Abstract
During the last few years, there has been considerable interest in devices of the spin transistor geometry such as those of M. Johnson. These devices consist of current flowing from a ferromagnet to a paramagnet, with voltage probes placed perpendicular to the current flow in the paramagnet. First theoretical calculations require anomalously large spin polarizations in order to explain the magnitude of the observed effect. To obtain a more detailed understanding of this and similar experiments, we have solved a set of coupled drift-diffusion equations for spin-up and spin-down electrons in two and three dimensions. This allows us to obtain the spin current, the charge current, and the electrochemical potentials for both spins of electrons in the same geometry as the actual experiments. As expected we can see the separation of spin and charge currents in these devices. More importantly, for large voltage probes the potential is not constant across the part of the voltage probe touching the sample. The actual voltage measured depends in detail on how the spin current and charge currents relax within the voltage probes. This may explain some of the discrepancy in understanding the spin transistor experiments. We have also examined a number of other nonplanar geometries. In several of them we are able to realize a spin-current source in which there is a net flow of magnetization, but no net charge current.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call