The Long-term Secular Mass Accretion Rate of the Recurrent Nova T Pyxidis*

Abstract

We present Hubble Space Telescope ultraviolet spectroscopy of the recurrent nova T Pyxidis obtained more than 5 years after its 2011 outburst indicating that the system might not have yet reached its deep quiescent state. The ultraviolet data exhibit a 20% decline in the continuum flux from the preoutburst deep quiescence state to the post-outburst near quiescent state. We suggest that a decline across each recurring nova eruption might help explain the proposed 2 mag steady decline of the system since 1866 (Schaefer et al. 2010). Using an improved version of our accretion disk model as well as International Ultraviolet Explorer ultraviolet and optical data, and the 4.8 kpc distance derived by Sokoloski et al. (2013) (and confirmed by De Gennaro Aquino et al. ( 2014)), we corroborate our previous findings that the quiescent mass accretion rate in T Pyx is of the order of 10 −6 M ⊙ yr −1 . Such a large mass accretion rate would imply that the mass of the white dwarf is increasing with time. However, with the just-release Gaia DR 2 distance of ∼ 3.3 kpc (after submission of the first version of this manuscript), we find a mass accretion more in line with the estimate of Patterson et al. (2017), of the order of 10 −7 M ⊙ yr −1 . Our results predict powerful soft X-ray or extreme ultraviolet emission from the hot inner region of the high accretion rate disk. Using constraining X-ray observations and assuming the accretion disk doesn't depart too much from the standard model, we are left with two possible scenarios. The disk either emits mainly extreme ultraviolet radiation which, at a distance of 4.8 kpc, is completely absorbed by the interstellar medium, or the hot inner disk, emitting soft X-rays, is masked by the bulging disk seen at a higher inclination.

Hubble Space Telescope Far-UV Spectroscopy of the Short Orbital Period Recurrent Nova CI Aql: Implications for White Dwarf Mass Evolution

“…Their white dwarf masses and accretion rates are, for T Pyx, 1.23 M ⊙ , 1.12 ×10 −7 M ⊙ /yr; for IM Nor, 1.21 M ⊙ , 4.8 ×10 −8 M ⊙ /yr; and for CI Aql, 1.21 M ⊙ , 1.12 ×10 −7 M ⊙ /yr. By comparison, Godon, et al (2018) derivedṀ ∼ 10 −7 M ⊙ /yr and M wd = 1.2M ⊙ for T Pyx, which is in good agreement. For CI Aql we derived 4 ×10 −8 M ⊙ /yr, which is nearly a factor of three smaller than Shara, et al (2015).…”

Novae heat their food: mass transfer by irradiation

“…At the intersection of the solid and dotted lines, is just enough to lift itself out through L1 using the accretion luminosity = / ( and are the white dwarf's mass and radius, respectively). We find that the self-sustaining rate is ∼ 10 −7 M ⊙ yr −1 (equivalently ∼ 10 36 erg s −1 ), providing a possible explanation for the puzzling high accretion rates of the recurrent novae T Pyx and IM Nor during quiescence (Patterson et al 2017(Patterson et al , 2020Godon et al 2018). Knigge et al (2000) proposed a different self-sustaining mechanism for T Pyx: an irradiation-driven wind carries away angular momentum from the binary, driving the companion to overflow its Roche lobe.…”

“…We find that under certain conditions, the decline halts and the mass transfer becomes self-sustaining: the white dwarf's accretion luminosity is enough to drive the accreted mass through the irradiated companion's Roche lobe. This 'bootstrapping' state may explain the high quiescent accretion rates (∼ 10 −7 M ⊙ yr −1 ) recently observed for the recurrent novae T Pyxidis (T Pyx) and IM Normae (IM Nor) (Patterson et al 2017(Patterson et al , 2020Godon et al 2018). If the self-sustaining rate is sufficiently high, the white dwarf can burn accreted hydrogen stably (Nomoto et al 2007;Shen & Bildsten 2007;Wolf et al 2013), potentially explaining long-lived supersoft X-ray sources with short orbital periods like RX J0537.7-7034 and 1E 0035.4-7230 as well (Kahabka & van den Heuvel 1997;King et al 2001).…”

Accrete, Accrete, Accrete… Bang! (and repeat): The remarkable Recurrent Novae

“…The majority of Galactic RNe follow the picture painted above, with 30% hosting a sub-giant donor and 50% with a red giant donor [10]. Yet two systems, T Pyxidis and IM Normae -the "recurrent unusual novae" -seemingly with main sequence donors, lower mass WDs, and short orbital periods, may buck this trend (for a non-complete selection of discussion articles about these unusual RNe see [32,33,34]). Figure 1 presents a colour-magnitude diagram (based on [10]) that illustrates how near-IR quiescent photometry is used to determine the donor type in a nova system.…”