolic-to-elliptical transition through canalization provides a single material whose polariton topology can be tuned simply by changing the illumination frequency.","Group velocities down to 0.0005c suggest that gypsum films could be used for slow-light devices and enhanced infrared light-matter interactions.","Because gypsum is exfoliable, its thin films can be stacked or integrated into heterostructures with other van der Waals materials.","The asymmetric intensity of the canalized wavefronts gives a new observable signature for shear behavior in low-symmetry crystals."],"supporting_citations":[{"why":"Supplies the bulk Lorentz-oscillator dielectric tensor of gypsum used in the transfer-matrix, analytical, and full-wave calculations.","marker":"[38]"},{"why":"Establishes the hyperbolic shear polariton concept and the frequency-dependent rotation method used to classify the regimes.","marker":"[26]"},{"why":"Provides the generalized transfer-matrix formalism used to compute polariton dispersions and isofrequency contours.","marker":"[54]"},{"why":"Reports real-space nanoimaging of hyperbolic shear polaritons in another monoclinic crystal, serving as the experimental baseline for shear signatures.","marker":"[28]"},{"why":"Theoretical shear metasurface work that identifies oscillator angle and losses as the parameters controlling shear asymmetry and the rotation of the propagation axis.","marker":"[29]"},{"why":"Establishes in-plane anisotropic polariton imaging in a van der Waals crystal and the fringe interpretation used for the gypsum line scans.","marker":"[7]"}],"fun_headline_variants":["Gypsum films show shear polaritons with three distinct shapes","Shear polaritons in gypsum transition from hyperbolic to elliptical","Ultra-slow light in gypsum films via shear polaritons","Gypsum: first exfoliable material for shear polariton regimes","Topological shear polaritons visualized in gypsum thin films"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes that the bulk infrared dielectric tensor of gypsum, including its Lorentz-oscillator parameters, describes the 75 nm and 150 nm exfoliated flakes at every frequency used.","fun_headline_variants_meta":{"raw":{"variants":["Gypsum films show shear polaritons with three distinct shapes","Shear polaritons in gypsum transition from hyperbolic to elliptical","Ultra-slow light in gypsum films via shear polaritons","Gypsum: first exfoliable material for shear polariton regimes","Topological shear polaritons visualized in gypsum thin films"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000677,"raw_usage":{"total_tokens":3094,"prompt_tokens":977,"completion_tokens":2117,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":593,"completion_tokens_details":{"reasoning_tokens":2027}},"tokens_in":593,"tokens_out":2117,"duration_ms":14247,"temperature":1.0,"reasoning_tokens":2027,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T20:49:43.329511+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the permittivity of an exfoliated gypsum flake independently, for example by infrared ellipsometry, and recompute the isofrequency contours; if the sign pattern of $\\Re\\{\\varepsilon_{mm}\\}$, $\\Re\\{\\varepsilon_{nn}\\}$, and $\\Re\\{\\varepsilon_{zz}\\}$ or the off-diagonal component differs from the bulk tensor, the assigned hyperbolic, canalized, and elliptical frequency bands would shift or disappear. A simpler check is near-field imaging at the nominal canalization frequency: the two parallel fringes should appear only there, and a closed elliptical contour at that frequency would rule out the reported transition.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the bulk Lorentz-oscillator dielectric tensor of gypsum used in the transfer-matrix, analytical, and full-wave calculations."},{"cited_title":"& Wang, D.-W","cited_arxiv_id":null,"evidence_quote":"Establishes the hyperbolic shear polariton concept and the frequency-dependent rotation method used to classify the regimes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the generalized transfer-matrix formalism used to compute polariton dispersions and isofrequency contours."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports real-space nanoimaging of hyperbolic shear polaritons in another monoclinic crystal, serving as the experimental baseline for shear signatures."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Theoretical shear metasurface work that identifies oscillator angle and losses as the parameters controlling shear asymmetry and the rotation of the propagation axis."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes in-plane anisotropic polariton imaging in a van der Waals crystal and the fringe interpretation used for the gypsum line scans."}],"review_version":1}