X-ray and UV awakening of a years old stellar tidal disruption event candidate AT2018fyk/ASASSN-18ul: an accretion instability, repeating partial disruption, or supermassive black hole binary?

ATel #15333; Dheeraj R. Pasham (MIT), Thomas Wevers (ESO), Ann Zabuldoff (U. of Arizona), Sixiang Wen (Radboud University), Peter Jonker (Radboud University), Eric Coughlin (Syracuse University), Zheng Cao (Radboud University), Dacheng Lin (Northeastern University)
on 15 Apr 2022; 22:57 UT
Distributed as an Instant Email Notice Transients
Credential Certification: Dheeraj Pasham (drreddy@mit.edu)

Tweet

AT2018fyk/ASASSN-18ul is a nuclear transient identified in the optical by the ASASSN survey on 8 September 2018 (Wevers et al. 2019). Based on a quiescent host galaxy Magellan spectrum and the lack of optical variability between 2014-2018, i.e., in the four years prior to the ASASSN detection, the source was classified as a stellar tidal disruption event (Wevers et al. 2019). The multi-wavelength datasets acquired between September 2018 and June 2020 support the TDE hypothesis (Wevers et al. 2021). The outburst reached a peak X-ray luminosity of (1.1+-0.1)e44 erg/s around July 2019 and eventually became undetectable with Chandra with a 0.3-10 keV luminosity upper limit of 1.5e40 erg/s on 29 June 2020 (Wevers et al. 2021).

Recent Neil Gehrels Swift XRT observations taken between January-April 2022 reveal a bright point source with a mean 0.3-8 keV count rate of 0.012+-0.001 cps. In these recent X-ray data, AT2018fyk is variable over days with count rates varying between 0.01 cps and 0.03 cps. A combined 0.3-8 keV energy spectral analysis of data taken between 24 March and 6 April 2022 (145 counts in total) is well fit with a powerlaw model (tbabs*zashift(pow) in XSPEC), yielding a c-stat/degree of freedom of 18.3/29. The column density of tbabs was fixed at the Galactic value in the source direction of 1.1e20 cm^-2. The best-fit power-law index is 2.6(+0.3,-0.2). The 0.3-8 keV observed flux is (3.9+-0.4)e-13 erg/s/cm^2. The corresponding unabsorbed flux and luminosity values are (4.9+-0.7)e-13 erg/s/cm^2 and (4.2+-0.6)e42 erg/s, respectively. These numbers imply that the source has brightened by more than a factor of 220 in X-rays since 29 June 2020. A thermal disk model (tbabs*zashift(diskbb) in XSPEC) gives a poor fit with c-stat/dof of 66/29. Concurrently, the UV emission, as inferred from the UVW1 filter data on Swift/UVOT, has increased by a factor of a few: from 7e41 erg/s during the last X-ray non-detection to 6e42 in the most recent data.

Such a re-brightening of a previously X-ray and UV bright TDE years after initial discovery has never been seen. At present, there is limited data to make a conclusive statement about the cause of this unprecedented X-ray and UV rebrightening, but we speculate that this could be due to an inner accretion disk instability predicted from theory (e.g., Shen and Matzner 2014), a repeating partial disruption event similar to ASASSN-14ko (e.g., Payne et al. 2021) and HLX-1 (Lasoto et al. 2011; Godet et al. 2014) but with a ~2 year stellar orbital period, or even a supermassive black hole binary (e.g., Coughlin et al. 2017). Variable obscuration seems unlikely here, because the UV emission remained above host galaxy levels during the previous Chandra X-ray upper limit measurement from 29 June 2020 (Wevers et al. 2021).

Follow-up observations with NICER and Swift are planned. A weekly updated NICER schedule can be found here: https://heasarc.gsfc.nasa.gov/docs/nicer/schedule/nicer_sts_current.html. To coordinate multiwavelength observations, please contact drreddy@mit.edu. We encourage follow-up observations especially in the optical and radio bands to further probe this unusual TDE phenomenon.