A magnetar is a neutron star whose observable activity is powered substantially by the evolution and dissipation of an exceptionally strong magnetic field. Its characteristic behavior includes persistent X-ray emission, sudden bursts, and occasional giant flares that release enormous amounts of energy within seconds. Most well-established magnetars have inferred large-scale surface fields of approximately – gauss, or – tesla, although magnetar-like activity also occurs in objects with much weaker inferred dipole fields.
Magnetars connect the physics of stellar collapse, dense nuclear matter, and extreme electromagnetism. Their magnetic fields can deform the stellar crust, alter the propagation of radiation through otherwise empty space, and store enough energy to sustain activity long after the star’s initial heat begins to fade. They are also directly implicated in at least some fast radio bursts. Understanding them requires a distinction between the large-scale field inferred from rotation and the stronger, more complicated fields that may be concealed within the star or concentrated near its surface.
Physical properties and energy supply
A magnetar has the basic structure of a neutron star: a compact remnant containing roughly a stellar mass or more within a radius of about 10–13 kilometres. Its outer layers include an atmosphere and a solid crust, beneath which lies a dense fluid interior. Much of its mass is supported against gravity by the properties of strongly interacting matter rather than by ordinary thermal gas pressure. Magnetars are not a separate category of material object from other neutron stars; they represent a distinctive magnetic and observational state.
The central physical distinction is the source of available energy. An ordinary rotation-powered pulsar draws its radiation and particle outflow primarily from rotational kinetic energy. An accreting neutron star draws energy from matter falling into its gravitational field. In a magnetar, the magnetic reservoir can supply persistent radiation and violent transients even when the measured loss of rotational energy is insufficient. The distinction is not absolute: a magnetar can also emit rotation-powered radiation, and some objects alternate between observational behaviors associated with different neutron-star classes.
For a star with moment of inertia , rotation period , and period derivative , the rate of rotational energy loss is
Using the customary estimate , many magnetars have a persistent X-ray luminosity exceeding this quantity. That mismatch was an important reason conventional rotation-powered models could not explain the class. Conversely, some faint magnetars have X-ray luminosities below their spin-down power; magnetic activity, rather than a single luminosity threshold, is the more fundamental identifying feature.
The magnetic energy contained within a roughly spherical volume provides a useful scale. In Gaussian units, a uniform field gives
This is only an order-of-magnitude estimate because real fields are neither uniform nor simply spherical. Nevertheless, it shows why an internal field of – gauss is a substantial energy reservoir. Only part of it is necessarily available for observable radiation. The field geometry, electrical conductivity, mechanical response of the crust, and coupling to the surrounding plasma determine how rapidly that reservoir can be tapped.
Discovery and recognition of the class
Magnetars were recognized by connecting two initially separate observational puzzles. Soft gamma repeaters produced repeated short bursts of hard X-rays and relatively soft gamma rays, unlike the usually nonrepeating cosmological gamma-ray bursts known at the time. Anomalous X-ray pulsars were persistent pulsating X-ray sources whose luminosities were difficult to explain through rotational energy loss and which generally lacked the evidence for companion-fed accretion seen in X-ray binaries.
An especially important event occurred on 5 March 1979. Detectors aboard several spacecraft recorded an intense initial flash followed by a pulsating tail with a period of about eight seconds. Its source, subsequently designated SGR 0526−66, lay in the direction of the supernova remnant N49 in the Large Magellanic Cloud. The event established that at least one repeating high-energy source could release an extraordinary amount of energy and then continue emitting with a neutron-star-like rotational modulation. Its interpretation remained uncertain for years.
Robert Duncan and Christopher Thompson developed the magnetar model in the early 1990s, with a foundational paper appearing in 1992. They proposed that very strong fields could arise in newly born neutron stars and subsequently power high-energy activity. Later work developed explanations for bursts, giant flares, and persistent emission through magnetic stresses and dissipation. The theory supplied a common physical framework for phenomena that had previously been grouped primarily by their observational appearance.
A decisive observational advance came in 1998, when Chryssa Kouveliotou and collaborators reported the approximately 7.5-second rotation period and rapid spin-down of SGR 1806−20. Interpreted with the standard magnetic braking estimate, the measurements implied a dipole field of order gauss. On 27 August of the same year, a giant flare from SGR 1900+14 displayed another long, periodically modulated tail. These findings tied repeating bursts, slow rotation, strong braking, and giant-flare phenomenology together.
The boundary between soft gamma repeaters and anomalous X-ray pulsars weakened further when sources in the latter group produced short bursts. In particular, the 2002 outburst of 1E 2259+586 demonstrated that a persistent anomalous X-ray pulsar could undergo a burst-rich episode closely resembling soft-gamma-repeater activity. The historical names remain useful descriptions, but they no longer imply fundamentally separate engines. Modern classification emphasizes the magnetic energy source and the range of activity that a single neutron star can display.
Birth and amplification of the magnetic field
Most magnetars are understood to originate in the collapse of massive stars, usually in association with a core-collapse supernova. The newborn neutron star is hot, rich in neutrinos, and subject to vigorous fluid motion. Magnetic flux inherited from the progenitor is compressed as the core contracts, but compression alone does not uniquely determine the final field. Rotation, convection, differential motion, and the progenitor’s evolutionary history can all influence the outcome.
One influential formation model invokes a rapidly rotating proto-neutron star. During its first seconds, convection and differential rotation can amplify seed fields through dynamo action. A birth rotation period of a few milliseconds or less makes rotational effects especially important and provides a large reservoir of mechanical energy. Such models can generate fields stronger than those typical of ordinary pulsars. However, they do not establish that every observed magnetar was born spinning at nearly the fastest physically possible rate.
Another possibility emphasizes strong magnetic flux already present in the progenitor, often called a fossil-field contribution. In a simplified flux-conserving picture, shrinking a magnetized core increases its field approximately as the inverse square of its characteristic radius. The difficulty is that stellar evolution redistributes both magnetic flux and angular momentum, while collapse involves instabilities and complex fluid motions. A progenitor’s measured surface field therefore cannot be translated directly into a unique neutron-star field.
Binary evolution provides additional routes to favorable initial conditions. Mass transfer, tidal interactions, and stellar mergers can change the rotation and internal structure of a massive star before collapse. The magnetar CXOU J164710.2−455216 in the massive cluster Westerlund 1 has been particularly important in discussions of progenitor evolution and binary interaction. Cluster membership constrains the environment and approximate age, but does not by itself reconstruct every stage of an individual star’s history.
Magnetar birth is also relevant to unusually energetic stellar explosions. A rapidly rotating neutron star can lose rotational energy into expanding ejecta and has been proposed as the central engine of some superluminous supernovae. Related models use a newborn magnetar to sustain emission in some gamma-ray bursts. In these applications, the chief early energy reservoir may be rotation rather than the magnetic energy that powers mature Galactic magnetar outbursts.
These ideas remain model-dependent. Explosion light curves can sometimes be explained by more than one power source, and observations of magnetar-associated remnants do not require every magnetar to have deposited an enormous rotational energy into its surroundings. More than one formation channel may lead to strong internal fields, and similar birth conditions may still produce different observable outcomes because field geometry matters as well as field strength.
Internal field evolution and crustal response
The field of a magnetar is not static. It evolves as electrical currents move and dissipate within the crust and as the charged and neutral constituents of the interior interact. This evolution couples magnetic structure to temperature, mechanical stress, and cooling. A star with a strong initial field can remain hotter than an otherwise comparable weak-field neutron star because magnetic energy is gradually converted into heat.
In the crust, ions occupy a lattice while mobile electrons carry much of the electrical current. Two important processes are Ohmic dissipation and Hall evolution. Ohmic dissipation converts current energy into heat because the electrical conductivity is finite. Its characteristic timescale across a length is
where is the conductivity and is the speed of light, expressed here in Gaussian units. Conductivity depends on temperature, lattice impurities, and the microscopic structure of the crust.
Hall evolution arises from the motion of the current-carrying electrons relative to the ionic lattice. A representative timescale is
where is the magnitude of the electron charge and is electron number density. The Hall term does not directly dissipate magnetic energy in the idealized limit, but it can transfer energy to smaller spatial scales, where resistive losses become faster. It can also rearrange an initially smooth field into localized structures, making the surface field much more complicated than a single dipole.
Magnetic stresses act on the solid crust. As the internal field evolves, the crust may respond elastically, deform plastically, or undergo rapid localized failure. These processes can move magnetic footpoints at the surface and disturb the external field. The familiar term “starquake” is useful shorthand, but the analogy with brittle earthquakes is incomplete: magnetar matter is under immense pressure, and a rapid event may involve plastic motion and magnetically coupled failure rather than the opening of ordinary cracks.
The core adds further uncertainty. Neutrons may exhibit superfluidity, while protons may exhibit superconductivity under some conditions. These phases change how magnetic flux and rotational vortices are organized and how they interact. Extremely strong fields may alter or suppress superconductivity in parts of the core. Slow relative motion of charged particles and neutrons can also transport magnetic flux, with rates sensitive to temperature and weak-interaction processes.
Coupled magnetic and thermal calculations therefore depend on uncertain initial conditions and microphysics. A strong buried toroidal field can heat the star and stress the crust without producing an equally strong external dipole. This provides a physical route to magnetic activity in stars whose timing measurements suggest only a moderate large-scale field.
Magnetosphere and persistent radiation
The magnetosphere is the region outside the star in which the magnetic field controls the motion of charged particles. In a magnetar it is not merely a passive extension of a dipole. Motions of the crust can twist external field lines, introducing currents along them. Those currents require a supply of charged particles and establish a direct connection between internal magnetic evolution and the radiation detected far from the star.
A persistent magnetar spectrum often includes a thermal-like soft X-ray component associated with the hot surface, together with a harder component produced or modified outside it. Typical fitted thermal energies are of order – kiloelectronvolt, although the inferred values depend on the spectral model and on interstellar absorption. The surface is not necessarily at one uniform temperature: magnetic insulation, localized heating, and returning particles can create hot regions whose visibility changes as the star rotates.
An important radiative mechanism is resonant cyclotron scattering. Surface photons encounter magnetospheric charges at locations where the photon energy in the particle’s rest frame matches the cyclotron resonance. Repeated scattering can redistribute photon energies and directions, transforming a thermal seed spectrum into a broader continuum. Some magnetars also exhibit hard X-ray emission extending to tens or hundreds of kiloelectronvolts, requiring energetic particles and more elaborate radiative models than a simple surface blackbody.
Strong electric fields parallel to the magnetic field can accelerate particles, but an unrestricted voltage is difficult to sustain. Energetic photons can create electron–positron pairs, and the resulting plasma helps screen the accelerating electric field. Particle creation, acceleration, radiation, and screening form a regulated discharge. Returning particles deposit energy at the surface, while escaping or circulating particles contribute to the magnetospheric current system.
Magnetic twist provides a useful interpretation of many outbursts. An abrupt disturbance can increase the current and heat a surface footprint. As the external field relaxes, the current-carrying region can shrink and the X-ray flux decline. Observed decays range from weeks to years and need not follow a single universal law. Heat deposited within the crust can produce a different cooling history, so an individual light curve may combine internal cooling with the relaxation of external currents.
Geometry complicates the interpretation. Strong gravitational light bending allows more than one hemisphere’s worth of surface to be visible at once, and the atmosphere emits anisotropically in a strong field. Consequently, pulse shape, spectrum, and emitting-area estimates must be modeled together. A small blackbody area inferred from a simple fit is not automatically a literal measurement of one uniformly heated patch on the surface.
Bursts, outbursts, and giant flares
Magnetar transients span a wide range of durations and energies. The most common recognized bursts last from milliseconds to fractions of a second and radiate chiefly in hard X-rays and soft gamma rays. Their energies are often approximately – erg, depending on distance and the band used to measure them. During an active episode, a source may produce many such events in a burst storm. Intermediate-duration events bridge the gap between these short flashes and the rare giant flares.
An outburst is a longer-lived elevation of the persistent emission, often accompanied by bursts, spectral changes, altered pulse profiles, or timing anomalies. A previously faint object can brighten by orders of magnitude and then fade over months or years. This behavior creates an important observational selection effect: some magnetars enter the known sample only when an outburst makes them conspicuous. The absence of recorded bursts from a source during a short observing interval does not establish that it is permanently inactive.
Giant flares are distinguished by an extremely bright initial spike followed by a longer, softer tail modulated at the stellar rotation period. Three landmark events occurred on 5 March 1979, 27 August 1998, and 27 December 2004, from SGR 0526−66, SGR 1900+14, and SGR 1806−20 respectively. The initial spike of the 2004 event released energy of order erg, with the estimate depending substantially on the source distance and radiation pattern. It saturated several spaceborne detectors and produced measurable ionospheric effects at Earth.
The leading physical picture involves a rapid reconfiguration of magnetic fields and the creation of an optically thick photon–pair plasma. Part of this plasma can remain confined on closed field lines, forming a trapped “fireball.” As it cools and contracts, rotation changes the observer’s view and produces the pulsating tail. Magnetic reconnection, which changes field-line connectivity and converts magnetic energy into particle and radiative energy, is a candidate process for the initial rapid release. Whether an event begins inside the star or through an instability outside it remains an active question.
Giant-flare tails have also shown quasi-periodic oscillations at multiple frequencies. These have been interpreted in terms of coupled magnetic and elastic oscillations of the star, rather than simple oscillations of an isolated crust. Their frequencies depend on crustal rigidity, field geometry, core coupling, and the properties of dense matter. The observations therefore offer a possible form of neutron-star seismology, but identifying every frequency with a unique physical mode remains difficult.
A giant flare in another galaxy can resemble a short gamma-ray burst because only its initial spike may be detectable. Nearby-galaxy associations and the absence of a detectable long tail can constrain this interpretation, but do not make every short gamma-ray burst a magnetar flare.
Rotation, population, and evolutionary connections
Most established magnetars rotate with periods of roughly 2–12 seconds, much more slowly than many young ordinary pulsars. Their strong magnetic braking can remove angular momentum rapidly. Under the idealized assumptions of a rotating vacuum dipole, a commonly used estimate of the equatorial surface field is
with in seconds. The numerical coefficient assumes conventional neutron-star parameters and a particular magnetic geometry; the corresponding polar dipole field is twice the equatorial value.
This estimate is valuable but is not a direct measurement of the complete field. Real neutron stars possess plasma-filled magnetospheres, non-dipolar surface components, and sometimes particle winds. Magnetospheric changes can alter the braking torque without requiring a comparable change in the underlying dipole. Timing therefore constrains a combination of magnetic structure and rotational dynamics. It cannot reveal the full internal magnetic energy reservoir.
The characteristic age,
also requires caution. It approximates a true age only under restrictive assumptions, including a stable braking law and a birth period much shorter than the current period. Field decay, variable torque, and an appreciable initial period can invalidate those assumptions. Associations with supernova remnants, stellar clusters, or measured motion can provide independent age information, although such associations themselves require careful positional and dynamical assessment.
An important challenge to a simple field-threshold definition came from SGR 0418+5729. Its bursting behavior established magnetar-like activity, while timing eventually indicated a dipole field of only about gauss. Other relatively low-dipole-field magnetars reinforced the conclusion that a modest external dipole can coexist with strong localized or internal fields. Conversely, high-field radio pulsars do not all exhibit frequent magnetar outbursts. The 2006 magnetar-like episode of PSR J1846−0258 and the 2016 outburst of PSR J1119−6127 demonstrated that observational categories can overlap.
Magnetars also display rotational irregularities. A glitch is a sudden increase in rotation rate, often interpreted through angular-momentum exchange between the crust and an interior superfluid. Some magnetar glitches accompany radiative changes, while others have little detectable electromagnetic activity. Sudden net spin-down events and complicated post-event recoveries may involve changes in external torque, internal coupling, or both.
The known population numbers only a few dozen confirmed or candidate objects, rather than constituting a complete census of Galactic magnetars. Absorption, limited monitoring, transient activity, and fading with age all affect discovery. Population estimates must therefore account for objects that spend long intervals below survey sensitivity or have evolved beyond conspicuous magnetar activity.
Radio emission and fast radio bursts
Magnetars were initially associated mainly with high-energy radiation, but some also produce pulsed radio emission. Radio pulsations were discovered from XTE J1810−197 in 2006 after its earlier X-ray outburst. Other examples include 1E 1547.0−5408 and the Galactic-centre magnetar SGR J1745−2900. Their radio emission can be highly variable, strongly polarized, and spectrally different from that of many ordinary radio pulsars. Radio visibility is neither universal nor a requirement for magnetar classification.
The radio pulses raise questions about where coherent emission occurs. Individual charged particles radiating independently cannot readily account for the enormous brightness temperatures of pulsar-like radio signals. A coherent mechanism, in which the radiation from many particles is organized collectively, is required. The detailed processes responsible for ordinary magnetar radio pulsations remain uncertain, and need not be identical to those producing their most extreme isolated radio flashes.
The strongest direct connection to fast radio bursts came on 28 April 2020, when SGR 1935+2154 emitted an exceptionally bright millisecond radio event during high-energy activity. The CHIME radio telescope and the STARE2 experiment detected the radio burst, while space observatories recorded associated X-rays. Often called FRB 200428, it demonstrated that a Galactic magnetar can produce a radio transient with properties linking it to the extragalactic fast-radio-burst population.
That observation did not establish that all fast radio bursts have identical origins. FRB 200428 was much less energetic than the most luminous known extragalactic events, although it overlapped the weaker end of the broader phenomenology. Fast radio bursts also occur in diverse environments and show widely varying repetition patterns. A successful population model must explain not only the existence of magnetar radio flashes but also their rates, energy distribution, lifetimes, and host-galaxy demographics.
Two broad classes of mechanism receive particular attention. Magnetospheric models place the coherent radio source close to the star, where a sudden magnetic disturbance accelerates particles and organizes their emission. Shock models place the source farther out: a relativistic outflow from an eruption encounters surrounding material and produces coherent radiation at a shock. The required conditions differ in magnetic geometry, density, efficiency, and the delay between radio and high-energy signals.
Simultaneous observations are especially constraining. The relative timing of X-ray and radio peaks, radio polarization, frequency structure, and the presence or absence of radio counterparts to other X-ray bursts test the models. Most magnetar high-energy bursts do not have comparably dramatic detected radio counterparts. The 2020 event therefore identifies a demonstrated physical capability, not a one-to-one correspondence between every magnetic eruption and a fast radio burst.
Extreme-field physics and observational tests
Magnetars provide access to magnetic conditions unavailable in terrestrial experiments. A useful reference scale in quantum electrodynamics is the electron critical field,
written in Gaussian units. At this field, the electron cyclotron energy is comparable to the electron rest energy. Many magnetars exceed this scale at their surfaces. The value is not a strict boundary between classical and quantum behavior, but it marks a regime in which magnetic quantization and strong-field radiative effects become especially important.
One prediction is vacuum birefringence. Quantum fluctuations make the strongly magnetized vacuum respond differently to distinct photon polarization modes, so that radiation can acquire mode-dependent propagation properties even without an ordinary material medium. In the surrounding plasma, vacuum and matter contributions interact and can produce resonances that modify polarization and spectral features. Magnetar atmospheres therefore require radiative-transfer calculations incorporating both strong fields and quantum effects.
Other relevant processes include photon splitting and magnetic conversion of a sufficiently energetic photon into an electron–positron pair. These affect how high-energy radiation escapes and how a magnetosphere maintains its particle supply. Their importance depends on photon energy, polarization, propagation angle, and local field strength. A field above does not by itself imply that every photon undergoes one of these processes.
X-ray polarimetry adds information unavailable from intensity spectra alone. Observations with the Imaging X-ray Polarimetry Explorer, including measurements of 4U 0142+61, have revealed strong energy-dependent polarization behavior. Such measurements constrain the surface-emission mechanism, scattering geometry, and magnetic orientation. Interpretation still requires distinguishing atmospheric or condensed-surface effects from magnetospheric processing; a polarization signature is not automatically a unique measurement of one quantum-electrodynamic process.
Spectral lines offer another possible field diagnostic. A proton cyclotron feature would occur at approximately
where is the surface gravitational redshift. Claimed absorption features can potentially probe localized surface fields, but continuum modeling, instrumental calibration, and phase dependence complicate their interpretation. Timing, spectroscopy, and polarimetry constrain different aspects of the system and are most informative when analyzed together.
Dense-matter physics is similarly entangled with magnetic uncertainty. Oscillation frequencies, cooling behavior, and rotational irregularities can in principle constrain the equation of state, crustal elasticity, and superfluid properties. In practice, unknown field geometry often produces effects comparable to those of changing the interior microphysics. More discriminating tests depend on combining long-term timing with broadband spectra, polarization measurements, and rapid observations of rare high-energy events.