The population of hyper-luminous X-ray sources as seen by XMM-Newton

¹ INAF – Osservatorio Astronomico di Roma, Via Frascati 33, I-00078 Monte Porzio Catone, (RM), Italy
² IRAP, CNRS, Université de Toulouse, CNES, 9 Avenue du Colonel Roche, 31028 Toulouse, France
³ European Space Agency (ESA), European Space Astronomy Centre (ESAC), Camino Bajo del Castillo s/n, 28692 Villanueva de la Cañada, Madrid, Spain
⁴ Institut de Ciêncies del Cosmos (ICCUB), Universitat de Barcelona (UB), c. Martí i Franquès, 1, 08028 Barcelona, Spain
⁵ Departament de Física Quàntica i Astrofísica (FQA), Universitat de Barcelona (UB), c. Martí i Franquès, 1, 08028 Barcelona, Spain
⁶ Institut d’Estudis Espacials de Catalunya (IEEC), c. Gran Capità, 2-4, 08034 Barcelona, Spain
⁷ School of Physics & Astronomy, University of Southampton, Southampton, Southampton, SO17 1BJ, UK
⁸ Cahill Center for Astrophysics, California Institute of Technology, 1216 East California Boulevard, Pasadena, CA, 91125, USA
⁹ Astronomical Institute of the Czech Academy of Sciences, Boční II 1401/1, 14100 Praha 4, Czech Republic

^⋆ Corresponding author: roberta.amato@inaf.it

Received: 11 April 2025
Accepted: 12 July 2025

Abstract

Context. Ultraluminous and hyper-luminous X-ray sources (ULXs and HLXs) are among the brightest astrophysical objects in the X-ray sky. While ULXs are most likely to host stellar-mass compact objects accreting at super-Eddington rates, HLXs are compelling candidates for accreting intermediate-mass black holes. However, HLXs are predominantly found in distant galaxies (d ≳ 100 Mpc), where the chances of source confusion and misidentification with active galactic nuclei (AGNs) or other transient X-ray phenomena are high.

Aims. Our goal is to produce a clean sample of HLXs, by removing possible contaminants, and to characterise the spectral properties of the remaining population. This sample can then be used to determine whether different types of astrophysical objects coexist within the same class.

Methods. Starting with a set of 115 HLXs detected by XMM-Newton, we identified and removed contaminants (AGNs, X-ray diffuse emission detected as point-like, and tidal disruption event candidates) and retrieved 40 sources for which XMM-Newton spectra are available. We fit them with an absorbed power law model and determined their unabsorbed luminosities and hardness ratios. We constructed the hardness-luminosity diagram and compared the results with the spectral properties of the HLX prototype, ESO 243-49 HLX-1. Then we conducted a deeper analysis on a selection of promising candidates.

Results. The resulting HLX population spans a luminosity range from 1 × 10⁴¹ erg s⁻¹ to nearly 10⁴³ erg s⁻¹ and is homogeneously spread in hardness between 0.5 and 5. Half of the population display hardness ratios higher than a typical AGN and could be considered an extension of the ULX population at higher energies. We found four very soft outliers, which are characterised by steep power law spectra and no X-ray emission above 1–2 keV, similarly to ESO 243-49 HLX-1. Those with multi-epoch archival data display changes in luminosity of up to nearly two orders in magnitude.

Conclusions. We show that sources currently identified as HLXs can be more diverse than ULXs and disentangling among these different types of objects is not trivial with the data currently available. New observations would be beneficial to expand the current sample and uncover the true nature of many objects of this class, some of which show very similar characteristics to ESO 243-49 HLX-1.