Three Methods of Determining Magnitudes of Individual Stars in Resolved Binary Systems

The only way to measure the mass of star is by analyzing its orbit with Kepler’s laws. Knowing the luminosity of a star in addition to its mass is very useful in theoretical astrophysics. For single stars with known distances, it is straightforward to calculate the luminosity. For binary stars determining the luminosity is often difficult, as most techniques measure the blended light from both and not that of the individual components. Three methods for measuring the differential magnitudes of resolved binary stars are investigated: adaptive optics (AO), aperture masking and bispectrum analysis. Of the three techniques used here adaptive optics performed the best, producing differential magnitudes in Johnson I, R, V and B filters for 36 stars. The observational technique was to take many exposures of the same object in an effort to increase the signal-to-noise ratio. It was discovered that AO frames show frame-to-frame intensity variations. These inter-frame variations showed up in AO data from both the 1.5-m telescope at the Starfire Optical Range and the 100-inch telescope at Mt. Wilson Observatory. These variations are most likely caused by the failure of the AO system to fully compensate for the atmospheric distortions. Aperture masking consists of dividing the telescope’s aperture into several subapertures. The light from each subaperture interferes and creates fringes. The brightness ratio of the binary can be determined from the resulting fringe pattern. The technique was hampered by the Intensified CCD (ICCD) detector used in these experiments and performed rather poorly, producing only one differential magnitude measurement. The main thrust of the bispectrum technique was to derive differential magnitudes from archival speckle data. The standard bispectrum technique is used to reconstruct images from speckle data, but it requires the observation of calibration stars, which are not needed in the standard speckle method. Instead, the method used here fits the phase of the bispectrum to a model binary star. This avoids having to use a calibrator star, because the phase of the bispectrum is not distorted by the atmosphere. Unfortunately this technique seems ill suited to ICCD speckle data, though it looks promising for photon-limited data