Circumstellar Matter and Interacting Binaries

Circumstellar matter is the dust, gas and plasma around stars, generally present in the form of stellar winds (blown off the star in a manner analogous to the solar wind) or nebulae (the result of the ejection of part of the envelope of a star, e.g. in a nova outburst). We study this matter to understand the effects of interacting winds, dust formation and development (astrochemistry), outflow physics and the late stages of star lifetimes.

Research into the matter surrounding binary and evolved stars is pursued at the JHI using spectroscopy and high-spatial resolution imaging. Three primary classes of objects are targets of this observational programme:

  1. Classical novae
  2. Symbiotic stars
  3. Helium flashers
  4. Interacting Binaries

Classical novae

Interacting binaries are systems in which the two stars orbit so closely that they can exchange matter and irradiate each other. When the accreting star is a white dwarf and the donor star a main sequence (Sun-like) star, then the build up of accreted matter on the white dwarf leads to a thermonuclear runaway (TNR), and a dramatic optical brightening perceived as a “new star” or “stella nova”. To distinguish this form of novae from many related objects (supernovae, dwarf novae, etc.) we adopt the term classical novae (CN). The explosion caused by the TNR is accompanied by the ejection of perhaps 0.0001 solar masses of material, and it is this which we study. Imaging has been carried out in the radio, where interferometers (MERLIN, VLA) can resolve the structure and development of the ejecta at very early times, without contamination from the central star. For example, MERLIN studies of three CN have shown that the irregular structures seen at late times begin to develop within months of the eruption. The various stages of the ejecta’s development are also followed using IR spectroscopy, where nebular lines, molecular features, and dust development can be traced. For example, extensive spectroscopy of V705 Cas (Nova Cas 1993) traced the growth of dust, and identified both oxygen- and carbon-rich material in the same object for the first time.

Symbiotic stars

In some interacting binaries, the orbits are relatively wide (with periods of years to decades), but one of the stars is a red giant. The wind from this giant interacts with the radiation field from the compact star (which may be a white dwarf or a main sequence star), leading to observable phenomena. These binaries are grouped into the rather heterogeneous group known as symbiotic stars. This class has been put forward as progenitors to a fraction of planetary nebulae (PNe) and Type Ia supernovae. In some cases, the compact component has nova-like outbursts, adding two-wind interaction to the irradiation effects on the red giant wind. In common with classical novae, we have studied symbiotic stars, and especially the erupting variety, using imaging and spectroscopy. Using radio interferometers (MERLIN, VLA) we have imaged the nebulosities of several symbiotic stars at scales approaching the binary separation over ten years, tracing developments of the structure which can only be understood as a combination of wind-interaction and irradiation. In 1999 to 2000, we supplemented this with HST WFPC2 imaging in multiple bands, allowing us to place the central stars relative to the nebulosities and each other, and to diagnose the nebular conditions. These HST observations are matched to the resolution of our MERLIN images, providing additional insight into, for example, the dust distribution in the nebulae. This is further complemented by ISO spectroscopy of a number of symbiotic stars in the HST programme, which can be modelled as two-shell dust distributions.

Helium flashers

When a star on the Asymptotic Giant Branch (AGB) begins to run out of fuel, it suffers one or more thermal pulses, also known as shell flashes. Eventually, the AGB star ejects its outer envelope, and the central core moves across the Hertzsprung-Russell (H-R) diagram to a significantly higher effective temperature before descending the white dwarf cooling track. In perhaps 15% of cases for intermediate mass stars (i.e. Sun-like ones) the residual hydrogen and helium atmosphere can re-ignite in a very late helium-shell flash.

These objects have been dubbed helium flashers, and undergo optical brightening as they reverse their route across the H-R diagram and become “born-again” giants.These objects are identified by the existence of the old PN (from the post-AGB phase) and extreme hydrogen deficiency (due to hydrogen ingestion during the flash). This stage of stellar evolution is very brief, taking perhaps a few years, a few centuries at the outside.

Consequently helium flashers are rarely observable, and indeed only one has been observed with non-optical instruments: V4334 Sgr, also known as Sakurai’s Object. This object, initially identified as a nova, shows both extreme H-deficiency an old PN.

Our observational program includes radio imaging of the PN and IR and mm-spectroscopy of both the born-again giant envelope and the ejecta. The former imaging provides the only reliable evidence of the structure and mass of the PN, due to the crowded nature of the field in the optical. The latter spectroscopy has followed dramatic changes in the star and its more recent ejecta, including: an early carbon star stage; early dust and molecule formation, including both 12CO and 13CO; the development of fast wind evidenced by an IR helium absorption line and the subsequent development of that line into emission only; dust obscuration and the subsequent development of the dust. As the star is currently optically obscured by the dust, our IR spectroscopic campaign is the only way of tracking the development.

Other candidate helium flashers include: V605 Aql (Nova Aql 1919), FG Sge (identified as brightening in the nineteenth century), and the PN A30 and A78 (H-deficient, and so perhaps the PN of helium flashers). One important distinction is that V4334 Sgr has apparently evolved across the H-R diagram much more rapidly than the other recent helium flashers; theorists can currently accommodate either the fast or the slow time-scale, but not both at once!

Interacting Binaries

Interacting binaries are pairs of stars orbiting around each other so closely that the two stars affect each other or ‘interact’ in some way. Gravitational interaction can lead to one or both stars being distorted from their normal spherical shape and can lead to gas (and hence mass) being transferred from one star to the other. Typically the mass-gaining star is a compact object such as a white dwarf, and as gas accretes onto it, gravitational energy is released, leading to a brightening of the system. Since accretion can occur spasmodically, these cataclysmic variable stars are characterised by dramatic variations in their brightness in many wavelengths. In non-magnetic binaries, the gas forms an accretion disc about the compact object. The study of cataclysmic variables allows us to test theories of mass transfer between stars, and the structure and behaviour of accretion discs. The latter in particular has wide applications in astrophysics, including in active galactic nuclei.

The study of cataclysmic variables at the Jeremiah Horrocks Institute uses multi-wavelength observations of eclipsing systems (where one star passes in front of the other). This allows us to constrain exactly where the emission is coming from. For example, when an eclipse is seen in the X-ray flux at the same time as in the optical lightcurve, it indicates that the X-ray emission is from the surface of the compact object.

Detailed modelling of the X-ray spectrum enables us to determine the temperature of the boundary layer between the accretion disc and the compact object. It is also possible to deduce how much gas lies above the accretion disc, partially obscuring the compact object and the inner disc. Simultaneous X-ray and ultraviolet observations have confirmed that the measured density of this so-called ‘iron curtain’ depends on which wavelength is being considered. The JHI’s involvement with XMM-Newton will open up new avenues in this area.

Lead researcher: Dr Stewart Eyres

Researchers: Dr Barbara Hassall

PDRA: Dr Mark Rushton