{
  "id": "graviton",
  "title": "Graviton",
  "lang": "en",
  "source_lang": "en",
  "summary": "A graviton is the hypothetical massless spin-2 quantum of gravity, predicted by the perturbative quantization of general relativity but not individually observed.",
  "category": "science",
  "subcategory": "physics",
  "type": "concept",
  "aliases": [
    "Gravitational field quantum",
    "quantum of the gravitational field"
  ],
  "generated_at": "2026-11-03",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/graviton/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/graviton/",
      "json": "https://aiwiki.page/zh/graviton.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "A graviton is the hypothetical elementary quantum of the gravitational field. In the standard theoretical description, it is a massless particle with spin 2, playing a role for gravity analogous to that of the photon for electromagnetism. Gravitons arise when weak disturbances of spacetime geometry are described using quantum mechanics. No individual graviton has been detected, and no experiment has yet established that the gravitational field possesses the quantum properties attributed to it.\n\nThe concept connects general relativity, which describes gravity through spacetime geometry, with quantum field theory, which describes particles as excitations of fields. At energies well below the Planck scale, these frameworks support a consistent, predictive description of quantum gravitational interactions. Extending that description to arbitrarily short distances remains an unresolved problem in quantum gravity.\n\nThe graviton is therefore both a specific theoretical object and a guide to deeper questions. Its proposed spin and interactions follow from strong consistency requirements, while its experimental status remains distinct from the firmly established existence of gravitational waves. Its role in high-energy theories, early-universe models, and proposed laboratory tests depends on which features of gravity are being quantized and on the physical regime under consideration."
    },
    {
      "heading": "Quantum numbers and physical meaning",
      "text": "In four-dimensional spacetime, the graviton associated with general relativity is a massless boson with spin quantum number 2. More precisely, freely propagating gravitons have two physical helicity states, conventionally denoted \\(+2\\) and \\(-2\\). Helicity is the component of angular momentum along the direction of motion; in units of the reduced Planck constant, these states carry angular-momentum projections \\(+2\\hbar\\) and \\(-2\\hbar\\). For a massless particle, helicity is the useful label because no rest frame exists in which an ordinary spin axis could be defined.\n\nThe spin-2 assignment is tied to the tensor character of gravity. Electromagnetism is represented by a vector potential, and its quanta have helicities \\(+1\\) and \\(-1\\). A small gravitational disturbance is represented by a symmetric rank-two tensor. After accounting for the redundancies and constraints of the theory, only two independent radiative degrees of freedom remain. Their classical counterparts are the two tensor polarizations of a gravitational wave, often called “plus” and “cross.” These linear polarizations can be combined to form circular polarizations corresponding to definite helicity.\n\nA massless graviton travels at the speed of light in vacuum and obeys\n\n\\[E=pc=\\hbar\\omega,\\]\n\nwhere \\(E\\) is its energy, \\(p\\) its momentum magnitude, and \\(\\omega\\) its angular frequency. Here \\(\\hbar=h/(2\\pi)\\), with \\(h\\) the Planck constant. A graviton at a frequency of \\(100\\) hertz would have an energy of approximately \\(4.14\\times10^{-13}\\) electronvolts. This exceptionally small energy illustrates one difficulty in associating an astronomical gravitational-wave measurement with the detection of a single quantum.\n\nIn ordinary perturbative gravity, the graviton is electrically neutral and is its own antiparticle. It carries no color charge and does not belong to the particle content of the Standard Model. Nevertheless, it couples to Standard Model particles because their energy and momentum contribute to gravity. It also couples to other gravitons: the gravitational field itself carries energy and therefore participates in gravitational interactions.\n\nThe word “graviton” sometimes appears in theories with massive spin-2 particles. Those objects differ from the massless graviton of general relativity. A massive spin-2 particle in four dimensions has five physical polarization states rather than two, and constructing consistent interactions for it requires additional care. Experimental bounds on a possible graviton mass test such extensions; they do not mean that the usual theoretical graviton is assumed to possess a small measured mass."
    },
    {
      "heading": "From spacetime geometry to a quantum field",
      "text": "General relativity represents gravity through the metric \\(g_{\\mu\\nu}\\), which determines distances, durations, and causal relations in spacetime. To identify graviton-like excitations in a weak gravitational field, the metric can be expanded around a chosen background. For a flat background, a common convention is\n\n\\[g_{\\mu\\nu}=\\eta_{\\mu\\nu}+\\kappa h_{\\mu\\nu}, \\qquad \\kappa^2=32\\pi G,\\]\n\nusing units in which \\(c=\\hbar=1\\). Here \\(\\eta_{\\mu\\nu}\\) is the Minkowski metric, \\(G\\) is Newton’s gravitational constant, and \\(h_{\\mu\\nu}\\) is the canonically normalized perturbation. The precise placement of constants is conventional; the distinction between background geometry and fluctuating field is the essential step.\n\nExpanding the gravitational action to quadratic order in \\(h_{\\mu\\nu}\\) gives the free massless spin-2 theory. Higher-order terms describe interactions. The perturbation initially has ten components because it is a symmetric four-by-four tensor. These do not represent ten independent particles or polarizations. Coordinate invariance produces a gauge redundancy whose linearized form is\n\n\\[h_{\\mu\\nu}\\longrightarrow h_{\\mu\\nu} +\\partial_\\mu\\xi_\\nu+\\partial_\\nu\\xi_\\mu,\\]\n\nwith normalization factors absorbable into \\(\\xi_\\mu\\). Gauge freedom and the field equations remove the unphysical components, leaving two propagating modes. A transverse-traceless description makes these modes especially transparent for waves in vacuum.\n\nQuantization promotes the amplitudes of the physical modes to operators. Each wave mode behaves mathematically like a quantum harmonic oscillator, with creation and annihilation operators that change its occupation number. A one-graviton state is a state containing one quantum in an appropriate mode. Multiparticle states, superpositions, and coherent states can then be constructed using the same broad framework employed for other quantum fields.\n\nThis construction does not require the background to be exactly flat, but curved backgrounds complicate the particle interpretation. In an expanding universe or near a black hole, different choices of time evolution can lead to different definitions of positive frequency and therefore of particles. A unique global graviton number need not exist. Local observables and field correlations may be more fundamental than a particle count in such situations.\n\nThe perturbation \\(h_{\\mu\\nu}\\) is also not itself invariant under a change of coordinates. Physical predictions must be expressed through suitable gauge-invariant quantities, relational observables, or scattering data. Thus a graviton is not simply a tiny localized deformation with a uniquely defined shape. It is a quantum excitation identified within a field description whose coordinate redundancies must be treated consistently."
    },
    {
      "heading": "Universal coupling and gravitational interactions",
      "text": "The defining interaction of the graviton is its coupling to energy and momentum. At leading order, the interaction is proportional to\n\n\\[\\mathcal{L}_{\\mathrm{int}}\\propto \\kappa h_{\\mu\\nu}T^{\\mu\\nu},\\]\n\nwhere the stress–energy tensor \\(T^{\\mu\\nu}\\) describes energy density, momentum density, stresses, and energy flow. This is more general than coupling only to rest mass. Light gravitates despite having no rest mass, and pressure contributes to gravity in relativistic systems. The coupling therefore reflects the relativistic structure of the gravitational source.\n\nThe universality of this coupling is closely related to the equivalence principle. In a relativistic quantum theory containing a massless spin-2 particle, unphysical polarizations must not affect measurable probabilities. Steven Weinberg’s 1964 analysis of low-energy, or “soft,” emission showed that consistency strongly constrains the interaction: the spin-2 particle must couple universally to the energy and momentum of the participating particles. Under the assumptions of the argument, different particle species cannot be assigned arbitrary independent gravitational charges.\n\nThe same reasoning helps explain why gravitons interact with one another. Once matter exchanges energy with the gravitational field, the matter stress–energy tensor alone is not the complete conserved source. The gravitational field’s own contribution must be included. Iterating this requirement generates nonlinear interactions. Under standard assumptions concerning locality, derivatives, and field content, a self-consistent interacting massless spin-2 theory leads to the structure of the Einstein field equations, possibly supplemented by a cosmological constant and higher-derivative corrections.\n\nIn perturbation theory, gravitational attraction can be represented through the exchange of virtual gravitons. The lowest-order exchange between slowly moving massive bodies reproduces the Newtonian potential,\n\n\\[V(r)=-\\frac{Gm_1m_2}{r}.\\]\n\nA virtual graviton in such a calculation is an internal element of an amplitude, not a directly detectable particle traveling between the objects along a definite trajectory. The division of a calculation into individual exchange processes can depend on gauge and computational choices, whereas the complete physical prediction does not.\n\nReal gravitons instead correspond to freely propagating radiation in settings where a particle interpretation is available. Accelerating sources can emit gravitational radiation, but conservation laws eliminate the leading monopole and dipole radiation familiar from simpler wave systems. For slowly moving isolated sources, the dominant contribution comes from changes in the mass quadrupole moment. Quantized radiation must reproduce that classical behavior in the appropriate limit.\n\nGravitational interactions are extraordinarily weak at ordinary particle energies. A characteristic dimensionless coupling grows as \\(GE^2/(\\hbar c^5)\\), or roughly as the squared ratio of the interaction energy to the Planck energy. Its smallness explains why graviton emission is negligible in most laboratory processes, even though gravity dominates the dynamics of planets, stars, and galaxies."
    },
    {
      "heading": "Historical development",
      "text": "The graviton concept emerged from attempts to apply quantum theory to the gravitational field after the development of general relativity in 1915. The discovery of quantum electrodynamics provided an influential model: if electromagnetic waves admitted a quantum description, gravitational waves might also possess elementary quanta. The analogy was suggestive but incomplete because gravity changes the geometry in which other fields propagate, rather than merely acting as another field within a fixed geometry.\n\nIn 1930, Léon Rosenfeld published an early treatment of the quantization of a linearized gravitational field and its coupling to electromagnetic radiation. Matvei Bronstein developed the subject further in work published in 1936. His analysis recovered key features of a quantized weak gravitational field and emphasized that extending the theory beyond the weak-field approximation would introduce profound conceptual difficulties. These studies established that the basic free-field graviton description was accessible before a complete quantum theory of gravity existed.\n\nMarkus Fierz and Wolfgang Pauli’s 1939 work on relativistic fields of arbitrary spin clarified the special structure of spin-2 theories, including the distinction between massive and massless fields. In later decades, Richard Feynman, Bryce DeWitt, and others developed methods for handling gravity as a quantum field theory with gauge symmetry. These efforts required careful treatment of unphysical modes, gauge fixing, and the additional fields used in covariant calculations to maintain consistency.\n\nDuring the 1960s, arguments based on soft-particle emission and the self-coupling of a spin-2 field sharpened the connection between gravitons and general relativity. They showed that the graviton’s proposed interactions were not simply chosen by analogy with electromagnetism. Much of their structure followed from Lorentz invariance, conservation laws, and the requirement that redundant descriptions not alter observable results. These conclusions remain conditional on the assumptions of the underlying field-theoretic framework.\n\nA major obstacle became precise in the 1970s and 1980s. Gerard ’t Hooft and Martinus Veltman showed in 1974 that pure Einstein gravity has a special one-loop cancellation on shell in four dimensions when the cosmological constant vanishes, although coupling to matter generally introduces divergences requiring additional terms. Marc Goroff and Augusto Sagnotti demonstrated in the mid-1980s that pure gravity develops a nonvanishing on-shell divergence at two loops. These results established the perturbative nonrenormalizability of Einstein gravity in its conventional formulation.\n\nFrom the 1990s onward, the effective-field-theory approach, notably developed for gravity by John Donoghue and others, clarified the significance of that failure. Nonrenormalizability does not prevent reliable low-energy quantum predictions. It limits the theory’s ability to describe arbitrarily high energies using only finitely many measured parameters. This distinction separated the well-defined low-energy graviton from the unresolved problem of gravity’s ultimate microscopic description."
    },
    {
      "heading": "Effective field theory and the Planck scale",
      "text": "The Einstein–Hilbert action contains two derivatives of the metric. Quantum loops generate additional terms involving higher powers of curvature and additional derivatives. Consequently, a general low-energy gravitational action takes the schematic form\n\n\\[S_{\\mathrm{eff}} = \\int d^4x\\,\\sqrt{-g} \\left[ \\frac{2}{\\kappa^2}(R-2\\Lambda) +c_1R^2 +c_2R_{\\mu\\nu}R^{\\mu\\nu} +\\cdots \\right],\\]\n\nagain in units with \\(c=\\hbar=1\\). Here \\(R\\) is the scalar curvature, \\(\\Lambda\\) is the cosmological constant, and the coefficients \\(c_i\\) encode information about short-distance physics. The omitted terms include further curvature invariants and matter interactions.\n\nThis is an example of effective field theory: a description organized by the scales relevant to a particular experiment. At sufficiently low energies and weak curvature, terms containing more derivatives are progressively suppressed. Only finitely many coefficients contribute at a specified accuracy, even though the complete action contains infinitely many possible terms. The theory therefore remains predictive order by order. Its inability to determine every coefficient from the Einstein–Hilbert term alone is not the same as an inability to calculate anything.\n\nThe characteristic quantum-gravity scales are the Planck length and Planck energy,\n\n\\[\\ell_{\\mathrm P}=\\sqrt{\\frac{\\hbar G}{c^3}} \\approx1.62\\times10^{-35}\\ \\mathrm m,\\]\n\n\\[E_{\\mathrm P}=\\sqrt{\\frac{\\hbar c^5}{G}} \\approx1.22\\times10^{19}\\ \\mathrm{GeV}.\\]\n\nThese conventional definitions differ by numerical factors from reduced Planck quantities often used in particle physics. They indicate where the ordinary perturbative expansion is expected to lose reliability, not an experimentally established minimum distance or an absolute boundary beyond which every conceivable description must fail.\n\nLow-energy quantum gravity can predict contributions insensitive to unknown short-distance details. Massless particles in loops produce characteristic nonanalytic dependence on momentum, which translates into long-range corrections in position space. Such terms cannot be reproduced merely by adjusting local higher-curvature coefficients. They provide an important example of genuine quantum gravitational predictions that do not require a complete theory at the Planck scale.\n\nFor two widely separated masses, quantum corrections to the Newtonian interaction include terms whose relative magnitude scales as \\(\\ell_{\\mathrm P}^2/r^2\\). Classical relativistic corrections instead involve ratios such as \\(Gm/(rc^2)\\). Separating these effects requires care because a coordinate-dependent potential is not itself a unique observable; scattering amplitudes and other measurable quantities supply the unambiguous predictions.\n\nThe practical obstacle is their size. At a separation of one metre, \\(\\ell_{\\mathrm P}^2/r^2\\) is approximately \\(2.6\\times10^{-70}\\). Graviton-loop corrections are consequently far beyond direct measurement in ordinary weak-field systems. Nevertheless, their calculability demonstrates that the absence of an experimentally verified high-energy completion does not invalidate the graviton as a controlled low-energy theoretical concept."
    },
    {
      "heading": "Gravitons in proposed fundamental theories",
      "text": "Different approaches to quantum gravity assign different microscopic meanings to the graviton. Some treat it as a fundamental excitation; others expect it to emerge only in an approximate, long-distance description. Agreement on a massless spin-2 sector at accessible energies can therefore coexist with deep disagreement about the underlying degrees of freedom.\n\nIn string theory, a graviton appears among the massless excitations of a closed string. Its polarization is described by the symmetric tensor part of the product of left-moving and right-moving string excitations. Other components commonly include an antisymmetric tensor field and a scalar called the dilaton. The appearance of an interacting massless spin-2 state was central to the reinterpretation of string theory as a candidate theory of quantum gravity during the 1970s. At energies small compared with the string scale, its interactions reproduce general relativity with additional fields and higher-derivative corrections.\n\nSupergravity combines gravity with supersymmetry. The graviton belongs to a multiplet that includes a spin-\\(3/2\\) partner called the gravitino; more extensive supersymmetry adds further fields. Symmetry can produce cancellations among quantum contributions, improving ultraviolet behavior relative to naive estimates. Such cancellations must be established at the relevant loop order, however, and do not by themselves constitute experimental evidence for gravitinos or supersymmetry.\n\nIn loop quantum gravity, the basic variables are not small metric disturbances on a fixed background. Quantum states are constructed using geometric and connection-based variables. A central task is demonstrating in sufficient detail how ordinary smooth spacetime and the graviton propagator emerge in an appropriate semiclassical regime. This illustrates a broader requirement: a microscopic theory need not begin with gravitons, but it must account for the successful weak-field predictions associated with them if it is to reproduce general relativity.\n\nOther programs investigate whether gravity becomes well behaved at very high energies through a nontrivial fixed point of the renormalization group. This approach, known as asymptotic safety, retains field-theoretic methods while seeking a predictive ultraviolet completion. Its implications for physical graviton states depend on the existence and properties of the proposed fixed point and on how the high-energy theory connects to measurable low-energy physics.\n\nIn certain versions of holographic duality, gravity in a higher-dimensional spacetime is related to a nongravitational quantum theory on its boundary. A bulk graviton corresponds to structures associated with the boundary stress–energy tensor. Such examples show that a graviton can emerge within a description whose fundamental formulation looks very different from quantized spacetime geometry. They are established theoretical constructions in particular settings, not evidence that the observed universe necessarily has the same boundary description."
    },
    {
      "heading": "Gravitational waves and the early universe",
      "text": "A gravitational wave is a propagating disturbance of spacetime curvature. Its detection does not, by itself, demonstrate the existence of gravitons, just as measuring a classical radio wave does not necessarily reveal individual photons. Classical general relativity predicts gravitational waves without introducing quantum states. A quantum account must reproduce those classical predictions while also allowing states and correlations unavailable to a purely classical field.\n\nThe signal GW150914, recorded on 14 September 2015 and announced in February 2016, provided the first direct detection of gravitational waves. It was produced by the merger of two black holes and measured by the two LIGO interferometers. Its observed waveform agreed with the predictions of general relativity for a strongly dynamical source. The detectors measured differential changes in arm length, expressed as a dimensionless strain; they did not resolve the absorption of individual gravitons.\n\nIn a quantum description, a classical-like wave is naturally represented by a coherent state or a related highly occupied state. For scale, one solar mass of energy carried entirely in \\(100\\)-hertz quanta would correspond to roughly \\(3\\times10^{78}\\) gravitons. This is only an illustrative conversion, not a particle count measured for an astronomical event. Real wave signals contain a range of frequencies, and the number attributed to a field state depends on its quantum description.\n\nThe early universe provides another setting in which gravitational quantum fluctuations may be important. Many models of cosmic inflation predict that initially microscopic tensor fluctuations are stretched to cosmological scales. Their amplitude depends on the expansion rate during inflation. In a common leading-order convention, the dimensionless tensor power is\n\n\\[\\mathcal P_t(k)\\simeq \\frac{2H_*^2}{\\pi^2\\overline M_{\\mathrm P}^{\\,2}},\\]\n\nwhere \\(H_*\\) is the expansion rate when the mode exits the inflationary horizon and \\(\\overline M_{\\mathrm P}=1/\\sqrt{8\\pi G}\\) in natural units. The expression assumes the usual Einstein-gravity treatment and an appropriate initial quantum state.\n\nPrimordial tensor perturbations can produce a distinctive component of the polarization of the cosmic microwave background, called B-mode polarization. Gravitational lensing also produces B modes, while Galactic dust and synchrotron emission complicate observations. A securely identified primordial tensor signal would constrain early-universe physics, but demonstrating its quantum origin would require further theoretical and observational arguments. A stochastic gravitational-wave background can also arise from classical processes, including astrophysical populations and early-universe phase transitions.\n\nInflationary tensor states are often highly squeezed rather than simple one-graviton states. Their observable properties concern correlations across modes, not isolated particle impacts. This makes cosmology a possible probe of quantum gravitational field fluctuations while emphasizing that “evidence for gravitons” can mean something more subtle than counting individual quanta."
    },
    {
      "heading": "Experimental searches and limits of inference",
      "text": "The most direct graviton experiment would identify a discrete quantum exchange with a gravitational radiation field and distinguish it from all nongravitational backgrounds. The difficulty is not merely the small energy of low-frequency gravitons. Their interaction probability with a detector is also exceptionally small. Increasing the target mass or observation time introduces practical obstacles, environmental backgrounds, and, in extreme thought experiments, constraints associated with gravitational collapse.\n\nFreeman Dyson and other researchers examined whether a single graviton could ever be detected under physically realizable conditions. Analyses of conventional absorption schemes found severe limitations involving detector size, source brightness, shielding, and backgrounds such as neutrinos. These arguments establish formidable obstacles for broad classes of detector designs. They are not universally accepted as a theorem forbidding every possible operational test of gravitational quantization.\n\nProposals involving resonant mechanical systems seek to exploit collective motion and quantum-limited measurements. A mechanical oscillator can exchange discrete amounts of energy even when its driving field is classical, however. Observing an energy jump in a quantum detector is therefore not automatically sufficient to prove that the incident gravitational field was quantized. A convincing test must specify what feature excludes a classical-field explanation, taking account of the detector’s own quantum behavior.\n\nAnother route uses quantum entanglement rather than radiation absorption. Proposals published in 2017 by Sougato Bose and collaborators, and independently by Chiara Marletto and Vlatko Vedral, considered two masses placed in spatial quantum superpositions. Their mutual gravitational interaction would generate branch-dependent phases and could entangle the masses. Under assumptions about local mediation and the absence of other interactions, observing such entanglement would imply that the mediator possesses nonclassical degrees of freedom. It would not constitute detection of a freely propagating single graviton or determine a complete theory of quantum gravity.\n\nGravitational-wave observations also constrain modifications of graviton propagation. A massive graviton would obey\n\n\\[E^2=p^2c^2+m_g^2c^4,\\]\n\ncausing frequency-dependent propagation speeds and potentially distorting waveforms over astronomical distances. The LIGO–Virgo analysis of the GWTC-3 catalog reported an upper bound of approximately\n\n\\[m_g<1.27\\times10^{-23}\\ \\mathrm{eV}/c^2\\]\n\nat 90 percent credibility within its adopted dispersive-propagation model. This is a constraint derived from waveform propagation, not a direct mass measurement or a model-independent exclusion of every theory containing massive spin-2 fields.\n\nThe joint observation of gravitational waves from GW170817 and a gamma-ray burst in 2017 tightly constrained the difference between gravitational-wave and light propagation speeds, subject to assumptions about their emission times. Such results restrict modified gravitational theories while remaining compatible with a classical description of the observed waves. The experimental questions—whether gravity is quantized, whether an isolated graviton can be identified, and which microscopic theory governs gravity—require related but distinct forms of evidence."
    }
  ],
  "links": [
    "gravity",
    "photon",
    "quantum-mechanics",
    "general-relativity",
    "quantum-field-theory",
    "quantum-gravity",
    "boson",
    "speed-of-light",
    "planck-constant",
    "standard-model",
    "spacetime",
    "action-physics",
    "quantum-harmonic-oscillator",
    "equivalence-principle",
    "einstein-field-equations",
    "perturbation-theory",
    "virtual-particle",
    "effective-field-theory",
    "string-theory",
    "loop-quantum-gravity",
    "renormalization-group",
    "gravitational-wave",
    "cosmic-inflation",
    "cosmic-microwave-background",
    "quantum-entanglement"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}