Ultraluminous X-ray sources
What powers X-ray sources that outshine the limit for neutron stars and stellar-mass black holes?
Summary
Why it is open
An ultraluminous X-ray source (ULX) is a point-like X-ray source in or near another galaxy, away from its nucleus, that appears very luminous if its emission is assumed to be the same in all directions [8], usually above 10³⁹ erg per second [14]. The Eddington limit, at which the pressure of radiation on infalling gas balances gravity [3], is about 1.3 × 10³⁸ erg per second for each solar mass of the accreting object [8]. A ULX must therefore hold an unusually massive black hole, exceed the limit for its mass, or send its light preferentially towards us [8, 9].
The problem has three parts: which compact object accretes, a neutron star, a stellar-mass black hole or an intermediate-mass black hole (IMBH) of 10² to 10⁵ solar masses, whose existence is itself an open question [8]; how neutron stars exceed their limit, which Andrew King, Jean-Pierre Lasota and Matthew Middleton called "still disputed" in 2023, essentially between geometrical beaming and very strong magnetic fields [14]; and whether ULXs are one population or several, which they call the main question [14].
No body sets a bar for a resolution. For a pulsating ULX, an isotropic luminosity implies an extreme rate of mass transfer from the companion star, while beaming allows a much lower one; Matteo Bachetti and colleagues write that "only an independent measurement of the mass transfer rate can help discriminate between these two scenarios" [13]. Cyclotron lines are, for Ling-Da Kong and colleagues, "the only way to directly and reliably measure the magnetic field near the surface" of a neutron star [12].
What is known
First models (1981–2001). The first ULXs were found once the Einstein Observatory resolved images of other galaxies [8]. In 1999 Edward Colbert and Richard Mushotzky estimated masses of about 10² to 10⁴ solar masses for compact X-ray sources near the centres of nearby spiral galaxies, assuming they were accreting black holes in a soft state [1]; such objects became known as IMBHs [14]. In 2001 King and colleagues argued instead that, with mild beaming, ULXs may be "a short-lived but extremely common stage" in the evolution of X-ray binaries, during thermal-timescale mass transfer, when a massive companion expands and overflows onto the compact object [2, 8].
Census. A catalogue published in 2021 lists 1,843 ULX candidates in 951 host galaxies [11]. Sources reaching 10⁴¹ erg per second are called hyperluminous (HLXs); contamination of HLX catalogues by foreground stars and background active galaxies reaches about 70% [8].
Pulsating ULXs (2014–2017). In 2014 Bachetti and colleagues found pulsations with a period of 1.37 seconds and a 2.5-day orbital modulation from a variable ULX in the galaxy M82, M82 X-2, that reaches 1.8 × 10⁴⁰ erg per second, "about 100 times the Eddington limit for a 1.4-solar-mass object" [4, 8]. Such pulsations come from a rotating neutron star and exclude a black hole [8]. NGC 7793 P13 [5] and NGC 5907 ULX-1 followed, the latter peaking at about 1,000 times a neutron star's Eddington limit if isotropic [7]. All three have nearly sinusoidal pulse profiles, which "suggests that the emission is not strongly beamed", Philip Kaaret, Hua Feng and Timothy Roberts write [8]. In 2023 King and colleagues counted 6 pulsating sources among about 1,800 ULXs, excluding a group with Be-star companions [14].
Winds. In 2016 Ciro Pinto, Middleton and Andrew Fabian found X-ray absorption lines blueshifted by about 0.2 times the speed of light in NGC 1313 X-1, with similar features in NGC 5408 X-1, and concluded that the compact object is surrounded by "powerful winds" [6]. Bubble nebulae and broad optical lines agree, and the outflows often need kinetic power comparable to or above the X-ray luminosity [8].
Magnetic fields. A cyclotron line is absorption by charged particles moving between quantised energy levels in a strong magnetic field [9]. In 2018 Murray Brightman and colleagues reported a 4.5 keV line (3.8σ) from a ULX in M51: it implies about 10¹¹ gauss if electrons scatter, or about 10¹⁵ gauss if protons do, which its narrow shape favours [9]. In 2022 Kong and colleagues found a line at about 146 keV from Swift J0243.6+6124, the first Galactic pulsating ULX, implying a surface field of about 1.6 × 10¹³ gauss, at least ten times the estimates of its dipole field; they argue that this "unambiguously proves" strong multipole components [12].
HLX-1. In 2009 Sean Farrell and colleagues reported a variable source in the galaxy ESO 243-49 reaching 1.2 × 10⁴² erg per second, with a "conservative lower limit" of about 500 solar masses for its black hole [3]. Optical spectroscopy confirmed that it belongs to the galaxy, about 95 megaparsecs away; unlike typical ULXs it shows the states of standard accretion discs, and disc models give a few times 10⁴ solar masses, "with a fair degree of uncertainty" [10]. Its outbursts first recurred about once a year, then at intervals that lengthened unpredictably [10].
Recent claims. In October 2022 Bachetti and colleagues reported in The Astrophysical Journal that the orbit of M82 X-2, followed for seven years, is decaying, and argued that the decay is driven by mass transfer more than 150 times the limit set by the Eddington luminosity, enough to explain the luminosity "with no need for beaming" [13]. King and Lasota had replied to the preprint in 2021; as their 2023 review summarises, period changes reflect mass transfer only when averaged over a time they estimate at about 1,000 years for this system, and the method gives rates that contradict three well-studied X-ray transients [14]. In 2024 Wen-Cong Chen proposed that magnetic braking by a strongly magnetised donor star of the Ap/Bp type could cause the decay instead [15].
Main approaches
Intermediate-mass black holes. Some ULXs are black holes of 10² to 10⁵ solar masses accreting below their limit [1, 8]. Strongest evidence: hyperluminous sources, whose luminosity is hard to explain by accretion onto stellar-mass black holes [8], above all HLX-1 [3, 10]. Main difficulty: neutron stars reach the same luminosities [4], and Jenny Greene, Jay Strader and Luis Ho conclude that "there is strong evidence that most ULXs are not IMBHs" [10]. HLX-1 itself may be the former central black hole of a small galaxy tidally stripped by ESO 243-49 [10], and King and colleagues cite disc-outburst models whose timescales they find incompatible with 10⁴ solar masses or more [14].
Geometrical beaming. When stellar-mass black holes or neutron stars are fed beyond the limit, disc winds block the X-rays except along narrow funnels near the disc axis, so an observer looking down a funnel overestimates the luminosity [2, 14]. Strongest evidence: King, Lasota and Middleton find the case "fairly strong" and the model the only one that applies straightforwardly to both black holes and neutron stars [14]; Pinto and colleagues read the 0.2c winds as matching predictions for very high accretion rates [6]. Main difficulty: strong beaming is hard to reconcile with sinusoidal pulse profiles [5, 8], and Bachetti and colleagues argue that M82 X-2 needs no beaming [13], which King and colleagues dispute [14].
Strongly magnetised neutron stars. Fields above about 10¹³ gauss reduce electron scattering for some photons, so that radiation can escape at more than the Eddington luminosity [8]. Strongest evidence: the Swift J0243.6+6124 line [12], and Gian Luca Israel and colleagues' finding that standard models fail for NGC 5907 ULX-1 "even assuming beamed emission", while a strong multipolar field can describe it [7]. Main difficulty: King and colleagues argue that the route works only up to about 3 × 10⁴⁰ erg per second, needs magnetar-strength fields of about 10¹⁴ gauss that "do not seem to appear in binary systems", and is not needed by the Be-star pulsating ULXs, which lack very strong fields [14]; the M51 line allows either field strength [9].
Several populations. ULXs may not be one class [14]. Strongest evidence: the number of sources per unit luminosity in star-forming galaxies breaks at (1–2) × 10⁴⁰ erg per second, and sources above the break could be a new class [8]. Main difficulty: with no observations dividing ULXs into distinct groups, King and colleagues see disc-wind beaming as "the only model open to test against all observations of ULXs" [14].
Open sub-questions
- Pulsar census. Among the 1,843 catalogued candidates [11], what fraction of bright ULXs pulse in deep timing searches, and is the count of 6 in about 1,800 [14] consistent with neutron stars making up most of the population?
- Electrons or protons. Can deeper spectra of the M51 source measure the width of its 4.5 keV line well enough to tell electron from proton scattering, the test Brightman et al. propose [9]?
- M82 X-2's orbit. Can longer timing separate orbital decay driven by mass transfer from magnetic braking by an Ap/Bp donor [13, 15]?
- HLX-1's mass. Can the timing of future HLX-1 outbursts be fitted by disc-instability models with a black hole above 10⁴ solar masses [10, 14]?
- Wind power. Do outflows measured in more ULXs reach speeds near 0.2c [6] and kinetic power comparable to their X-ray luminosity [8]?
- Early black holes. Can short episodes of accretion beyond the Eddington rate, as inferred in ULXs, grow seeds into black holes of 2 × 10⁹ solar masses within 0.78 billion years of the Big Bang [8]? See the Papex page Early supermassive black holes.
Work on this in Papex
If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.
Key references
- Colbert, E. J. M. & Mushotzky, R. F. (1999). The nature of accreting black holes in nearby galaxy nuclei. The Astrophysical Journal 519(1), 89–107. 10.1086/307356 · Checked 16 Oct 2026
- King, A. R., Davies, M. B., Ward, M. J., Fabbiano, G. & Elvis, M. (2001). Ultraluminous X-ray sources in external galaxies. The Astrophysical Journal 552(2), L109–L112. 10.1086/320343 · Checked 16 Oct 2026
- Farrell, S. A., Webb, N. A., Barret, D., Godet, O. & Rodrigues, J. M. (2009). An intermediate-mass black hole of over 500 solar masses in the galaxy ESO 243-49. Nature 460(7251), 73–75. 10.1038/nature08083 · Checked 16 Oct 2026
- Bachetti, M., Harrison, F. A., Walton, D. J., Grefenstette, B. W., Chakrabarty, D., Fürst, F. et al. (2014). An ultraluminous X-ray source powered by an accreting neutron star. Nature 514(7521), 202–204. 10.1038/nature13791 · Checked 16 Oct 2026
- Fürst, F., Walton, D. J., Harrison, F. A., Stern, D., Barret, D., Brightman, M. et al. (2016). Discovery of coherent pulsations from the ultraluminous X-ray source NGC 7793 P13. The Astrophysical Journal Letters 831(2), L14. 10.3847/2041-8205/831/2/L14 · Checked 16 Oct 2026
- Pinto, C., Middleton, M. J. & Fabian, A. C. (2016). Resolved atomic lines reveal outflows in two ultraluminous X-ray sources. Nature 533(7601), 64–67. 10.1038/nature17417 · Checked 16 Oct 2026
- Israel, G. L., Belfiore, A., Stella, L., Esposito, P., Casella, P., De Luca, A. et al. (2017). An accreting pulsar with extreme properties drives an ultraluminous x-ray source in NGC 5907. Science 355(6327), 817–819. 10.1126/science.aai8635 · Checked 16 Oct 2026
- Kaaret, P., Feng, H. & Roberts, T. P. (2017). Ultraluminous X-ray sources. Annual Review of Astronomy and Astrophysics 55, 303–341. 10.1146/annurev-astro-091916-055259 · Checked 16 Oct 2026
- Brightman, M., Harrison, F. A., Fürst, F., Middleton, M. J., Walton, D. J., Stern, D. et al. (2018). Magnetic field strength of a neutron-star-powered ultraluminous X-ray source. Nature Astronomy 2(4), 312–316. 10.1038/s41550-018-0391-6 · Checked 16 Oct 2026
- Greene, J. E., Strader, J. & Ho, L. C. (2020). Intermediate-mass black holes. Annual Review of Astronomy and Astrophysics 58, 257–312. 10.1146/annurev-astro-032620-021835 · Checked 16 Oct 2026
- Walton, D. J., Mackenzie, A. D. A., Gully, H., Patel, N. R., Roberts, T. P., Earnshaw, H. P. & Mateos, S. (2022). A multimission catalogue of ultraluminous X-ray source candidates. Monthly Notices of the Royal Astronomical Society 509(2), 1587–1604 (published online 20 October 2021). 10.1093/mnras/stab3001 · Checked 16 Oct 2026
- Kong, L.-D., Zhang, S., Zhang, S.-N., Ji, L., Doroshenko, V., Santangelo, A. et al. (2022). Insight-HXMT discovery of the highest-energy CRSF from the first Galactic ultraluminous X-ray pulsar Swift J0243.6+6124. The Astrophysical Journal Letters 933(1), L3. 10.3847/2041-8213/ac7711 · Checked 16 Oct 2026
- Bachetti, M., Heida, M., Maccarone, T., Huppenkothen, D., Israel, G. L., Barret, D. et al. (2022). Orbital decay in M82 X-2. The Astrophysical Journal 937(2), 125. 10.3847/1538-4357/ac8d67 · Checked 16 Oct 2026
- King, A., Lasota, J.-P. & Middleton, M. (2023). Ultraluminous X-ray sources. New Astronomy Reviews 96, 101672. 10.1016/j.newar.2022.101672 · Checked 16 Oct 2026
- Chen, W.-C. (2024). An alternative model for the orbital decay of M82 X-2: the anomalous magnetic braking of a Bp star. The Astrophysical Journal 973(1), 38. 10.3847/1538-4357/ad66ce · Checked 16 Oct 2026
Source
Open, as listed by Wikipedia · Unsolved problems in astronomy. Checked 08 Oct 2026.
The status is the source’s. Papex does not decide whether a problem is solved.
Records on this problem
No public record addresses this problem yet.