Spectroscopic Signature of a Re-established Accretion Disk in Symbiotic-like Recurrent Nova RS Ophiuchi

Among the recurrent novae, one small subgroup erupts within the wind of a close red giant companion (see, e.g., Bode 2010; Anupama 2013; Darnley & Henze 2020). These systems are very different than the classical, compact novae for which the ejecta are expanding freely once the explosion terminates. In these symbiotic-like binaries, the accretion is, at least in part, driven by capture of the imbedding wind by the companion white dwarf. A disk forms and the accretion proceeds similarly to the compact, cataclysmic type systems. The main difference is in the post-eruption behavior of the expelled mass. The ejecta plow into the ambient stellar wind, a dense, stratified environment in which a strong leading shock is formed that also modifies the ionization and density structure of the wind. An important question, as yet unresolved, is how the system relaxes back to the pre-outburst accretion state and when the disk—that disrupts during the explosion, reforms.

Zamanov et al. (2022) and Romanov (2022) have recently reported that flickering—taken as the signature of vigorous accretion in cataclysmic binaries—has reappeared in the post-outburst symbiotic-like recurrent nova RS Ophiuchi. The current outburst has been followed across the spectrum. The previous outburst, in 2006, was also extensively covered, especially well with the UVOT grism on the Neil Gehrels Swift Observatory from day 30 for about one year in the range 1700–7000 Å with a resolution of about 150 (Azzollini 2021). The 2021 outburst of RS Oph began on August 8 (MJD 59,674), corresponding to orbital phase 072, with zero being inferior conjunction of the red giant (Brandi et al. 2009). The 2006 outburst began on February 12, at orbital phase 026. Thus, views through the wind of the giant of the white dwarf and its environs were completely different, in the 2006 event the system was seen without the intervening obscuration, around 065 and in this latest the orbital phase was about 011. Thus, we concentrate on that spectra sequence.

The spectrum was composite, consisting of contributions from the shock and associated ejecta from the nova, emission lines from he wind of the red giant produced by the UV precursor generated by the shock, and the additional ionization following the appearance of a soft X-ray source below 1 keV due to emission from the pointlike white dwarf. The latter appeared only after the combined neutral column density of the ejecta and wind was sufficiently reduced by the advance of the ionization front. The He ii 4686 Å and Balmer lines were especially well observed throughout the interval and we show the variations of Hδ and He ii in Figure 1. Overplotted we display the latest flux measurements from the 2021 eruption that follow precisely the same pattern.

Zoom In Zoom Out Reset image size

The variations can be phenomenologically separated into four developmental states. The first is from the shock during its free expansion stage and its subsequent Sedov-Taylor expansion. There was also an accumulation of mass from the environment, a peculiarity of these systems in contrast to classical nova explosions that expand ballistically from the start. This snowplow increased the mass of the ejecta and slowed them rapidly. The second stage, from about day 75, is breakout, when the further mass loading of the ejecta is negligible but the gas is continually ionized by the residual nuclear burning on the white dwarf (the so-called supersoft source). The third stage, from around day 100, is when following the turnoff of the central source, the ballistic expansion rate competes with the recombinations (Shore et al. Shore et al. 1996). The final, fourth stage is when the accretion disk has reformed and re-establishes the boundary layer and again photoionizes the ejecta.

It is this last stage that is signalled by a renewed increase in the strength of the higher Balmer and ionized permitted lines, especially He ii 4686 Å. In the 2006 sequence, this last stage began with minimum strength for the emission lines at after around day 150 for the higher Balmer sequence and about day 160 for He ii. In the current outburst that corresponds to after about 2022 February 15, when Marchev et al. (Marchev et al. 2022) did not detect optical flickering. The formation of a disk is assured in these systems by the continued accretion from the companion's wind but the photometric flickering requires the formation of a steady state disk and boundary layer. Therefore, we propose that the beginning stages of disk formation were consistent with the last stage of emission line decline, thus providing an independent means for establishing how and when the white dwarf recommences its buildup to the next explosion. The delay between the photometric and spectroscopic indicators can be attributed to the time required for viscosity to bring the accretion disk to a steady state.

We thank Kim Page, Jordi José,and Marco Bellomo for discussions. NPMK acknowledges support by the UKSA.

Facilities: Swift (XRT and UVOT) - , AAVSO. -