A broadband X-ray spectral study of the intermediate-mass black hole candidate M82 X-1 with NuSTAR, Chandra and Swift

M82 X-1 is one of the brightest ultraluminous X-ray sources (ULXs) known, which, assuming Eddington-limited accretion and other considerations, makes it one of the best intermediate-mass black hole (IMBH) candidates. However, the ULX may still be explained by super-Eddington accretion onto a stellar-remnant black hole. We present simultaneous NuSTAR, Chandra and Swift/XRT observations during the peak of a flaring episode with the aim of modeling the emission of M82 X-1 and yielding insights into its nature. We find that thin-accretion disk models all require accretion rates at or above the Eddington limit in order to reproduce the spectral shape, given a range of black hole masses and spins. Since at these high Eddington ratios the thin-disk model breaks down due to radial advection in the disk, we discard the results of the thin-disk models as unphysical. We find that the temperature profile as a function of disk radius ($T(r)\propto r^{-p}$) is significantly flatter ($p=0.55^{+ 0.07}_{- 0.04}$) than expected for a standard thin disk ($p=0.75$). A flatter profile is instead characteristic of a slim disk which is highly suggestive of super-Eddington accretion. Furthermore, radiation hydrodynamical simulations of super-Eddington accretion have shown that the predicted spectra of these systems are very similar to what we observe for M82 X-1. We therefore conclude that M82 X-1 is a super-Eddington accretor. Our mass estimates inferred from the inner disk radius imply a stellar-remnant black hole ($M_{\rm BH}=26^{+9}_{-6} M_{\odot}$) when assuming zero spin, or an IMBH ($M_{\rm BH}=125^{+45}_{-30} M_{\odot}$) when assuming maximal spin.