Photonica

Four-wave mixing (FWM)

A third-order nonlinear process in which three input optical waves generate a fourth wave at a new frequency. The basis of wavelength conversion, parametric amplification, and a source of inter-channel crosstalk in WDM systems.

Four-wave mixing is the process by which three input photons interact through the χ(3)\chi^{(3)} nonlinearity to produce a fourth photon, conserving energy and momentum:

ω4  =  ω1+ω2ω3,k4  =  k1+k2k3.\omega_4 \;=\; \omega_1 + \omega_2 - \omega_3, \qquad \vec{k}_4 \;=\; \vec{k}_1 + \vec{k}_2 - \vec{k}_3.

The amplitude of the generated wave scales as the product of the three input amplitudes — quadratically with intensity for degenerate cases (ω1=ω2\omega_1 = \omega_2).

Phase matching is essential for efficient FWM. The momentum conservation condition translates to a refractive-index matching requirement:

n4ω4  =  n1ω1+n2ω2n3ω3.n_4 \omega_4 \;=\; n_1 \omega_1 + n_2 \omega_2 - n_3 \omega_3.

In a dispersive medium, this is satisfied only over a narrow frequency span. FWM efficiency drops off rapidly when phase matching is lost — making FWM both useful (for selective wavelength conversion) and tolerable as a system penalty (where dispersion limits inter-channel mixing in WDM systems).

Applications.

ApplicationConfiguration
Wavelength conversionPump at ωp\omega_p and signal at ωs\omega_s produce idler at 2ωpωs2\omega_p - \omega_s — converts signal to a different channel
Parametric amplificationStrong pump and weak signal produce idler; signal can be amplified (gain >1> 1) with proper phase matching
Phase-conjugationParticular FWM geometry produces a phase-reversed copy of the input — used for distortion compensation
Squeezed-light generationSub-shot-noise quantum states for quantum-enhanced metrology and communication
Frequency comb generationCascaded FWM in microresonators generates octave-spanning combs (see Kerr effect)
WDM inter-channel crosstalkAdjacent channels mix to produce spurious tones at neighboring frequencies — system impairment

Phase-matching in fiber. In single-mode fiber, FWM phase matching requires near-zero dispersion at the operating wavelength. Standard SMF-28 at 1550 nm has D=17D = 17 ps/(nm·km), strongly mismatching FWM and suppressing inter-channel mixing in DWDM systems. Dispersion-shifted fiber (DSF) with zero dispersion at 1550 nm was once attractive for transmission but produced severe FWM crosstalk in DWDM — eventually abandoned for non-zero dispersion-shifted fiber (NZDSF) that has small but nonzero dispersion to suppress FWM while remaining suitable for long-haul transmission.

Highly nonlinear fiber (HNLF). Specialty fiber with small effective area (15–20 μm² vs 80 μm² for SMF) and engineered zero-dispersion wavelengths offers Kerr coefficients \sim 10× higher than SMF. Used as FWM mixer for wavelength conversion and parametric amplification in research and telecom subsystems.

Silicon waveguide FWM. SOI provides intense optical confinement and high Kerr nonlinearity, achieving useful FWM efficiencies over << 1 cm of waveguide length. Used for on-chip wavelength conversion and frequency-comb generation.