Researchers create super-small transistor, artificial atom powered by single electrons

April 18, 2011, University of Pittsburgh
An atomic-scale depiction of the SketchSET shows three wires (green bars) converging on the central island (center green area), which can house up to two electrons. Electrons tunnel from one wire to another through the island. Conditions on the third wire can result in distinct conductive properties. Credit: U. Pittsburgh

A University of Pittsburgh-led team has created a single-electron transistor that provides a building block for new, more powerful computer memories, advanced electronic materials, and the basic components of quantum computers.

The researchers report in that the transistor's central component—an island only 1.5 nanometers in diameter—operates with the addition of only one or two electrons. That capability would make the transistor important to a range of computational applications, from ultradense memories to quantum processors, powerful devices that promise to solve problems so complex that all of the world's computers working together for billions of years could not crack them.

In addition, the tiny central island could be used as an artificial atom for developing new classes of artificial electronic materials, such as exotic superconductors with properties not found in natural materials, explained lead researcher Jeremy Levy, a professor of physics and astronomy in Pitt's School of Arts and Sciences. Levy worked with lead author and Pitt physics and astronomy graduate student Guanglei Cheng, as well as with Pitt physics and astronomy researchers Feng Bi, Daniela Bogorin, and Cheng Cen. The Pitt researchers worked with a team from the University of Wisconsin at Madison led by materials science and engineering professor Chang-Beom Eom, including research associates Chung Wun Bark, Jae-Wan Park, and Chad Folkman. Also part of the team were Gilberto Medeiros-Ribeiro, of HP Labs, and Pablo F. Siles, a doctoral student at the State University of Campinas in Brazil.

Levy and his colleagues named their device SketchSET, or sketch-based single-electron transistor, after a technique developed in Levy's lab in 2008 that works like a microscopic Etch A SketchTM, the drawing toy that inspired the idea. Using the sharp conducting probe of an atomic force microscope, Levy can create such electronic devices as wires and of nanometer dimensions at the interface of a crystal of strontium titanate and a 1.2 nanometer thick layer of lanthanum aluminate. The electronic devices can then be erased and the interface used anew.

The SketchSET—which is the first single-electron transistor made entirely of oxide-based materials—consists of an island formation that can house up to two electrons. The number of electrons on the island—which can be only zero, one, or two—results in distinct conductive properties. Wires extending from the transistor carry additional electrons across the island.

One virtue of a single-electron transistor is its extreme sensitivity to an electric charge, Levy explained. Another property of these oxide materials is ferroelectricity, which allows the transistor to act as a solid-state memory. The ferroelectric state can, in the absence of external power, control the number of electrons on the island, which in turn can be used to represent the 1 or 0 state of a memory element. A computer memory based on this property would be able to retain information even when the processor itself is powered down, Levy said. The ferroelectric state also is expected to be sensitive to small pressure changes at nanometer scales, making this device potentially useful as a nanoscale charge and force sensor.

Since August 2010, Levy has led a $7.5 million, multi-institutional project to construct a semiconductor with properties similar to SketchSET, he said. Funded by the U.S. Air Force Office of Scientific Research's Multi-University Research Initiative (MURI) program, the five-year effort is intended to overcome some of the most significant challenges related to the development of quantum information technology. Levy works on that project with researchers from Cornell, Stanford, the University of California at Santa Barbara, the University of Michigan, and UW-Madison.

Explore further: Team develops nanoscale light sensor compatible with 'Etch-a-Sketch' nanoelectronic platform

Related Stories

New STM Microscope To Study Propeties of Electron Spin

June 23, 2004

University of Arkansas scientists seek to harness an electron's spin to create tiny machines with large memories. To do this, they have built a microscope that may allow them to be the first researchers to measure the properties ...

Recommended for you

Optical nanoscope images quantum dots

January 23, 2018

Physicists have developed a technique based on optical microscopy that can be used to create images of atoms on the nanoscale. In particular, the new method allows the imaging of quantum dots in a semiconductor chip. Together ...

Quantum dot ring lasers emit colored light

January 22, 2018

Researchers have designed a new type of laser called a quantum dot ring laser that emits red, orange, and green light. The different colors are emitted from different parts of the quantum dot—red from the core, green from ...


Adjust slider to filter visible comments by rank

Display comments: newest first

5 / 5 (1) Apr 18, 2011
this stuff is neat
5 / 5 (1) Apr 18, 2011
I can't wait to see this used in concrete product. However the news of new technologies like this outrun the real implementation 1000 to 1.. even more.
4 / 5 (1) Apr 18, 2011
I always wondered: since these technologies are getting smaller and smaller, aren't they becoming more and more vulnerable too?

Are processes depending on single electrons even stable?
not rated yet Apr 18, 2011
I always wondered: since these technologies are getting smaller and smaller, aren't they becoming more and more vulnerable too?

Are processes depending on single electrons even stable?

I'm no expert in this particular subject, but I think that you do have some instabilities occurring. It seems that researchers are discovering that computations can be made with some errors with little or no perceivable effect. You gotta realize that this one electron isn't the single condition for "nuclear vents open" or "nuclear vent closed", but rather a small operation. When doing millions of operations per millisecond, you can have some errors. You just make sure that you expect these errors and plan accordingly. Especially when you're dealing with quantum tunneling. With quantum mechanics, it's all probabilities and no definite "yes" or "no's". Will the electron tunnel given x,y and z? Maybe.
not rated yet Apr 18, 2011
Multi-terabyte pen drives, here we come !!
5 / 5 (1) Apr 19, 2011
Those spam messages are starting to become annoying!
not rated yet Apr 19, 2011
1 (and now apparently 2) electron SETs are actually rather stable (electrically) if they can be made small enough. One condition that needs to be met to get an electron into a SET is that the applied Voltage (V) is greater than abs(q/(2C)). q electric charge, C capacitance. For nano scale capacitors, C is small so your voltage needs to be large. Once you get an electron on, it will be really hard to get another on (or the current one off).
Regarding thermal energy, the relation is q^2/(2*C)>>1/2*kb*T. So again, if C is reeeeeeally small, thermal energy won't get an electron to jump the gap between wire and island.

The main problem is reading the values off of these devices which is another problem entirely.
not rated yet Apr 19, 2011
So what happens when electron transistors become entangled?
not rated yet Apr 19, 2011
So what happens when electron transistors become entangled?

Babies of course.
not rated yet Apr 19, 2011

That's not exactly my field, but I would expect that would be a desired effect, as it leads to many possible states in the system simultaneously, allowing for quantum computation.

~10 q bits is supposedly enough to have quantum computing compete with classical computing in some applications.
not rated yet Apr 20, 2011
ultra dense memory - ultra dense CPU's --

memory that is on the order of 100000x denser... basically a terabyte of memory on an iPhone is the equivalent. -- but what are the read write and seek speeds??

but imagine a processor with a cache of 1TB -- what is the point of RAM at that point??

video memory GPU's with 100's of MBs of memory per core

lovely -- oh the supercomtputing possiblities -- of course the average person has no need for such things --- but who else will subsidize the cost of these advancements but by making people buy improvements in technology they will ever need.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.