Wednesday, March 23, 2011

Templated Growth Technique Produces Graphene Nanoribbons With Metallic Properties

"We can now make very narrow, conductive nanoribbons that have quantum ballistic properties," said Walt de Heer, a professor in the School of Physics at the Georgia Institute of Technology."These narrow ribbons become almost like a perfect metal. Electrons can move through them without scattering, just like they do in carbon nanotubes."

De Heer was scheduled to discuss recent results of this graphene growth process March 21st at the American Physical Society's March 2011 Meeting in Dallas. The research was sponsored by the National Science Foundation-supported Materials Research Science and Engineering Center (MRSEC).

First reported Oct. 3 in the advance online edition of the journal Nature Nanotechnology, the new fabrication technique allows production of epitaxial graphene structures with smooth edges. Earlier fabrication techniques that used electron beams to cut graphene sheets produced nanoribbon structures with rough edges that scattered electrons, causing interference. The resulting nanoribbons had properties more like insulators than conductors.

"In our templated growth approach, we have essentially eliminated the edges that take away from the desirable properties of graphene," de Heer explained."The edges of the epitaxial graphene merge into the silicon carbide, producing properties that are really quite interesting."

The"templated growth" technique begins with etching patterns into the silicon carbide surfaces on which epitaxial graphene is grown. The patterns serve as templates directing the growth of graphene structures, allowing the formation of nanoribbons and other structures of specific widths and shapes without the use of cutting techniques that produce the rough edges.

In creating these graphene nanostructures, de Heer and his research team first use conventional microelectronics techniques to etch tiny"steps" -- or contours -- into a silicon carbide wafer whose surface has been made extremely flat. They then heat the contoured wafer to approximately 1,500 degrees Celsius, which initiates melting that polishes any rough edges left by the etching process.

Established techniques are then used for growing graphene from silicon carbide by driving off the silicon atoms from the surface. Instead of producing a consistent layer of graphene across the entire surface of the wafer, however, the researchers limit the heating time so that graphene grows only on portions of the contours.

The width of the resulting nanoribbons is proportional to the depth of the contours, providing a mechanism for precisely controlling the nanoribbon structures. To form complex structures, multiple etching steps can be carried out to create complex templates.

"This technique allows us to avoid the complicated e-beam lithography steps that people have been using to create structures in epitaxial graphene," de Heer noted."We are seeing very good properties that show these structures can be used for real electronic applications."

Since publication of the Nature Nanotechnology paper, de Heer's team has been refining its technique."We have taken this to an extreme -- the cleanest and narrowest ribbons we can make," he said."We expect to be able to do everything we need with the size ribbons that we are able to make right now, though we probably could reduce the width to 10 nanometers or less."

While the Georgia Tech team is continuing to develop high-frequency transistors -- perhaps even at the terahertz range -- its primary effort now focuses on developing quantum devices, de Heer said. Such devices were envisioned in the patents Georgia Tech holds on various epitaxial graphene processes.

"This means that the way we will be doing graphene electronics will be different," he explained."We will not be following the model of using standard field-effect transistors (FETs), but will pursue devices that use ballistic conductors and quantum interference. We are headed straight into using the electron wave effects in graphene."

Taking advantage of the wave properties will allow electrons to be manipulated with techniques similar to those used by optical engineers. For instance, switching may be carried out using interference effects -- separating beams of electrons and then recombining them in opposite phases to extinguish the signals.

Quantum devices would be smaller than conventional transistors and operate at lower power. Because of its ability to transport electrons with virtually no resistance, epitaxial graphene may be the ideal material for such devices, de Heer said.

"Using the quantum properties of electrons rather than the standard charged-particle properties means opening up new ways of looking at electronics," he predicted."This is probably the way that electronics will evolve, and it appears that graphene is the ideal material for making this transition."

De Heer's research team hopes to demonstrate a rudimentary switch operating on the quantum interference principle within a year.

Epitaxial graphene may be the basis for a new generation of high-performance devices that will take advantage of the material's unique properties in applications where higher costs can be justified. Silicon, today's electronic material of choice, will continue to be used in applications where high-performance is not required, de Heer said.

"This is an important step in the process," he added."There are going to be a lot of surprises as we move into these quantum devices and find out how they work. We have good reason to believe that this can be the basis for a new generation of transistors based on quantum interference."


Source

Thursday, March 17, 2011

Room-Temperature Spintronic Computers Coming Soon? Silicon Spin Transistors Heat Up and Spins Last Longer

"Electronic devices mostly use the charge of the electrons -- a negative charge that is moving," says Ashutosh Tiwari, an associate professor of materials science and engineering at the University of Utah."Spintronic devices will use both the charge and the spin of the electrons. With spintronics, we want smaller, faster and more power-efficient computers and other devices."

Tiwari and Ph.D. student Nathan Gray report their creation of room-temperature, spintronic transistors on a silicon semiconductor this month in the journalApplied Physics Letters. The research -- in which electron"spin" aligned in a certain way was injected into silicon chips and maintained for a record 276 trillionths of a second -- was funded by the National Science Foundation.

"Almost every electronic device has silicon-based transistors in it," Gray says."The current thrust of industry has been to make those transistors smaller and to add more of them into the same device" to process more data. He says his and Tiwari's research takes a different approach.

"Instead of just making transistors smaller and adding more of them, we make the transistors do more work at the same size because they have two different ways {electron charge and spin} to manipulate and process data," says Gray.

A Quick Spin through Spintronics

Modern computers and other electronic devices work because negatively charged electrons flow as electrical current. Transistors are switches that reduce computerized data to a binary code of ones or zeros represented by the presence or absence of electrons in semiconductors, most commonly silicon.

In addition to electric charge, electrons have another property known as spin, which is like the electron's intrinsic angular momentum. An electron's spin often is described as a bar magnet that points up or down, which also can represent ones and zeroes for computing.

Most previous research on spintronic transistors involved using optical radiation -- in the form of polarized light from lasers -- to orient the electron spins in non-silicon materials such as gallium arsenide or organic semiconductors at supercold temperatures.

"Optical methods cannot do that with silicon, which is the workhorse of the semiconductor and electronics industry, and the industry doesn't want to retool for another material," Tiwari says.

"Spintronics will become useful only if we use silicon," he adds.

The Experiment

In the new study, Tiwari and Gray used electricity and magnetic fields to inject"spin polarized carriers" -- namely, electrons with their spins aligned either all up or all down -- into silicon at room temperature.

Their trick was to use magnesium oxide as a"tunnel barrier" to get the aligned electron spins to travel from one nickel-iron electrode through the silicon semiconductor to another nickel-iron electrode. Without the magnesium oxide, the spins would get randomized almost immediately, with half up and half down, Gray says.

"This thing works at room temperature," Tiwari says."Most of the devices in earlier studies have to be cooled to very low temperatures" -- colder than 200 below zero Fahrenheit -- to align the electrons' spins either all up or all down."Our new way of putting spin inside the silicon does not require any cooling."

The experiment used a flat piece of silicon about 1 inch long, about 0.3 inches wide and one-fiftieth of an inch thick. An ultra-thin layer of magnesium oxide was deposited on the silicon wafer. Then, one dozen tiny transistors were deposited on the silicon wafer so they could be used to inject electrons with aligned spins into the silicon and later detect them.

Each nickel-iron transistor had three contacts or electrodes: one through which electrons with aligned spins were injected into the silicon and detected, a negative electrode and a positive electrode used to measure voltage.

During the experiment, the researchers send direct current through the spin-injector electrode and negative electrode of each transistor. The current is kept steady, and the researchers measure variations in voltage while applying a magnetic field to the apparatus

"By looking at the change in the voltage when we apply a magnetic field, we can find how much spin has been injected and the spin lifetime," Tiwari says.

A 328 Nanometer, 276 Picosecond Step for Spintronics

For spintronic devices to be practical, electrons with aligned spins need to be able to move adequate distances and retain their spin alignments for an adequate time.

During the new study, the electrons retained their spins for 276 picoseconds, or 276 trillionths of a second. And based on that lifetime, the researchers calculate the spin-aligned electrons moved through the silicon 328 nanometers, which is 328 billionths of a meter or about 13 millionths of an inch.

"It's a tiny distance for us, but in transistor technology, it is huge," Gray says."Transistors are so small today that that's more than enough to get the electron where we need it to go."

"Those are very good numbers," Tiwari says."These numbers are almost 10 times bigger than what we need {for spintronic devices} and two times bigger than if you use aluminum oxide" instead of the magnesium oxide in his study.

He says Dutch researchers previously were able to inject aligned spins into silicon using aluminum oxide as the"tunneling medium," but the new study shows magnesium oxide works better.

The new study's use of electronic spin injection is much more practical than using optical methods such as lasers because lasers are too big for chips in consumer electronic devices, Tiwari says.

He adds that spintronic computer processors require little power compared with electronic devices, so a battery that may power an electronic computer for eight hours might last more than 24 hours on a spintronic computer.

Gray says spintronics is"the next big step to push the limits of semiconductor technology that we see in every aspect of our lives: computers, cell phones, GPS (navigation) devices, iPods, TVs."


Source

Friday, March 4, 2011

New Kinds of Superconductivity? Physicists Demonstrate Coveted 'Spin-Orbit Coupling' in Atomic Gases

In the researchers' demonstration of spin-orbit coupling, two lasers allow an atom's motion to flip it between a pair of energy states. The new work, published inNature, demonstrates this effect for the first time in bosons, which make up one of the two major classes of particles. The same technique could be applied to fermions, the other major class of particles, according to the researchers. The special properties of fermions would make them ideal for studying new kinds of interactions between two particles -- for example those leading to novel"p-wave" superconductivity, which may enable a long-sought form of quantum computing known as topological quantum computation.

In an unexpected development, the team also discovered that the lasers modified how the atoms interacted with each other and caused atoms in one energy state to separate in space from atoms in the other energy state.

One of the most important phenomena in quantum physics, spin-orbit coupling describes the interplay that can occur between a particle's internal properties and its external properties. In atoms, it usually describes interactions that only occur within an atom: how an electron's orbit around an atom's core (nucleus) affects the orientation of the electron's internal bar-magnet-like"spin." In semiconductor materials such as gallium arsenide, spin-orbit coupling is an interaction between an electron's spin and its linear motion in a material.

"Spin-orbit coupling is often a bad thing," said JQI's Ian Spielman, senior author of the paper."Researchers make 'spintronic' devices out of gallium arsenide, and if you've prepared a spin in some desired orientation, the last thing you'd want it to do is to flip to some other spin when it's moving."

"But from the point of view of fundamental physics, spin-orbit coupling is really interesting," he said."It's what drives these new kinds of materials called 'topological insulators.'"

One of the hottest topics in physics right now, topological insulators are special materials in which location is everything: the ability of electrons to flow depends on where they are located within the material. Most regions of such a material are insulating, and electric current does not flow freely. But in a flat, two-dimensional topological insulator, current can flow freely along the edge in one direction for one type of spin, and the opposite direction for the opposite kind of spin. In 3-D topological insulators, electrons would flow freely on the surface but be inhibited inside the material. While researchers have been making higher and higher quality versions of this special class of material in solids, spin-orbit coupling in trapped ultracold gases of atoms could help realize topological insulators in their purest, most pristine form, as gases are free of impurity atoms and the other complexities of solid materials.

Usually, atoms do not exhibit the same kind of spin-orbit coupling as electrons exhibit in gallium-arsenide crystals. While each individual atom has its own spin-orbit coupling going on between its internal components (electrons and nucleus), the atom's overall motion generally is not affected by its internal energy state.

But the researchers were able to change that. In their experiment, researchers trapped and cooled a gas of about 200,000 rubidium-87 atoms down to 100 nanokelvins, 3 billion times colder than room temperature. The researchers selected a pair of energy states, analogous to the"spin-up" and"spin-down" states in an electron, from the available atomic energy levels. An atom could occupy either of these"pseudospin" states. Then researchers shined a pair of lasers on the atoms so as to change the relationship between the atom's energy and its momentum (its mass times velocity), and therefore its motion. This created spin-orbit coupling in the atom: the moving atom flipped between its two"spin" states at a rate that depended upon its velocity.

"This demonstrates that the idea of using laser light to create spin-orbit coupling in atoms works. This is all we expected to see," Spielman said."But something else really neat happened."

They turned up the intensity of their lasers, and atoms of one spin state began to repel the atoms in the other spin state, causing them to separate.

"We changed fundamentally how these atoms interacted with one another," Spielman said."We hadn't anticipated that and got lucky."

The rubidium atoms in the researchers' experiment were bosons, sociable particles that can all crowd into the same space even if they possess identical values in their properties including spin. But Spielman's calculations show that they could also create this same effect in ultracold gases of fermions. Fermions, the more antisocial type of atoms, cannot occupy the same space when they are in an identical state. And compared to other methods for creating new interactions between fermions, the spin states would be easier to control and longer lived.

A spin-orbit-coupled Fermi gas could interact with itself because the lasers effectively split each atom into two distinct components, each with its own spin state, and two such atoms with different velocities could then interact and pair up with one other. This kind of pairing opens up possibilities, Spielman said, for studying novel forms of superconductivity, particularly"p-wave" superconductivity, in which two paired atoms have a quantum-mechanical phase that depends on their relative orientation. Such p-wave superconductors may enable a form of quantum computing known as topological quantum computation.


Source