In physics, the graviton is a hypothetical elementary particle that transmits the force of gravity in the framework of quantum field theory. If it exists, the graviton must be massless (because the gravitational force has unlimited range) and must have a spin of 2 (because gravity is a second-rank tensor field).

Gravitons are postulated because of the great success of the quantum field theory (in particular, the Standard Model) at modeling the behavior of all other forces of nature with similar particles: electromagnetism with the photon, the strong interaction with the gluons, and the weak interaction with the W and Z bosons. However, attempts to extend the Standard Model with gravitons run into serious theoretical difficulties at high energies (processes with energies close or above the Planck scale) because of infinities arising due to quantum effects (in technical terms, gravitation is nonrenormalizable.) Some proposed theories of quantum gravity (in particular, string theory) address this issue. In string theory, gravitons (as well as the other particles) are states of strings, whose behavior at low energies can be approximated by a quantum field theory of point particles.

Since gravity is very weak, there is little hope of detecting single gravitons experimentally in the foreseeable future^{[citation needed]}.

Some question the analogy which motivates the introduction of the graviton. Unlike the other forces, gravitation plays a special role in general relativity in defining the spacetime in which events take place. Because it does not depend on a particular spacetime background, general relativity is said to be background independent. In contrast, the Standard Model is not background independent. In other words, general relativity and the standard model are incompatible. A theory of quantum gravity is needed in order to reconcile these differences. Whether this theory should itself be background independent, or whether the background independence of general relativity arises as an emergent property is an open question. The answer to this question will determine whether or not gravity plays a "special role" in this underlying theory similar to its role in general relativity.

While the classical theory (i.e. the tree diagrams) and semiclassical corrections (one-loop diagrams) behaved as expected, the Feynman diagrams with two (or more) loops led to ultraviolet divergences; that is, infinite results that could not be removed because the quantized general relativity was not renormalizable, unlike quantum electrodynamics. In popular terms, the discreteness of quantum theory is not compatible with the smoothness of Einstein's general relativity. These problems, together with some conceptual puzzles, led many physicists to believe that a theory more complete than just general relativity must regulate the behavior near the Planck length. Superstring theory finally emerged as the most promising solution; it is the only known theory in which the quantum corrections of any order to graviton scattering are finite.

String theory predicts the existence of gravitons and their well-defined interactions which represents one of its most important triumphs. A graviton in perturbative string theory is a closed string in a very particular low-energy vibrational state. The scattering of gravitons in string theory can also be computed from the correlation functions in conformal field theory, as dictated by the AdS/CFT correspondence, or from Matrix theory.

An interesting feature of gravitons in string theory is that, as closed strings without endpoints, they would not be bound to branes and could move freely between them; this "leakage" of gravitons from our brane into higher-dimensional space could explain why gravity is such a weak force, and gravitons from other branes adjacent to our own could provide a potential explanation for dark matter. See brane cosmology for more details.

Some proposed quantum theories of gravity do not predict a graviton. For instance, loop quantum gravity has no analogous particle.

Detecting a graviton, if it exists, would prove rather problematic. Because the gravitational force is so incredibly weak, as of today, physicists are not even able to directly verify the existence of gravitational waves, as predicted by general relativity. (Many people are surprised to learn that gravity is the weakest force. The dominance of gravity at large scales is due to the fact that the nuclear forces have a limited range, and the electromagnetic force often largely cancels due to the existence of positive and negative charges. In contrast, gravitational charge -- i.e., mass -- is positive or zero for all known forms of matter.)

Gravitational waves may be viewed as coherent states of many gravitons, much like the electromagnetic waves are coherent states of photons. Projects that should find the gravitational waves, such as LIGO and VIRGO, are just getting started.

Many believe the graviton does not exist, at least in the simplistic manner in which it is envisioned. Superficially speaking, quantum gravity using the gauge interaction of a spin-2 field (graviton) fails to work like the photon and other gauge bosons do.

But more importantly the spin-2, linear wave (classical gravitational wave) is only a perturbation on certain, highly restrictive metrics. In general there are wave-like fluctuations, but they are non-linear, as is often the case in General Relativity. Maxwell's equations always admit a spin-1, linear wave, but Einstein's equations rarely admit a spin-2, linear wave, and when they do it is only perturbative and not exact.

The more analogous gravitational object to the electromagnetic wave is actually the Weyl curvature. In classical electromagnetism the fields are determined by sources along with source-free electromagnetic waves. In gravitation, the Ricci curvature is determined by the stress-energy tensor along with the source-free Weyl tensor which contains the gravitational waves.

• Gravitomagnetism

Template:Elementary