Researchers in the US have built the first silicon spin field-effect transistor (spinFET), which uses an applied voltage to control a current of spin-polarized electrons. Although this is not the first spinFET -- a device was built two years ago using a carbon nanotube -- it is the first to be made of silicon. The component is an an important step towards the first commercial "spintronic" devices, which could use the spin -- as well as the charge -- of electrons to store and process information more quickly and efficiently than conventional transistors (Appl. Phys. Lett. 91 072501).
The spinFET was made by Biqin Huang and Ian Appelbaum of the University of Delaware, along with Douwe Monsma of Cambridge Nanotech in Massachusetts, who earlier this year worked out a way to transport a spin-polarized current of electrons 10 µm through a piece of silicon. Their new device builds on this work, taking advantage of the fact that the direction of the spin-polarization can be rotated as the electrons move through the silicon by applying a magnetic field.
The device was set up so that the current entering the silicon was spin-polarized in one direction – “up”, for example. As the electrons move through the silicon, they are exposed to the magnetic field for a relatively long time and therefore experience significant spin rotation. However, if an electric field is applied along the direction of travel, the electrons move more quickly, spend less time in the magnetic field and are rotated less.
Once the current has traversed the silicon, it goes through a spin filter, which only passes the portion of the current that is still polarized in the up direction. As a result, if no electric field is applied, very little current is detected at the collector. However, if a voltage is applied along the silicon to create an electric field, more current is passed through to the collector. In one experiment, the current at the collector increased by a factor of about seven when the voltage is increased from zero to about 3 V.
The device is similar to a conventional FET because it uses an electric field to control an output current. While it is not the first spinFET – that honour goes to Christian Schönenberger and colleagues at the University of Basel, Switzerland, who built a carbon nanotube-based device two years ago — Appelbaum told physicsworld.com that theirs is the first spinFET made from silicon. This is significant because silicon-based spintronics should be compatible with today’s commercial chip-making processes.
Indeed, the spintronics pioneer David Awschalom of the University of California, Santa Barbara told physicsworld.com that the silicon spinFET is “an important step driving the transition from research to practical applications in spintronics”.
However, Appelbaum remains cautious, citing two significant challenges that remain before a commercial spinFET becomes a reality. Their device is based on the “ballistic” transport of electrons through thin magnetic films, which results in very small output currents on the order of tens of picoamps. Also, their device must be operated a very low temperatures – about 85 K – which would be impractical for a commercial device. “This [temperature] can easily be increased substantially”, says Appelbaum, “but design changes are most likely necessary for demonstration of spin transport at room temperature”.
