Pith. sign in

REVIEW 3 major objections 6 minor 37 references

Stealthy-Hyperuniform Wave Dynamics in Two-Dimensional Photonic Crystals

T0 review · 3 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

Pith's one-line read Photonic band linewidths directly measure scattering by stealthy-hyperuniform disorder, with a sharp transition at |k| = K/2 and residual scattering from the complex effective mass.

desk verdict A strong experimental-theory paper on linewidth as a probe of stealthy-hyperuniform scattering; the central K/2 transition holds up, and the main weakness is the unmeasured fabricated spectral density. read the letter →

arxiv 2507.05253 v1 pith:YRHOC6DO submitted 2025-07-07 physics.optics cond-mat.dis-nncond-mat.mes-hallcond-mat.soft

classification physics.opticscond-mat.dis-nncond-mat.mes-hallcond-mat.soft PACS 42.70.Qs42.25.Dd
keywords stealthyhyperuniformityphotoniccrystalslablinewidthdisorderscatteringboundstateinthecontinuumcomplexeffectivemassFourierfilteringmultiple
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Photonic crystal slabs whose hole radii carry stealthy-hyperuniform disorder should suppress single scattering for wavevectors whose iso-frequency contour lies entirely inside the momentum exclusion region. The paper uses the excess linewidth of a quadratic photonic band—the difference in linewidth between disordered and clean samples—as a direct, angle-resolved experimental measure of that scattering. It reports a sharp transition at $|\mathbf{k}| = K/2$, where $K$ is the exclusion-region cutoff, with steeply rising linewidth above and almost none below. It also finds that residual linewidth inside the stealthy region is not an artifact of imperfect stealthiness but an intrinsic non-Hermitian effect: the band's complex effective mass, caused by radiative loss out of the slab, allows scattering to states with different real energy. If correct, this makes photonic-crystal linewidths a quantitative probe of hyperuniform wave transport and identifies non-Hermiticity as essential to interpreting 'transparent' stealthy systems.

What carries the argument

The machinery is the leading-order self-energy of a quadratic band under a weak random potential whose spectral density $\tilde{\rho}(\mathbf{q})$ vanishes in the exclusion region $|\mathbf{q}|<K$. Its imaginary part, $\mathrm{Im}\,\Sigma_{\mathbf{k}} = (1/N)\sum_{\mathbf{q}} \mathrm{Im}[\tilde{\rho}(\mathbf{q})/(\mathrm{Re}\,E_{\mathbf{k}} - E_{\mathbf{k}+\mathbf{q}} + i0^+)]$, equals the excess linewidth and is evaluated analytically by integration over the iso-frequency contour; the contour geometry yields the $k=K/2$ transition. The second load-bearing element is the complex effective mass $m$ inherited from the symmetry-protected bound state in the continuum: the radiative decay rate $O(k^2)$ appears as $\mathrm{Im}(1/2m)$, which converts the infinitely thin iso-frequency contour into a ring of finite $\mathbf{k}$-space width, producing the non-Hermitian residual scattering. Period-tripling of the disorder pattern (correlating hole radii over $3\times 3$ plaquettes, $b=3a$) enlarges the effective stealthiness parameter and creates nine exclusion circles that produce a second transition at $k_x = \pi/(3a) - K/2$ along $k_y=0$.

What would settle it

Measure the excess linewidth inside the stealthy region ($|\mathbf{k}|<K/2$) for a slab whose band is purely guided below the light line, where radiative loss and hence $\mathrm{Im}(m)$ vanish; if a finite residual excess linewidth remains, the non-Hermitian explanation is wrong. Alternatively, directly characterize the dielectric profile of a fabricated sample and compute $\tilde{\rho}(\mathbf{q})$ to verify that the exclusion region is genuinely empty.

Watch

Extended reading notes

Core claim

The central claim is that the transition between stealthy and non-stealthy scattering behavior is directly observable in the linewidths of photonic bands of large silicon slab samples, and that the transition condition is geometric: single scattering is forbidden below $|\mathbf{k}| = K/2$ because no allowed scattering vector can connect two points on the same iso-frequency contour while staying inside the excluded spectral-density region. Above that threshold the excess linewidth grows from zero as an arccos function of $K/2|\mathbf{k}|$, in quantitative agreement with a leading-order self-energy calculation. The same measurements show that the excess linewidth in the stealthy regime is finite and proportional to the imaginary part of the effective mass times the square of the disorder, establishing that out-of-plane radiative loss—through a symmetry-protected bound state in the continuum at the band tip—qualitatively changes the scattering phase diagram: it broadens the iso-frequency contour in $\mathbf{k}$-space and permits scattering that a Hermitian theory forbids. Higher-order disorder scaling reveals additional multiple-scattering contributions in the stealthy regime.

Load-bearing premise

The load-bearing assumption is that a hole-radius change acts as a weak local potential that shifts only the band-edge energy while leaving the quadratic dispersion and the complex effective mass fixed, and that the fabricated dielectric pattern is effectively stealthy because its in-region spectral density is below $10^{-2}$ of the out-of-region value, as estimated by simulation rather than direct measurement of the fabricated samples.

Editorial extensions

If this is right

  • Excess linewidth can serve as a general experimental observable for scattering by correlated disorder in photonic systems, complementing direct transmission measurements.
  • The transition at $|\mathbf{k}| = K/2$ is a direct spectroscopic signature of stealthy-hyperuniformity: below it, leading-order single scattering is suppressed; above it, scattering grows sharply.
  • Residual transparency in the stealthy regime is set by non-Hermitian radiative loss, not by fabrication imperfections, so suppressing out-of-plane loss would sharpen the transition and reduce residual linewidth.
  • Multiple scattering sets in as disorder increases even inside the stealthy region, meaning the suppression of scattering holds only to leading order.
  • Period-tripling provides a practical way to reach large stealthiness parameters at small accessible wavevectors, with a predictable second transition from the nine-fold exclusion geometry.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the linewidth probe is as direct as claimed, the same method could be inverted: angle-resolved excess-linewidth data could reconstruct the spectral density $\tilde{\rho}(\mathbf{q})$ of other correlated-disorder patterns, effectively imaging the scattering kernel.
  • The complex-mass scattering mechanism is generic for any open wave system with a BIC-pinned quadratic band, so similar residual scattering should appear in acoustic, plasmonic, or polaritonic analogs with radiative loss.
  • A multiple-scattering theory beyond $\tilde{\rho}^2$ is the clear next step; linewidth measurements already show the need for it at $w_0>0.2$.
  • The fixed-effective-mass assumption could be relaxed by letting disorder renormalize $\mathrm{Im}(m)$; this likely explains part of the quantitative mismatch at the largest disorder values.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper reports experiments on large-area silicon photonic crystal slabs in which stealthy-hyperuniform disorder is imposed by Fourier-filtered, site-dependent hole-radius variations on a square lattice. Using angle-resolved reflection spectroscopy, the authors extract the linewidths of a quadratic photonic band and define an excess linewidth relative to a clean periodic sample. They observe a sharp increase in excess linewidth at |k| = K/2, which they identify as the transition from the stealthy to the non-stealthy scattering regime. They further report a finite excess linewidth inside the stealthy region, which they attribute to the imaginary part of the complex effective mass arising from out-of-plane radiative loss, and a super-quadratic disorder dependence, which they attribute to multiple scattering. The theoretical framework is a leading-order self-energy expression using the spectral density of the imposed potential, evaluated for Hermitian and non-Hermitian quadratic bands, with closed-form analytical results verified against numerical summation and compared with the experimental phase diagrams.

Significance. If the central interpretation holds, this is a valuable experimental advance: it uses photonic linewidths as a direct probe of the scattering kernel in stealthy-hyperuniform media, provides an analytic formula for the K/2 transition, and demonstrates a non-Hermitian contribution to disorder scattering through a parameter-free prediction based on the clean-sample imaginary effective mass. The large system sizes (10^6-10^7 sites), the closed-form self-energy results, and the explicit quantitative comparison between theory and experiment are notable strengths. The manuscript is also unusually candid about its assumptions, particularly in Supplemental Section 2, where the authors state that the photonic crystal is only 'effectively stealthy' and quantify the leakage by a simulated estimate. However, the load-bearing premise -- that the fabricated samples have negligible spectral density inside the exclusion circle -- is never directly measured, and this same unmeasured quantity competes with both the non-Hermitian and multiple-scattering attributions.

major comments (3)
  1. [Supplemental Section 2, Eqs. (S34)-(S38) and Fig. S3] The premise that the fabricated samples are effectively stealthy is not directly tested. The only support for the in-stealthy-region leakage factor F <= 10^-2 is a calculation on the designed dielectric profile that assumes independent circular holes and a linear radius-to-potential mapping (Eqs. S34-S37), not a measurement of the fabricated samples. Etch bias, sidewall roughness, and radius errors (acknowledged in Fig. 4 for w0 < 0.05) can only add spectral weight inside the exclusion circle. Because the transition at |k| = K/2, the non-Hermitian residual, and the multiple-scattering term are all interpreted through this scattering kernel, the paper should either measure the spectral density of the actual fabricated patterns (for example, from SEM images) or demonstrate robustness of the conclusions to F at fabrication-realistic levels. As it stands, the central 'effectively stealthy' premise is an unverified assumption.
  2. [Fig. 4 and Supplemental Eq. (S51)] The attribution of the O(w0^4) excess linewidth in the stealthy regime to multiple scattering is not uniquely supported. The authors' own estimate in Supplemental Section 2 gives F proportional to w0^2, and Eq. (S51) for k < K/2 yields an excess linewidth proportional to F * w0^2, i.e., O(w0^4) -- the same scaling as the fitted higher-order term in Fig. 5(c). Thus the observed super-quadratic growth is equally consistent with residual non-stealthy single scattering from imperfect stealthiness. The paper needs a diagnostic that separates these two mechanisms, such as measuring F for each fabricated sample or comparing samples with intentionally different F, before the multiple-scattering interpretation can be accepted.
  3. [Main text Fig. 5 and Supplemental Eq. (S51)] The headline claim that residual single scattering inside the stealthy region is an intrinsically non-Hermitian effect is not uniquely established by the data. A Hermitian quadratic band with a non-zero stealthiness leakage F already produces a finite, roughly k-independent excess linewidth for k < K/2 (Eq. S51), with magnitude proportional to F. Therefore the observed finite excess linewidth in the stealthy region can be explained without any complex effective mass if F is large enough. The quantitative match in Fig. 5(b) is encouraging, but it relies on the simulated F ~ 10^-3; without a fabricated-sample measurement of F or a control experiment, the non-Hermitian attribution is not fully separated from imperfect stealthiness.
minor comments (6)
  1. [Main text Figs. 3 and 4] The color maps for the excess linewidth in Figs. 3 and 4 are presented without colorbars or explicit units; adding a shared colorbar would make the phase diagrams substantially easier to interpret.
  2. [Main text Eq. (2)] The sum over q in Eq. (2) would benefit from stating explicitly that it runs over the first Brillouin zone and that the spectral density is the disorder-averaged quantity defined in Supplemental Section 1.
  3. [Main text Fig. 2(f)] The caption of Fig. 2(f) states that arrows marked with an x denote forbidden scattering events, but no x marks appear in the panel; please clarify the caption or the figure.
  4. [Supplemental Section 6, Fig. S12(e)] The quadratic fit used to extract Im(1/2m) and Im(E0) is described only by its resulting values; reporting the number of fitted k points and a residual or goodness-of-fit measure would strengthen confidence in these extracted parameters.
  5. [Supplemental Section 5, Fig. S9(e)] The second transition at kx = pi/(3a) - K/2 is demonstrated with a single sample and only in the Supplemental Material; the main text should either mention this second transition explicitly or point the reader to the supplement more prominently.
  6. [General] The manuscript does not include a data availability statement; given the large experimental dataset and the centrality of the extracted linewidths, a statement about data and code availability would be helpful.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's predictions are forward calculations from the designed disorder spectrum and clean-sample band parameters, compared against independently measured disordered-sample linewidths.

full rationale

The central derivation chain is self-contained: the paper imposes a stealthy-hyperuniform spectral density by Fourier filtering (zero inside |q|<K), computes the single-scattering self-energy from that spectral density and from clean periodic-sample band parameters (E0 and complex effective mass), and then compares the resulting excess-linewidth predictions to measurements on separately fabricated disordered samples. The K/2 transition follows mathematically from the imposed zero spectral density, but the measured linewidth is an independent observable, so the agreement is a genuine experimental check rather than an identity with the input. The non-Hermitian residual-scattering prediction in the stealthy regime uses the complex effective mass extracted from the clean lattice only, and the disordered-sample data are not used in that prediction; the comparison is therefore not a fit masquerading as a prediction. The only explicitly fitted term is the O(w0^4) multiple-scattering correction in Fig. 5(c), which is labeled as fitted and is not the paper's central prediction. Self-citations (e.g., Ref. [16]) are contextual and not load-bearing for the main claims. Concerns about the fabricated spectral density not being directly measured, or about the radius-to-potential linearization, are validity and systematic-error risks rather than circular reductions: the paper itself quantifies the higher-order leakage with an estimated F factor and shows it is small in the relevant parameter range. Therefore no step in the derivation chain reduces, by definition or by fit, to its own inputs.

Assumptions & free parameters 7 free parameters · 5 assumptions · 0 invented entities

The prediction uses measured clean-band parameters (E0 and 1/(2m) with real and imaginary parts) plus the simulated calibration V0 and the imposed K and w0. These are external inputs, not fits to the disordered-sample excess linewidth, except for the O(w0^4) term. No new physical entities are invented; the complex effective mass is a parametrization of measured radiative loss.

free parameters (7)
  • Real part of inverse effective mass Re(1/(2m)) = 0.579 (experimental clean sample)
    Extracted from quadratic fit to the clean periodic sample band dispersion; used as input to the disorder theory.
  • Imaginary part of inverse effective mass Im(1/(2m)) = 0.121 (experimental clean sample)
    Extracted from the quadratic fit to the clean-sample linewidth; controls the non-Hermitian residual scattering in the stealthy region.
  • Band tip energy E0 = 7.11 a^-2 (experimental clean sample, real part)
    Fitted to the clean periodic band structure; enters the self-energy and excess linewidth formulas.
  • Imaginary part of tip energy Im(E0) = -0.0075 a^-2
    Fitted to the clean-sample linewidth offset; accounts for uniform absorption loss.
  • Potential calibration V0 = 2.03 a^-2
    Derived from Legume simulations of band edge energy versus hole radius; converts hole-radius disorder into a local potential.
  • O(w0^4) multiple-scattering coefficient = not given in text
    Added to the leading-order theory in Fig. 5(c) to match the measured disorder dependence; no microscopic value is derived.
  • Fano fit parameters (F0, c0, c1, c2, A, phi, omega0, gamma) = fit per spectrum
    Eight free parameters used to extract the linewidth and resonance position from each reflection spectrum; they are data-reduction parameters rather than physics parameters.
assumptions (5)
  • domain assumption The photonic band near Gamma is isotropic and quadratic over the measured wavevector range.
    Used in Eq. (1) and throughout Supplemental Section 3; the band deviates near the Brillouin zone edge, but only small k is probed.
  • domain assumption Disorder is weak enough for the first-order Born approximation (single-scattering self-energy, Eq. (2)/S7) to dominate at w0 = 0.2.
    The central predictions use this expansion; higher-order terms are invoked only qualitatively for larger disorder.
  • ad hoc to paper Hole-radius changes act as a site-local scalar potential proportional to E0 shift, with fixed effective mass.
    Established by a linear fit in Fig. S2(d); the radius dependence of the effective mass is not included in the model.
  • domain assumption The photonic crystal is effectively stealthy because the in-stealthy-region spectral density F is below about 10^-2.
    Quantified by simulation in Fig. S3, not measured on the fabricated samples; the effect of nonzero F is estimated to be smaller than 0.08 nm in linewidth.
  • domain assumption The clean periodic sample provides the intrinsic linewidth baseline that is subtracted from disordered samples.
    The excess linewidth definition assumes the fabrication disorder in the clean sample is negligible or identical in both samples.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Stealthy-Hyperuniform Wave Dynamics in Two-Dimensional Photonic Crystals." pith.science (2026). https://pith.science/paper/YRHOC6DO

@misc{pith2026250705253,
  author       = {Pith},
  title        = {Pith review of: Stealthy-Hyperuniform Wave Dynamics in Two-Dimensional Photonic Crystals},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YRHOC6DO}},
  note         = {Machine review of arXiv:2507.05253}
}
abstract

Hyperuniform structures are spatial patterns whose fluctuations disappear on long length scales, making them effectively homogeneous when observed from afar. Mathematically, this means that their spectral density, $\tilde{\rho}({\bf k})$, approaches zero for low wavenumber, $|\textbf{k}|$. Crystalline lattices are hyperuniform, as are certain quasicrystals, maximally random jammed packing of spheres, and electrons in the fractional quantum Hall state. Stealthy-hyperuniformity is an even stronger constraint on the spectral density: it requires that $\tilde{\rho}({\bf k})$ is strictly zero in a finite range of wavevectors around $\mathbf{k}=\mathbf{0}$, called the stealthy regime, or exclusion region. Since the degree of scattering by disorder is, to leading order, proportional to $\tilde{\rho}({\bf k})$, waves propagating through such structures may do so without scattering for sufficiently long wavelengths and short distances. Here, we measure scattering by disorder in photonic crystal slabs with stealthy-hyperuniform disorder by measuring the linewidths of the photonic bands. We observe the transition between the stealthy and non-stealthy regimes, marked by a sharp increase in linewidth. We also observe the effects of multiple scattering in the stealthy regime, which implies diminishing transparency. Moreover, we show that residual single scattering in the stealthy regime arises from an intrinsically non-Hermitian effect: propagating light has a complex effective mass due to radiative loss out of the slab.

Figures

Figures reproduced from arXiv: 2507.05253 by the authors.

Figure 1
Figure 1. Properties of the clean (i.e., non-disordered) periodic photonic crystal slab. (a) SEM image of photonic crystal slab [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Description of the stealthy-hyperuniform photonic crystal. (a) Schematic diagram of disorder manifested by varying [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Dependence of excess linewidth on cutoff wavenumber [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Dependence of excess linewidth on disorder [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Non-Hermitian effects on excess linewidth due to scattering by the disorder. Experimental (blue dots) and theoretical [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

37 extracted references · 37 canonical work pages

  1. [1]

    intrinsic loss

    due to higher- order scattering. Upon incorporating a fitted O(w4 0) term (next order in the scattering expansion, scaling as 5 O(˜ρ2)) within the leading-order theory, we again observe very good quantitative agreement between experimental data (blue dots) and theory (black line). In conclusion, we used large silicon photonic crystal slabs exhibiting stea...

  2. [2]

    The transition point still occurs at k = K 2 in the case of complex effective mass and band tip energy

  3. [3]

    Compared to Eq. (S48), the excess linewidth before the transition k < K 2 is no longer zero, but a finite value proportional to mi, which clearly shows that the origin of the non-Hermitian effect is the imaginary part of the effective mass

  4. [4]

    (S58) 25

    In the stealthy region, the k dependence in ∆ λk is not very sensitive, so it can be estimated by ∆ λk|k=0: ∆λk|k=0 = − 2w2 0V 2 0 a2 3 1 − π K a 2π 2 E 3 2 r mi · ln qmax K . (S58) 25

  5. [5]

    thickness

    With the presence of non-Hermiticity, the transition at k = K 2 becomes smoother compared to the Hermitian case. The transition in the numerical summation without approximation is even smoother than the non-Hermitian theory. kx ky E (a) -0.5 0.0 0.5 kx [2 a 1] -0.5 0.0 0.5 ky [2 a 1] K (b) -0.5 0.0 0.5 kx [2 a 1] -0.5 0.0 0.5 ky [2 a 1] K (c) Figure S7. T...

  6. [6]

    Compared to the result without period tripling in Eq

    The first excess linewidth transition occurs at kx = K 2 , but the stealthiness parameter increases to χ = π K 3a 2π 2 . Compared to the result without period tripling in Eq. (S25), we successfully obtain a small transition kx (which means small cutoff wavenumber K) with large stealthiness parameter χ

  7. [7]

    The second transition is induced due to the fact that ˜ ρ(q) and Ek have different periodicity

    Other than the first transition at kx = K 2 , we observe a second transition at kx = π 3a − K 2 . The second transition is induced due to the fact that ˜ ρ(q) and Ek have different periodicity. The second transition is generally smoother than the first transition, but is still possible to be observed in our experiment

  8. [8]

    (S69) Compared to Eq

    In the stealthy region, ∆ λ(kx,0) is not very sensitive to kx, so we can use the ∆ λ(kx,0) kx=0 to estimate the excess linewidth: ∆λ(kx,0) kx=0 = − 2w2 0V 2 0 a2 3 1 − π K 3a 2π 2 E 3 2 r mi ·  ln qmax K − ln   1 1 − K 3a 2π 2 2 1 − 1 2 K 3a 2π 2 2     . (S69) Compared to Eq. (S58) in the non-period-tripled case, the extra term of − ln 1 1−(K ...

Show all 37 references
  1. [9]

    & Stillinger, F

    Torquato, S. & Stillinger, F. H. Local density fluctuations, hyperuniformity, and order metrics. Physical Review E 68, 041113 (2003)

  2. [10]

    Hyperuniform states of matter

    Torquato, S. Hyperuniform states of matter. Physics Reports 745, 1–95 (2018)

  3. [11]

    & Carminati, R

    Leseur, O., Pierrat, R. & Carminati, R. High-density hyperuniform materials can be transparent. Optica 3, 763–767 (2016)

  4. [12]

    & Romero-Garc ´ ıa, V

    Ch´ eron,´E., F´ elix, S., Groby, J.-P., Pagneux, V. & Romero-Garc ´ ıa, V. Wave transport in stealth hyperuniform materials: The diffusive regime and beyond. Applied Physics Letters121 (2022)

  5. [13]

    & Torquato, S

    Kim, J. & Torquato, S. Theoretical prediction of the effective dynamic dielectric constant of disordered hyperuniform anisotropic composites beyond the long-wavelength regime. Optical Materials Express14, 194–215 (2023)

  6. [14]

    & Torquato, S

    Kim, J. & Torquato, S. Effective electromagnetic wave properties of disordered stealthy hyperuniform layered media beyond the quasistatic regime. Optica 10, 965–972 (2023)

  7. [15]

    & Steinhardt, P

    Florescu, M., Torquato, S. & Steinhardt, P. J. Designer disordered materials with large, complete photonic band gaps. Proceedings of the National Academy of Sciences106, 20658–20663 (2009)

  8. [16]

    Man, W. et al. Isotropic band gaps and freeform waveguides observed in hyperuniform disordered photonic solids. Pro- ceedings of the National Academy of Sciences110, 15886–15891 (2013)

  9. [17]

    & Scheffold, F

    Muller, N., Haberko, J., Marichy, C. & Scheffold, F. Silicon hyperuniform disordered photonic materials with a pronounced gap in the shortwave infrared. Advanced Optical Materials2, 115–119 (2013)

  10. [18]

    Froufe-P´ erez, L. S. et al. Role of short-range order and hyperuniformity in the formation of band gaps in disordered photonic materials. Physical review letters117, 053902 (2016)

  11. [19]

    & Florescu, M

    Gkantzounis, G., Amoah, T. & Florescu, M. Hyperuniform disordered phononic structures. Physical Review B95, 094120 (2017)

  12. [20]

    Aubry, G. J. et al. Experimental tuning of transport regimes in hyperuniform disordered photonic materials. Physical review letters 125, 127402 (2020)

  13. [21]

    A., Steinhardt, P

    Klatt, M. A., Steinhardt, P. J. & Torquato, S. Wave propagation and band tails of two-dimensional disordered systems in the thermodynamic limit. Proceedings of the National Academy of Sciences119, e2213633119 (2022)

  14. [22]

    Vynck, K. et al. Light in correlated disordered media. Reviews of Modern Physics95, 045003 (2023)

  15. [23]

    Siedentop, L. et al. Stealthy and hyperuniform isotropic photonic band gap structure in 3d. PNAS nexus 3, pgae383 (2024)

  16. [24]

    F., Gopalakrishnan, S

    Karcher, J. F., Gopalakrishnan, S. & Rechtsman, M. C. Effect of hyperuniform disorder on band gaps. Physical Review B 110, 174205 (2024)

  17. [25]

    Dikopoltsev, A. et al. Observation of anderson localization beyond the spectrum of the disorder. Science Advances 8, eabn7769 (2022)

  18. [26]

    & Goetschy, A

    Monsarrat, R., Pierrat, R., Tourin, A. & Goetschy, A. Pseudo-gap and localization of light in correlated disordered media. arXiv preprint arXiv:2110.12034(2021)

  19. [27]

    & Dal Negro, L

    Sgrignuoli, F., Torquato, S. & Dal Negro, L. Subdiffusive wave transport and weak localization transition in three- dimensional stealthy hyperuniform disordered systems. Physical Review B105, 064204 (2022)

  20. [28]

    K., Kim, J., Steinhardt, P

    Morse, P. K., Kim, J., Steinhardt, P. J. & Torquato, S. Generating large disordered stealthy hyperuniform systems with ultrahigh accuracy to determine their physical properties. Physical Review Research5, 033190 (2023)

  21. [29]

    See Attached Supplemental Information Section

  22. [30]

    Minkov, M. et al. Inverse design of photonic crystals through automatic differentiation. ACS Photonics 7, 1729–1741 (2020)

  23. [31]

    W., Zhen, B., Stone, A

    Hsu, C. W., Zhen, B., Stone, A. D., Joannopoulos, J. D. & Soljaˇ ci´ c, M. Bound states in the continuum. Nature Reviews Materials 1, 1–13 (2016)

  24. [32]

    & Kivshar, Y

    Koshelev, K., Lepeshov, S., Liu, M., Bogdanov, A. & Kivshar, Y. Asymmetric metasurfaces with high-q resonances governed by bound states in the continuum. Physical review letters121, 193903 (2018)

  25. [33]

    A., Havlin, S., Schwartz, M

    Makse, H. A., Havlin, S., Schwartz, M. & Stanley, H. E. Method for generating long-range correlations for large systems. Physical Review E53, 5445 (1996)

  26. [34]

    D., Johnson, S

    Joannopoulos, J. D., Johnson, S. G., Winn, J. N. & Meade, R. D. Photonic Crystals: Molding the Flow of Light (Second Edition) (Princeton University Press, 2008), 2 edn

  27. [35]

    Lee, P. A. & Ramakrishnan, T. V. Disordered electronic systems. Reviews of modern physics57, 287 (1985)

  28. [36]

    & Joannopoulos, J

    Fan, S. & Joannopoulos, J. D. Analysis of guided resonances in photonic crystal slabs. Physical Review B 65, 235112 (2002)

  29. [37]

    & Wankerl, H

    Luce, A., Mahdavi, A., Marquardt, F. & Wankerl, H. Tmm-fast, a transfer matrix computation package for multilayer thin-film optimization: tutorial. J. Opt. Soc. Am. A39, 1007–1013 (2022)

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.