Hidden black hole or neutron star found driving a massive star's decade of flares

Astronomers say they have finally identified what has been causing a massive, unstable star 30 million light-years away to repeatedly flare up and fade without ever destroying itself. Observations from NASA's Chandra X-ray Observatory indicate the star, known as AT 2016blu, has a hidden companion — most likely a stellar-mass black hole or a neutron star — that swings close enough every 113 days to strip material from it, producing the periodic outbursts that have puzzled observers for more than a decade. The findings, published this week in The Astrophysical Journal, mark the first time this type of eruptive star has been caught emitting X-rays, and the first time one has been shown to have a compact companion.

AT 2016blu sits in the spiral galaxy NGC 4559, a system bright enough to be a familiar target for amateur telescopes. The star itself is a luminous blue variable, a class of extremely massive, short-lived stars that periodically shed outer layers in violent eruptions. Some luminous blue variables brighten so dramatically that they are initially mistaken for supernovae — hence the nickname "supernova impostors" — before astronomers realize the star has survived the outburst. AT 2016blu has done this at least 27 times since it was first flagged in 2012, according to the NASA account of the research, brightening and dimming on a cycle so regular that astronomers could set a calendar by it.

A pulse every 113 days

The new result comes from comparing fresh Chandra data with the observatory's own archive. Chandra had looked at the same patch of sky in 2001 and 2002, years before the outbursts began, and saw nothing at the star's location. When the team scheduled a set of "target of opportunity" observations timed to catch a predicted outburst peak in March 2026, Chandra detected X-rays across five closely spaced observing windows. According to the Chandra X-ray Center's release, the source was at least 100 times brighter in X-rays during that 2026 campaign than the strict non-detection from two decades earlier.

The paper, led by Mojgan Aghakhanloo of the University of Virginia, calculates an X-ray luminosity of roughly 4 x 10^38 ergs per second — bright enough that the system qualifies as a candidate ultraluminous X-ray source, a category usually associated with black holes or neutron stars feeding voraciously on a companion. The inferred mass-transfer rate, at least about 8 x 10^-8 solar masses per year, points to accretion rather than any process confined to the luminous blue variable alone. Combined with the clockwork 113-day optical cycle, the authors conclude that AT 2016blu is orbited by a compact object in an eccentric orbit, and that each close pass — periastron, in orbital terms — triggers a burst of mass transfer that lights up both the optical outburst and the X-ray emission. That makes the system, the paper argues, a previously unrecognized high-mass X-ray binary, with the luminous blue variable itself, estimated at more than 33 times the mass of the sun, as the uncomfortable donor star.

A decade of watching and waiting

Pinning down the 113-day period took years of patient monitoring, much of it contributed by a global network of professional and amateur observers whose photometry is credited throughout the preprint version of the study, posted to the arXiv server in August. That long observational baseline is what let the team forecast the March 2026 outburst precisely enough to request Chandra time in advance, rather than waiting for a chance detection. Earlier work on the same object, including spectroscopic follow-up published in Monthly Notices of the Royal Astronomical Society, had already shown the outbursts were "quasiperiodic" and suspected they were linked to an eccentric binary; this is the first study to catch the predicted X-ray signature that would confirm it.

Luminous blue variables are rare and short-lived by stellar standards, and astronomers care about them disproportionately because they are considered likely progenitors of some supernovae and may shed enough mass before dying to influence how — or whether — they later collapse into black holes or neutron stars. A star that already has a compact companion siphoning off its outer layers adds a new wrinkle to that picture, since mass loss driven by a binary companion is a very different process from mass loss driven by the star's own instability.

"For 15 years, astronomers have been watching this star get really bright and then dim again on a regular basis," said lead author Mojgan Aghakhanloo of the University of Virginia.

What the researchers say

Co-author Nathan Smith of the University of Arizona described timing as the central challenge of the project, saying "we needed to plan the Chandra X-ray observations carefully to be simultaneous with an optical outburst." That requirement is also why the result has taken so long to nail down — Chandra observing time is competitive and must be requested specifically for predicted events, and the earlier non-detections in 2001 and 2002 only became informative once researchers had a specific, testable prediction to check them against.

The team has not yet been able to say definitively whether the compact companion is a black hole or a neutron star; the current X-ray data constrain the accretion behavior but do not include the kind of rapid timing signal — a pulsation period, for instance — that would distinguish a spinning neutron star from a black hole. The researchers note that AT 2016blu's behavior resembles that of SN 2010da, another system in which a compact object accretes from a massive companion, though in that case the donor star is a red supergiant rather than a luminous blue variable, making AT 2016blu's exact configuration unusual.

Limitations and what comes next

As with most results built on sparse, hard-won observing windows, caveats remain. The X-ray detection spans five observations clustered around a single predicted outburst in 2026, not the full 113-day cycle, so researchers cannot yet confirm the accretion signature repeats identically at every periastron passage, nor rule out that the current outburst behaved somewhat differently from the 26 that preceded it. The 113-day period itself is derived from fourteen years of optical photometry gathered by instruments and observers of varying quality, which can introduce timing uncertainties over such a long baseline. And because the companion's nature is still inferred rather than directly measured, the "black hole or neutron star" question is likely to require additional X-ray spectroscopy or timing campaigns, ideally spanning more than one outburst cycle, before it can be settled.

The composite image released alongside the study shows AT 2016blu as little more than a faint pink dot nestled against the disk of NGC 4559 — a reminder of how much work went into isolating a single, periodically flickering point of light from a crowded host galaxy 30 million light-years away. For now, the star remains under watch, both from Chandra and from the same network of ground-based observers whose years of monitoring made the discovery possible in the first place; its next predicted outburst will offer another chance to test whether the accretion picture holds.