A basic discovery in magnetic semiconductors could result in a new generation of devices for sensors and memory applications — and perhaps, ultimately, quantum computation — physicists from the California Institute of Technology and the University of California at Santa Barbara have announced. The new phenomenon, called the giant planar Hall effect, has to do with what happens when the spins of current-carrying electrons are manipulated. For several years scientists have been engaged in exploiting electron spin for the creation of a new generation of electronic devices –hence the term “spintronics” — and the Caltech-UCSB breakthrough offers a new route to realizing such devices.
From Caltech:
Discovery of giant planar Hall effect could herald a generation of “spintronics” devices
A basic discovery in magnetic semiconductors could result in a new generation of devices for sensors and memory applications — and perhaps, ultimately, quantum computation — physicists from the California Institute of Technology and the University of California at Santa Barbara have announced.
The new phenomenon, called the giant planar Hall effect, has to do with what happens when the spins of current-carrying electrons are manipulated. For several years scientists have been engaged in exploiting electron spin for the creation of a new generation of electronic devices –hence the term “spintronics” — and the Caltech-UCSB breakthrough offers a new route to realizing such devices.
The term “spintronics” is used instead of “electronics” because the technology is based on a new paradigm, says Caltech physics professor Michael Roukes. Rather than merely using an electric current to make them work, spintronic devices will also rely on the magnetic orientation (or spin) of the electrons themselves. “In regular semiconductors, the spin freedom of the electrical current carriers does not play a role,” says Roukes. “But in the magnetic semiconductors we’ve studied, the spin polarization — that is, the magnetism — of electrical current carriers is highly ordered. Consequently, it can act as an important factor in determining the current flow in the electrical devices.”
In the naturally unpolarized state, there is no particular order between one electron’s spin and its neighbor’s. If the spins are aligned, the result can be a change in resistance to current flow.
Such changes in resistance have long been known for metals, but the current research is the first time that semiconductor material has been constructed in such a way that spin-charge interaction is manifested as a very dramatic change in resistivity. The Caltech-UCSB team managed to accomplish this by carefully preparing a ferromagnetic semiconductor material made of gallium manganese arsenide (GaMnAs). The widely-used current technology employs sandwiched magnetic metal structures used for magnetic storage.
“You have much more freedom with semiconductors than metals for two reasons,” Roukes explains. “First, semiconductor material can be made compatible with the mainstream of semiconductor electronics; and second, there are certain phenomena in semiconductors that have no analogies in metals.”
Practical applications of spintronics will likely include new paradigms in information storage, due to the superiority of such semiconductor materials to the currently available dynamic random access memory (or DRAM) chips. This is because the semiconductor spintronics would be “nonvolatile,” meaning that once the spins were aligned, the system would be as robust as a metal bar that has been permanently magnetized.
The spintronics semiconductors could also conceivably be used in magnetic logic to replace transistors as switches in certain applications. In other words, spin alignment would be used as a logic gate for faster circuits with lower energy usage.
Finally, the technology could possibly be improved so that the quantum states of the spins themselves might be used for logic gates in future quantum computers. Several research teams have quantum logic gates, but the setup is the size of an entire laboratory, rather than at chip scale, and therefore still unsuitable for device integration. By contrast, a spintronics-based device might be constructed as a solid-state system that could be integrated into microchips.
A full description of the Caltech-UCSB team’s work appeared in the March 14 issue of Physical Review Letters [Tang et al, Vol 90, 107201 (2003)]. The article is available by subscription, but the main site can be accessed at http://prl.aps.org/. This discovery is also featured in the “News and Views” section of the forthcoming issue of Nature Materials.
Contact: Robert Tindol (626) 395-3631
