Throughout history man has been fascinated by the wandering stars in the sky, the planets, whose godliness seemed to afford them the ability to travel freely from night to night through the background of stars. Though it has only been very recently that we have been able to detect these heavenly bodies in stellar systems other than our own, the idea of extrasolar planets itself is fairly fundamental and has been around for centuries. It is not a far stretch to imagine that since our star has planets, than some of the others must as well. However, the technology needed for detection of these distant stellar companions simply was not available until towards the end of the twentieth century. In the last ten years, the extrasolar planet count has grown from a few to 200 known planets. This relatively new area of research is booming and brimming with possibilities with billions of stars yet to have a telescope trained on them for the purpose of planet hunting. Although the number of exoplanets is growing large, only a handful of those planets are members of a binary or multiple star system. And since the majority of stars in the galaxy exist in systems of two or more, it is to these systems that we will focus the scope of our search.

(image credit: David A. Aguilar, Harvard-Smithsonian Center for Astrophysics.)

Technique

Using the transit method of planet detection, planets passing between their parent star and earth will produce a decrease in brightness of the star during the eclipse. Thus, if we take a series of images of the star for a period of time, we can use photometry to determine the magnitude of the star over that period and produce a light (magnitude vs. time) curve. Planets will appear on the light curve as periodic dips in magnitude. The disadvantage of this method is that the planetary plane of orbit must be nearly coincidental with the line of sight to the star, or else the planet will not eclipse the star as seen by the Earth. We can, however, use the nature of binary stars to increase our chances of success with this method. Planetary companions in a binary star system are likely to occupy the same plane of orbit as the stars themselves. Therefore, if we pick known eclipsing binary systems, any planets in the system are likely to at some point pass over our line of sight to the star.

(image credit: Steven Simpson, Sky and Telescope.)

Theory

When dealing with planetary orbits in binary star systems, there are two fundamental types of orbits that we must consider. A planetary, or P-type, orbit is defined by a planet orbiting outside both stars, with the stellar barycenter providing the planet's center of orbit. The other type is the satellite, or S-type orbit, in which a planet orbits around one of the stars. There are two questions that arise when discussing the theoretical existence of planets in binary systems. First is the question of whether or not planets can form in such systems. Second, can planets achieve stable orbits. These issues have been addressed in the literature, yielding favorable results. Planets should be able to form and stay stable for a wide range of initial conditions in both the P-type and S-type situations.

Candidate Systems

As mentioned above, I chose eclipsing binary systems so that there exists a higher probability of planetary orbits coinciding with our line of sight to the system. Detectability also has a great deal to do with how much the system will dim when the planet passes in front of the star. In other words, the dip in the light curve must be larger than the noise of the signal. This is where binary systems are a disadvantage. A planet passing in front of one companion will leave the other unmolested and therefore the dip will be less than in a single star system. To account for this, I chose systems with deep primary eclipses. The advantage of this is twofold. First, a deep eclipse signifies that most of the luminosity of the system is from one star, so a planet eclipse of this star would block a higher percentage of light. Second, a deep eclipse may signify that the system is nearly edge on, giving a higher probability of detecting a planet transit. Low luminosity systems such as M dwarfs and K dwarfs are of particular interest. A planet transiting a low luminosity star will block a much larger percentage of the system's light than a similar mass planet around a brighter star. I also included systems where one of the components is a white dwarf. The reasoning here is that if a planet passes in front of a luminous yet small white dwarf, it will eclipse most of the light in the system and produce a large dip in brightness. Systems with solar type stars were also included despite their relatively high luminosity, because the majority, 58 percent, of known extrasolar planets orbit around G type stars. However, this could be do to an observational bias, i.e. astronomers may overwhelmingly prefer to search solar type stars for planets simply because Earth orbits a G star, and Sol is the only star known to have a habitable terrestrial planet in orbit around it.

During our observations of CM Dra, in which we were looking for the presence of orbiting planets, six stellar flares were observed over the course of one week in May 2006, one of which was very intense. Check out our publication on these events in the Information Bulletin on Variable Stars by clicking the link below.