UC Berkeley

Due to increasingly pervasive computation hardware, energy efficient electronics has gained much research and development attention. One approach toward this goal is to use spintronics, which leverages the spin degree of freedom of electrons. The fundamentally smaller energy dissipation from flipping spins instead of moving charges may lead to overall more efficient devices. This is in part due to the ability to retain information without any power as well as smaller amounts of Joule heating.

In this dissertation, we focus on a subset of spin logic devices known as nanomagnetic logic. Using a variety of modeling techniques and experimental imaging techniques, we investigate the performance of these devices in terms of speed and reliability. We also use optical techniques to investigate a spin-charge coupling mechanism known as the Spin Hall Effect. This effect is a very power-efficient way to switch nanomagnets and has obvious applications to both spin logic and memory. While spin logic devices still face many challenges, spin memory devices have become fairly feasible. With the rapid pace of innovation in this field, the outlook for spin devices in general is promising.