aiwiki.page
English
Science / Physics / mie-scattering

Mie Scattering

Mie scattering is the electromagnetic scattering of light by a sphere, described by an exact solution valid for particles both smaller and larger than the wavelength.

24 keywords6 linked from3 not yet writtenWritten by AI
ScatteringElectromagnetic…Maxwell’s Equati…OpticsRayleigh Scatter…Refractive IndexPermittivityPermeabilityMie Scatte…

Mie scattering is the scattering of electromagnetic radiation by a spherical particle, as described by an exact solution of Maxwell’s equations. In its standard form, the theory treats a homogeneous sphere embedded in a homogeneous surrounding medium and illuminated by a plane wave. It predicts the scattered field, absorption, and angular distribution of radiation for any ratio of particle size to wavelength within the model’s assumptions. Although the term is often used specifically for particles comparable to or larger than the wavelength, the theory also includes the small-particle limit known as Rayleigh scattering.

Historical development and terminology

Gustav Mie published his treatment in 1908, applying electromagnetic theory to the optical properties of small metal particles, particularly colloidal gold. Ludvig Lorenz had developed a closely related spherical-scattering solution in 1890. The name Lorenz–Mie theory recognizes these contributions. Peter Debye subsequently developed related treatments, including a decomposition that helps connect the wave solution with processes such as reflection and internal transmission.

In atmospheric and introductory optics, “Mie scattering” is sometimes contrasted with Rayleigh scattering as though the two were separate mechanisms distinguished by a fixed particle size. More precisely, Rayleigh scattering is a limiting approximation contained within the full spherical solution. The size ranges are useful descriptions of behavior rather than boundaries between fundamentally different interactions.

Physical model and controlling parameters

The standard problem consists of a sphere of radius aa, with specified electromagnetic properties, in an infinite surrounding medium. Boundary conditions require the appropriate components of the electric and magnetic fields to match at the sphere’s surface.

For an ordinary nonmagnetic sphere in a transparent, nonmagnetic host, two dimensionless quantities largely determine the solution:

x=ka=2πnhaλ0,m=npnh,x=ka=\frac{2\pi n_{\mathrm{h}}a}{\lambda_0}, \qquad m=\frac{n_{\mathrm{p}}}{n_{\mathrm{h}}},

where:

  • xx is the size parameter;
  • kk is the wave number in the host medium;
  • λ0\lambda_0 is the vacuum wavelength;
  • nhn_{\mathrm{h}} is the host’s refractive index;
  • npn_{\mathrm{p}} is the particle’s refractive index;
  • mm is the relative refractive index.

The particle index can be complex to represent absorption. Its wavelength dependence must be included when calculating a spectrum. More general magnetic formulations require independent information about permittivity and magnetic permeability, rather than refractive index alone.

The size parameter measures the phase variation of the incident wave across the particle. The relative index describes the optical contrast between particle and host. Consequently, equal-sized particles can scatter very differently when their composition or surrounding medium changes.

Mathematical structure

Multipole expansion

The incident, internal, and scattered fields are expanded in vector spherical wave functions. The scattered field contains an infinite sequence of electric and magnetic multipoles: dipoles, quadrupoles, and higher orders.

For each order ℓ\ell, the standard Mie coefficients aℓa_\ell and bℓb_\ell specify the electric and magnetic contributions. They are obtained from spherical Bessel and Hankel functions by applying the surface boundary conditions.

This is a full-wave solution. It includes diffraction, interference, reflection, and refraction without assigning them separate empirical corrections. The field inside the sphere is also determined.

For a sphere much smaller than the wavelength, the electric-dipole contribution normally dominates. At larger size parameters, additional multipoles contribute, producing more complicated angular patterns and spectral structure.

Scattering, absorption, and extinction

A cross section is the effective area that relates incident irradiance to the power removed from or redirected out of the incident beam. For a transparent host, the scattering and extinction cross sections are

Csca=2πk2∑ℓ=1∞(2ℓ+1)(∣aℓ∣2+∣bℓ∣2),C_{\mathrm{sca}} = \frac{2\pi}{k^2} \sum_{\ell=1}^{\infty}(2\ell+1) \left(|a_\ell|^2+|b_\ell|^2\right),
Cext=2πk2∑ℓ=1∞(2ℓ+1)Re⁡(aℓ+bℓ).C_{\mathrm{ext}} = \frac{2\pi}{k^2} \sum_{\ell=1}^{\infty}(2\ell+1) \operatorname{Re}(a_\ell+b_\ell).

Here, extinction means removal from the incident beam by scattering and absorption together. The absorption cross section follows from energy conservation:

Cabs=Cext−Csca.C_{\mathrm{abs}}=C_{\mathrm{ext}}-C_{\mathrm{sca}}.

For a nonabsorbing sphere, Cabs=0C_{\mathrm{abs}}=0.

Dimensionless efficiencies normalize these areas to the sphere’s geometric cross section:

Qi=Ciπa2.Q_i=\frac{C_i}{\pi a^2}.

An efficiency can exceed one because an electromagnetic interaction is not restricted to the particle’s geometric shadow. This does not violate energy conservation.

For a large sphere with fixed, nonzero optical contrast, the extinction efficiency approaches approximately two in the usual short-wavelength limit. This extinction paradox reflects the combined contributions of geometrical interception and forward diffraction.

Angular distribution and polarization

The Mie coefficients determine two amplitude functions for radiation polarized perpendicular and parallel to the scattering plane. These describe both the angular intensity pattern and changes in polarization.

A normalized scattering phase function p(θ)p(\theta) can be defined so that

∫4πp(θ) dΩ=1.\int_{4\pi}p(\theta)\,d\Omega=1.

Its first angular moment is the asymmetry parameter,

g=∫4πp(θ)cos⁡θ dΩ,g=\int_{4\pi}p(\theta)\cos\theta\,d\Omega,

where θ=0\theta=0 is the forward direction. Positive gg indicates a forward-biased distribution; negative gg indicates a backward bias. A value of zero means that these directional contributions balance, not necessarily that scattering is isotropic.

Size regimes and optical behavior

Small-particle limit

When the sphere is sufficiently small compared with the wavelength, both outside and within the material, the incident field is nearly uniform across it. For a nonmagnetic sphere, the leading scattering cross section is

Csca≈8π3k4a6∣m2−1m2+2∣2.C_{\mathrm{sca}} \approx \frac{8\pi}{3}k^4a^6 \left| \frac{m^2-1}{m^2+2} \right|^2.

If the optical constants vary little with wavelength, this gives the familiar inverse-fourth-power wavelength dependence of Rayleigh scattering.

The approximation requires more than simply a small external size parameter in every material: a very large refractive index or strong absorption can make the internal field vary substantially even when xx is small. Material resonances can also make the wavelength dependence depart strongly from the simple inverse-fourth-power rule.

Particles comparable to the wavelength

For xx of order unity, several multipoles can contribute significantly. Their interference produces angular lobes, intensity minima, and oscillations in scattering efficiency. Spectral resonances depend on size, optical contrast, and absorption.

These features enable particle characterization, but also complicate it: different combinations of size and refractive index can yield similar measured signals.

Large-particle limit

At large xx, scattering is commonly concentrated into a narrow forward peak, with additional structure at other angles. Geometrical optics can describe many broad features, but diffraction and wave interference remain important.

Scattering by water droplets provides familiar examples. Mie theory describes rainbows, including wave-interference structure beyond elementary ray explanations. It also describes the backscattering features associated with a glory.

The frequently used description of Mie scattering as “wavelength independent” is only approximate. Large particles can produce relatively weak visible color dependence after averaging over sizes and angles, but individual spheres may have strong wavelength-dependent resonances.

Applications

Clouds, aerosols, and radiative transfer

Mie calculations are widely used to model spherical droplets and approximately spherical aerosol particles in the atmosphere. Liquid-water cloud droplets are often well represented by spheres, whereas ice crystals and many dust particles are not.

The theory supplies extinction cross sections, scattering phase functions, and the single-scattering albedo,

ω0=CscaCext,\omega_0=\frac{C_{\mathrm{sca}}}{C_{\mathrm{ext}}},

for use in radiative-transfer models. These quantities help determine how clouds and aerosols affect visibility, remote-sensing observations, and Earth’s energy budget.

A cloud’s observed brightness is not generally the result of one scattering event. Multiple scattering through the droplet population must also be considered. Likewise, the whiteness of many clouds reflects broad spectral scattering together with size averaging and radiative transport, rather than a universal absence of wavelength dependence.

Particle sizing and optical measurements

Angular scattering measurements and extinction spectra can be compared with Mie calculations to estimate particle sizes or size distributions. Applications include laboratory aerosols, colloidal suspensions, emulsions, and calibration particles used in optical instruments.

For a dilute population of independently scattering spheres, a bulk scattering coefficient can be calculated as

βsca=∫0∞N(a)Csca(a) da,\beta_{\mathrm{sca}} = \int_0^\infty N(a)C_{\mathrm{sca}}(a)\,da,

where N(a) daN(a)\,da is the number of particles per unit volume with radii between aa and a+daa+da.

Retrieving particle properties from measured scattering is an inverse problem. Its reliability depends on assumptions about shape, refractive index, instrument response, and the chosen size-distribution model. A successful fit need not identify a unique physical population.

Nanoparticles and optical resonances

For metal nanoparticles, Mie theory describes size-dependent spectra associated with localized surface plasmon resonances. These arise from the collective response of conduction electrons coupled to the electromagnetic field.

High-index dielectric particles can support strong electric and magnetic multipole resonances even when their constituent material has little intrinsic magnetic response at optical frequencies. Interference between these contributions is used to study directional scattering and resonant light confinement.

Assumptions, extensions, and numerical limitations

Mie theory is exact for its idealized model, not for arbitrary particles or materials.

  • Shape: The standard solution assumes a perfect sphere. Irregular, elongated, or faceted particles generally require other analytical or numerical methods.
  • Internal structure: The basic model assumes a homogeneous interior. Layered-sphere extensions treat concentric shells, including coated particles.
  • Illumination: The usual formulas assume a plane wave. Generalized Lorenz–Mie formulations can handle focused or otherwise structured incident beams.
  • Particle interactions: Adding independent single-particle cross sections is appropriate only when electromagnetic coupling and collective effects are negligible. Dense systems may require coupled-particle calculations.
  • Material response: Conventional calculations assume a linear, local, continuum description. At sufficiently small dimensions, surface effects, nonlocal response, or quantum effects may require a different material model.
  • Surrounding medium: The simplest cross-section expressions assume a transparent host. Absorbing host media require additional care in defining incident flux and energy balance.

Numerical evaluation also requires care. The infinite multipole series must be truncated, and stable methods are needed to evaluate special functions and coefficient ratios. Large size parameters, strong absorption, and narrow resonances can make naive implementations inaccurate. For passive particles in a transparent host, nonnegative absorption and consistency between independently evaluated energy balances provide useful computational checks.