Behind the buzz and beyond the hype:
Our Nanowerk-exclusive feature articles
Posted: Jun 13, 2011
Nano-PacMan caught on video - observing nanoparticle channelling in graphene in real-time
(Nanowerk Spotlight) Graphene has two distinct types of edges produced when it is cut – armchair type or zigzag type – which correspond to the two crystal axis of graphene. These edge types have distinct electronic, magnetic, and chemical properties, and being able to pattern graphene along particular crystallographic directions to leave edges consisting of a single chirality is crucial for the fabrication of graphene nanoribbon and nanoelectronics devices.
"All previous studies of channelling behavior have been limited by the need to perform the experiment ex situ, i.e. comparing single 'before' and 'after' images," Peter Bøggild, an associate professor at DTU Nanotech, explains to Nanowerk. "In these and other ex situ experiments the dynamic behavior must be inferred from the length of channels and heating time after completion of the experiment, with the rate of formation of the channel assumed to be consistent over the course of the experiment."
In new work, reported in the June 9, 2011 advance online edition of Nano Letters ("Discrete dynamics of nanoparticle channelling in suspended graphene"), Bøggild and his team report the nanoscale observation of this channelling process by silver nanoparticles in an oxygen atmosphere in-situ on suspended mono- and bilayer graphene in an environmental transmission electron microscope, enabling direct concurrent observation of the process, impossible in ex-situ experiments.
A previously undescribed stepwise oxidation of mono- and few layer suspended graphene by silver nanoparticles is observed in situ at subnanometer scale in an environmental transmission electron microscope. Over the range of 600–850 K, crystallographically oriented channelling with rates in the range 0.01–1 nm/s are observed and an activation energy of 0.557 ± 0.016 eV is calculated. (Reprinted with permission from American Chemical Society)
"For the first time, we've watched the catalytic oxidation by silver nanoparticles of suspended graphene monolayers and few-layers," says Tim Booth, a postdoc at DTU Nanotech and the paper's first author. "The catalytic particles cut relatively long, neat and narrow tracks in the suspended graphene, and the tracks are exactly aligned to important symmetry directions in graphene. In contrast to previous work, we have watched this process as it happens, with near atomic resolution, and on suspended monolayer graphene, and have therefore discovered some really interesting things about the dynamics of the process."
Although these tracks and the channel-etching behavior have been observed many times before, the DTU team are the first to see this process happen in situ, in real time.
"Previous studies relied on the equivalent of taking before and after pictures of a live process" says Booth. "Although this can tell you a huge amount about what has happened, like inferring dinosaur behavior from the fossil record, you'd much rather watch it happen instead."
As a result of watching this process occur live in a transmission electron microscope, the researchers say they have seen many details that were hidden before, and video really brings the "nano pacman" behavior to life:
Booth notes that the long, smooth straight tracks are actually made in quite a jerky and discontinuous way by the particles.
"This is because we're dealing with the discrete removal of a relatively small number of carbon atoms from the graphene" he says. "In any one instant, the silver particle might remove a few atoms, only one or two, or possibly none at all, dependent on the temperature and the Poisson probability distribution – which is the same distribution which governs, for example, the number of heads you would flip on a coin in one minute."
Carefully measuring the rate at which atoms in the graphene are removed – and also knowing that the individual measurements are Poisson distributed – gives the researchers a very good idea of the activation energy for the process, and this in turn allows them to compare the results of DFT simulations and models of the process. Booth says that they have found really good agreement between the model and experiments.
"With a deeper understanding of the fine details we hope to one day use this nanoscale channelling behavior to directly cut desired patterns out of suspended graphene sheets, with a resolution and accuracy that isn't achievable with any other technique," says Bøggild. "A critical advantage here is that the graphene crystal structure guides the patterning, and in our case all of the cut edges of the graphene are 'zigzag' edges."
Because the channel widths and edges are so well defined, this patterning technique could be the ideal way to make graphene-based electrodes for molecular electronics. However, before this can properly be considered a patterning or lithographic technique, there are some crucial challenges to overcome. Some are technical, such as how to position the particles in the correct starting place, and could be relatively easy to solve. More fundamental challenges include how to get the channelling nanoparticles to turn when and where it is required; and how to balance the inherently grainy, stochastic nature of the process against the precision of the structures defined.
"To overcome challenges like this we need to really understand the physics of the channelling behavior" says Booth.
Bøggild and Booth point out that transmission electron microscopy and related techniques have been really important in elucidating the structure and properties of graphene, and in situ experiments are becoming more common.
"We think the more experimentation we can do while directly observing the effects of our actions at the atomic scale, the better able we will be to make rapid progress in nanotechnology."