Mark Hersam Writes Review Piece for Nature Nanotechnology

June 4, 2008

When single-walled carbon nanotubes were discovered 15 years ago, researchers had high hopes: the tiny hollow cylinders of carbon had great electrical, optical, thermal, and mechanical properties and inspired a host of possible applications, including transistors, sensors, scanning probes, and hydrogen storage elements.

But while researchers have demonstrated these applications in laboratory-scale experiments, the material has struggled to find wide-spread use commercially. No technique for creating the nanotubes creates an identical population, and researchers have yet to find the best way to sort out nanotubes of different sizes and properties.

In order to realize these applications in an industrial setting, engineers either need to find a way to create identical populations of nanotubes or discover the best way to sort them.

That’s what Mark Hersam, professor of materials science and engineering and professor of chemistry in the McCormick School of Engineering at Northwestern University, writes in his review article for Nature Nanotechnology, now available online. In the article, Hersam reviews the current technologies for separating nanotubes and offers suggestions as to how to move forward.

Nanotubes need to be sorted since their properties are dependent on their diameter as well as their chiral angle, which describes the arrangement of carbon atoms along the length of the nanotube.

Currently available strategies for sorting nanotubes can be classified into five categories: selective chemistry, selective destruction, electrophoretic separation, chromatography, and ultracentrifugation.

In selective chemistry, scientists utilize molecules that react specifically with the nanotube they want. Once that molecule has reacted with the nanotube, it gives the nanotube a property difference, like solubility in a solvent, so scientists can separate out the chosen nanotube species. Though this method is potentially inexpensive and easy to implement, the chemical selectively is typically less than 100 percent, which limits nanotube purity.

In selective destruction, nanotubes are placed on a conductive wafer that is coated with an insulating film. When a voltage is applied to the substrate, it selectively reduces the electrical conductivity of the semiconducting nanotubes. When large electrical currents are then passed through the nanotube film, the metallic nanotubes are selectively destroyed. This method is effective at producing films of nanotubes that are predominantly semiconducting, but the inverse process has not yet been established, and this approach destroys the metallic nanotubes instead of just sorting them out.

In electrophoresis, a technique that is also used to sort DNA, an electric field is applied to nanotubes to exploit differences in the nanotubes’ polarizability, or their relative tendency to have their electronic structure distorted by an external electric field. While effective in laboratory-scale experiments, this method has not yet been implemented on a large scale.

In chromatography, scientists load a column with a porous medium, inject a nanotube solution, and then apply a force — often an electric field — which causes the nanotubes to move through the column at different rates. This strategy has enabled separation by diameter, length, and electronic type for DNA-encapsulated nanotubes. Unfortunately, the high cost of DNA creates an economic barrier for this approach.

Hersam’s research involves the last method – ultracentrifugation. In this method, researchers load a centrifuge tube with a fluid that varies in density — high density at the bottom and low density on top. Then they load the nanotubes into the tube and spin it at high rotational frequencies in a centrifuge. Under these conditions, the nanotubes sediment through the density gradient until they reach their isopycnic point – the point where the nanotube density matches the fluid density. The relationship between the nanotube density and its structure and properties can be engineered through the use of appropriate surfactant chemistries, so ultracentrifugation possesses great flexibility in sorting.

Though some would argue that this method would be difficult to use on a large scale, Hersam disagrees.

“I think that you can produce enough sorted single-walled carbon nanotubes with ultracentrifugation to satisfy the current worldwide demand,” he says.

But Hersam ultimately argues in the paper that the best way to solve the problem would be to develop a way to grow identical nanotubes. Until that problem is solved, however, researchers need to agree upon standards that will prove how high a purity each separation method creates.

“There have been many developments in this field,” he says. “Researchers have demonstrated impressively high purities. But the challenge now is to produce sufficiently large quantities to enable the industrial-scale applications that have been promised for the past 15 years. None of these separation methods is a clear winner, and unless there is an unanticipated breakthrough, it is probably going to require a clever combination of techniques to achieve this goal.”