By Candace Stuart
Small Times Senior Writer
Aug. 6, 2001 — Spintronics? No, it’s not the name of a ’60s pop band, but it is a small tech approach that is expected to eliminate the need to boot up mobile phones, computers and other electronic devices while making make them faster and more powerful.
Motorola Inc., a leader in the
 |
| Equipment at Pacific Northwest National Laboratory in Richland, Wash., allow scientists to control the filmmaking process atom by atom, varying amounts and temperatures subtly to produce a superior material. |
On another front, scientists at IBM and Pacific Northwest National Laboratory (PNNL) are expected to report Saturday at a spintronics conference in Washington, D.C., that they produced a magnetic semiconducting material that could open the door to quantum computing. Quantum computing is another promising application for spintronics. Unlike other existing magnetic semiconductors, the material made at PNNL retains its magnetism at room temperature.
Electrons, the subatomic particles that revolve around an atom’s nucleus, have spin as well as charge. The binary code used to record computer information is based on positive and negative charges, which disappear when the power source is turned off. That’s why information must be saved and stored on a hard drive before the computer is shut off, and reloaded to the local memory when the computer is revived.
Scientists are looking at spin as a solution. If the majority of electrons spin in the same direction, they create a magnetic force. If the spin can be controlled, it can be used in a similar binary system — as north and south. Because some magnets hold their magnetism indefinitely, data need not get stored before power is turned off, and information is available the instant the “on” button is pushed.
“You can, in principal, create a whole new generation of electronics,” said Sam Bader, leader of the magnetic film group at Argonne National Laboratory near Chicago. Argonne is focusing on basic properties of magnetic films and ways to make powerful, nanoscale magnets.
“They’ll have instant boot up and less of a power drain because (the memory) won’t need to be refreshed. They’ll be faster and smaller. … This could have major repercussions.”
Earlier this year, Motorola announced it developed a new memory chip dubbed MRAM, for magnetoresistive random access memory, a technology the company claims could replace existing memory parts in wireless products. Motorola expects to introduce the chips in its mobile phone and automotive sectors in about three years, according to Saied Tehrani, the MRAM technology manager at Motorola’s research labs in Tempe, Ariz.
“We feel we’re in the lead position,” he said. IBM and Honeywell are also developing MRAM devices, but Motorola is the first to unveil a 256-kilobit prototype. “We’ve demonstrated the most memory.”
The heart of Motorola’s MRAM system is a transistor that accesses information connected to a three-layered storage part: a free magnetic layer, a fixed magnetic layer with an insulating barrier sandwiched in between. The fixed layer maintains its north or south direction, while the free layer can vary as either north or south, depending on what bit of information is stored within it.
When both magnetic layers hold the same direction, they create a low field of resistance. If they are opposite, resistance is high. The transistor interprets that resistance, measured as voltage when a current passes through the insulating layer, as information.
Information remains stored in a single memory chip, allowing instant start-up and potentially a faster and more energy-efficient computing device. Motorola will design the chip with standard packaging techniques that require no infrastructure changes.
“There are no additional costs,” Tehrani said.
Scientists at IBM’s Almaden Research Center in San Jose, Calif., and PNNL in Richland, Wash., hope to use spin not for storage but for the actual computing parts. The area is known as quantum computing, a development that would vastly improve memory and speed. They moved that dream one step closer to reality by making a magnetized semiconducting film that stays magnetic at room temperature.
Other scientists have succeeded in synthesizing magnetic semiconductors, but the materials need to be cooled to impractical temperatures to keep their magnetism, said Scott Chambers, a chemist at PNNL and the project’s lead scientist. “No one wants a computer that operates at minus 240 degrees Fahrenheit,” he said.
Chambers’ team created the film in a controlled vacuum that allows them to deposit atoms individually on a substrate. They used forms of titanium and oxygen peppered with the metal cobalt for magnetism to grow the film.
Robin Farrow, a physicist and research staff member at IBM’s Almaden site and a collaborator with PNNL, said he is interested in this film-growing technique as a preliminary point for making a real product. But both scientists cautioned that the existing film, while a leap ahead for spintronics, was far from technologically practical.
“We’re nowhere close to having the ideal growth formula,” Chambers said.
The PNNL-IBM research builds on studies performed in Japan, where scientists used a scattershot approach to test numerous chemical mixes as potential magnetic semiconductors. “With this method, you can find a needle in a haystack, but it is not a very sharp needle,” Chambers said.
Chambers advanced that work by tweaking one of the more promising recipes, using the much more precise vacuum apparatus that he designed. The equipment allowed his team to control the filmmaking process atom by atom, varying amounts and temperatures subtly to produce a superior material.
Farrow analyzed the film to gauge its magnetic properties and said he was surprised that it was five to 10 times more magnetic than the Japanese film. Even better, it retained its magnetism when heated.
“We want these things to work way above room temperature,” he said, since most real world semiconductors must withstand hot electronic or mechanic environments. Neither he nor Chambers could explain why the film is so strongly magnetic, a question they hope to address with more research.
They also want to further explore the film’s semiconductor properties and substitute their present substrate, strontium titanate, for the types preferred by industry. The substrate, a flat surface that allows the film to grow into a consistently thin crystalline sheet, affects the properties of the film itself, making substitutions a tricky proposition.
“We don’t know if it can be grown as a semiconductor,” Farrow said. “The substrate is an important issue.”
Farrow said the ingredients themselves are readily available and inexpensive, a plus for industry. And he pointed out that silicon, which is ubiquitous in today’s semiconductors, itself was riddled with quality and fabrication problems at one time. He suggested applying some of the techniques developed to make silicon a better semiconductor to the new film, called anatase titanium dioxide.
“There are a lot of tricks in those (semiconductor) materials,” Farrow said.
Chambers, Farrow and their colleagues have submitted a paper on their findings to the journal Applied Physical Letters for review. They will discuss their work Saturday in a poster presentation at the Spintronics Workshop at George Washington University in Washington, D.C.
null
CONTACT THE AUTHOR:
Candace Stuart at [email protected] or call 734-528-6290.