Quasi-periodic eruptions are mysterious X-ray bursts originating near supermassive black holes, offering a unique probe of extreme gravitational environments. These black hole X-ray bursts were first serendipitously detected in 2018 during follow-up observations of an X-ray flare in the Seyfert 2 galaxy GSN 069, with a second event soon identified in a similar galaxy. Quasi-periodic eruptions manifest as rhythmic flares that recur over hours to days, with each eruption lasting between 10³–105 seconds and a typical duty cycle of ~10%.1
Their peak X-ray luminosity ranges from 104¹ to 104³ erg s?¹, and their spectra are well-described by thermal emission with temperatures of 100–200 eV. Recent studies indicate that some Quasi-periodic eruptions occur in galactic nuclei a few years after a tidal disruption event, suggesting that these eruptions may arise when a star orbiting a supermassive black hole interacts with an accretion disk formed by the tidal disruption event. Quasi-periodic eruptions thus provide new insights into accretion physics and black hole astrophysics.1
Image Credit: Triff/Shutterstock.com
The Origins of the Discovery
While the origin of periodic signals from galactic nuclei remains uncertain, they are assumed to be associated with supermassive black holes in those regions. Several theoretical models have been proposed, including interactions between a star/compact object and an accretion disk, mass transfer from a star to the black hole, disk-instability-driven limit-cycle oscillations, gravitational self-lensing by the black hole, and Lense–Thirring precession of an outflow. Among these, the interaction between an orbiting star and the accretion disk has received particular attention. It is known as the “extreme mass-ratio inspiral (EMRI)+disk” model and suggests that a star in an EMRI repeatedly plunges through the disk, producing recurring X-ray flares. This scenario explains complex variations naturally in recurrence time through orbital eccentricity and general relativistic precession, features that challenge other models. Additionally, the EMRI+disk model offers promising connections to future multi-messenger astronomy as EMRIs are prime targets for upcoming space-based gravitational-wave observatories, potentially linking X-ray periodicity with gravitational-wave signals.1
What Do QPEs Tell Us About Black Holes?
Recent observations indicate that some quasi-periodic eruptions are closely linked to tidal disruption events, in which a star is torn apart by the tidal forces of a supermassive black hole during a close encounter. Tidal disruption events appear as bright, multiwavelength flares in galactic centers, and several quasi-periodic eruptions have been detected in X-ray follow-up campaigns of events believed to be tidal disruption events. The host galaxies of quasi-periodic eruptions and tidal disruption events also share notable similarities, including a tendency to occur in post-starburst environments, strengthening the case for a physical connection. The most prominent example is the quasi-periodic eruption detection following the optical tidal disruption event AT2019qiz.1
Within the EMRI+disk framework, a tidal disruption event involving a different star provides the gas that forms an accretion disk, enabling periodic interactions with an inspiraling compact object/star. The resulting disk is expected to have evolved into a radiatively efficient, geometrically thin, standard accretion disk as quasi-periodic eruptions are generally observed several years after the associated tidal disruption event. Thus, current research focuses on whether interactions between the EMRI and this standard disk can explain the timing, luminosity, and spectral characteristics of observed quasi-periodic eruptions.1
Recent Breakthrough Studies
A paper published in The Astrophysical Journal reported that periodic collisions between a stellar EMRI and a supermassive black hole’s accretion disk naturally produced flares matching observed quasi-periodic eruptions. The researchers examined how a star interacted with a gaseous disk of fixed accretion rate and concluded that such a star could not survive long in a persistent active galactic nuclei disk. They then focused on a transient disk formed by a tidal disruption event and applied their model to GSN 069, also noting its relevance to black hole–disk collisions like those proposed for OJ 287.2
The findings showed that star–disk collisions created shock-heated debris clouds that produced flares matching quasi-periodic eruption properties, with solar-type stars favored over compact objects. A mildly eccentric stellar EMRI naturally generated alternating long–short flare intervals, though brightness–interval correlations remained uncertain. The model reproduced GSN 069’s timing behavior and predicted hard flare spectra, while flare temperature depended sensitively on gas density. Quasi-periodic eruption lifetimes were limited by rapid stellar ablation, consistent with transient tidal disruption event-generated disks. Rates implied that a noticeable fraction of tidal disruption events should host quasi-periodic eruptions. Mass stripping could trigger secondary tidal disruption event-like flares, as seen in GSN 069. The model extended to higher-mass supermassive black holes and supermassive black hole binaries like OJ 287, though reproducing all observed properties there remained challenging.2
Another paper published in The Astrophysical Journal presented a model for coupled star–disk evolution that incorporated mass and thermal energy injected into the disk by stellar collisions and ablation. Under weak heating, the disk became thermally unstable, triggering limit-cycle oscillations that led to accretion-powered outbursts on timescales of years to decades, with an average accretion rate of approximately 0.1???dd. In contrast, stronger heating stabilized the disk, enabling steady accretion at the rate set by EMRI stripping. Stellar destruction through ablation, and thus the maximum quasi-periodic eruption lifetime, was about 10²–10³ years. The resulting disks could be self-sustaining and explain observed secular variability and recurrence-pattern changes, such as those in GSN 069.3
In a recent study published in ArXiv*, researchers reviewed the quasi-periodic eruption emission process in the EMRI-plus-disk model, based on Linial & Metzger (2023) (LM23). They estimated quasi-periodic eruption duration, luminosity, and temperature, showing that eruptions arise when a star orbiting a supermassive black hole crosses its accretion disk supersonically, generating shocks that compress and eject gas, with photons eventually diffusing out to produce the observed flares.1
The findings showed that during the slim-disk phase, the disk scale height approached the radius, surface density fell, quasi-periodic eruption durations shortened, luminosity stayed constant, and temperatures rose rapidly until the accretion rate dropped below the Eddington limit. Predicted eruptions were brief, often under 10³ seconds and sometimes near 100 s, with temperatures reaching 10–50 keV. Although detectable in principle, their very small duty cycle and obscuration by thick disk winds made observation difficult except from polar angles.1
Observational and Theoretical Challenges
Modeling quasi-periodic eruptions is challenging because the early-time evolution of the tidal disruption event disk is unclear, limiting precise predictions of quasi-periodic eruption behavior. Although late-time disks are observed, their super-Eddington accretion complicates modeling. Additionally, the connection between quasi-periodic eruptions and tidal disruption events raises questions about whether they represent repeatable tidal disruption events or a distinct class of black hole behavior, highlighting uncertainties in interpreting their origin and variability.1-3
Commercial and Technological Relevance
Recent work has exploited machine learning to scan archival X-ray data and reliably identify rare quasi-periodic eruptions, demonstrating over 98?% classification accuracy on real observations. Simultaneously, physical models like the EMRI + TDE disk model showed how stellar collisions with a post-tidal disruption event disk produce periodic X-ray flares. These advances indicate the need for more sensitive, high-cadence space-based X-ray telescopes and improved satellite instrumentation. They also illustrate the growing role of artificial intelligence, particularly machine learning, for signal detection and pattern recognition in astrophysical data, a trend with potential commercial impact for satellite builders and data-analysis platforms.1-4
What’s Next in QPE Research?
Future missions like the Athena X-ray observatory and the proposed Laser Interferometer Space Antenna (LISA) gravitational-wave detector could enhance quasi-periodic eruption studies by providing higher-sensitivity, high-cadence observations, linking X-ray flare properties with EMRI gravitational-wave signals, and refining models of star–disk interactions near supermassive black holes.
Want to learn more about singularities? It's all here
References and Further Reading
- Suzuguchi, T., Matsumoto, T. (2025). Quasi-Periodic Eruptions as a Probe of Accretion Disk in Tidal Disruption Events. ArXiv. DOI: 10.48550/arXiv.2509.01663, https://arxiv.org/abs/2509.01663
- Linial, I., Metzger, B. D. (2023). EMRI+ TDE= QPE: periodic X-ray flares from star–disk collisions in galactic nuclei. The Astrophysical Journal, 957(1), 34. DOI: 10.3847/1538-4357/acf65b, https://iopscience.iop.org/article/10.3847/1538-4357/acf65b/meta
- Linial, I., Metzger, B. D. (2024). Coupled disk-star evolution in galactic nuclei and the lifetimes of QPE sources. The Astrophysical Journal, 973(2), 101. DOI: 10.3847/1538-4357/ad639e, https://iopscience.iop.org/article/10.3847/1538-4357/ad639e/meta
- Webbe, R., Young, A. J. (2023). Searching for quasi-periodic eruptions using machine learning. RAS Techniques and Instruments, 2(1), 238-255. DOI: 10.1093/rasti/rzad015, https://academic.oup.com/rasti/article/2/1/238/7172876
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.