REVIEW 3 major objections 5 minor 45 references
Tunable Nanophotonic Devices and Cavities based on a Two-Dimensional Magnet
T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash
Pith's one-line read The paper demonstrates that CrSBr photonic crystal slabs support magnetically tunable exciton-polaritons, with a 0.4 T field shifting resonances by up to 25 nm and switching a guided mode from elliptic to hyperbolic without changing the…
desk verdict Solid, useful demonstration of magnetically tunable photonic resonances in CrSBr slabs, but the hyperbolic-switching and strong-coupling claims outrun the permittivity evidence. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing mechanism is the combination of an unusually large, highly anisotropic, magnetically sensitive permittivity in CrSBr near its excitonic resonances with a photonic crystal slab geometry. The b-axis permittivity is described by a Lorentzian oscillator model with three oscillators X1, X2, and X* plus a background epsilon_b = 11.1; the oscillator energies shift when the magnetic ground state switches from antiferromagnetic to ferromagnetic (X1 from 1.3637 to 1.3481 eV, X2 from 1.76 to 1.67 eV). This permittivity model is fed into RCWA and finite-element solvers to reproduce the measured reflectivity and to label the modes (TE21, TE41). The flattening of the photonic bands in momentum space follows from the large in-plane index, and the etch pattern (elliptical holes in a 50-nm SiN mask) brightens the dark guided modes.
What would settle it
Measure the reflectivity of a patterned CrSBr slab before and after removing the 50-nm SiN cap (for example, by a reactive-ion etch that stops at CrSBr) and compare mode positions and linewidths with RCWA simulations built from the unpatterned-flake permittivity; if the observed mode shifts or widths deviate by more than the simulated values, or if the 66 meV Rabi splitting cannot be recovered by fitting the measured spectra, the central claim would be falsified.
Extended reading notes
Core claim
The central discovery is that photonic crystal slabs fabricated from the van der Waals antiferromagnet CrSBr support guided-mode resonances that self-hybridize with the material's excitons, forming exciton-polaritons, and that these modes are highly tunable in situ by external magnetic fields. Because the real part of the b-axis permittivity reaches values between 100 and 450 near the X1 exciton, the modes are spectrally dense, have small mode volumes, and Q-factors above 500. Applying a 0.4 T field along the b-axis abruptly flips the spins from antiferromagnetic to ferromagnetic ordering, shifting the resonances by as much as 10 nm near X1 and 25 nm near X2, and converting a TE21 guided resonance into a hyperbolic exciton-polariton resonance without altering the geometry. The authors report a Rabi splitting of 66 meV for the TE21 mode, an order of magnitude larger than the Fabry-Perot mode's coupling, and attribute patterning-induced features in the epsilon-below-zero region to hyperbolic exciton-polaritons.
Load-bearing premise
The fitted permittivity model extracted from unpatterned CrSBr flakes is assumed to remain valid for the etched, SiN-capped CrSBr slabs, so all simulated mode identities, the 66 meV Rabi splitting, and the hyperbolic classification rest on that transfer.
Editorial extensions
If this is right
- A 0.4 T field can repeatedly and reversibly shift guided resonances by up to 25 nm, and because the origin is an electronic transition, there is no limit on the number of switching operations.
- The same design principles apply to other layered magnets such as CrOCl, NbOCl2, and CrPS4, raising the possibility of room-temperature nanophotonic devices with similar tunability.
- Much thinner slabs, down to lambda/150, are feasible, and higher Q-factors could be reached through bound states in the continuum, enabling stronger light-matter interactions and potentially lasing.
- The permittivity contrast across the AFM-to-FM transition exceeds the contrast between silicon nitride and vacuum over a wavelength range of more than 180 nm, so CrSBr slabs could act as magnetic-field-activated dielectric switches even without patterning.
Reading between the lines
- If the central claim holds, the same slab should exhibit an anticrossing when the guided-mode resonance is tuned through the exciton by temperature or magnetic field; re-examining the published spectra for such an anticrossing would test the 66 meV Rabi splitting.
- The authors' claim that the photonic bands are flat in momentum space could be checked directly by angle-resolved reflectivity, which they argue is unnecessary; this is a straightforward independent verification.
- A near-field optical experiment imaging the mode at 0 T and 0.4 T would show the predicted change from an elliptic guided resonance to a hyperbolic polariton as a change in the spatial pattern of the local density of states.
- Because the SiN cap has permittivity near 4 while CrSBr's is near 20 at X2, the cap may perturb the X2 modes more than the X1 modes; removing or thinning the cap in a future device would isolate the intrinsic CrSBr response.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript reports the fabrication of photonic crystal slabs patterned directly into the layered antiferromagnetic semiconductor CrSBr, with a 50-nm SiN etch mask left in place, and characterizes them by reflectivity and photoluminescence at 5 K. The authors observe sharp guided-mode resonances near the X1 exciton (~910 nm) and a broader resonance near X2 (~710 nm), and show that applying a 0.4-T magnetic field along the b-axis reversibly shifts these features by up to 10 nm and 25 nm, respectively, as the spin ground state switches from antiferromagnetic to ferromagnetic. Using a Lorentzian permittivity model fitted to unpatterned CrSBr flakes, RCWA and FEM simulations are used to assign mode labels (TE21, TE41), to identify epsilon-below-zero/hyperbolic modes, and to claim intrinsic strong coupling with a 66-meV Rabi splitting. The central claims are the demonstration of monolithic, in situ magnetically tunable nanophotonic cavities and the elliptic-to-hyperbolic mode switching without geometry change.
Significance. If the interpretation holds, this is a significant advance: it would establish CrSBr as a monolithic platform in which magnetic fields tune both the spectral position and the character (elliptic vs hyperbolic) of high-Q photonic resonances in a reversible, geometry-independent way. The direct experimental observations—reflectivity dips and PL peaks shifting by 10–25 nm under 0.4 T—are credible and well matched by the reported simulations. The paper also benefits from a systematic thickness series of unpatterned flakes (16 flakes, 13–208 nm) used for the permittivity fit and from an original stencil-lithography fabrication route. However, the strongest physics claims (mode labels, hyperbolic polaritons, 66-meV Rabi splitting) rest entirely on the validity of transferring the unpatterned permittivity model to patterned, SiN-capped slabs, and on an X* oscillator that is introduced specifically to produce the hyperbolic region. These load-bearing assumptions are not yet independently verified.
major comments (3)
- [Methods, Permittivity Fits; Results, epsilon-below-zero region] The hyperbolic-polariton and epsilon-below-zero claims are circular in their present form. The third oscillator X* is introduced in Methods with the explicit motivation "as we are also interested in the hyperbolic permittivity region," and the same model is then used to classify the observed ~860–908 nm dips as hyperbolic exciton-polaritons. X* is not independently constrained by any measurement (e.g., ellipsometry, angle-resolved reflectance, or a fit residual analysis), so the sign of Re(epsilon_b) in that region, and hence the elliptic-to-hyperbolic switching in Fig. 3a, is an input rather than an output of the analysis. Please provide an independent determination of the permittivity in the 860–908 nm window or show that the data cannot be fitted without X*.
- [Methods, Nanofabrication; Results, Fig. 4] The transfer of the unpatterned-flake permittivity model to the patterned devices is unverified and is load-bearing for all mode assignments. The model was fit to unpatterned CrSBr on sapphire, but the devices are 20–30 nm CrSBr on 285-nm SiO2 capped by a 50-nm SiN mask that is deliberately left in place. Near X2 the b-axis permittivity is only ~20, so the SiN layer (epsilon ≈ 4, 50 nm thick) is not negligible compared with the active layer; strain or etching damage could further shift the exciton energies. Because the TE21/TE41 labels, the hyperbolic classification, and the 66-meV splitting are all derived from simulations using this model, the conclusion is not established by the data alone. Please include sensitivity tests (e.g., simulations with the SiN removed, with SiN thickness varied by ±20%, or with the exciton energies shifted by the observed flake-to-flake spread) and, if possible, an angle-resolved measurement to constrain the guided-mode dispersion.
- [Results, Rabi splitting] The strong-coupling claim and the value ℏΩ_R = 66 meV are stated without the supporting evidence normally required for a Rabi splitting. The text reports that "the PL maximum follows the reflectivity dip," which is equally consistent with a purely photonic resonance tracking the exciton redshift. No anticrossing is shown as a function of detuning or in-plane momentum, and no coupled-oscillator fit is presented. Please show the dispersion or a detuning series demonstrating an avoided crossing, and state the cavity and exciton linewidths to justify that the splitting exceeds them.
minor comments (5)
- [Methods, Permittivity Fits] The Lorentzian formula in the Methods is typeset incorrectly (the denominator appears as "i E Gamma_i i"); please fix and define all symbols (E, E_i, f_i, Gamma_i).
- [Methods, Simulations] The definition of TEml says "m antinodes in the electric field profile in the a-axis and l antinodes in the b-axis"; please clarify whether m and l count antinodes or nodes and provide the field profiles with the TE21 and TE41 labels in the main text or a Supplementary figure.
- [References] References [28]–[30] are to a manuscript under review and two preprints; please update with published versions if available, and ensure reference [28] is accessible to readers.
- [Introduction and Results] The phrases "unprecedented in situ control" and "the first instance of achieving on-demand operational mode switching" are strong claims; please temper them or support them with a direct comparison to prior tunable photonic-crystal demonstrations.
- [Results, paragraph after Fig. 2] The term "epsilon-below-zero region ~860–908 nm" should specify which component of the permittivity tensor is meant (b-axis), since the in-plane a-axis permittivity is positive and ~11.
Circularity Check
No circularity: the magnetic tunability is directly measured, and the simulated mode classifications are forward-model consistency checks rather than reductions of the claims to their fitted inputs.
full rationale
The central demonstration — the 10 nm (15 meV) shift near X1 and 25 nm (50 meV) shift near X2 upon a 0.4 T AFM-to-FM spin flip, and the resulting elliptic-to-hyperbolic mode switching — is a direct reflectivity/PL measurement on patterned devices, not a quantity derived from the fitted permittivity model. The simulations do use the unpatterned-flake permittivity as input (Results: 'The permittivity tensor extracted from the experimental data and prior work are then used as input to RCWA and finite element method solvers for reflectivity simulations'), but they are then compared against measured spectra of the patterned slabs; this is a forward consistency test, and the observed mode shifts are independently established by experiment. The X* oscillator was added in Methods 'as we are also interested in the hyperbolic permittivity region', but negative real permittivity below the X1 resonance is already implied by the X1 oscillator itself, so the hyperbolic-polariton classification is not created solely by the added oscillator; moreover, the dips in that region are observed in the patterned samples and classified as hyperbolic exciton-polaritons by reference to an external work (ref. 34), not by an identity with the fit. The cited raw permittivity data in ref. 28 (same-group, under review) is a verifiability and availability weakness, and the assertion that the SiN cap is inconsequential (Methods: 'As SiN has a very small relative permittivity compared to CrSBr... leaving the membrane on CrSBr after the patterning is inconsequential') is an unvalidated modeling assumption. These are evidence/correctness concerns, not circularity: no equation, fitted parameter, or self-citation is shown to reduce to the claim it supports.
Assumptions & free parameters
free parameters (4)
- b-axis background permittivity (epsilon_b) =
11.1
- X1 oscillator strength, energies, damping =
f=1.7 eV^2; E_AFM=1.3637 eV, E_FM=1.3481 eV; Gamma=0.0008 eV
- X2 oscillator strength, energies, damping =
f=1.2 eV^2; E_AFM=1.76 eV, E_FM=1.67 eV; Gamma=0.025 eV
- X* oscillator strength, energies, damping =
f=0.5 eV^2; E_AFM=1.3771 eV, E_FM=1.361 eV; Gamma=0.0075 eV
assumptions (5)
- domain assumption The b-axis complex permittivity of CrSBr is described by a Lorentzian oscillator sum with the fitted parameters (X1, X2, X*, background 11.1).
- domain assumption The a-axis and c-axis permittivities of CrSBr are 11 and 4, taken from prior studies.
- domain assumption The 50-nm SiN mask left on top of CrSBr does not significantly alter the optical response because its permittivity is much smaller than CrSBr's near X1.
- standard math Maxwell solvers (RCWA via S4, FEM via COMSOL) accurately model normal-incidence reflectance of the patterned stack.
- domain assumption The photonic modes in the patterned slab are strongly TE-like and can be labeled by antinode counts in the a and b axes.
invented entities (1)
-
X* oscillator
Cite this review
Pith. "Pith review of Tunable Nanophotonic Devices and Cavities based on a Two-Dimensional Magnet." pith.science (2026). https://pith.science/paper/7M6HMSTK
@misc{pith2026250709720,
author = {Pith},
title = {Pith review of: Tunable Nanophotonic Devices and Cavities based on a Two-Dimensional Magnet},
year = {2026},
howpublished = {\url{https://pith.science/paper/7M6HMSTK}},
note = {Machine review of arXiv:2507.09720}
}
read the original abstract
Central to the field of nanophotonics is the ability to engineer the flow of light through nanoscale structures. These structures often have permanent working spectral ranges and optical properties that are fixed during fabrication. Quantum materials, with their correlated and intertwined degrees of freedom, offer a promising avenue for dynamically controlling photonic devices without altering their physical structure. Here, we fabricate photonic crystal slabs from CrSBr, a van der Waals antiferromagnetic semiconductor, and demonstrate unprecedented in situ control over their optical properties. Leveraging the combination of the exceptionally large refractive index of CrSBr near its excitonic resonances and its tunability via external fields, we achieve precise manipulation of photonic modes at near-visible and infrared wavelengths, showcasing a new paradigm for nanophotonic device design. The resulting guided resonances of the photonic crystal are tightly packed in the spectrum with very small mode volumes, are highly tunable via external magnetic fields, and exhibit high Q-factors exceeding 500. These resonances self-hybridize with the excitonic degrees of freedom, resulting in intrinsic strong light-matter coupling. Our findings underscore the potential of quantum materials for developing in situ tunable photonic elements and cavities.
Figures
Figures from the paper (1 more)
Reference graph
Works this paper leans on
-
[1]
Zhang, X. et al. Guiding of visible photons at the ångström thickness limit. Nat. Nanotechnol. 14, 844–850 (2019)
work page 2019
-
[2]
Wu, S. et al. Monolayer semiconductor nanocavity lasers with ultralow thresholds. Nature 520, 69–72 (2015)
work page 2015
-
[3]
Ye, Y. et al. Monolayer excitonic laser. Nat. Photonics 9, 733–737 (2015)
work page 2015
-
[4]
Munkhbat, B. et al. Transition metal dichalcogenide metamaterials with atomic precision. Nat. Commun. 11, 4604 (2020)
work page 2020
-
[5]
Lin, H. et al. Diffraction-limited imaging with monolayer 2D material-based ultrathin flat lenses. Light Sci. Appl. 9, 137 (2020)
work page 2020
-
[6]
Klembt, S. et al. Exciton-polariton topological insulator. Nature 562, 552–556 (2018)
work page 2018
- [7]
-
[8]
Laitz, M. et al. Uncovering temperature-dependent exciton-polariton relaxation mechanisms in hybrid organic-inorganic perovskites. Nat. Commun. 14, 2426 (2023)
work page 2023
Show all 45 references
-
[9]
Byrnes, T., Kim, N. Y. & Yamamoto, Y. Exciton–polariton condensates. Nat. Phys. 10, 803–813 (2014)
2014
-
[10]
& Yamamoto, Y
Deng, H., Weihs, G., Santori, C., Bloch, J. & Yamamoto, Y. Condensation of Semiconductor Microcavity Exciton Polaritons. Science 298, 199–202 (2002)
2002
-
[11]
M., Khitrova, G
Gibbs, H. M., Khitrova, G. & Koch, S. W. Exciton–polariton light–semiconductor coupling effects. Nat. Photonics 5, 273–273 (2011)
2011
-
[12]
Liu, X. et al. Strong light–matter coupling in two-dimensional atomic crystals. Nat. Photonics 9, 30–34 (2015)
2015
-
[13]
Dirnberger, F. et al. Spin-correlated exciton–polaritons in a van der Waals magnet. Nat. Nanotechnol. 17, 1060–1064 (2022)
2022
- [14]
-
[15]
Maggiolini, E. et al. Strongly enhanced light–matter coupling of monolayer WS2 from a bound state in the continuum. Nat. Mater. 22, 964–969 (2023)
2023
-
[16]
Weber, T. et al. Intrinsic strong light-matter coupling with self-hybridized bound states in the continuum in van der Waals metasurfaces. Nat. Mater. 22, 970–976 (2023)
2023
-
[17]
Dirnberger, F. et al. Magneto-optics in a van der Waals magnet tuned by self-hybridized polaritons. Nature 620, 533–537 (2023)
2023
-
[18]
Jarc, G. et al. Cavity-mediated thermal control of metal-to-insulator transition in 1T-TaS2. Nature 622, 487–492 (2023)
2023
-
[19]
Latini, S. et al. The ferroelectric photo ground state of SrTiO3: Cavity materials engineering. Proc. Natl. Acad. Sci. 118, e2105618118 (2021)
2021
-
[20]
& Rubio, A
Viñas Boström, E., Sriram, A., Claassen, M. & Rubio, A. Controlling the magnetic state of the proximate quantum spin liquid α-RuCl3 with an optical cavity. Npj Comput. Mater. 9, 1–10 (2023)
2023
-
[21]
N., Averitt, R
Basov, D. N., Averitt, R. D. & Hsieh, D. Towards properties on demand in quantum materials. Nat. Mater. 16, 1077–1088 (2017)
2017
-
[22]
Wilson, N. P. et al. Interlayer electronic coupling on demand in a 2D magnetic semiconductor. Nat. Mater. 20, 1657–1662 (2021)
2021
-
[23]
Jiang, S., Li, L., Wang, Z., Mak, K. F. & Shan, J. Controlling magnetism in 2D CrI3 by electrostatic doping. Nat. Nanotechnol. 13, 549–553 (2018)
2018
-
[24]
Lee, K. et al. Magnetic Order and Symmetry in the 2D Semiconductor CrSBr. Nano Lett. 21, 3511–3517 (2021)
2021
-
[25]
Ziebel, M. E. et al. CrSBr: An Air-Stable, Two-Dimensional Magnetic Semiconductor. Nano Lett. 24, 4319–4329 (2024)
2024
-
[26]
Rizzo, D. J. et al. Visualizing Atomically Layered Magnetism in CrSBr. Adv. Mater. 34, 2201000 (2022)
2022
-
[27]
Klein, J. et al. The Bulk van der Waals Layered Magnet CrSBr is a Quasi-1D Material. ACS Nano 17, 5316–5328 (2023)
2023
-
[28]
A., Demir, A
Nessi, L., Occhialini, C. A., Demir, A. K., Powella, L. & Comin, R. Magnetic field tunable polaritons in the ultra-strong coupling regime in CrSBr. Under Review. (2024)
2024
- [29]
-
[30]
Shi, J. et al. Giant Magneto-Exciton Coupling in 2D van der Waals CrSBr. arXiv.org https://arxiv.org/abs/2409.18437v1 (2024)
2024 arXiv
-
[31]
Barcons Ruiz, D. et al. Engineering high quality graphene superlattices via ion milled ultra-thin etching masks. Nat. Commun. 13, 6926 (2022)
2022
-
[32]
Demir, A. K. et al. Transferable Optical Enhancement Nanostructures by Gapless Stencil Lithography. Nano Lett. 24, 9882–9888 (2024)
2024
-
[33]
Wang, T. et al. Magnetically-dressed CrSBr exciton-polaritons in ultrastrong coupling regime. Nat. Commun. 14, 5966 (2023)
2023
-
[34]
Ruta, F. L. et al. Hyperbolic exciton polaritons in a van der Waals magnet. Nat. Commun. 14, 8261 (2023)
2023
-
[35]
Kodigala, A. et al. Lasing action from photonic bound states in continuum. Nature 541, 196–199 (2017)
2017
-
[36]
Hsu, C. W. et al. Observation of trapped light within the radiation continuum. Nature 499, 188– 191 (2013)
2013
-
[37]
& Joannopoulos, J
Fan, S. & Joannopoulos, J. D. Analysis of guided resonances in photonic crystal slabs. Phys. Rev. B 65, 235112 (2002)
2002
-
[38]
Cerjan, A. et al. Observation of bound states in the continuum embedded in symmetry bandgaps. Sci. Adv. 7, eabk1117 (2021)
2021
-
[39]
A., Cerjan, A
Vaidya, S., Benalcazar, W. A., Cerjan, A. & Rechtsman, M. C. Point-Defect-Localized Bound States in the Continuum in Photonic Crystals and Structured Fibers. Phys. Rev. Lett. 127, 023605 (2021)
2021
-
[40]
Zhang, T. et al. Magnetism and Optical Anisotropy in van der Waals Antiferromagnetic Insulator CrOCl. ACS Nano 13, 11353–11362 (2019)
2019
-
[41]
Guo, Q. et al. Ultrathin quantum light source with van der Waals NbOCl2 crystal. Nature 613, 53–59 (2023)
2023
-
[42]
Guo, Q. et al. Colossal in-plane optical anisotropy in a two-dimensional van der Waals crystal. Nat. Photonics 1–6 (2024) doi:10.1038/s41566-024-01501-3
2024 doi
-
[43]
Gu, P. et al. Photoluminescent Quantum Interference in a van der Waals Magnet Preserved by Symmetry Breaking. ACS Nano 14, 1003–1010 (2020)
2020
-
[44]
Tabataba-Vakili, F. et al. Doping-control of excitons and magnetism in few-layer CrSBr. Nat. Commun. 15, 4735 (2024)
2024
-
[45]
& Fan, S
Liu, V. & Fan, S. S4 : A free electromagnetic solver for layered periodic structures. Comput. Phys. Commun. 183, 2233–2244 (2012)
2012
Reviewed August 6, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.