Friday, March 14, 2008

Atomic-layer-deposited nanostructures for graphene-based nanoelectronics

Atomic-layer-deposited nanostructures for graphene-based nanoelectronics

Y. Xuan, Y. Q. Wu, T. Shen, M. Qi, M. A. Capano, J. A. Cooper, and P. D. Ye

Appl. Phys. Lett. 92, 013101 (2008)


As people have realized that the race to well-made graphene devices is pretty much a gold rush, the literature has seen people try to do what they do best in their own discipline- but do it on graphene (more on this interdisciplinary trend later). In science, the best discoveries often come from an alert scientist realizing they got something different than they expected, and being curious enough not to throw it in the trash. In this paper, the authors try to grow uniform layers of aluminum oxide and hafnium oxide on fancy graphite and stumble upon a way to make some metal nanoribbons in addition to patterning graphite/graphene.

The authors use atomic layer deposition (ALD) to try to make some films of metal oxides on the surface of highly ordered graphite. ALD is a variant of chemical vapor deposition (CVD). In CVD, reactive gasses (such as silane, SiH4, and O2) are exposed to a substrate (such as silica), and the reaction of these two creates a thin layer on the surface of the substrate (for example, SiO2 on silica; the overall reaction is SiH4 + O2 = SiO2 + H2). This is good for thin films; what if you want really thin films? In ALD you expose the substrate to these two gasses one at a time, allowing only one layer of product per cycle to be formed. As you expose the substrate with only one gas, it can only absorb on the surface, since it doesn't have anything to react with; when you put in the second gas, the only thing the second gas has to react with is the molecules absorbed on the surface. At the end of the day, ALD gives you very thin films that are partially controlled by how well your reactants can absorb on the surface.

When ALD was attempted to deposit thin layers of metal oxides on the (fancy ultrasmooth) graphite, the authors found that their reactive gasses did not absorb onto most of the graphite, which was probably quite a bummer. However, they found that they got very thin (1.5 nm) ribbons of their metal oxides on certain sections of the graphite. Further investigations showed that the ribbons only grew on the edges (steps?) of individual graphene sheets. These edges are typically more reactive than the bulk material, since you get C-H or other bonds on the edge instead of the C-C aromatic bonds, which everyone who's taken sophomore organic knows you can't really mess with. They point out that the edges of graphene have absorbed other small molecules in the past, and that the two kinds of edges possible (armchair and zigzag; perhaps I'll write that up some day) might give preferential growth for their nanoribbons, an assertion they say they're looking into for future papers. The lateral size of the ribbons is determined by the number of ALD cycles and temperature, while the vertical size of the ribbons is determined only by the number of ALD cycles.

The authors say that the metal oxide layers could serve as etching/decomposition masks, giving a nice handle on a way to get very thin graphite or graphene ribbons. Unfortunately, you couldn't really pattern with these masks, since the orientation is entirely dependent on where your graphene edges are. Nice paper overall, and a great example of making some delicious lemonade when life gives you lemons (or perhaps realizing that life really gave you lemonade after all).

Like the last post, these guys are electrical engineers that have become interested in jumping on the graphene bandwagon. The field of "graphene science" is quite interdisciplinary; in the few posts I've done so far (which didn't include any theoretical papers), we've had contributions from electrical engineers, physicists, chemical engineers, mechanical engineers, materials scientists, chemists, and three guys who work at the "Grenoble High Magnetic Field Laboratory", which I guess makes them magneticists (or perhaps just magicians). The research world is getting more and more interconnected, and graphene is a perfect example of how a problem can attract a wide array of scientists and engineers, who all try to tackle a problem using their unique skills. For example, the last post had a bunch of nanoimprint guys say, "hey, we know how to do nanoimprint stuff and we've got the equipment; why not use it to contribute to the new gold rush?" The authors of this paper know how to deposit aluminum oxide and hafnium oxide with ALD, and their attempt to use their knowledge in graphene engineering gives us a neat handle on making more nanostructures.

Does this mean that we should all be getting interdisciplinary degrees, as in "PhD in cancer research"? I would be inclined to say no. Conducting research in an interdisciplinary manner is essential to an area such as graphenes, but most of the good papers I've read in the area are done by people who are (or were) experts in something other than graphene research. Learning to be a great chemist in grad school allows you to do great chemistry in a lot of areas; a degree that teaches you small snippets of everything probably won't make you an expert in any of the disciplines, while leaving your knowledge base less flexible. It warms my heart to see the graphene problem being approached competently from so many angles.

Xuan, Y., Wu, Y.Q., Shen, T., Qi, M., Capano, M.A., Cooper, J.A., Ye, P.D. (2008). Atomic-layer-deposited nanostructures for graphene-based nanoelectronics. Applied Physics Letters, 92(1), 013101. DOI: 10.1063/1.2828338