Luisa Morales-Rueda: Research interests

Binary star evolution

In the last couple of years I have worked in a particular subset of binary systems, the subdwarf B binary stars, which consist of a subdwarf B star and a companion. Subdwarf B stars (sdB) are thought to be core helium burning stars with a very thin hydrogen envelope. The companion can either be a white dwarf of a main sequence star. These subset of binaries are extremely important from the point of view of binary star evolution because, a) after common envelope phase they are detached systems and their binary parameters (e.g orbital period) do not change considerably and b) because the selection effects are not significant like in other binary systems that are detected because they change in brightness. When we observed a sdB binary and determine its orbit we are sampling directly the end product of binary evolution. This is not true for interacting binaries like cataclysmic variables where mass transfer plays an important part on their evolution. In collaboration with groups in Britain and Germany I have measured the orbits of more than 30 sdB binaries (Maxted et al 2002, Morales-Rueda et al 2003a, 2003b, 2004) where only 10 were known before we started our project). See Fig.1 for the orbital period distribution of sdB binaries known up to now (updated June 2003).

Morales-Rueda et al 2003
Han et al. 2003 The first result from this study was the realisation that evolutionary theories were missing some important formation channel as they could not explain the orbital period distribution found in our data. Han et al (2002,2003) made use of our observations to justify three formation channels for sdB binaries. In one of them, the binary would go through common envelope phase only once giving as a result an sdB star with a main sequence companion. We found at least two such systems in our sample. Most of the targets in our sample seem to have formed through a channel where they system went through two common envelope phases resulting in a white dwarf companion to the sdB star. They also predicted a formation channel that had not been considered in detail before. For this channel there would not be a common envelope phase but stable Roche lobe overflow. The system that would result from this formation channel would be a very long period binary, i.e. a few hundred days. We have found one such example in our sample, which only now starts to have a long enough time baseline. Fig. 2 shows the predicted orbital period distribution resulting from the three formation channels discussed. Solid line: binaries resulting from the first common envelope phase, dotted line: binaries resulting from the second common envelope phase, dashed line: binaries resulting from the Roche lobe overflow channel. Our data points are represented by ticks at the bottom of the figure. Notice that the very long orbital period system we have found is not plotted in the figure. this figure has been taken from Han et al (2003).
We are also interested in studying the companions to these sdB stars as their masses and metallicities give us clues to their evolution. In particular we can distinguish between white dwarf and main sequence companions by looking for reflection effect in their lightcurves. This is easily done when their orbital periods are less a a couple of days. Binaries with periods longer than that will not show reflection effect even if the companion is a large main sequence star.

Interacting binary stars

I am also interested in the field of interacting binary stars. An interacting binary system (as opposed the detached systems I have described above) is one in which there are two stars close enough to feel each other's gravitational pull. This leads to mass being transferred from the less massive star to the more massive one. In particular I am interested in cataclysmic variable stars, in which one of the stars is a white dwarf and the companion is a main sequence star (sometimes it can be a slightly evolved star actually). They are classified in various ways depending on different factors like the magnetic field of the white dwarf or their photometric behaviour. Accretion processes are common in the universe, in small systems like cataclysmic variables and in large systems like active galaxies. In cataclysmic variables most of the accretion is thought to happen by means of the formation of an accretion disk around the accreting object. This is also the accepted model for accretion onto a massive black hole in the centre of an active galaxy. Although we could learn much about accretion from the study of active galactic nuclei the timescales in which these accrete are too large for us to get a hold on what is happening. The timescales in which we see the results of accretion in binaries are of the order of hours to years. This makes binaries the key to understanding mass transfer and accretion processes. The next figure shows what a cataclysmic variable looks like (thanks to Rob Hynes for his visualisation code)

We can learn much about accretion by studying cataclysmic binaries in which the accretion rates change periodically. This is exactly the behaviour of dwarf novae. These cataclysmic variables show periodical increases in brightness usually of several magnitudes. These increases in luminosity, referred to as outbursts, occur every few days or even decades depending on the particular binary. They are thought to be caused by an increase in the accretion rate of the system. The favoured explanation for this increase in the accretion rate is given by the Disk Instability model. An increase in the viscosity of the gas in the accretion disk around the accreting object causes the disk to empty itself onto the white dwarf in a very short timescale. A great deal is known about these objects but there are still fundamental questions which we do not have an answer for. For example, although it is known that transport in the disk is driven by some kind of viscosity, it has been impossible to pinpoint its origin. It is extremely difficult to develop a theory of plasma dynamics that includes all the parameters that we think fundamental for accretion disk dynamics, i.e magnetic fields, tidal instabilities, gravitational radiation amongst others.

I am also interested in the study of tidal instabilities as a possible source of viscosity in the disc. This is a plausible explanation for the presence of spiral structure in the discs of some dwarf novae. I have done some work on the first dwarf novae that showed these spiral structure (Morales-Rueda et al 2000) and intend to monitor the evolution of this structure during outburst and quiescence. This monitoring will help up distinguish between the different mechanisms that have been proposed to explain spiral shocks, i.e. brightness enhancements caused by tidal stresses in the disc and shocks. Fig. 4 shows the dwarf novae IP Peg during outburst. This picture is in velocity coordinates and is produced using Doppler tomography, an indirect imaging technique.

We have compiled a sample of dwarf novae that show strong He II emission in their optical spectrum (Morales-Rueda & Marsh, 2002). It has been suggested that this emission would be the result of reprocessed light from regions of the disk that have thickened as a result of tidal instabilities. We intend to follow up these systems and answer the question of how often this spiral structure is present.

We can also learn a great deal about accretion physics by studying discs composed mainly of helium instead of hydrogen (accretion discs in cataclysmic variables are composed mainly of hydrogen). This can achieved by looking at other class of binaries called AM~CVn systems where the mass transferring star is hydrogen deficient. The higher ionisation rate of helium causes these discs to be in steady state at very low accretion rates, lower than if they were made of hydrogen. I am involved in several projects concerning the study of AM~CVn systems.

The Faint Sky Varibility Survey

The Faint Sky Variability Survey (FSVS) is aimed at finding photometric and/or astrometric variable objects between 16th and 24th mag on time-scales between tens of minutes and years with photometric precisions ranging from 3 millimag to 0.2 mag. An area of 23 squared degrees, located at mid and high Galactic latitudes, was covered using the Wide Field Camera (WFC) on the 2.5-m Isaac Newton Telescope (INT) on La Palma.

In the following colour-colour diagrams we show the point sources from the FSVS that show no variability (top left panel), the point sources that show variability (top right panel) and the ratio between variable and non-variable point sources (bottom panel). Variability is determined in each case by calculating the $\chi^2$ of the light curve with respect to its average value. Objects with $\chi^2$ above the 5-$\sigma$ variability level are considered variable (Groot et al. 2003). The ratio is presented in percentages. The main difference between the distribution of variable and non-variable objects is the excess of variable systems at colours B$-$V$\sim$0 and V$-$I$\sim$1. These sources will be mostly QSOs but we also expect any cataclysmic variables (CVs) in the field to appear in that colour-colour region. The fraction of variable to non-variable objects along the main sequence seems fairly uniform, but more rigorous calculations are required before we can draw any conclusions.

We also display the results in colour-colour diagrams for different ranges of variability timescales. The ranges have been selected so the first would include orbital periods corresponding to CVs (variability scales up to 6 hours), the second RR Lyr stars (scales from 6 hours to 1 day), the third longer variability trends of CVs (from 1 to 4 days), and anything else (above 4 days). We have also plotted the 3-$\sigma$ upper limit of the main sequence for clarity. We find no obvious correlation between the time scales of the variability and their amplitudes.

In a preliminary analysis of the FSVS we find that, down to 24 mag, there are of the order of 500 objects in the variability range corresponding to CVs and 300 in the range corresponding to RR Lyr stars. We find 62 sources showing longer variability periods.