Nonlinear Optics: How Intense Light Bends Its Own Rules

Q: What makes an optical effect "nonlinear"?

An effect is nonlinear when the material's induced polarization is not simply proportional to the driving electric field. At everyday light levels the response is linear. At the high field strengths reached by focused laser light, the electron cloud's restoring force becomes anharmonic, adding terms proportional to the square, cube, and higher powers of the field, producing new frequencies, intensity-dependent focusing, and beam-to-beam interactions absent in linear optics.

Q: Why do some crystals produce second-harmonic light and others don't?

Second-order nonlinear response is forbidden by symmetry in any material with a centre of inversion symmetry, such as glass, liquids, gases, and cubic crystals. Crystals such as quartz, lithium niobate, KTP, and BBO lack a centre of symmetry, so their second-order susceptibility is nonzero and they can efficiently double an input frequency provided the beam is properly phase matched.

Q: What is phase matching and why does it matter?

Phase matching keeps the fundamental and harmonic waves travelling through a nonlinear crystal in step, so that the harmonic light generated at every point adds up constructively instead of cancelling. It is achieved through birefringent angle tuning or periodic poling (quasi-phase matching), and can raise conversion efficiency from a fraction of a percent to over 50 percent.

3D Simulations

🔦 Optics · Photonics

📅 July 2026⏳ 11 min read🟡 Intermediate · Last updated: 3 July 2026

Nonlinear Optics: How Intense Light Bends Its Own Rules

Shine a dim flashlight through glass and the light passes through unchanged in colour, direction and phase relative to any other beam nearby. Focus a high-power laser pulse through the same glass and everything changes: new colours appear that were never in the original beam, the beam can focus itself into a needle of light, and one beam can switch another on and off. This is nonlinear optics — the physics of light so intense that a material's response is no longer simply proportional to the driving field.

Written by MySimulator Team · Reviewed by MySimulator Editorial Review

1. From Linear to Nonlinear Response

In ordinary "linear" optics — the physics behind lenses, mirrors, and everyday refraction — a material's induced polarization P (the collective response of its bound electrons to an electric field) is directly proportional to the driving field E. Double the field amplitude and the polarization doubles too. This proportionality is why linear optics obeys superposition: two light beams crossing in glass pass through each other completely unaffected, each behaving as if the other were not there.

That approximation holds beautifully for sunlight, light bulbs, and even most laser pointers because the electric field of the light is minuscule compared with the internal atomic field binding an electron to its nucleus — typically around 5×10¹¹ V/m. But focus a pulsed laser to a tight spot and the local field can climb to a meaningful fraction of that atomic field. At that point, the electron cloud's restoring force is no longer well approximated by a simple spring (Hooke's law); it becomes anharmonic, and the polarization response picks up extra terms that depend on the square, cube, and higher powers of the field.

Historical note: Nonlinear optics as an experimental field began almost immediately after the invention of the laser. In 1961, Peter Franken and colleagues at the University of Michigan focused a ruby laser (694 nm) into a quartz crystal and detected a faint ultraviolet signal at 347 nm — exactly half the wavelength. This first observation of second-harmonic generation is often cited as the birth of nonlinear optics; famously, an editor at Physical Review Letters nearly removed the tiny UV spot from the published photograph, mistaking it for a speck of dust.

2. The Polarization Expansion and Susceptibility Tensors

The standard way to describe nonlinear response is to expand the induced polarization as a power series in the applied field:

P = ε₀( χ(1)E + χ(2)E² + χ(3)E³ + ⋯ ) χ(1) — linear susceptibility (ordinary refraction, absorption) χ(2) — second-order susceptibility (frequency doubling, mixing) χ(3) — third-order susceptibility (Kerr effect, four-wave mixing)

Each successive order is dramatically weaker than the last: typical values are χ⁽¹⁾ ~ 1, χ⁽²⁾ ~ 10⁻¹² m/V, and χ⁽³⁾ ~ 10⁻²² m²/V². Nonlinear effects therefore only become visible with the enormous instantaneous field strengths delivered by focused, often pulsed, laser beams.

A crucial symmetry rule follows directly from this expansion: χ⁽²⁾ vanishes identically in any material with inversion symmetry (a centrosymmetric medium, such as glass, liquids, gases, or cubic crystals like silicon). Reversing the sign of E must reverse the sign of P in a centrosymmetric medium, but the E² term does not change sign when E does — so it must be zero. Second-order effects such as frequency doubling therefore require non-centrosymmetric crystals: quartz, lithium niobate (LiNbO₃), beta-barium borate (BBO), potassium titanyl phosphate (KTP), and potassium dihydrogen phosphate (KDP) are all workhorses of the field precisely because their crystal structure lacks a centre of symmetry. Third-order effects (χ⁽³⁾) have no such restriction and occur in every material, including ordinary glass fibre and air.

In real anisotropic crystals, χ⁽²⁾ and χ⁽³⁾ are not single numbers but tensors relating vector components of the fields to vector components of the polarization — this tensor structure is what allows crystal cut and orientation to control which nonlinear process is efficient and which polarizations are involved.

3. Second-Harmonic Generation and Frequency Mixing

The simplest and most widely used second-order process is second-harmonic generation (SHG), also called frequency doubling. When a strong field at frequency ω drives a χ⁽²⁾ medium, the E² term in the polarization expansion produces a component oscillating at 2ω — light at exactly twice the frequency (half the wavelength) of the input.

Input field: E(t) = E₀ cos(ωt) E²(t) = E₀² cos²(ωt) = (E₀²/2)[1 + cos(2ωt)] \to a DC (rectification) term + a new oscillation at 2ω \to second-harmonic power scales as P 2ω \propto (χ (2))² \cdot I ω ² \cdot L²

This is the effect that turns the invisible 1064 nm infrared beam of a Nd:YAG laser into the familiar bright green 532 nm beam of a green laser pointer, using a KTP crystal. More generally, the same χ⁽²⁾ mechanism enables:

Sum-frequency generation (SFG): two input beams at ω₁ and ω₂ combine to produce a new beam at ω₁+ω₂.
Difference-frequency generation (DFG): two beams combine to produce a beam at ω₁−ω₂, often used to reach mid-infrared wavelengths with no direct laser source.
Optical rectification: the DC term generates a static or slowly varying field, exploited to generate terahertz radiation from ultrafast pulses.

Notice that second-harmonic power grows with the square of the input intensity — doubling the pump power quadruples the green output, which is why efficient SHG demands tightly focused, high-peak-power pulsed lasers rather than continuous low-power beams.

4. Phase Matching: Making Nonlinear Conversion Efficient

Generating second-harmonic light is easy; generating it efficiently is the real engineering challenge. As the fundamental wave propagates through the crystal, it continuously creates new second-harmonic light at every point along its path. For that light to build up coherently, the newly generated wavelets must stay in phase with light generated earlier in the crystal — otherwise the wave created near the front of the crystal destructively interferes with the wave created further in, and net conversion stalls after a short distance called the coherence length.

Phase mismatch: Δk = k 2ω - 2k ω = (2ω/c)(n 2ω - n ω) Coherence length: L c = π/Δk Perfect phase matching requires: n 2ω = n ω

Because normal dispersion means n(ω) increases with frequency, n_2ω is normally larger than n_ω, and the two waves drift out of phase within microns to tens of microns. Two practical strategies solve this:

Birefringent phase matching: using a birefringent crystal (see our companion article on crystal optics), the fundamental and second-harmonic waves are launched with different polarizations so that the extraordinary index of the harmonic wave, tuned by choosing the propagation angle relative to the optical axis, is made to equal the ordinary index of the fundamental. Rotating the crystal precisely tunes this angle-matching condition.
Quasi-phase matching (QPM): in materials like periodically poled lithium niobate (PPLN), the sign of χ⁽²⁾ is periodically flipped (poled) along the propagation direction with a period equal to twice the coherence length. Each time the two waves are about to interfere destructively, the crystal's nonlinear sign flips, resetting the phase relationship and allowing energy to keep flowing from fundamental to harmonic over the entire crystal length.

With good phase matching, conversion efficiencies from fundamental to second harmonic can exceed 50% in a single pass through a well-designed nonlinear crystal — a remarkable feat given how weak χ⁽²⁾ is at the level of a single atom.

5. The Kerr Effect, Self-Focusing and Self-Phase Modulation

The third-order term χ⁽³⁾ is present in every optical material and gives rise to the optical Kerr effect: the refractive index itself becomes intensity-dependent.

n(I) = n₀ + n₂I n₀ — ordinary linear refractive index n₂ — nonlinear (Kerr) index, typically 10⁻²ⁿ⁻²₀ m²/W in glass I — local light intensity (W/m²)

Because a real laser beam is more intense at its centre than at its edges (a roughly Gaussian transverse profile), the Kerr effect makes the centre of the beam experience a higher refractive index than the edges. The medium therefore behaves like a lens that the beam itself creates — a phenomenon called self-focusing. If the input power exceeds a critical threshold (a few megawatts in typical glass), self-focusing can overcome natural diffractive spreading entirely, causing the beam to collapse toward an intense filament. In practice this collapse is eventually arrested by other effects such as ionization-induced defocusing, producing stable, self-guided optical filaments that can propagate for metres through air — the basis of laser lightning-guidance and remote atmospheric sensing experiments.

Self-Phase Modulation

The same intensity-dependent index acts in time as well as space. As an ultrashort pulse passes through a Kerr medium, the leading edge (rising intensity) sees a rapidly increasing index, while the trailing edge (falling intensity) sees a rapidly decreasing index. Because the instantaneous frequency shift is proportional to the time-derivative of the phase, this produces a "chirp" — new frequencies are generated across the pulse, red-shifted at the leading edge and blue-shifted at the trailing edge. This process, called self-phase modulation (SPM), is what turns a narrow-linewidth femtosecond pulse into a spectrally broadened pulse spanning tens to hundreds of nanometres, and it is the essential first step in supercontinuum and frequency-comb generation.

6. Parametric Processes and Optical Amplifiers

A photon at frequency ω_p (the pump) travelling through a χ⁽²⁾ crystal can spontaneously split into two lower-energy photons, called the signal (ω_s) and idler (ω_i), subject to energy conservation:

Energy conservation: ω p = ω s + ω i Momentum conservation (phase matching): k p = k s + k i

This is the reverse process of sum-frequency generation and is called optical parametric generation. Seeded with a weak signal beam, the same interaction becomes an optical parametric amplifier (OPA): the pump transfers energy coherently into the signal (which is amplified) while simultaneously creating a matching idler beam — with no population inversion or excited-state storage required, unlike a conventional laser gain medium. Enclosing the crystal in a resonant cavity for the signal and/or idler produces an optical parametric oscillator (OPO), which can be continuously tuned across a wide range of wavelengths simply by adjusting the phase-matching angle or crystal temperature — making OPOs indispensable sources of tunable infrared and visible light for spectroscopy where no direct laser gain medium exists.

A closely related and technologically important effect is the linear electro-optic (Pockels) effect, a degenerate case of a χ⁽²⁾ interaction in which one of the two input fields is a slowly varying (or DC) applied voltage rather than another light wave. The applied voltage linearly shifts the crystal's refractive index, which is exploited in Pockels cells to build extremely fast optical shutters and modulators used for Q-switching lasers and encoding data onto light in fibre-optic communication systems.

7. Four-Wave Mixing and Supercontinuum Generation

In centrosymmetric media where χ⁽²⁾ vanishes, the lowest-order nonlinear frequency-mixing process is third-order four-wave mixing (FWM), in which three input photons combine to generate a fourth photon at a frequency satisfying ω₄ = ω₁ + ω₂ − ω₃. FWM is the dominant nonlinear interaction in optical fibres — silica glass is centrosymmetric and has no usable χ⁽²⁾, but its χ⁽³⁾ is more than sufficient at the high intensities confined within a fibre's small core.

In a specific degenerate case of FWM, two photons from an intense pump beam are converted into one signal photon and one idler photon (ω_signal + ω_idler = 2ω_pump), giving fibre-based parametric amplifiers and oscillators an all-glass alternative to bulk-crystal OPAs.

Combining self-phase modulation, four-wave mixing, and higher-order dispersion, an intense femtosecond pulse launched into a specially engineered fibre (such as a photonic-crystal fibre with a tailored dispersion profile) can spread its spectrum from a narrow input line into a continuous band spanning from the ultraviolet through the visible and into the near-infrared — a supercontinuum. Supercontinuum sources underpin optical coherence tomography light sources, ultra-precise frequency combs used in atomic clocks and exoplanet-hunting spectrographs, and broadband spectroscopy instruments.

8. Applications in Science and Technology

Nonlinear optics has moved from a laboratory curiosity to an essential toolkit across science, medicine, and telecommunications.

Green and other visible lasers: nearly every green laser pointer contains an infrared diode-pumped Nd:YAG or Nd:YVO₄ crystal followed by a KTP frequency-doubling crystal — the everyday consumer face of second-harmonic generation.
Multiphoton microscopy: two-photon fluorescence microscopy exploits the fact that simultaneous absorption of two lower-energy (infrared) photons only occurs at the tight focal point of a laser, giving inherent three-dimensional sectioning and much deeper tissue penetration than single-photon confocal microscopy, without the phototoxicity of ultraviolet light.
Optical parametric oscillators: widely tunable OPOs provide coherent light at wavelengths from the visible to the mid-infrared where no efficient direct-gain laser exists, serving spectroscopy, remote gas sensing, and mid-IR imaging.
Fibre-optic telecommunications: the Kerr effect enables soliton pulse propagation (balancing self-phase modulation against dispersion) and four-wave mixing enables all-optical wavelength conversion and parametric amplification, but it can also be a source of unwanted crosstalk between wavelength channels that engineers must design around.
Ultrafast science and frequency combs: supercontinuum generation and difference-frequency generation underlie optical frequency combs, which won the 2005 Nobel Prize in Physics and now define the world's most precise atomic clocks and enable direct comparison of optical frequencies across the electromagnetic spectrum.
Terahertz generation: optical rectification and difference-frequency mixing in nonlinear crystals such as ZnTe and GaP are standard laboratory methods for generating and detecting terahertz pulses used in security screening, materials characterisation, and non-destructive testing.

9. Key Takeaways

Summary

Nonlinear optics begins where proportionality ends: at high field strength, a material's polarization response includes terms in E², E³, and higher powers, not just the linear term that describes ordinary refraction.
χ⁽²⁾ requires broken inversion symmetry: second-order effects like frequency doubling only occur in non-centrosymmetric crystals (quartz, LiNbO₃, BBO, KTP); third-order effects occur in every material, including glass and air.
Second-harmonic generation converts ω → 2ω, powering green laser pointers and countless spectroscopy systems, with power scaling as the square of input intensity.
Phase matching is essential for efficiency: birefringent angle-tuning or periodic poling (quasi-phase matching) keeps the fundamental and harmonic waves in step so conversion builds up coherently over the whole crystal.
The Kerr effect (n₂) makes the refractive index intensity-dependent, producing self-focusing in space and self-phase modulation in time — the seed of laser filamentation and supercontinuum generation.
Parametric processes conserve photon energy and momentum without needing an excited-state population, enabling widely tunable OPOs and the closely related Pockels electro-optic effect used in fast optical modulators.
Four-wave mixing dominates in centrosymmetric media such as optical fibre, driving supercontinuum generation, frequency combs, and all-optical signal processing in telecommunications.

Frequently Asked Questions

What makes an optical effect "nonlinear"?

An effect is nonlinear when the material's response — its induced polarization, and hence the light it emits or the way it bends incoming light — is not simply proportional to the strength of the driving electric field. At everyday light levels, atoms behave like simple harmonic oscillators and the response is linear (ordinary refraction and absorption). At the very high field strengths reached by focused laser light, the electron cloud's restoring force becomes anharmonic, adding terms proportional to E², E³, and higher powers of the field, which is what produces new frequencies, intensity-dependent focusing, and beam-to-beam interactions that are absent in linear optics.

Why do some crystals produce second-harmonic light and others don't?

Second-order nonlinear response (χ⁽²⁾) is forbidden by symmetry in any material with a centre of inversion symmetry, because reversing the sign of the driving field must reverse the sign of the response, and an E² term cannot satisfy that condition unless it is zero. Glass, liquids, gases, and cubic crystals like silicon are centrosymmetric and therefore cannot produce second-harmonic light at all via χ⁽²⁾. Crystals such as quartz, lithium niobate, KTP, and BBO lack a centre of symmetry, so their χ⁽²⁾ is nonzero and they can efficiently double an input frequency, provided the beam is properly phase matched.

What is phase matching and why does it matter?

As a fundamental laser beam travels through a nonlinear crystal, it generates second-harmonic light continuously along its path. For that generated light to add up constructively rather than cancel itself out, the harmonic wave created at each point must stay in phase with harmonic light generated earlier in the crystal — a condition that requires the refractive indices at the fundamental and harmonic frequencies to be equal. Because normal dispersion makes these indices differ, engineers use birefringent angle tuning or periodic poling (quasi-phase matching) to force the indices into alignment, which can raise conversion efficiency from a fraction of a percent to over 50%.

What is the Kerr effect and how does it cause self-focusing?

The optical Kerr effect is the intensity dependence of a material's refractive index, n(I) = n₀ + n₂I, arising from the third-order susceptibility χ⁽³⁾ present in every optical material. Because a real laser beam is brighter at its centre than at its edges, the Kerr effect makes the centre of the beam see a higher refractive index than the edges, so the medium acts as a lens the beam creates for itself. Above a critical power, this self-focusing can overcome the beam's natural diffractive spreading and collapse it into an intense, self-guided filament.

How is nonlinear optics used in everyday technology?

Green laser pointers use second-harmonic generation to convert infrared laser light to visible green. Two-photon microscopes use nonlinear absorption for deep, gentle imaging of living tissue. Optical fibre communication systems rely on the Kerr effect for soliton pulse shaping and on four-wave mixing for wavelength conversion. Optical frequency combs, built from supercontinuum and difference-frequency generation, define the world's most precise atomic clocks. Even everyday displays and sensors increasingly rely on nonlinear crystal components for wavelength conversion and ultrafast optical switching.