The thesis presents a study of the temporal variability of X-rays from black hole systems to probe the geometry and physical processes of the accretion flow close to the black hole. The results of such analysis can determine the spin and mass of the black hole.

The stars of the night sky can to the naked eye appear to be steady and unchanging, apart from the twinkling created by air moving in the atmosphere. However, when viewed in X-rays, the sky is far from constant, with detectable changes occurring on very short timescales.

Black hole X-ray binaries are strong sources of X-rays. These systems contain a star and a black hole in orbit around each other. As matter from the companion star is accreted by the black hole, large amounts of gravitational energy are released, giving rise to strong X-ray emission. The accretion flow close to a black hole is characterised by strong gravity, high-energy radiation and variability on timescales down to milliseconds. These systems allow us to probe physics under conditions we cannot recreate in a laboratory, and provide some of the strongest observational indications of the existence of black holes. Temporal analysis is a powerful diagnostic of the geometry and physical processes of this environment.

The bulk of the thesis concerns studies of the rapid variability of perhaps the most well-known of all black hole binaries: Cygnus X-1. By tapping into the large amount of archival data available, a systematic study of the variability, in the form of the power spectrum, is conducted. The results show that timing studies can indeed give valuable information on the emission mechanisms and accretion geometry. Tying characteristic frequencies to effects predicted by general relativity directly gives information about the parameters of the compact object. Using these results, the past evolution of the binary system is studied.

Time:

Thursday 30 October, 14.00
Place: Observatory Museum, Drottninggatan 120

Link:
Abstract Contacts:

Magnus Axelsson +46-8-5537 8515, magnusa@astro.su.se