Five years ago, two groups of theorists made a bold proposal: If you apply a voltage to the ends of a semiconducting strip, spin-up electrons will accumulate along one edge, while spin-down electrons will accumulate along the other.
Spin segregation had already been predicted to occur when electrons scatter off impurities, an effect known as the extrinsic spin Hall effect. What made the new proposal intriguing is the mechanism: Under the right conditions, a material’s intrinsic band structure, not its extrinsic impurities, would sort the spins.
Whether this intrinsic spin Hall effect could be observed was controversial. Even in that paragon of purity, epitaxially grown gallium arsenide, scattering seemed likely to smother the effect (see Physics Today, Physics Today 0031-9228 58 2 2005 17 https://doi.org/10.1063/1.1897513 February 2005, page 17 ).
Despite the initial controversy—and perhaps because of it—the two papers set off an explosion of interest in band-based...
