Table of Contents
Fetching ...

Universality and anisotropy of the Photonic Urbach Tail

M. Menéndez, Lan Hoang Mai, Nazifa Tasnim Arony, Henry Carfagno, Lauren N. McCabe, Joshua M. O. Zide, Cefe López, Matthew F. Doty, P. D. García

TL;DR

This work demonstrates that near-band-edge states in disordered photonic-crystal waveguides follow a universal Urbach tail, characterized by the cumulative form $F(\Delta)=\exp[-(\Delta/\alpha)^{\beta}]$ with $\beta \approx 1$ regardless of disorder strength or orientation. By combining controlled anisotropic disorder in GaAs photonic-crystal waveguides with full-vector simulations and careful fitting, the authors reveal a directional Urbach energy: $\alpha_{\parallel}$ remains nearly constant while $\alpha_{\perp}$ grows with disorder, indicating anisotropic disorder–mode coupling. The Urbach energy thus serves as a practical metric to quantify and compare anisotropic disorder, with broad implications for characterizing disorder in photonic devices and potentially in other wave systems such as cold-atom lattices and acoustic metamaterials. Together, these results establish Urbach physics as a robust framework for exponential near-edge tails and provide a directional diagnostic for disorder effects in structured photonic media.

Abstract

Disorder in photonic crystals and waveguides creates states inside the photonic band gap. These states are often described as Lifshitz tails despite exhibiting energy distributions inconsistent with Lifshitz statistics near the band edge. Here we show that in photonic-crystal waveguides with intentionally engineered anisotropic disorder, the band-edge tail accessible experimentally follows an Urbach law universally, with cumulative statistics $F(Δ)=\exp[-(Δ/α)^β]$, where $Δ$ is the spectral detuning from the band edge, and an exponent $β\approx 1$ independent of disorder strength and orientation. In contrast to Lifshitz behavior, the density of states is maximal at the band edge and decays into the gap. Crucially, we find that the Urbach energy $α$ is anisotropic, with a pronounced directional splitting and qualitatively different scaling for disorder parallel and perpendicular to the waveguide axis. These conclusions are supported by quantitative agreement between optical measurements of GaAs photonic-crystal waveguides and full-vector simulations. The anisotropic Urbach energy emerges as a sensitive probe of disorder-mode coupling and a practical metric to characterize structural disorder in photonic devices.

Universality and anisotropy of the Photonic Urbach Tail

TL;DR

This work demonstrates that near-band-edge states in disordered photonic-crystal waveguides follow a universal Urbach tail, characterized by the cumulative form with regardless of disorder strength or orientation. By combining controlled anisotropic disorder in GaAs photonic-crystal waveguides with full-vector simulations and careful fitting, the authors reveal a directional Urbach energy: remains nearly constant while grows with disorder, indicating anisotropic disorder–mode coupling. The Urbach energy thus serves as a practical metric to quantify and compare anisotropic disorder, with broad implications for characterizing disorder in photonic devices and potentially in other wave systems such as cold-atom lattices and acoustic metamaterials. Together, these results establish Urbach physics as a robust framework for exponential near-edge tails and provide a directional diagnostic for disorder effects in structured photonic media.

Abstract

Disorder in photonic crystals and waveguides creates states inside the photonic band gap. These states are often described as Lifshitz tails despite exhibiting energy distributions inconsistent with Lifshitz statistics near the band edge. Here we show that in photonic-crystal waveguides with intentionally engineered anisotropic disorder, the band-edge tail accessible experimentally follows an Urbach law universally, with cumulative statistics , where is the spectral detuning from the band edge, and an exponent independent of disorder strength and orientation. In contrast to Lifshitz behavior, the density of states is maximal at the band edge and decays into the gap. Crucially, we find that the Urbach energy is anisotropic, with a pronounced directional splitting and qualitatively different scaling for disorder parallel and perpendicular to the waveguide axis. These conclusions are supported by quantitative agreement between optical measurements of GaAs photonic-crystal waveguides and full-vector simulations. The anisotropic Urbach energy emerges as a sensitive probe of disorder-mode coupling and a practical metric to characterize structural disorder in photonic devices.
Paper Structure (15 sections, 3 equations, 10 figures, 2 tables)

This paper contains 15 sections, 3 equations, 10 figures, 2 tables.

Figures (10)

  • Figure 1: Photonic-crystal waveguides with controlled anisotropic disorder. (a) Waveguide geometry (triangular lattice of air holes with $a = 500$ nm the lattice constant) with hole displacements along $\parallel$ and $\perp$ directions, and the intensity profile of an Anderson-localized mode. (b) Simulated spectrum showing Fabry--Perot-like modes and cutoff frequency at $189.5~\text{THz}$ (dashed line) when disorder is introduced in the $\parallel$ direction ($\sigma_\parallel = 0.01a$). (c,d) SEM image and positional disorder map for $\parallel$ disorder ($\sigma_\parallel = 0.03a$). (e,f) SEM image and positional disorder map for $\perp$ disorder ($\sigma_\perp = 0.03a$). Color scale indicates hole displacement from ideal lattice position. Strong contrast in (d) and (f) confirms disorder is predominantly along the intended direction.
  • Figure 2: Experimental observation of Urbach tails. (a),(b) Representative optical reflection spectra from GaAs photonic-crystal waveguides with intentional disorder along $\parallel$ (a) and $\perp$ (b) at $\sigma \approx 0.03a$. Each curve shows an individual Anderson-localized resonance detected via evanescent fiber coupling. (c)--(f) Eigenfrequency histograms for all detected localized modes; shaded regions indicate the band gap. At weak disorder $\sigma = 0.03a$ (c,d), $\parallel$ and $\perp$ disorder distributions are similar due to comparable intrinsic isotropic disorder ($\sigma \simeq 0.007a$). At stronger disorder $\sigma = 0.05a$ (e,f), clear anisotropy emerges: the $\perp$ disorder tail extends $\sim 1$ THz deeper into the gap, confirming $\alpha_\perp > \alpha_\parallel$.
  • Figure 3: Anisotropy of the Urbach tail. (a,b) Urbach fits to experimental data: $\ln F$ versus $\Delta$ for $\parallel$ disorder (a, teal) and $\perp$ disorder (b, orange) at $\sigma = 0.03a$. Black lines show fits to $F(\Delta) = \exp[-(\Delta/\alpha)^\beta]$. Steeper decay for $\perp$ disorder confirms $\alpha_\parallel < \alpha_\perp$. (c) Experimental Urbach exponent $\beta$ versus disorder amplitude for both directions, showing $\beta \approx 1$ across all cases, confirming universal Urbach behavior. (d) Experimental Urbach energy $\alpha$ versus disorder amplitude. $\alpha_\parallel$ (teal) remains nearly flat while $\alpha_\perp$ (orange) increases linearly with $\sigma_{\perp}$, in quantitative agreement with simulations (Fig. \ref{['fig4']}f).
  • Figure 4: Numerical analysis of the Urbach tails. (a),(b) Ensemble-averaged spectra for $\parallel$ (a, teal) and $\perp$ (b, orange) disorder at $\sigma = 0.03a$, showing sub-edge states below the band edge at $\omega_c = 189.5~\text{THz}$. The faint orange overlay curve in (a) shows the perpendicular disorder direction for comparison; the $\perp$ disorder tail extends deeper into the gap. (c),(d) Urbach analysis: $\ln F(\Delta)$ versus detuning $\Delta$ for $\parallel$ disorder (c, teal) and $\perp$ disorder (d, orange), with $F$ the cumulative DOS. The almost linear decay confirms exponential band-edge tails; the steeper slope for $\perp$ disorder implies $\alpha_\parallel < \alpha_\perp$. (e) Urbach exponent $\beta$ versus disorder amplitude, with $\beta \approx 1$ for all $\sigma$ and both directions (dashed line indicates $\beta = 1$), consistent with Urbach-type near-edge behavior. (f) Urbach energy $\alpha$ versus disorder amplitude. $\alpha_\parallel$ (teal) is nearly constant, while $\alpha_\perp$ (orange) grows with $\sigma_{\perp}$, revealing directional sensitivity.
  • Figure S1: Comparison between the qualitative behavior of (a) a Lifshitz tail and (b) an Urbach tail. Panel (a) illustrates a one-dimensional Lifshitz form $\rho(\Delta)\propto \exp[-A\,\Delta^{-1/2}]$, where the exponent varies as an inverse power of the detuning and the density of states therefore increases as one moves deeper into the gap. This captures the key qualitative feature of the Lifshitz regime predicted for uncorrelated disorder, whose general $d$-dimensional form is $\rho(\Delta)\sim\exp[-A\,\Delta^{-d/2}]$. Panel (b) shows an Urbach tail $\rho(\Delta)\propto\exp(-\Delta/E_U)$, where the exponent is linear in $\Delta$ and the DOS decreases into the gap. The purpose of the figure is to highlight the opposite monotonicity of Lifshitz and Urbach behavior. Only the Urbach regime is observed in the band-edge spectra of photonic-crystal waveguides, consistent with the correlated-disorder theory of John et al.john1986.
  • ...and 5 more figures