Research reported this week shows that electrical resistance in nanoribbons of epitaxial graphene changes in discrete steps following quantum mechanical principles. The research shows that the graphene nanoribbons act more like optical waveguides or quantum dots, allowing electrons to flow smoothly along the edges of the material. In ordinary conductors such as copper, resistance increases in proportion to the length as electrons encounter more and more impurities while moving through the conductor.

The ballistic transport properties, similar to those observed in cylindrical carbon nanotubes, exceed theoretical conductance predictions for graphene by a factor of 10. The properties were measured in graphene nanoribbons approximately 40 nanometers wide that had been grown on the edges of three-dimensional structures etched into silicon carbide wafers.

“This work shows that we can control graphene electrons in very different ways because the properties are really exceptional,” said Walt de Heer, a Regent’s professor in the School of Physics at the Georgia Institute of Technology. “This could result in a new class of coherent electronic devices based on room temperature ballistic transport in graphene. Such devices would be very different from what we make today in silicon.”

The research, which was supported by the National Science Foundation, the Air Force Office of Scientific Research and the W.M. Keck Foundation, was reported February 5 in the journal Nature. The research was done through a collaboration of scientists from Georgia Tech in the United States, Leibniz Universität Hannover in Germany, the Centre National de la Recherche Scientifique (CNRS) in France and Oak Ridge National Laboratory – supported by the Department of Energy – in the United States.

For nearly a decade, researchers have been trying to use the unique properties of graphene to create electronic devices that operate much like existing silicon semiconductor chips. But those efforts have met with limited success because graphene – a lattice of carbon atoms that can be made as little as one layer thick – cannot be easily given the electronic bandgap that such devices need to operate.

De Heer argues that researchers should stop trying to use graphene like silicon, and instead use its unique electron transport properties to design new types of electronic devices that could allow ultra-fast computing – based on a new approach to switching. Electrons in the graphene nanoribbons can move tens or hundreds of microns without scattering.

“This constant resistance is related to one of the fundamental constants of physics, the conductance quantum,” de Heer said. “The resistance of this channel does not depend on temperature, and it does not depend on the amount of current you are putting through it.”

What does disrupt the flow of electrons, however, is measuring the resistance with an electrical probe. The measurements showed that touching the nanoribbons with a single probe doubles the resistance; touching it with two probes triples the resistance.

“The electrons hit the probe and scatter,” explained de Heer. “It’s a lot like a stream in which water is flowing nicely until you put rocks in the way. We have done systematic studies to show that when you touch the nanoribbons with a probe, you introduce a method for the electrons to scatter, and that changes the resistance.”

The nanoribbons are grown epitaxially on silicon carbon wafers into which patterns have been etched using standard microelectronics fabrication techniques. When the wafers are heated to approximately 1,000 degrees Celsius, silicon is preferentially driven off along the edges, forming graphene nanoribbons whose structure is determined by the pattern of the three-dimensional surface. Once grown, the nanoribbons require no further processing.

The advantage of fabricating graphene nanoribbons this way is that it produces edges that are perfectly smooth, annealed by the fabrication process. The smooth edges allow electrons to flow through the nanoribbons without disruption. If traditional etching techniques are used to cut nanoribbons from graphene sheets, the resulting edges are too rough to allow ballistic transport.

“It seems that the current is primarily flowing on the edges,” de Heer said. “There are other electrons in the bulk portion of the nanoribbons, but they do not interact with the electrons flowing at the edges.”

The electrons on the edge flow more like photons in optical fiber, helping them avoid scattering. “These electrons are really behaving more like light,” he said. “It is like light going through an optical fiber. Because of the way the fiber is made, the light transmits without scattering.”

The researchers measured ballistic conductance in the graphene nanoribbons for up to 16 microns. Electron mobility measurements surpassing one million correspond to a sheet resistance of one ohm per square that is two orders of magnitude lower than what is observed in two-dimensional graphene – and ten times smaller than the best theoretical predictions for graphene.

“This should enable a new way of doing electronics,” de Heer said. “We are already able to steer these electrons and we can switch them using rudimentary means. We can put a roadblock, and then open it up again. New kinds of switches for this material are now on the horizon.”

Theoretical explanations for what the researchers have measured are incomplete. De Heer speculates that the graphene nanoribbons may be producing a new type of electronic transport similar to what is observed in superconductors.  

“There is a lot of fundamental physics that needs to be done to understand what we are seeing,” he added. “We believe this shows that there is a real possibility for a new type of graphene-based electronics.”

Georgia Tech researchers have pioneered graphene-based electronics since 2001, for which they hold a patent, filed in 2003. The technique involves etching patterns into electronics-grade silicon carbide wafers, then heating the wafers to drive off silicon, leaving patterns of graphene.

In addition to de Heer, the paper’s authors included Jens Baringhaus, Frederik Edler and Christoph Tegenkamp from the Institut für Festkörperphysik, Leibniz Universität, Hannover in Germany; Edward Conrad, Ming Ruan and Zhigang Jiang from the School of Physics at Georgia Tech; Claire Berger from Georgia Tech and Institut Néel at the Centre National de la Recherche Scientifique (CNRS) in France; Antonio Tejeda and Muriel Sicot from the Institut Jean Lamour, Universite de Nancy, Centre National de la Recherche Scientifique (CNRS) in France; An-Ping Li from the Center for Nanophase Materials Sciences at Oak Ridge National Laboratory, and Amina Taleb-Ibrahimi from the CNRS Synchotron SOLEIL in France.

This research was supported by the National Science Foundation (NSF) Materials Research Science and Engineering Center (MRSEC) at Georgia Tech through award DMR-0820382; the Air Force Office of Scientific Research (AFOSR); the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy, and the Partner University Fund from the Embassy of France. Any conclusions or recommendations are those of the authors and do not necessarily represent the official views of the NSF, DOE or AFOSR.

CITATION: Jens Baringhaus, et al., “Exceptional ballistic transport in epitaxial graphene nanoribbons,” (Nature 2013). (http://dx.doi.org/10.1038/nature12952).

Research News
Georgia Institute of Technology
177 North Avenue
Atlanta, Georgia  30332-0181  USA

Media Relations Contacts: John Toon (404-894-6986) (jtoon@gatech.edu) or Brett Israel (404-385-1933) (brett.israel@comm.gatech.edu).

Writer: John Toon

Graphene electronics has motivated much of graphene science for the past decade. A primary goal was to develop high mobility semiconducting graphene with a band gap that is large enough for high performance applications. Graphene ribbons were thought to be semiconductors with these properties, however efforts to produce ribbons with useful bandgaps and high mobility has had limited success. We show here that high quality epitaxial graphene nanoribbons 40 nm in width, with annealed edges, grown on sidewall SiC are not semiconductors, but single channel room temperature ballistic conductors for lengths up to at least 16 micrometers. Mobilities exceeding one million corresponding to a sheet resistance below 1 Ohm have been observed, thereby surpassing two dimensional graphene by 3 orders of magnitude and theoretical predictions for perfect graphene by more than a factor of 10. The graphene ribbons behave as electronic waveguides or quantum dots. We show that transport in these ribbons is dominated by two components of the ground state transverse waveguide mode, one that is ballistic and temperature independent, and a second thermally activated component that appears to be ballistic at room temperature and insulating at cryogenic temperatures. At room temperature the resistance of both components abruptly increases with increasing length, one at a length of 160 nm and the other at 16 micrometers. These properties appear to be related to the lowest energy quantum states in the charge neutral ribbons. Since epitaxial graphene nanoribbons are readily produced by the thousands, their room temperature ballistic transport properties can be used in advanced nanoelectronics as well.

The development of graphene electronics requires the integration of graphene devices with Si-CMOS technology. Most strategies involve the transfer of graphene sheets onto silicon, with the inherent difficulties of clean transfer and subsequent graphene nano-patterning that degrades considerably the electronic mobility of nanopatterned graphene. Epitaxial graphene (EG) by contrast is grown on an essentially perfect crystalline (semi-insulating) surface, and graphene nanostructures with exceptional properties have been realized by a selective growth process on tailored SiC surface that requires no graphene patterning. However, the temperatures required in this structured growth process are too high for silicon technology. Here we demonstrate a new graphene to Si integration strategy, with a bonded and interconnected compact double-wafer structure. Using silicon-on-insulator technology (SOI) a thin monocrystalline silicon layer ready for CMOS processing is applied on top of epitaxial graphene on SiC. The parallel Si and graphene platforms are interconnected by metal vias. This method inspired by the industrial development of 3d hyper-integration stacking thin-film electronic devices preserves the advantages of epitaxial graphene and enables the full spectrum of CMOS processing.

Epitaxial graphene mesas and ribbons are investigated using terahertz (THz) nearfield microscopy to probe surface plasmon excitation and THz transmission properties on the sub-wavelength scale. The THz near-field images show variation of graphene properties on a scale smaller than the wavelength, and excitation of THz surface waves occurring at graphene edges, similar to that observed at metallic edges. The Fresnel reflection at the substrate SiC/air interface is also found to be altered by the presence of graphene ribbon arrays, leading to either reduced or enhanced transmission of the THz wave depending on the wave polarization and the ribbon width.

Patterning of graphene is key for device fabrication. We report a way to increase or reduce the number of layers in epitaxial graphene grown on the C-face (000-1) of silicon carbide by the deposition of a 120 nm to 150nm-thick silicon nitride mask prior to graphitization. In our process we find that areas covered by a Si-rich SiN mask have three more layers than non-masked areas. Conversely N-rich SiN decreases the thickness by three layers. In both cases the mask decomposes before graphitization is completed. Graphene grown in masked areas show good quality as observed by Raman, AFM and transport data. By tailoring the growth parameters selective graphene growth has been obtained.

Spin information processing is a possible new paradigm for post-CMOS (complementary metal-oxide semiconductor) electronics and efficient spin propagation over long distances is fundamental to this vision. However, despite several decades of intense research, a suitable platform is still wanting. We report here on highly efficient spin transport in two-terminal polarizer/analyser devices based on high-mobility epitaxial graphene grown on silicon carbide. Taking advantage of high-impedance injecting/detecting tunnel junctions, we show spin transport efficiencies up to 75%, spin signals in the mega-ohm range and spin diffusion lengths exceeding 100 {\mu}m. This enables spintronics in complex structures: devices and network architectures relying on spin information processing, well beyond present spintronics applications, can now be foreseen.

A method is proposed to extract pure Raman spectrum of epitaxial graphene on SiC by using a Non-negative Matrix Factorization. It overcomes problems of negative spectral intensity and poorly resolved spectra resulting from a simple subtraction of a SiC background from the experimental data. We also show that the method is similar to deconvolution, for spectra composed of multiple sub- micrometer areas, with the advantage that no prior information on the impulse response functions is needed. We have used this property to characterize the Raman laser beam. The method capability in efficient data smoothing is also demonstrated