Ultraluminous X-ray pulsar P13 in the faint state

ATel #13751; Roberto Soria (University of the Chinese Academy of Sciences; University of Sydney), Christian Motch (Strasbourg Observatory), Manfred W. Pakull (Strasbourg Observatory)
on 22 May 2020; 11:16 UT
Credential Certification: Roberto Soria (rsoria@physics.usyd.edu.au)

Tweet

We report on the X-ray brightness evolution of the ultraluminous X-ray pulsar P13 (Israel et al. 2017, MNRAS, 466, L48; Fuerst et al. 2016, ApJ, 831, L14; Motch et al. 2014, Nature, 514, 198) in NGC 7793 (d = 3.5 Mpc). After reaching a 0.3-10 keV luminosity of ~1E40 erg/s during 2016-2018 (with an additional superorbital modulation at Px = 66.8 +/- 0.4 d; Hu et al. 2017, ApJ, 835, L9), the source started to decline last year. It was down to a luminosity of 1.3E39 erg/s at the end of its Swift visibility window, in early January 2020. When the Swift monitoring resumed (2020 April 20), P13 was no longer detectable in individual observations, nor in the stacked ones so far (total effective exposure time of 8.0 ks from seven observations between April 20 and May 18). Most recent light curve at this link.

For the binned 2020 April-May XRT observations, the upper limit from the online XRT lightcurve generator (credit: Phil Evans/Swift/UKSSDC) is ~2E-3 XRT ct/s, corresponding to an upper limit of ~2.5E38 erg/s for the unabsorbed 0.3-10 keV luminosity. In fact, from a more detailed photometric analysis of the same stacked data, we determine a 90% upper limit of ~4E-4 ct/s (~5E37 erg/s) in the same band.

A 50-ks XMM-Newton observation from 2019 November 22 (Motch et al., in prep.) caught P13 way into its declining phase, but still just above the 1E39 erg/s ULX threshold. The continuum spectrum was still dominated by a hard power-law-like component with the same slope as seen at peak luminosity (photon index = 1.15 +/- 0.10, luminosity ~ 1.4E39 erg/s), but an additional thermal plasma component was also noticeable around 1 keV (kT = 0.9 +/- 0.1 keV, mekal luminosity ~2E37 erg/s). The thermal plasma component would have been comparably too faint to be detectable when P13 was at ~1E40 erg/s.

Historically, the previous faint phase of P13 was in 2011-2012, when it was also below the Swift/XRT detection limit. Click here for the 2010-2020 XRT lightcurve. Chandra and XMM-Newton observations from 2011-2012 showed a luminosity of a few E37 erg/s (Walton et al. 2018, MNRAS, 473, 4360; Motch et al., in prep.), with a similarly strong contribution from the thermal plasma component.

At least three scenarios have been suggested for the faint state: (i) a state transition to the low/hard state in the accretion disk; (ii) a transition from accretor to propeller state; (iii) a precession effect of the accretion disk, which is now blocking our direct view of the X-ray emission from the innermost region, so that we only see photons scattered in an extended outflow.

Optical monitoring of the V ~ 20 mag counterpart during the faint state are strongly encouraged (especially in the U,B,V bands). Changes in the luminosity of the irradiated face of the donor star will help us distinguish between an intrinsic decline of the X-ray luminosity, and our preferred scenario (Motch et al. in prep) of a change in the orientation of the emission beam. Coordinates of P13: RA = 23:57:51.01, Dec = -32:37:26.62 (J2000).