The spin-orbit effect is at the heart of efforts to merge spintronics-where information is carried and stored by spin, rather than by charge-with semiconductor technology.
Subject Areas: SpintronicsThe 2007 Nobel Prize in Physics (awarded to Albert Fert and Peter Grünberg) highlighted the remarkably rapid transition of "spintronics" from fundamental studies of spin-dependent transport in metallic ferromagnetic multilayers [1] to a device technology critical to the magnetic storage industry. While the mainstream of spintronics continues to expand the scientific and technological frontiers of ferromagnetic metal devices [2], a parallel effort in semiconductor spintronics is growing vigorously [3]. This incarnation of spintronics has the same general motivation as metallic spintronics: to understand and control the transport of spin-polarized currents and to eventually apply this knowledge in information technologies. However, semiconductor spintronics also brings with it the promise of integrating the best qualities of two disparate worlds: the unparalleled storage capacity of magnetic memory and the impressive computing power of semiconductor logic. Add to this the possibility of exploiting the relatively long-lived quantum coherence of spin states in semiconductors [4] and one can even imagine unleashing the full power of the quantum world in truly revolutionary devices that exploit both the amplitude and phase of wave functions. Needless to say, the realization of this potential requires concerted efforts aimed at both understanding the mechanisms for spindependent transport in semiconductors as well as critically comparing spin-based device schemes with existing technologies [5].At first glance, it seems almost a given that ferromagnetism would be a necessary and integral component of any scheme for semiconductor spintronic devices. For instance, a semiconductor spintronic device generically requires an imbalance between spin "up" and "down" populations of electrons (or holes). We can imagine this imbalance being created by the injection of spin-polarized charge carriers from a ferromagnet, which acts as a spin polarizer. Alternatively, we could build devices from ferromagnetic semiconductors that have an intrinsic spin imbalance. Indeed, important advances have been made in semiconductor spintronics by using these very notions, with a number of interesting proof-of-concept semiconductor spintronic device demonstrations that incorporate ferromagnetic elements for injecting, detecting, and manipulating spins [6]. But, discoveries in recent years have inspired a completely different avenue to semiconductor spintronics-one that does not involve any ferromagnetism whatsoever [7][8][9].The development of this alternate track-"spintronics without magnetism"-relies on our ability to manipulate carrier spins in semiconductors through the spin-orbit interaction. This is an attractive pathway for designing semiconductor spintronic devices because spin-orbit coupling enables the generation and manipulation of spins solel...