Quasiclassical methods for spin-charge coupled dynamics in low-dimensional systems

Spintronics is a new field of study concerned with the manipulation of the spin degrees of freedom in solid state systems. One of its main goals is the realization of a new generation of devices capable of making full use of, besides the charge, the electronic – and possibly nuclear – spin. At least in principle all-electrical control of the spin, which would be preferable from the point of view of applications, can be achieved through spin-orbit interaction. In recent years a number of spin-orbit related effects in low-dimensional fermionic systems have made this possibility much more concrete. We cite as prominent example the spin Hall effect, first proposed in 2003 for a twodimensional hole gas and soon after observed. In more general terms, the description of spin-charge coupled dynamics in mesoscopic, disordered low-dimensional systems is a non-trivial matter. For the task we relied on the quasiclassical formalism, an approximate quantum-field theoretical formulation ideally suited for tackling transport problems on the mesoscopic scale, with a solid microscopic foundation and a resemblance to standard Boltzmann transport theory that makes for physical transparency. The goal of this thesis has been two-fold: on one hand the generalization of the quasiclassical equations so as to make them capable of describing spincharge coupled dynamics, on the other the application of such equations to particular physical problems. Among them the spin Hall effect and voltageinduced spin polarization in two-dimensional, disordered electron gases under a variety of conditions – static or time dependent, in infinite systems, in confined geometries with different boundary conditions.