A polariton is a hybrid excitation formed when an electromagnetic field couples to an electrically or magnetically active excitation in matter. It is described as a quasiparticle: a collective mode of an interacting system rather than a new elementary particle. Its electromagnetic component can transport energy and couple to external light, while its material component contributes properties such as interactions, confinement, and sensitivity to the surrounding medium. Polaritons occur in bulk materials, at interfaces, in thin layers, and in optical resonators. Major examples include exciton-polaritons, phonon-polaritons, and surface plasmon-polaritons.
Physical basis
Light propagating through matter drives charges or magnetic moments, whose motion in turn generates an electromagnetic field. The resulting normal modes belong to the combined field–matter system. Near a material resonance, treating light and matter as independent excitations can therefore become inadequate.
In a quantum description, a polariton combines a photon with a material excitation. That excitation may be a bound electron–hole pair, a lattice vibration, or a collective oscillation of mobile electrons. A polariton is not generally a photon permanently attached to one particular particle: its material component can extend coherently over many atoms or molecules.
The same mode structure can often be derived classically from Maxwell’s equations and a frequency-dependent permittivity. Quantization then assigns discrete excitations to these coupled modes. Consequently, observing a polariton dispersion does not by itself demonstrate a nonclassical state of light.
The relative contributions of light and matter depend on frequency, wave vector, geometry, and coupling strength. A more photon-like polariton usually inherits greater propagation speed and easier optical access. A more matter-like polariton more strongly inherits the interactions and dissipation of the material. These are tendencies rather than universal rules.
Coupled-mode description
A minimal model contains one electromagnetic mode and one material mode. Within the rotating-wave approximation, its Hamiltonian is
where creates a photon in the selected mode, creates a material excitation, and are their uncoupled angular frequencies, and is the coupling rate. Here is the reduced Planck constant. Treating the material excitation as a harmonic mode is appropriate only within the relevant low-excitation regime.
For a lossless system, diagonalization gives two polariton frequencies:
These define the upper and lower polariton branches. At zero detuning, where , their angular-frequency separation is . As the uncoupled frequencies are tuned through one another, the hybrid branches avoid crossing.
A polariton creation operator can be written schematically as
In the simple lossless model, . These quantities are often called Hopfield coefficients and describe the photonic and material fractions. With several material resonances, there can be more than two branches.
Strong coupling and losses
In an open system, photons escape and material excitations decay or lose coherence. Well-resolved hybrid modes require coupling sufficiently strong relative to these processes. This is the strong-coupling regime.
There is no single numerical strong-coupling criterion that applies independently of definitions and measurement conditions. Conditions for separated complex eigenfrequencies, resolved spectral peaks, and observable coherent energy exchange are related but not identical. A double peak alone is not conclusive evidence: uncoupled resonances, absorption, or interference can sometimes produce similar spectra.
In the ultrastrong-coupling regime, is a substantial fraction of a bare resonance frequency. Terms omitted by the rotating-wave approximation and additional electromagnetic self-interaction terms then matter. The elementary two-mode expression no longer provides a complete description.
Main types
Exciton-polaritons
An exciton is a bound state of an electron and an electron hole. Coupling an optically active exciton to an electromagnetic mode produces an exciton-polariton.
Exciton-polaritons occur in bulk crystals and, especially prominently, in optical microcavities containing quantum wells, organic materials, or atomically thin semiconductors. A cavity confines light between reflecting structures and allows its resonance to be brought close to the exciton energy.
The photonic component can give a cavity polariton a very small in-plane effective mass, while the excitonic component supplies interactions and optical nonlinearities. Exciton-polaritons are therefore widely studied as interacting, driven quantum fluids. At high excitation densities, exciton saturation, screening, and ionization can invalidate a dilute bosonic description.
Phonon-polaritons
A phonon-polariton couples an electromagnetic field to an infrared-active optical phonon, a collective lattice vibration that carries an oscillating electric dipole. These modes are important in polar crystals, commonly at infrared or terahertz frequencies.
In a simple polar dielectric, the frequency interval between transverse and longitudinal optical phonon resonances forms a Reststrahlen band, where the real part of the permittivity can be negative. Surfaces and thin films can support tightly confined phonon-polariton modes in this region.
Anisotropic crystals may have principal permittivity components with opposite signs. Their hyperbolic phonon-polaritons have strongly directional propagation and can support wave vectors much larger than those of free-space light at the same frequency. Hexagonal boron nitride is a prominent example. Such large-wave-vector descriptions ultimately encounter limits imposed by losses, nonlocal response, and atomic structure.
Plasmon-polaritons
A plasmon is a collective oscillation of mobile charge carriers. When this motion couples to an electromagnetic field at a suitable interface, it can form a surface plasmon-polariton.
Surface plasmon-polaritons propagate along interfaces, typically between a conducting material and a dielectric, while their fields decay away from the interface. They can confine electromagnetic energy below the free-space wavelength, but dissipation in the conductor limits their propagation.
Not every plasmon is a polariton in the same sense. A bulk longitudinal plasma oscillation is distinct from a surface mode containing coupled charge motion and transverse electromagnetic fields. Localized resonances in metal nanoparticles are also related but have a different geometry from propagating surface modes.
Other material excitations
Polariton terminology also extends to electromagnetic modes coupled to molecular vibrations, atomic transitions, magnetic excitations such as magnons, and intersubband electronic transitions. The prefix identifies the principal material component; it does not imply that all polaritons have the same dispersion, interactions, or statistical behavior.
Dispersion, confinement, and propagation
The relation between frequency and wave vector—the dispersion relation—is central to polariton physics. Hybridization changes both the energies and the character of the modes near a resonance.
For a sufficiently well-defined, weakly damped wave packet, the group velocity along a propagation direction is
Strongly varying dispersion can slow propagation or redirect energy flow. In anisotropic media, energy flow need not be parallel to the wave vector. In strongly absorbing regions, group velocity alone is not a reliable measure of energy or information transport.
For an ideal planar interface between a metal and a dielectric, the surface plasmon-polariton wave number is
where is the vacuum speed of light and and are the relative permittivities. This expression assumes local, isotropic, nonmagnetic media. Real materials generally have complex permittivities, so the wave number also describes attenuation.
Many surface and guided polaritons possess more in-plane momentum than a freely propagating photon of the same frequency. Direct illumination of a perfectly flat interface may therefore fail to excite them. Prisms, gratings, edges, or near-field probes can supply the required momentum.
Collective behavior and condensation
At low excitation densities, many polariton modes behave approximately as bosons. Exciton-polaritons can accumulate macroscopically in a low-energy state and develop long-range or extended coherence, depending on geometry and conditions. This connects their study to Bose–Einstein condensation.
Most semiconductor microcavity experiments involve finite polariton lifetimes and continuous or pulsed pumping. Their condensates are therefore usually driven–dissipative systems, rather than isolated gases in thermal equilibrium. Condensation depends on the competition between relaxation, interactions, pumping, and loss.
Polariton lasing describes coherent emission associated with a macroscopically occupied polariton state while strong coupling is retained. It differs from conventional photon lasing, in which stimulated emission into an optical mode is normally supported by population inversion. Distinguishing the regimes experimentally requires more than an emission threshold: persistence of strong coupling, dispersion, coherence, and density dependence are important.
Experiments have also investigated vortices and superfluid-like flow. Claims of superfluidity require attention to the system’s nonequilibrium character and to the particular response being measured.
Experimental study
Polariton modes are commonly investigated through angle-resolved reflection, transmission, and emission spectroscopy. In planar structures, emission angle provides information about in-plane wave vector, allowing the dispersion branches to be mapped.
Time-resolved measurements probe lifetimes, relaxation, and coherent dynamics. Near-field optical microscopy accesses strongly confined modes that cannot radiate directly into the far field. Scanning probes can launch polaritons and reveal interference fringes produced by their reflection from edges or defects.
Reliable identification generally combines several observations: dispersion consistent with hybridization, systematic tuning through resonance, linewidth behavior, and agreement with an appropriate electromagnetic or coupled-mode model.
Historical development
The modern theoretical framework emerged from mid-twentieth-century studies of electromagnetic waves in crystals. Work by S. I. Pekar and J. J. Hopfield established influential descriptions of coupled excitonic and electromagnetic excitations. Hopfield’s 1958 quantum treatment helped establish the polariton picture and the coefficients used to express its light–matter composition.
A major experimental milestone was the observation of exciton–photon normal-mode splitting in semiconductor microcavities in 1992. Reports of exciton-polariton condensation in 2006 helped expand the field from linear optical spectroscopy into nonequilibrium many-body physics.
Research subsequently broadened to organic materials, two-dimensional semiconductors, strongly confined phonon and plasmon modes, and ultrastrong light–matter coupling.
Applications and limitations
Polaritons provide ways to combine optical transport with material interactions. Active research areas include compact coherent emitters, optical switching, infrared sensing, nanoscale imaging, directional energy transport, and simulation of interacting quantum systems. Surface and guided modes are particularly useful for concentrating electromagnetic fields into small volumes.
The same material component that supplies useful interactions also introduces absorption, dephasing, disorder, and sensitivity to fabrication. Greater confinement often comes with reduced propagation distance or more demanding excitation conditions. The balance differs substantially between excitonic, phononic, and plasmonic platforms.
In molecular systems, vibrational strong coupling has also been investigated as a means of modifying energy transfer or chemical behavior. The existence of optical mode hybridization does not by itself establish a change in reaction pathways or rates; such claims require separate chemical measurements and controls. More generally, demonstrating a polariton mode and demonstrating a practical advantage from that mode are distinct experimental tasks.