This thesis addresses ultrafast magnetization dynamics from a theoretical perspective. The manipulation of magnetization using the inverse Faraday effect has been studied, as well as magnetic relaxation processes in quantum dots.The inverse Faraday effect -the generation of a magnetic field by nonresonant, circularly polarized light -offers the possibility to control and reverse magnetization on a timescale of a few hundred femtoseconds. This is important both for the technological advantages and for the deeper fundamental understanding of magnetization dynamics that can be gained. However, several aspects of the inverse Faraday effect have remained poorly understood. The question of whether light can manipulate magnetization alone or whether an additional angular momentum reservoir is needed, in particular, remains unanswered.This question is answered here: the light beam that causes the inverse Faraday effect provides the angular momentum required for the magnetization to precess. No other reservoir is needed. This implies that manipulation of the magnetization occurs on the timescale of a laser pulse, which can be made extremely short. Even magnetization reversal on this timescale could be possible, provided a material with a sufficiently strong magnetooptical response can be found. This is a technical challenge, not a fundamental obstacle.The Faraday effect in the presence of optical birefringence has also been analyzed. This effect has been used for imaging magnetization dynamics in transparent media on an ultrafast timescale, but transparent magnetic materials usually have a complex crystal structure and complicated optical properties, which render the relationship between Faraday rotation and magnetization unclear. We have shown that the Faraday effect can be used to measure the instantaneous magnetization, even in the presence of birefringence, provided certain experimental conditions are met. Suggestions concerning these experimental conditions are made.The relaxation of magnetization, particularly the relaxation of a spin in a quantum dot, has been studied. This problem is relevant to the fields of quantum computing and highly-multiplexed optical memory, both of which are of great current interest. We have investigated the interaction of the spin with a metallic electrode and calculated the dephasing and dissipation rates. We found that under current experimental conditions, this relaxation pathway is negligible compared to spin-phonon scattering, but as systems are miniaturized, interactions with electrodes become more important.The methods developed to study the relaxation of a spin were also applied to the relaxation of a charge in a double quantum dot, another important problem in quantum computing. Again, we found that the interaction with a gate electrode is generally much weaker than the interaction with phonons but may be important for smaller systems.
Kurzfassung