Spectra from accretion disks of compact stellar remnants

We are computing spectral models of gaseous accretion disks in various environments. In essence, our method assumes a radial disk structure (e.g. an α-disk) and we compute in detail the vertical disk structure. The code is an offspring of our non-LTE stellar atmosphere modeling package. In addition, we are developing radiation transfer codes to model the spectra of accretion disk winds

Cataclysmic Variables and Symbiotic Stars

We are investigating disks around WD accretors in close binary systems (CVs). CVs are the origin of nova and dwarf-nova phenomena. Particularly interesting are AM CVn systems, which have (almost) pure helium disks. Determining their metal abundances allows us to conclude on the nature of the donor star, which could be a helium white dwarf.  Such white-dwarf binaries are of general interest because they are potential SN Ia progenitors and primary targets for future gravitational wave detectors.

Artist’s concept of an accretion disk in a cataclysmic variable. (This image is copyrighted and provided courtesy of <a title="Öffnet externen Link in neuem Fenster" class="external-link-new-window" href="http://www.markgarlick.com/">Mark A. Garlick</a> / <a title="Öffnet externen Link in neuem Fenster" class="external-link-new-window" href="http://www.space-art.co.uk/index.php">space-art.co.uk</a>. No unauthorised use.)
Accretion-disk spectrum of AM CVn compared with a model spectrum. (From Nagel et al. 2004)

We also model the temporal evolution of dwarf-nova outbursts of CVs, which are thermal instabilities in the disk, resulting in a strong increase of the mass-accretion rate.


Many CV spectra exhibit prominent P Cygni line profiles in the UV, indicative for strong mass-loss away from the disk. We are modeling the disk-wind spectra in order to conclude on mass-loss rates, wind structure and wind-acceleration processes. Depending on the mass-loss rate, the evolution of the binary system can be significantly affected.

P Cygni profile of the N V resonance line in AM CVn (black, HST spectrum). The three panels display disk-wind models with increasing mass-loss rate: 0.5, 1, and 5∙10<sup>-9</sup> solar masses per year. (From Kusterer 2008, PhD thesis)

Like CVs, Symbiotic Stars also consist of WD accretors in binaries, however, with long orbital periods. In contrast to CVs, mass-transfer does not occur by Roche-lobe overflow but through accretion of wind-matter from the red-giant companion. Our disk plus WD atmosphere models can be used to compute the spectral energy distribution of such systems.

Single white dwarfs

Dust- and even gas-debris disks were discovered around numerous single WDs. They can be interpreted as remains of planetary systems of the progenitor stars. We are working on analyses of optical und UV spectra of gaseous disks. Their composition possibly reflects that of tidally disrupted asteroids, allowing us to study in detail the chemistry of extrasolar planetary systems.

False-colour infrared image of the planetary nebula NGC 7293 (Helix nebula) taken with the Spitzer space telescope. The red colour indicates 24&mu; infrared emission from dust around the white dwarf central star.

Low-mass X-ray binaries (LMXBs)

X-ray binaries contain accreting NSs of black holes. The donor stars can be either low- or high-mass stars. The former (= LMXBs) can be rather compact, i.e., their orbital periods are less than 80 minutes. In these systems the donor star must be a stripped WD such that the disk is free of hydrogen and helium. C and O rich matter from the former WD interior is being transferred, giving us the unique possibility to study the interior WD composition. We analyse UV and optical spectra of such systems.

Supernova-fallback disks

Some of the matter that is ejected in core-collapse supernova explosions cannot escape the gravitational well of the neutron star or black hole, falls back towards the compact remnant and can build up an accretion disk. The existence of such disks is debated. Their chemistry must exotic, being primarily built of iron or iron-rich silicon-burning ash. We compute the spectra of such disks which can be compared to observations of potential candidates (e.g., NSs with observed IR excess). The models can also be used to infer the maximum size of such disks around particular systems.

SN1987A (HST image). Emission from the central compact object has not yet been discovered.
Emission spectrum from an iron-accretion-disk model. Such models were used to estimate the maximum extent of such a disk in SN1987A to 70,000 km. (From Werner et al. 2007)

More details on our work on accretion disks can be found on the Accretion Disks Project web pages.