Luminescence is the emission of light by matter through processes other than those responsible for ordinary thermal radiation. It occurs when energy supplied by radiation, an electric current, a chemical reaction, mechanical action, or another source creates excited states that subsequently release some of their energy as photons. Familiar examples include fluorescent dyes, light-emitting diodes, fireflies, glow sticks, and phosphorescent safety markings. Although often called “cold light,” luminescence does not require a low temperature or imply that no heat is produced.
The distinction between luminescence and incandescence concerns the mechanism of emission rather than its color or brightness. A hot filament emits mainly because of its temperature, whereas a fluorescent molecule emits because an electronic excitation has been created and then decays radiatively. Both kinds of emission can occur in the same object. Luminescence connects quantum mechanics, materials science, chemistry, and biology, and provides the operating basis of many lighting technologies, analytical assays, radiation detectors, and dating methods.
Excited states and the origin of emitted light
Luminescence begins with the creation of an excited state in an atom, molecule, crystal, or other material. An electron may be promoted to a higher-energy orbital, or an electron and a positively charged vacancy may be created in a solid. The excited system is not in thermal equilibrium with its surroundings: its electronic populations cannot generally be described by the same temperature that characterizes the material’s vibrations. Relaxation toward a lower-energy state can release a photon whose energy is approximately the difference between the initial and final states:
where is the Planck constant, is frequency, is the speed of light in vacuum, and is the vacuum wavelength. A photon at 500 nanometres carries about 2.48 electronvolts.
Photon emission competes with nonradiative relaxation. In molecules, electronic energy can become vibrational energy and then pass into the surroundings. In crystals, it may be transferred to phonons, which are quantized lattice vibrations. Other competing processes include charge separation, energy transfer to neighboring species, and chemical change. Consequently, strong absorption does not necessarily imply strong luminescence: an excited state may be produced efficiently but lose nearly all its energy without emitting light.
For a simple population of identical, independently decaying excited states, the population follows
where and are the radiative and total nonradiative rate constants. The corresponding luminescence quantum yield is
This expression applies when the state has first-order decay channels and when formation of the emitting state is accounted for separately. Real materials often contain multiple environments, energy-transfer pathways, or traps, producing multiexponential or nonexponential decays.
The spectrum and rate of emission depend on electronic structure and transition probabilities. Atomic emission can consist of narrow lines, whereas molecules commonly produce broader bands because electronic transitions are accompanied by changes in vibrational and rotational states. Solids add interactions with the lattice, disorder, and collective electronic states. Rules involving symmetry and spin angular momentum determine whether a transition is strongly allowed or weakly allowed. A weak transition may produce a long-lived excited state, but the observed lifetime also depends on how rapidly competing nonradiative processes remove its energy.
Photoluminescence: fluorescence, phosphorescence, and delayed emission
Photoluminescence is luminescence initiated by the absorption of photons. It includes fluorescence, phosphorescence, and several forms of delayed emission. In a common organic fluorophore, absorption promotes the molecule from its electronic ground state to an excited singlet state. Rapid vibrational relaxation usually brings it to the lowest vibrational levels of an excited electronic state before emission occurs. The emitted photon therefore commonly has less energy, and a longer wavelength, than the absorbed photon. The separation between absorption and emission bands is called the Stokes shift.
Fluorescence generally involves a transition between states of the same spin multiplicity, most often from an excited singlet state to a singlet ground state. Typical organic fluorescence lifetimes are on the order of nanoseconds, although much shorter and longer values occur. Phosphorescence involves states of different spin multiplicity, commonly an excited triplet state and a singlet ground state. Population reaches the triplet state through intersystem crossing, a nonradiative change in spin character. Because the return transition is often weakly allowed, phosphorescence can persist from microseconds to seconds or longer, depending on the material and its environment. Duration alone, however, is not a rigorous way to distinguish the two processes.
The molecular surroundings strongly affect both forms of emission. Dissolved oxygen often quenches triplet states, making room-temperature phosphorescence difficult to observe in many fluid solutions. Rigid hosts can suppress molecular motion and reduce nonradiative relaxation. Heavy atoms may strengthen spin–orbit coupling, facilitating intersystem crossing and increasing the probability of phosphorescent emission. These effects explain why the same molecular species can show very different emission in solution, a frozen glass, and a crystalline solid.
Delayed fluorescence produces a fluorescence spectrum even though the light appears well after the initial excitation. In thermally activated delayed fluorescence, thermal energy assists reverse intersystem crossing from a triplet state to an emissive singlet state when their energy separation is sufficiently small. Another route is triplet–triplet annihilation: two triplet excitations interact, and their combined energy can produce one higher-energy singlet excitation. These mechanisms differ from direct phosphorescence even when their decay times overlap.
Not all photoluminescence is shifted to lower photon energy. Anti-Stokes emission can draw additional energy from thermal motion, and photon upconversion can combine sequential excitation events to generate a higher-energy photon. Such processes obey energy conservation because the emitted photon receives energy from more than the single lower-energy photon against which it is compared. Photoluminescence also differs from scattering: Raman scattering changes photon energy through interaction with molecular or lattice excitations, but is not normally classified as luminescence.
Electrical excitation, energetic particles, and radiation
Electroluminescence is produced through electrical excitation. In a light-emitting diode, an applied voltage injects electrons and holes into a semiconductor junction. A hole is a missing electron in an otherwise occupied electronic band and behaves as a positive charge carrier. When an electron recombines with an electron hole, energy can be released as light. The emission energy depends chiefly on the semiconductor’s band gap, although confinement, alloy composition, temperature, and defects also influence the spectrum.
Direct-gap semiconductors, including many gallium arsenide and gallium nitride compounds, support efficient radiative recombination because the relevant electronic states can exchange energy and momentum without requiring an additional lattice interaction. In indirect-gap materials such as crystalline silicon, a phonon is generally needed to conserve momentum during band-edge recombination, making this radiative route less probable. Semiconductor heterostructures confine carriers within selected layers, increasing their likelihood of recombining in the intended emitting region. Blue and ultraviolet devices based on gallium nitride and related alloys made efficient phosphor-converted white lighting possible.
An organic light-emitting diode uses organic emitting layers between electrodes. Injected carriers form bound electron–hole excitations called excitons. Simple spin statistics give approximately one singlet for every three triplets when uncorrelated carriers combine, although actual formation and conversion processes can be more complex. Conventional fluorescent emitters primarily use singlet excitations. Phosphorescent metal complexes and thermally activated delayed-fluorescence materials provide routes for utilizing triplet excitations as well. Device performance also depends on charge balance, exciton confinement, electrode losses, and how effectively generated light escapes the layered structure.
Cathodoluminescence results from bombardment by energetic electrons. It powered the phosphor screens of cathode-ray-tube displays and remains useful in electron microscopy, where emission reveals crystal defects, compositional variations, and growth structures. Luminescence caused by ionizing radiation is often termed radioluminescence. Within that broad category, scintillation usually refers to flashes produced when deposited radiation energy is rapidly converted into light. Sodium iodide activated with thallium, inorganic oxide crystals, and organic plastics are established scintillator classes.
Gas discharges provide a related route to excited-state emission. An electric field accelerates charged particles, collisions excite or ionize gas atoms and molecules, and subsequent transitions generate characteristic radiation. Neon signs and low-pressure mercury lamps illustrate this process. Some lamp terminology distinguishes discharge radiation from the luminescence of a solid phosphor that absorbs it; the physical boundary is less important than identifying the actual excitation and emission steps. Radiation generated by particle acceleration, such as synchrotron radiation, is not generally classified as luminescence merely because electrical equipment supplies its energy.
Chemical light and biological light production
Chemiluminescence occurs when a chemical reaction forms an electronically excited product or transfers reaction energy to an emitting species. The chemical energy does not first have to heat the bulk material to incandescence. Efficient emission requires a reaction pathway that channels an appreciable fraction of available energy into an excited electronic state rather than distributing it entirely among molecular vibrations, translations, and other products. Many energetic reactions emit very little light because that branching fraction is small.
Luminol is a well-known chemiluminescent reagent. Under suitable alkaline, oxidizing conditions, its reaction produces electronically excited 3-aminophthalate, which emits blue light as it relaxes. Metal ions and several biological substances can catalyze the reaction, so emission is not uniquely specific to a single analyte. Peroxyoxalate systems, familiar from glow sticks, use an oxalate ester and peroxide to generate chemical excitation that is transferred to a fluorescent dye. Changing the dye changes the color without requiring the principal energy-producing chemistry to change.
Bioluminescence is chemically generated light produced by living organisms. It has evolved independently in many groups, including bacteria, marine protists, fungi, insects, and numerous marine animals. The names luciferin and luciferase refer respectively to light-producing substrates and the enzymes that catalyze their reactions; they do not designate a single universal substrate or enzyme family. Different organisms use chemically and evolutionarily distinct systems.
In fireflies, luciferase first activates luciferin using adenosine triphosphate, followed by oxygen-dependent chemistry that generates an excited oxyluciferin product. The emission color depends on the emitter’s molecular state and its environment within the enzyme. Bacterial luciferase instead uses reduced flavin mononucleotide, oxygen, and a long-chain aldehyde, producing blue-green emission. In the jellyfish Aequorea victoria, the calcium-responsive photoprotein aequorin generates blue light that can transfer energy to green fluorescent protein. The protein then fluoresces green; it does not itself supply the underlying chemical energy.
Biological light serves several documented functions: attracting mates or prey, startling predators, signaling, and camouflage through counterillumination. In the ocean, an animal’s downward-directed emission can reduce its silhouette against light arriving from the surface. Some animals contain symbiotic luminous bacteria rather than producing light entirely with their own biochemical machinery. Bioluminescence must also be distinguished from biological fluorescence: a fluorescent coral or protein generally requires an external source of excitation light, whereas a bioluminescent reaction generates its excitation chemically.
The same mechanisms support laboratory assays. Luciferase reporters connect light output to gene expression, enzyme activity, or cellular processes. ATP-dependent light production is used in assays of ATP abundance. The interpretation nevertheless depends on reaction conditions: substrate delivery, oxygen availability, enzyme concentration, and optical absorption can alter the signal independently of the biological quantity being investigated.
Stored excitation, persistent emission, and mechanical stimulation
Some materials store part of the energy deposited by earlier irradiation and release it later as luminescence. In an insulating or semiconducting crystal, excitation can separate charge carriers that become trapped at defects or impurity sites. These trapped charges occupy metastable configurations rather than remaining continuously in the final light-emitting state. Subsequent release allows migration or tunneling to a recombination center, where radiative emission can occur. This distinction is important: a long afterglow need not correspond to an exceptionally long lifetime of one emitting excited state.
Persistent luminescence continues after the excitation source has been removed. Europium-activated strontium aluminate, often co-doped with dysprosium, is a prominent green-emitting persistent phosphor used in luminous markings. Its performance depends on host composition, defects, dopants, and processing conditions. Trap distributions and the detailed role of co-dopants vary among compositions, and not every aspect of charge storage in these materials is captured by a single universal model. Commercial descriptions often call such materials “phosphorescent,” although their afterglow can involve trap-mediated recombination rather than direct molecular triplet-state phosphorescence.
Thermoluminescence is light emitted when heating releases previously trapped charge. Heating supplies the energy needed for detrapping; the resulting emission is therefore not simply the ordinary thermal radiation associated with the sample’s temperature. In a basic activated model, the release rate contains the factor
where is a trap activation energy, is the Boltzmann constant, and is absolute temperature. Measured glow curves show intensity as temperature rises. Their peaks depend on trap depth, release frequency, heating rate, retrapping, and recombination kinetics.
Optically stimulated luminescence releases stored charge through illumination rather than heating. Quartz and feldspar grains can retain signals created by natural environmental radiation after burial. Laboratory stimulation measures part of the accumulated signal, allowing an absorbed dose to be estimated. These measurements support luminescence dating when the dose rate and the history of signal resetting are adequately constrained.
Mechanical action can also generate light. Triboluminescence is associated with rubbing, fracture, or related mechanical disturbance, while mechanoluminescence is a broader term encompassing mechanically induced emission. Crushing some molecular crystals separates charge, and electrical discharge through surrounding gas can contribute to the observed flash. Other materials emit under elastic deformation through mechanisms involving internal electric fields, defects, or trapped carriers. Thus, “mechanically generated light” identifies an initiating stimulus, not one microscopic mechanism.
A further category, sonoluminescence, occurs during the collapse of acoustically driven bubbles in liquids. Compression can create extremely energetic conditions within a bubble and produce brief optical flashes. Its microscopic description differs substantially from the simple picture of a fluorescent molecule relaxing after absorption. The example illustrates why broad classifications by excitation source must be supplemented by an account of the actual emitting system.
Luminescent materials and control of emission
A luminescent material combines an emitting electronic structure with a physical environment that governs excitation, energy transport, and loss. In an inorganic phosphor, a host lattice commonly contains a small concentration of activator ions. The activator supplies useful electronic transitions, while the host controls its local coordination, crystal field, vibrational environment, and chemical stability. The word phosphor describes function and does not imply that the material contains phosphorus or necessarily emits by phosphorescence.
Different activators produce characteristic spectral behavior. Many trivalent rare-earth ions have relatively narrow emission lines because their emitting 4f electrons are partly shielded by outer electron shells. Europium in the trivalent state commonly produces red emission, and terbium in the trivalent state often produces green emission. Divalent europium and trivalent cerium frequently yield broader bands involving 5d and 4f states; the energies of these bands depend strongly on the host environment. Cerium-doped yttrium aluminum garnet is an established yellow-emitting phosphor for blue-pumped white LEDs.
Organic emitters provide a different set of design variables. Conjugated molecular structures permit visible-light absorption and emission, while substituents and molecular geometry tune electronic energies. Rigid structures can suppress vibrational relaxation, whereas uncontrolled aggregation can create nonemissive pathways. Aggregation-induced emission describes systems in which clustering instead strengthens emission, often because molecular motions that dissipate excitation are restricted. The behavior is therefore not determined by molecular formula alone: packing and the surrounding matrix can be decisive.
Quantum dots are semiconductor nanostructures whose carrier confinement modifies their optical properties. In a size regime where confinement is important, smaller dots of a given composition generally have larger effective optical gaps and emit at shorter wavelengths. Surface atoms represent a substantial fraction of the material, so surface defects can introduce efficient nonradiative losses. Core–shell structures and surface passivation reduce such losses and improve stability. Composition, size distribution, and interactions between particles all influence the final emission spectrum.
Defects can either impair or enable luminescence. Unwanted defects often quench emission, but deliberately engineered centers can provide useful optical transitions. The nitrogen–vacancy center in diamond, for example, combines optical emission with a spin-dependent response that supports sensing and quantum experiments. In other hosts, vacancies and impurities serve as charge traps for persistent emission.
Concentration and temperature introduce additional trade-offs. Increasing activator concentration can improve excitation absorption, but may also promote energy migration to quenching sites or interactions that dissipate excitation. Thermal quenching occurs when higher temperatures make nonradiative pathways more accessible. Material development consequently balances brightness, emission color, decay time, chemical durability, environmental sensitivity, and manufacturing constraints rather than maximizing a single property.
Measurement, efficiency, and interpretation
Luminescence measurements describe several distinct quantities: the emission spectrum, total photon yield, decay kinetics, polarization, and spatial distribution. Spectroscopy commonly records the emission spectrum at a fixed excitation wavelength or an excitation spectrum while monitoring a selected emission band. An excitation spectrum can resemble an absorption spectrum, but the two need not coincide when different absorbing species have different emission efficiencies or when energy transfer links absorption to emission elsewhere in the sample.
Quantum yield is the ratio of emitted photons to a specified number of excitation events, usually absorbed photons for photoluminescence. It is not the same as energy-conversion efficiency. If one photon is emitted for each absorbed photon but the emitted photons have lower energies, the photon quantum yield can approach unity while the optical energy yield remains lower. Electrical devices require further distinctions: internal quantum efficiency concerns photons generated within the device relative to injected charge, while external quantum efficiency counts photons that actually emerge. Wall-plug efficiency compares emitted optical power with electrical input power.
Absolute photoluminescence quantum yields are often measured with an integrating sphere, which collects light over a large range of directions and helps determine absorbed and emitted photon numbers. Relative methods compare a sample with a reference of known yield under controlled conditions. Both approaches require corrections for detector sensitivity, excitation intensity, background, and optical geometry. Calibration is especially important when comparing emission at different wavelengths, because detector response can vary substantially across a spectrum.
Time-resolved measurements reveal mechanisms that steady-state intensity alone cannot separate. Time-correlated single-photon counting determines the distribution of photon arrival times following repeated excitation pulses. Other methods use gated detection, fast photodetectors, or phase and modulation measurements. The observed decay must be interpreted together with the instrument response. A fitted lifetime is not automatically the natural radiative lifetime: it generally includes all competing decay processes and can be affected by energy transfer, reabsorption, or multiple emitting populations.
Quenching analysis is another major tool. For a simple dynamic quencher, the Stern–Volmer relationship is
where and are intensity and lifetime without quencher, and are their values with quencher, is the bimolecular quenching rate constant, and is quencher concentration. Static complex formation, heterogeneous accessibility, and simultaneous mechanisms can produce different relationships. Comparing intensity with lifetime helps discriminate among these possibilities.
Optical artifacts can otherwise resemble changes in molecular behavior. The inner-filter effect reduces excitation reaching the emitting volume or reabsorbs emitted light before detection. Photobleaching irreversibly destroys or alters emitters during illumination. Sample scattering and background fluorescence also distort signals. Reliable interpretation therefore requires an explicit distinction between changes in excitation, intrinsic emission efficiency, and light collection.
Lighting, analysis, imaging, and dating
Lighting and display technologies exploit different ways of creating and controlling luminescence. Fluorescent lamps use ultraviolet radiation from a mercury discharge to excite phosphors on the inside of a tube. Many white LEDs instead combine a blue semiconductor emitter with one or more wavelength-converting phosphors. Residual blue light mixes with longer-wavelength emission to produce a white appearance. Phosphor selection controls color balance, while absorption, conversion, and light extraction determine efficiency. Displays use direct emission from inorganic LEDs or organic emitters, or incorporate luminescent materials to improve the spectrum of a backlight.
In analytical chemistry, luminescence can provide high sensitivity because emitted light is observed against a relatively dark background rather than as a small reduction in a strong transmitted beam. Fluorescent labels make molecules or cellular components optically distinguishable. Flow cytometry measures emission from individual particles or cells passing through an illuminated region, while fluorescence microscopy maps labeled structures in space. Multiplex measurements use spectrally distinct labels, although overlapping spectra require optical separation or mathematical correction.
Energy transfer between fluorophores supplies information on molecular proximity. Förster resonance energy transfer is a nonradiative transfer process whose efficiency depends strongly on donor–acceptor separation, commonly over distances of a few nanometres. Its use requires suitable overlap between donor emission and acceptor absorption, as well as consideration of molecular orientation and donor quantum yield. It can report conformational changes or interactions, but does not translate into a unique distance without assumptions about the labeling system.
Radiation detection uses scintillators to convert deposited energy into measurable light. A photomultiplier tube or semiconductor photosensor converts that light into an electrical signal. Brightness affects energy measurement, decay time affects timing, and density and elemental composition affect interaction with incoming radiation. Scintillator detectors are used in laboratory instruments, security systems, particle experiments, and medical imaging, including positron emission tomography. Thermoluminescent and optically stimulated luminescent dosimeters instead store information about accumulated exposure for later readout.
Luminescence dating is important in archaeology and Quaternary geology. Heating can reset a pottery sample’s stored signal, while sufficient daylight exposure can reset mineral grains before sediment burial. The basic age relation is
The equivalent dose is the laboratory radiation dose needed to reproduce the relevant natural signal. The environmental dose rate includes contributions from radioactive elements and cosmic radiation, with corrections for factors such as water content. Dating requires evidence that the relevant resetting event occurred and that signal behavior and dose history are adequately characterized.
Luminescent sensors exploit changes in emission intensity, spectrum, or lifetime caused by oxygen, acidity, temperature, or binding interactions. Lifetime-based sensing can reduce sensitivity to probe concentration and illumination fluctuations. Its interpretation still depends on the probe’s local environment, because several variables may alter the same emission pathway.
Historical development and changing explanations
Light from organisms, glowing minerals, and electrical phenomena was observed long before a unified concept of luminescence existed. Early accounts did not reliably distinguish chemical emission, fluorescence, reflected light, and heat-induced glow. Around 1602, the Bologna shoemaker and alchemist Vincenzo Cascariolo prepared a light-storing material from barite. The material became known as the Bologna stone and is associated with barium sulfide produced by processing the mineral. Its afterglow attracted attention because illumination appeared to leave a temporary store of light in matter.
The experimental study of fluorescence advanced substantially in the nineteenth century. In 1852, George Gabriel Stokes described changes in the refrangibility of light and introduced the term fluorescence, drawing the name from fluorite. His work helped establish that absorbed radiation could be re-emitted at a different wavelength rather than merely reflected or scattered unchanged. The commonly observed shift toward longer wavelengths became associated with his name, although later research identified processes that do not follow this ordinary pattern.
The German physicist Eilhard Wiedemann introduced “luminescence” in 1888 to group emissions not attributable simply to the temperature of the emitting body. This terminology organized phenomena by their nonthermal origin while allowing subdivisions based on excitation source. At that stage, the microscopic nature of excited electronic states was still unknown. The emergence of quantum theory, atomic spectroscopy, and molecular electronic-state models during the early twentieth century supplied the concepts needed to explain characteristic spectra, transition probabilities, and emission lifetimes.
Aleksander Jabłoński’s work in the 1930s helped establish the energy-level diagrams now used to describe absorption, vibrational relaxation, fluorescence, intersystem crossing, and phosphorescence. These diagrams made it possible to distinguish pathways that could look similar in a simple observation of glowing material. Research on inorganic phosphors developed alongside electric lighting, X-ray screens, cathode-ray tubes, and radiation detection. Improvements increasingly depended on controlled impurities and defects rather than on host composition alone.
Semiconductor light emission moved from early observations toward practical devices during the twentieth century. H. J. Round reported electroluminescence from silicon carbide in 1907, and Oleg Losev investigated related effects in the 1920s. Practical semiconductor LEDs emerged in the 1960s. The development of efficient blue LEDs by Isamu Akasaki, Hiroshi Amano, and Shuji Nakamura enabled major advances in white solid-state lighting and was recognized by the 2014 Nobel Prize in Physics.
Biological luminescence research also transformed experimental methods. The isolation and characterization of green fluorescent protein from Aequorea victoria, followed by its use as a genetically encoded marker, allowed fluorescent labeling without attaching a synthetic dye to each target molecule. Osamu Shimomura, Martin Chalfie, and Roger Tsien received the 2008 Nobel Prize in Chemistry for the discovery and development of this protein. Subsequent engineering produced proteins with altered colors, brightness, maturation, and environmental responses, extending luminescence research from the study of emitted light to the optical tracking of molecular and cellular processes.