At first glance, a free-standing 2D metal seems impossible. This is because the bonding between atoms in a metal is mediated by conduction electrons, which are free to move in any direction. As a result, metals tend to have 3D crystal structures and no tendency to form planar sheets. This is unlike crystalline carbon, which is held together by highly directional covalent bonds that allow free-standing atom-thick sheets of graphene to exist. While single epitaxial layers of metal atoms can be created on a substrate, these are not true 2D materials because the atoms are bonded to the underlying structure.

Plugging a gap

In the new research, Mark Rümmeli and colleagues at the Leibniz Institute for Solid State and Materials Research in Dresden, Germany, and at institutes in Poland and Korea studied the behaviour of metal atoms at the edges of holes in graphene. They grew a sheet of graphene by chemical vapour deposition on a surface and detached it by etching the substrate with an iron-chloride solution. This left trace amounts of iron on the surface of the graphene. Irradiating the graphene with an electron beam created small holes and also encouraged the iron atoms to move around. The edge atoms of graphene are the most reactive because they contain dangling bonds; so when the mobile iron atoms encounter the edge of a hole, they bond to it. This continues with iron atoms bonding to the other iron atoms around the edge, until the hole is completely sealed with a 2D square lattice of iron.

The group's theoretical calculations show that the largest thermodynamically stable sheet would be about 12 atoms across – or just 3 nm – wide. The largest sheets observed in the experiment were only 10 atoms wide. Beyond this, the tendency of iron to form a 3D structure wins out over the bonding between the iron and carbon atoms at the edges. "The atoms usually form a tiny crystal that sticks to one of the edges," explains Rümmeli, who is now at the Institute for Basic Science in Korea.

History has shown that when someone comes up with an unexpected new material, someone else usually comes up with an unexpected use for it
Mark Rümmeli, Institute for Basic Science, Korea

Other calculations suggest that changes in the electronic band structure of the iron when it forms a 2D lattice should give it a substantially larger magnetic moment than bulk iron. This, the researchers speculate, could make it useful for magnetic memories. Rümmeli stresses, however, that more basic science must be done before the membranes could be considered for practical applications. "History has shown that when someone comes up with an unexpected new material," he says, "someone else usually comes up with an unexpected use for it." The team plans to try to make other 2D metals by the same method and investigate their properties.

Stability problems

Pietro Gambardella, a materials scientist at ETH Zurich, says the work is very interesting from a fundamental perspective, but he remains sceptical of potential engineering applications. "If you have something that breaks down when it becomes larger than 12 atoms across, it's clearly quite unstable," he says, "so it's difficult to see how you could use it in a device without it breaking down."

Arkady Krasheninnikov, an electronic-structure theorist at Aalto University and the University of Helsinki, both in Finland, is more optimistic. "At present, it's clearly too unstable to be useful outside the laboratory," he says, "but it's quite astounding that such a 2D structure can even exist. Now people can start looking for ways to make it more stable." He suggests that it might potentially be stabilized by sandwiching it between two layers of graphene. "Hopefully, it will keep its peculiar magnetic properties," he says.

The research is published in Science.

Further reading

Graphene goes magnetic (Nov 2011)
Analysing graphene sheets (Mar 2011)
Silicon atoms seen ‘dancing’ in graphene (apr 2013)