The conductivity of a single graphene nanoribbon can be measured by lifting the nanoribbon off a surface with the tip of a scanning tunnelling microscope.
Graphene — a single sheet of carbon atoms arranged in a honeycomb lattice — has generated a frenzy of research activity in recent years1,2,3 and in turn considerable speculation on its potential applications1,2,3. For applications in nanoelectronics, a form of graphene known as graphene nanoribbons4,5, which are long, narrow strips of the material, are of particular interest. However, if devices based on graphene nanoribbons are to be built, it will be important to know how they conduct, and such measurements have so far proved challenging. Writing in Nature Nanotechnology, Leonhard Grill and colleagues at the Fritz-Haber-Institute of the Max-Planck-Society, the Institute of Materials Research and Engineering in Singapore and CEMES-CNRS in Toulouse have now shown the conductivity of a single graphene nanoribbon can be directly measured using a scanning tunnelling microscope (STM)6.
One of the reasons why graphene has attracted such interest is the simplicity with which it can be made: the material can be cleaved from bulk graphite using only duct tape1. Producing graphene nanoribbons is, however, more demanding and requires innovative strategies such as cutting open carbon nanotubes or the use of lithographic patterning4,5. To produce nanoribbons with atomically well-defined edges is more challenging still, but a technique to assemble atomically precise nanoribbons from basic molecular building blocks on a surface has recently been developed7. Grill and colleagues used a similar method and starting molecules to assemble their graphene nanoribbons on a gold surface (Fig. 1). They first deposit the precursor molecules on the surface that is held at 200 °C. The molecules undergo a polymerization reaction that links them together to form long molecular chains. Subsequent heating to 400 °C leads to the formation of graphene nanoribbons with atomically well-defined edges.
Measuring the conductivity of a macroscopic wire is simple: attach two electrodes at opposite ends of the wire and then apply a voltage across it. Because of the challenges associated with manipulating and probing wires with miniscule dimensions, conductivity measurements at the nanoscale are inevitably not as straightforward. To perform these measurements, Grill and colleagues used the tip of the STM to lift one end of the nanoribbon off the surface while the other end remains attached (Fig. 1). In this geometry, the STM tip and the gold surface act as the two electrodes. Then by applying different voltages across the nanoribbon, the researchers can directly measure its conductivity.
Graphene nanoribbons can have two types of edges, which are termed 'armchair' and 'zig-zag' because of their shapes3,4,5,7,8,9. The zig-zag edge has a special electronic structure known as the Tamm state, which originates from non-bonding molecular orbitals near the Fermi level8. The existence of the Tamm state was theoretically predicted over a decade ago8 but was only recently experimentally observed9. The graphene nanoribbons assembled by Grill and colleagues have armchair edges along the long axis of the ribbon and zig-zag edges along the short axis. Nevertheless, the Tamm states of these narrow ends are clearly evident in the STM images, where they appear as protrusions (lobes). The electronic structures of the nanoribbons derived from the STM images are also confirmed by theoretical simulations.
From the measurements, the researchers show that the conductivity of a nanoribbon depends critically on its electronic states such as the highest occupied molecular orbitals (HOMO) and lowest unoccupied molecular orbitals (LUMO). It also depends on the distance between the electrodes. In the experiments, the conductivity of the nanoribbon is measured by pulling it off different distances from the surface. At small tip-to-surface distances, the conductivity is dominated by electron tunnelling but at higher tip heights — that is, at larger distances between electrodes — the conductivity decreases exponentially, as expected.
A key finding in these measurements is that the conductivity of the nanoribbon rapidly increases when the voltage applied across the electrodes matches the energies of the molecular orbitals, that is, the HOMO and LUMO of the nanoribbon. Although this behaviour was also expected, the measurements on individual nanoribbons unambiguously prove that the conductivity is directly related to the electronic states. Because the orbital energies or electronic states of the nanoribbon can be tuned by varying the shape and size of it, this finding highlights an important way to tailor its conductivity. Furthermore, the geometry of the nanoribbon is also shown to play a role. In particular, a bent nanoribbon exhibits an exponential decay in conductivity, whereas pseudo-ballistic transport would be expected for a flat graphene nanoribbon. This geometric effect could be used, for example, to vary the conductivity of the nanoribbon between two states like a switch. Moreover, it might be possible to tune the current passing through the nanoribbon by varying the degree of bending. It could then be used as a gated field-effect transistor.
The atomically precise assembly of graphene on a surface offers a potential route to graphene nanoribbons with tailored shapes and sizes7. The technique developed by Grill and colleagues now provides a means to measure the conductivity of these assembled nanoribbons with atomic precision6. The fabrication of nanoelectronic devices will of course require more than this. Reliable and reproducible nanoribbon–electrode junctions will, for example, have to be developed. Moreover, a reliable strategy to assemble these nanoribbons into device geometries is still required. Nevertheless, conductivity measurements of single graphene nanoribbons are an important step on the way to functional nanoelectronic devices.