The Best Constraints on A Super-Eddington Accretion Flow: XMM-Newton Observations of An Intermediate-mass Black Hole

Abstract

RX J1140.1+0307 (hereafter RX1140) is a Narrow Line Seyfert 1 (NLS1) with one of the lowest black hole masses known in an AGN (M ${\le} 10^6$ M$_{\odot}$). We show results from two new {\it XMM-Newton} observations, showing soft 2-10~keV spectra, a strong excess at lower energies, and fast X-ray variability as is typical of this class. The soft excess can be equally well fit by either low temperature Comptonisation or highly smeared, ionised reflection models, but we use a covariance analysis of the fast X-ray variability as well as lag and coherence spectra to show that the low temperature Comptonisation model gives a better description of the break in variability properties between soft and hard X-rays. Both models also require an additional component at the softest energies, as expected from the accretion disc. However, this inner disc spectrum does not join smoothly onto the variable optical and far UV emission (which should be produced in the outer disc) unless the mass is underestimated by an order of magnitude. The variable optical and far UV emission instead suggests that $L/L_{Edd}\sim 10$ through the outer disc, in which case advection and/or wind losses are required to explain the observed broadband spectral energy distribution. However, the similarity of the X-ray properties of RX1140 to other simple NLS1 such as PG 1244+026, RE J1034+396 and RX~J0136 means it is likely that these are also super-Eddington sources. This means their spectral energy distribution cannot be used to determine black hole spin despite appearing to be disc dominated. It also means that the accretion geometry close to the black hole is unlikely to be a flat disc as assumed in the new X-ray reverberation mapping techniques.