The intermediate–mass black hole 2XMM J123103.2+110648: A varying disc accretion rate during possible X–ray quasi–periodic eruptions?

¹ SRON, Netherlands Institute for Space Research, Niels Bohrweg 4, 2333 CA Leiden, The Netherlands
² Department of Astrophysics/IMAPP, Radboud University, P.O. Box 9010 6500 GL Nijmegen, The Netherlands
³ National Astronomical Observatories, Chinese Academy of Sciences, 20A Datun Road, Beijing 100101, China
⁴ Department of Astronomy, University of Wisconsin, Madison, WI 53706, USA
⁵ Racah Institute of Physics, The Hebrew University, Jerusalem 91904, Israel
⁶ The University of Arizona, 933 N. Cherry Ave., Tucson, AZ 85721, USA

^⋆ Corresponding author: z.cao@sron.nl

Received: 12 December 2024
Accepted: 8 June 2025

Abstract

We fit the evolving X-ray spectra of the variable and fading source 2XMM J123103.2+110648 (J1231), which is an intermediate-mass black hole (IMBH) candidate. Recent X-ray timing studies propose that the quasi–periodic oscillation (QPO) of J1231, observed at the peak of its X-ray light curve, is a variant of the quasi-periodic eruptions (QPEs) observed in other sources. Here, we fit X–ray spectra from XMM-Newton, Swift, and Chandra using a slim disc model for the black hole accretion disc, obtaining a best-fit black hole mass of (6 ± 3)×10⁴ M_⊙ and a spin of > 0.6 at 2σ confidence. This mass is consistent with previous estimates, supporting the IMBH interpretation, and the spin measurement is new. Yet, the nature of J1231 remains uncertain: its long-term variability (decade-long continuum evolution) could signal a tidal disruption event or active galactic nucleus variability. We find that the spectral evolution within the first three years after the source’s detection can be well explained by either a varying disc accretion rate, ṁ, or a varying disc inclination, θ. Meanwhile, we find that during the short-term variability (the QPO with a ∼3.8 h period), each oscillation does not show the ‘hard-rise-soft-decay’ typical of QPEs. We fit the average spectrum at the QPO light curve maxima and the average spectrum at its minima, finding that the spectral difference is well explained by ṁ decreasing from peaks to valleys if θ < 30° and remaining constant between all data epochs. This result suggests that the short–term QPO behaviour might also be driven by a varying disc ṁ.