Obscuring Supersoft X-ray Sources in Stellar Winds

As shown in van den Heuvel et al. (1992), a massive white dwarf (WD) accreting material from a companion star at the steady-burning rate given in Nomoto (1982) will emit X-rays with a spectrum consistent with that observed for supersoft X-ray sources (SSS). If steady burning WDs do indeed look like SSS and the single degenerate (SD) progenitor scenario is the dominant contributor to the SN Ia rate, then we should expect to see a corresponding population of SSS large enough to produce the observed SN Ia rate. However, as shown by Gilfanov & Bogdán (2010) and Di Stefano (2010), the number of observed SSS in both the Milky Way and external galaxies appears to be at least 1-2 orders of magnitude too low to account for the number of observed SNe Ia, both in spiral galaxies and ellipticals. This dearth of SSS may mean that the missing SSS are simply not there, and hence that they are not the dominant contributors to the SN Ia rate. An alternative possibility is that they do in fact exist and produce a significant fraction of the total SN Ia rate, but are somehow hidden from view of our current observational capabilities during their supersoft phase, only to become visible when they explode as SNe Ia.

In the following we wish to explore the latter option. The object has been to make a simple analytical model of a massive, accreting WD with a giant companion star which is losing matter into the circumstellar region (in addition to the matter it transfers to the accretor) to determine which combination of parameters renders the WD undetectable as a SSS. It should be noted that we do not necessarily claim that systems with the required parameters exist, or that they exist in large enough numbers to make up for the deficit in SSS, as compared to SN Ia rate. All we aim to do is get a general handle on how much circumstellar matter is needed to render a canonical SSS unobservable to modern X-ray telescopes. 2 Mikkel T.B. Nielsen, Carsten Dominik, Gijs Nelemans

We consider a massive (∼ 1 M ⊙ ) WD steadily accreting and burning material from a close companion star. The companion can be a main sequence, red giant or asymptotic giant branch star. The accretion mechanism is intentionally left unspecified, but can be envisioned to be any mechanism capable of supplying the WD with the required amount of matter, such as Roche-Lobe overflow (RLOF), Bondi-Hoyle wind accretion, tidally enhanced wind accretion (see Chen et al. (2011)), and wind-RLOF accretion (see Mohamed & Podsiadlowski (2007)).

The binary is assumed to be embedded in a spherical distribution of matter that has been lost from the binary. In our simulations we have assumed the mechanism behind this mass loss to be a radiation driven wind from the donor star. However, other ways may be envisioned in which the system can lose material into the surroundings, such as binary interactions or thermal pulses from the donor. We note that there is observational support for the existence of circumstellar material in early type Ia spectra, see for example Sternberg’s talk (Sternberg et al 2011, in prep.). The material is assumed to be distributed spherically symmetrically around the donor star, somewhat analoguous to the wind bubble described in the talk by Chiotellis, see also Chiotellis et al. (2011), except that our model system is considered stationary with respect to the ISM.

The code is one-dimensional and assumes a line of sight that minimizes the amount of material between the accretor and the observer. The density of material at the surface of the WD, as well as the obscuring column from there to the observer, is derived by integrating a simple r -2 density profile, where r is the distance from the surface of the companion star emitting the obscuring wind.

We assume solar chemical abundances and use the parametrized model of Wilms et al. (2000). The contribution to the obscuration from hydrogen and helium in the photon energy range relevant to Chandra (i.e. above ∼ 100 eV) is completely negligible. Also, the abundances and cross sections of iron group elements are too small to play a role. Therefore, what matters are K-shell ionizations of intermediate mass elements, in particular C, N and O.

As a concrete application we consider what the SSS will look like when located in the Large Magellanic Cloud (LMC) and observed with the ACIS-I detector on the Chandra X-ray Observatory.

Given an orbital separation between the WD and the companion we want to calculate what mass loss rate is required to obscure the system from X-ray observations. By the mass loss rate we mean the rate of mass that is lost from the companion star without being accreted by the WD, i.e. Ṁtotal = Ṁaccr + Ṁlost . This material will be present as a circumstellar wind bubble, potentially absorbing the X-rays from the SSS.

The black body curve of the unabsorbed and absorbed SSS is folded with the effective area function of ACIS-I detector on the Chandra satellite and the result is the relative num

This content is AI-processed based on open access ArXiv data.