NQPI researchers weigh in on spintronic supercomputers

by Ben White

 10 May 2012

As anorexic iPods grow leaner each year, scientists work daily to improve electronics, making them faster, smaller and dependent on less energy. In the past decade, technology has seen the biggest leap in storage capacity per square millimeter in the history of electronics. Consumers can easily find Micro SD cards the size of a pinky fingernail that hold 64 GB, the equivalent of 14,000 songs. But how small can technology go?

Recently, computer research has shifted toward a strange, small domain with its own world of laws which even Einstein called “spooky.” Researchers dub the next generation of computing spintronics, named after an abstract property of quantum physics that man still has yet to fully grasp.

In 2007, European physicists Albert Fert and Peter Grünberg received the Nobel Prize for independently discovering a phenomena dubbed the Giant Magnetoresistive (GMR) effect a decade earlier. First, they stacked multiple thin strips of magnetic metals between layers of nonmagnetic film. When scientists introduced a magnetic field over the Kit-Kat-like semiconductor, its electrical resistance drastically decreased. This discovery led to virtually immediate and profitable results in computer memory. Sensors that read magnetic information in hard drives became much more delicate, allowing memory to be stored in smaller pieces.

Hard drives with GMR read heads hit the market quickly, creating an explosion of small, high-capacity media players and hard drives.

GMR works because it exploits spin, a characteristic of the electron. This abstract property, intrinsic only to subatomic particles, can best be described as angular momentum – a magnetic force either pointed up or down. Dr. Eric Stinaff, a physicist working on quantum information processing at Ohio University's Department of Physics and Astronomy, compares spin to a penny, either face up or face down.

Stinaff and other quantum physicists aim to use spintronics (short for spin transport electronics, also known as magnetoelectronics) in the realm of logic devices. In other words, scientists want to use an electron's spin as the bit in binary code. Current technology manipulates an electron's charge to direct it to a sensor. Whether an electron represents a “one” or “zero” depends on if the sensor recognizes a group of electrons in a certain slot. Since an electron's spin has the potential for two states (spin-up and spin-down), computers could store at least twice as much information with the same number of electrons. This would make computers faster, smaller and more energy-efficient.

Some scientists believe that D-Wave Systems, a Canadian technology company, achieved this in 2010, though much of the scientific community met their research with stiff criticism. D-Wave claims to have produced several quantum computers, using spintronics to create working spin manipulation systems. The company's original computer, Orion, drew a great amount of press when Google programmers used it to train their newest image classifying system. Now, D-Wave markets a $10,000 supercomputer which can only perform discrete optimization, a complex and limited function used in computer science and mathematics. Several scientists publicly doubted the reliability and impact of D-Wave's machine, and the consensus remains that a true leap in quantum physics stands in the way of creating a true quantum computer.

In the last week, two separate publications appeared that may signify the quick-moving nanoscience community planting its feet and leaning over its haunches in preparation for the immanent quantum leap.
In the April 26 edition of Nature, a group of scientists from the National Institute of Science and Technology explained how they created the largest-to-date quantum simulator, a simplified skeleton of spin manipulation technology. In a device known as a Penning Trap, researchers in effect created a two-dimensional plane onto which they inserted 300 beryllium ions. After cooling the entire “special-purpose analogue processor” to near-absolute zero, they manipulated the spin of the ions' outermost electron, catalyzing interactions between the quantum bits (qubits) that followed the accepted model of quantum behavior. Other laboratories had produced similar experiences, but on a much smaller and less applicable scale.

The most obvious hurdle for a commercially-viable quantum leap remains the fact that no one has ever conducted such an experiment near room temperature. However, an article published in the May 4th edition of Physical Review Letters suggests a new nanomaterial could solve this problem in the future. A team of theoretical physicists from the Universities of South Florida and Kentucky discovered through complex computations that graphene, a honeycomb-like sheet of carbon only one atom thick, could be used in combination with cobalt to achieve a surface which holds a net spin at room temperature.

The theorists believe that if physicists replace single carbon atoms at certain points in the lattice with cobalt atoms, the majority of the graphene sheet's spin could be controlled and manipulated at room temperature. Graphene excites many nanoscience researchers because of its high conductivity and toughness (graphene's breaking strength is 200 times greater than steel's). But all scientists believe a revolution remains far on the horizon.

Dr. Ian Appelbaum, an authority on spintronics from the University of Maryland, believes “there is a lot of hyperbole about how spintronics will change technology. I tend not to subscribe to this.”

Appelbaum maintains that a complete upgrade to quantum computing will pose too much of a technical challenge for tomorrow's scientists. In the short term, he believes spintronics will have a more dramatic impact on lasers, which research has shown can operate under half their current power when manipulated with spin technology.
Stinaff also doubts the “quantum leap” will extend to all electronics, but remembers modern computers evolved over a century. Instead of an overnight revolution in computing, he expects processors, video cards and other electronic components will soon incorporate spintronics with existing technology on a hybrid silicon chip.

While he admits that science requires a revolution in the theory and practice of quantum mechanics, Stinaff believes “if you can harness quantum properties of the spin, you have a whole new set of possibilities.”