Magnetic contacts on the surface of an n-type silicon layer enable generation of a spin current that flows separately from a charge current. The spin orientation is electrically detected as a voltage at secondary magnetic contacts. The contacts' relative magnetizations allow full control over the spin orientation in the silicon channel. We recently accomplished this in a lateral transport geometry using lithographic techniques compatible with existing device geometries and fabrication methods.1,2 This achievement is a key enabling step towards developing devices that rely on electron spin rather than charge. Progress in this emergent field of ‘semiconductor spintronics’ is expected to lead to devices that exhibit better performance with lower power consumption and heat dissipation.
The electronics industry relies predominantly on controlling charge flows. It has continuously been improving the performance of existing electronics through physical-size reductions of elements such as transistors. However, size scaling cannot continue indefinitely as atomic length scales are reached, and new approaches must be developed. Basic research has shown that spin angular momentum, a fundamental property of electrons, can be used to store and process information in metal and semiconductor-based devices.
The 2007 Nobel Prize in physics was awarded for the discovery of giant magnetoresistance, a phenomenon based on spin-polarized electron currents in metals. This research went from discovery in 1988 to commercialization in approximately 10 years. It is credited with the ubiquitous availability of low-cost, high-density hard-disk drives in consumer products ranging from computers to video games and handheld electronics. The spin angular momentum of electrons can be used to store and process information in semiconductor devices similarly as in metals. In fact, the International Technology Roadmap for Semiconductors has identified use of the electron spin as a new state variable that should be explored as an alternative to its charge. Using pure-spin currents to process information is regarded as the ‘holy grail’ of semiconductor spintronics, since it liberates one from the constraints of capacitive-time constants, resistive-voltage drops, and heat buildup that accompany charge motion.
Much of the initial successes in this field were achieved in type III-V semiconductors with a direct band gap, such as gallium arsenide, where powerful optical spectroscopic techniques can be applied relatively easily to enable detailed insights into spin-system behavior.3 We recently made significant strides towards using spin transport in silicon, an indirect-gap material, by demonstrating efficient injection of spin-polarized electrons from a ferromagnetic metal contact.4 We have now taken an important step towards realization of a functional silicon spintronic device. We first injected a spin-polarized electric current from a ferromagnetic iron/aluminum oxide tunnel-barrier contact into silicon, which generates a pure-spin current flowing in the opposite direction (see Figure 1). This spin current produces shifts in the spin-dependent electrochemical potential, which can be electrically detected as a voltage outside the charge path at a second magnetic contact.
Figure 1. Schematic of the lateral device with nonlocal detection that was used to demonstrate the electrical injection, detection, and modulation of spin current in silicon (Si). A charge current (I) of spin-polarized electrons follows the applied voltage (V) and flows to the right, while a pure spin current flows to the left. B: Magnetic field. Au: Gold. Fe: Iron. Al2O3: Aluminum oxide.
Our team demonstrated that this voltage is sensitive to the relative spin orientation in the silicon and the degree of magnetization of the second contact. We further showed that the spin orientation in the silicon could be uniformly rotated by an applied magnetic field, referred to as ‘coherent precession,’ demonstrating that information could be successfully imprinted onto the spin system and read out as a voltage. Generation of spin currents, coherent spin precession, and electrical detection using magnetic tunnel-barrier contacts, as well as a simple lateral device geometry compatible with ‘back-end’ silicon processing, will greatly facilitate development of silicon-based spintronic devices, which represents our continuing research direction.
Olaf van 't Erve, A. T. Hanbicki, M. Holub, C. H. Li, C. Awo-Affouda, G. Kioseoglou, P. E. Thompson, B. T. Jonker
US Naval Research Laboratory
1. O. M. J. van 't Erve, A. T. Hanbicki, M. Holub, C. H. Li, C. Awo-Affouda, P. E. Thompson, B. T. Jonker, Electrical injection and detection of spin-polarized carriers in silicon in a lateral transport geometry, Appl. Phys. Lett. 91, pp. 212109, 2007.
2. O. M. J. van 't Erve, C. Awo-Affouda, A. T. Hanbicki, C. H. Li, P. E. Thompson, B. T. Jonker, Information processing with pure spin currents in silicon: spin injection, extraction, manipulation, and detection, IEEE Trans. Electron Dev. 56, pp. 2343, 2009.