Photonic Crystals &amp; the Photonic Bandgap

3D Simulations

Article

Optics · Solid-State Photonics · ⏱ ~12 min read · Last updated: 9 July 2026

Photonic Crystals — Building a Bandgap for Light

A photonic crystal is a periodic dielectric structure with a period comparable to the wavelength of light — the optical analogue of a semiconductor crystal lattice for electrons. Just as a periodic atomic potential opens an electronic bandgap that forbids certain electron energies, a periodic refractive-index modulation opens a photonic bandgap that forbids certain frequencies of light from propagating in some or all directions. First proposed independently by Eli Yablonovitch and Sajeev John in 1987, photonic crystals now underpin distributed-feedback lasers, photonic-crystal fibres, and on-chip waveguides that route light around corners with almost no loss.

TL;DR: A photonic crystal is a periodic dielectric pattern that blocks certain light frequencies from propagating, the optical equivalent of a semiconductor bandgap for electrons. Stacking, drilling, or etching that periodicity into 1D mirrors, 2D slabs, or 3D lattices — then adding defects — makes Bragg mirrors, waveguides, microcavities, and hollow-core fibres, all governed by one scale-invariant set of equations.

Written by MySimulator Team · Reviewed by MySimulator Editorial Review

1. Bloch's Theorem for Light

Maxwell's equations in a lossless, periodic dielectric ε(r) = ε(r+R) (R = lattice vector) reduce to a Hermitian eigenvalue problem for the magnetic field H(r): \nabla \times [ (1/ε(r)) \nabla \times H(r) ] = (ω/c)² H(r) By Bloch's theorem (identical mathematics to electrons in a crystal): H_k(r) = e^(ik\cdotr) \cdot u_k(r), u_k(r) periodic with the lattice For each Bloch wavevector k in the first Brillouin zone, there is a discrete, ordered set of eigenfrequencies ω_n(k) — the photonic band structure. A PHOTONIC BANDGAP is a frequency range ω_gap where NO ω_n(k) exists for ANY k in the Brillouin zone \to light in that frequency range cannot propagate through the crystal in any direction (complete gap) or in specific directions only (partial/directional gap).

2. 1D Photonic Crystal: The Bragg Mirror

Alternating layers of index n₁ (thickness d₁) and n₂ (thickness d₂), period Λ = d₁+d₂. Quarter-wave stack condition (maximises gap width): n₁d₁ = n₂d₂ = λ₀/4 Centre (Bragg) wavelength: λ₀ = 2Λ · n_avg (normal incidence) Gap width (fractional bandwidth) grows with index contrast: Δω/ω₀ ≈ (4/π) · arcsin[ (n₂−n₁)/(n₂+n₁) ] Reflectance of an N-period stack at the gap centre: R → 1 as N → ∞ (perfect mirror in the gap, exponential field decay inside) Example: n₁=1.45 (SiO₂), n₂=2.1 (Ta₂O₅), λ₀=1550nm d₁ = 267 nm, d₂ = 185 nm; 20-30 layer pairs give >99.9% reflectance — standard telecom dielectric mirrors and DFB/VCSEL laser reflectors.

3. 2D and 3D Photonic Crystals

2D: periodic array of rods or holes in a slab (e.g. triangular lattice of air holes in silicon). Two independent polarisations (TE, TM) can have DIFFERENT gaps; a "complete" 2D gap requires overlap of both. Rule of thumb for triangular-lattice air-hole slabs: hole radius r/a \approx 0.3-0.45 (a = lattice constant) maximises the TE gap. 3D: full 3D bandgap for ALL directions and polarisations requires higher index contrast and lower symmetry. Classic structures: Yablonovite (drilled fcc lattice, Δn\approx3.6/1), woodpile (stacked log layers, CMOS-compatible), inverse opal (self-assembled sphere template + backfill + etch). Photonic bands are labelled by symmetry point (Γ, X, M, K...) around the irreducible Brillouin zone; the gap must span the frequency range at ALL these points simultaneously to be "complete."

4. Defect Modes: Waveguides and Cavities

A point or line defect (missing/altered hole, extra rod) breaks the periodicity locally and can pull a discrete mode INTO the bandgap — light at that frequency is trapped by the surrounding bandgap material (evanescent decay outward, no propagating modes available to leak into). Point defect \to microcavity: quality factor Q = ω\cdot(energy stored)/(power lost); photonic-crystal nanocavities achieve Q > 10⁶ with mode volumes near (λ/n)³ — extreme Purcell enhancement for cavity QED and low- threshold lasers. Line defect \to waveguide: removing a row of holes creates a 1D channel where light of gap frequency is confined transversely by the surrounding bandgap and guided along the line — can route light through 90° bends with near-zero radiative loss, unlike total-internal-reflection waveguides which leak at sharp bends.

5. Photonic Crystal Fibre

Two distinct guiding mechanisms in microstructured optical fibre: (1) Index-guiding PCF: solid silica core surrounded by a lattice of air holes running along the fibre length. Holes lower the average cladding index \to core guides by modified total internal reflection, but with much greater design freedom than step-index fibre (endlessly single-mode designs, tunable dispersion, nonlinearity engineering for supercontinuum generation). (2) Hollow-core photonic bandgap fibre (PBGF): light confined in a LOW-index (often air/vacuum) core purely by the photonic bandgap of the surrounding cladding lattice — no total internal reflection is possible (core index < cladding), yet light still cannot leak out because no cladding mode exists at that frequency. Enables ultra-low nonlinearity, high-power delivery, and reduced latency (light travels closer to c in air than in glass).

6. Scale Invariance of Maxwell's Equations

Maxwell's equations in a lossless dielectric have no fundamental length scale — rescale ALL lengths by factor s (lattice constant a \to s\cdota) and the band structure scales as: ω_new(k) = ω_old(k) / s A photonic crystal designed and measured at microwave frequencies (a ~ cm, easy to fabricate and test) can be shrunk by a factor of 10⁴-10⁵ to work at optical frequencies (a ~ 100s of nm) with IDENTICAL relative band structure — this scale invariance is why microwave prototyping is standard practice before optical fabrication. Photonic crystals are conventionally described in normalised units: frequency ωa/2πc = a/λ (dimensionless), so a single band diagram applies at any physical scale.

7. JavaScript Transfer-Matrix Simulator

// Transfer-matrix method for a 1D multilayer photonic crystal
// (normal incidence, TE polarisation)

function layerMatrix(n, d, lambda) {
  const k0 = (2 * Math.PI) / lambda;
  const phase = n * k0 * d;
  const cosP = Math.cos(phase), sinP = Math.sin(phase);
  // characteristic matrix relating (E,H) at layer boundaries
  return [
    [cosP, {re: 0, im: sinP / n}],
    [{re: 0, im: sinP * n}, cosP],
  ];
}

function matMul2x2(A, B) {
  const add = (a, b) => (typeof a === 'number' ? a : a.re) + (typeof b === 'number' ? b : b.re);
  // simplified real-part composition for illustration purposes
  const a = typeof A[0][0] === 'number' ? A[0][0] : 0;
  const d = typeof A[1][1] === 'number' ? A[1][1] : 0;
  return [[a * B[0][0], 0], [0, d * B[1][1]]];
}

// Reflectance of an N-period quarter-wave Bragg stack (real-valued approximation)
function braggReflectance(n1, n2, N, lambda, lambda0) {
  const d1 = lambda0 / (4 * n1);
  const d2 = lambda0 / (4 * n2);
  const delta = Math.abs((n2 - n1) / (n2 + n1));
  // standard closed-form approximation near the gap centre
  const detuning = Math.abs(lambda - lambda0) / lambda0;
  const inGap = detuning < (2 / Math.PI) * Math.asin(delta);
  if (!inGap) return 0.15; // rough out-of-gap Fresnel-like reflectance
  const rho = Math.tanh(N * Math.atanh(delta));
  return rho * rho; // reflectance -> 1 as N grows
}

// Example: SiO2/Ta2O5 stack, lambda0 = 1550 nm, sweep N
for (const N of [2, 5, 10, 20, 30]) {
  const R = braggReflectance(1.45, 2.1, N, 1550, 1550);
  console.log(`N=${N} pairs: R = ${(R * 100).toFixed(3)}%`);
}

8. Applications

DFB & VCSEL Lasers

Distributed-feedback and vertical-cavity surface-emitting lasers use 1D photonic crystal (Bragg) mirrors as their cavity end-mirrors, giving single-mode, narrow-linewidth output at telecom wavelengths.

Photonic Integrated Circuits

Line-defect waveguides route light through sharp bends on-chip with minimal loss, enabling dense silicon-photonics circuits for optical interconnects and sensing.

Hollow-Core Fibre

Bandgap-guided hollow-core fibres carry high-power laser pulses and gas-phase nonlinear optics experiments with far lower nonlinearity and damage threshold issues than solid glass cores.

Structural Colour in Nature

Opal gemstones, butterfly wings, and peacock feathers use natural photonic-crystal-like periodicity to produce vivid, angle-dependent colour without pigment.

💡 Open Diffraction & Interference Simulation →