Indian Astronomers Combined 39 Chandra, XMM-Newton and Swift Observations Taken Over 24 Years of NGC 4490 ULX-8, a Bright X-ray Source Roughly 25 Million Light Years Away, and Find It Fits a Stellar-Mass Black Hole Feeding at or Near the Eddington Limit

A team of researchers at Indian institutes has put 24 years of X-ray observations of one bright source into a single study. The source, NGC 4490 ULX-8, sits in a galaxy about 7.8 megaparsecs away (the distance the paper adopts, roughly 25 million light years), and the team finds that its glow fits a stellar-mass black hole feeding at or near its Eddington limit. The work uses 14 Chandra, 6 XMM-Newton and 19 Swift-XRT observations taken between 2000 and 2024. The Ministry of Science and Technology announced it on 2 November 2026 and says it was published in Monthly Notices of the Royal Astronomical Society.

The authors are Tarang Vashisht, Aru Beri, Tanuman Ghosh, Aman Upadhyay and Vikram Rana, working from IISER Mohali, the Indian Institute of Astrophysics (IIA), the Inter-University Centre for Astronomy and Astrophysics (IUCAA) and the Raman Research Institute (RRI).

A source that outshines its own limit

An ultraluminous X-ray source, or ULX, is a point-like X-ray emitter outside the centre of a galaxy that radiates more than 10³⁹ ergs per second. The paper notes that this is roughly the Eddington limit of a black hole of 10 Suns. The release describes a ULX as a neutron star or a black hole pulling gas off a companion star, with the infalling gas heating up and giving off X-rays.

The Eddington limit is the brightness at which the outward push of light balances gravity’s inward pull on the gas. It rises with the mass of the object. A physics review published by the American Physical Society calls it a rule of thumb that comes from an idealised spherical model, and lists thicker discs, jets and mergers among the ways a black hole could exceed it.

For ULXs the paper gives two readings. One is a heavier black hole than usual. The other, which it says growing evidence favours, is a stellar-mass black hole or neutron star accreting at near or above the limit, helped by outflows of gas or by beaming. Pulsations settle the question when they appear: the paper counts more than 10 pulsating ULXs, each a neutron star. ULX-8 shows none. Other routes to black holes are covered in our pieces on the Event Horizon Telescope and on gravitational waves from colliding black holes.

Why this galaxy pair, and why this source

NGC 4490 is a spiral galaxy interacting with a smaller irregular galaxy, NGC 4485. The paper cites a steady star formation rate of about 4.7 solar masses a year in the pair. A 20-kilosecond Chandra observation reported in 2002 by Timothy Roberts and colleagues found six ULXs there, five of them probable black hole binaries and the sixth a probable supernova remnant, and noted that all lie in star-forming regions. A 2008 catalogue later numbered the system’s ULXs from 1 to 8.

ULX-8 suited a long-term study because it is relatively isolated. ULX-1, 3 and 4 lie so close together that source contamination complicates XMM-Newton analysis, while ULX-8 allowed the team to extract its light from a circle up to 19 arcseconds across. Earlier work on it, including a 2009 study by James Gladstone and Roberts, used four Chandra and XMM-Newton observations from 2000 to 2004. The paper lists 22 of the 39 observations in the new study (two from XMM-Newton, one from Chandra and all 19 from Swift) as unexplored for this source before.

Thirty-nine looks from three telescopes

Chandra launched on 23 July 1999, XMM-Newton on 10 December 1999 and Swift on 20 November 2004. The paper says some of the Swift pointings, and one Chandra observation, were aimed at SN 2008ax, a supernova in NGC 4490, and caught ULX-8 in the same field. The team checked that the supernova’s position showed no X-ray emission that could contaminate the Swift data.

The first observation used was Chandra’s on 3 November 2000. The last was Swift’s on 4 May 2024. The team used every available Chandra and XMM-Newton observation except two with too little usable exposure, and kept the Swift observations in which the source stood out above three times the noise, which left 19.

Aru Beri, a co-author whom the release describes as an IIA faculty member, called the result “one of the longest X-ray studies of this source to date”. The paper’s own wording is that it is the most extensive investigation of ULX-8 so far. The paper lists Beri’s IISER Mohali address first, with IIA third. Vashisht, the first author, is a student at IISER Mohali, and the paper’s acknowledgements say the work began in a visiting student programme at the RRI.

All the data come from public archives: NASA’s HEASARC, ESA’s XMM-Newton archive and the Chandra Data Archive. India’s own X-ray observatory, AstroSat, was not part of this analysis.

What 24 years of light looked like

The source stayed ultraluminous throughout. Its brightness changed substantially over multi-year timescales and only modestly within any single observation. On shorter timescales the paper measures swings by factors of 3.73 (plus or minus 1.04) in a Chandra series taken at roughly monthly intervals from November 2020 to August 2021 and 2.59 (plus or minus 1.37) in Swift data from March and April 2008.

The size of the change depends on the model used to fit the spectrum. The disc model gives luminosities of about 2 to 5 times 10³⁹ ergs per second, and the power-law model gives about 3 to 20 times 10³⁹. Two recent Swift observations, on 1 December 2022 and 4 May 2024, show a sharp rise: about 2.4 and 2.8 counts per hundred seconds, against 0.65 to 1.5 in the 17 earlier Swift observations. The authors say this may reflect a change in the rate at which gas is falling in, rather than a switch between accretion states.

The spectra are a smooth single curve peaking at 1 to 2 kiloelectronvolts, with no separate hard or soft component. The photon index, which describes how quickly the count of X-ray photons falls with energy, runs from 0.9 to 2.7. The inner edge of the disc comes out at temperatures of 1.0 to 1.6 kiloelectronvolts.

Many black hole binaries in our own galaxy swing between distinct hard and soft states, which shows up as a looping track on a plot of hardness against brightness. ULX-8 shows no clear transition between the two. The release describes this as an absence of dramatic transitions between accretion states.

How the team reads it as a black hole

Two lines of argument point the same way. First, the source’s hardness ratio of 0.84 to 3.04 and its photon indices place it outside the harder region of the diagram where neutron-star ULXs sit, and the paper says its spectrum matches disc-dominated emission expected from a super-critical flow. Second, the team converted the fitted disc size into a mass. At an assumed disc tilt of 60 degrees, a common choice when the true angle is unknown, the result is about 17 Suns for a non-spinning black hole and about 81 for a rapidly spinning one. The authors call these illustrative, model-dependent values.

Two cautions sit in the paper itself. The relation between disc temperature and luminosity is weakly constrained and fits both a thin-disc and a slim-disc picture, so it cannot tell the accretion regimes apart. And luminosity depends on distance, which is itself uncertain here: the paper notes that Hubble measurements of the stars in each galaxy give 6.5 megaparsecs for NGC 4490 and 8.8 for NGC 4485, inconsistent with a close interaction, so it keeps the Tully-Fisher figure of 7.8 megaparsecs to stay comparable with earlier X-ray studies.

At, near or beyond the limit

The release words its conclusion in two ways. One paragraph calls the results strong evidence for a black hole accreting at rates exceeding the classical Eddington limit. A later paragraph says ULX-8 is most likely a stellar-mass black hole accreting at or near the limit. The paper’s abstract uses the second form: the results are consistent with a stellar-mass black hole operating at or near the Eddington limit. IBG News, which repeats the release, also gives “at or near”. This article follows the paper.

What comes next

The release names two instruments that could extend the record: the European Space Agency’s NewAthena X-ray observatory and India’s proposed XSat, described as a next-generation timing and spectroscopy mission. No further detail on XSat appeared in the sources checked for this article. The IIA and the RRI, where much of the work was done, are both autonomous institutes of the Department of Science and Technology. The NewAthena mission page gives a planned launch in 2037, with adoption by ESA targeted for 2027.