PHILADELPHIA -- Major advances in materials science and nanotechnology promise to revolutionize electronic devices with unprecedented strength and conductivity, but those promises can’t be fulfilled if the devices can’t be consistently manufactured. Working on the nanoscale means the effects of even the smallest imperfections are magnified, but University of Pennsylvania physicists have developed a new printing technique that is effective at making carbon nanotube devices with minimal impurities.          

The research was conducted by A. T. Charlie Johnson, graduate student Samuel M. Khamis and undergraduate researcher Ryan A. Jones of the Department of Physics and Astronomy in the School of Arts and Sciences.

Their work was published in the new open access journal AIP Advances, where Johnson is an editor.

The main manufacturing process used in making electrical circuitry with current materials, such as silicon, is photolithography. The process involves making a pattern of the intended circuit, then transferring the pattern to the silicon wafer using light. The material on which the pattern is drawn is called a “photoresist,” and it needs to be removed before the circuit can be used.

Carbon-nanotube-based circuits can be made in the same way, but, when the photoresist is removed, molecules in it tend to stick to the nanotube exteriors, which degrades the properties of the circuit. The physicists needed a way to block this binding.

“We developed a method where we have the usual photoresist,” Johnson said, “but below it we have another layer that does not interact very strongly with the nanotubes. Not only does it stick less, we have a way of removing that layer after the fact; we heat it until it vaporizes. By doing that properly, we can improve the characteristics — the electrical resistance of these transistors when they’re turned on — of these devices by a factor of 100.”

The problem was one of nano-geometry.

Both carbon nanotubes and graphene are composed of carbon atoms arranged in a lattice of flat, “pi-bonded,” hexagonal rings. The sought-after electrical properties of both nanotubes and graphene come from those hexagons being arranged in single plane, but those hexagons also like to nestle into stacks. When the rings stack, they form graphite. That is fine for making pencils but not for making high-performance nanotube circuitry.

Traditional photolithography techniques produced subpar nanotube devices because commonly used photoresists also contain molecules consisting of hexagonal rings. Though not made exclusively of carbon, the photoresist rings would also stack and bind to the exterior of the nanotubes.

Finding a buffer with a different molecular pattern was the first step toward a solution. Selecting one that was compatible with standard processing represents a significant cost savings for the mass production of carbon nanotubes and potentially graphene.   

“The layer we use to protect the nanotubes is another material that’s already used in the semiconductor industry,” Johnson said. “It is a polymer that contains rings of atoms, but the rings don’t interact with the nanotubes pi-bonds. That means they don’t stick to the nanotubes as well, and that makes all the difference.”

Published in an open-access journal, the study is available to the public on the AIP Advances Web site.

Beyond the importance of increasing the amount of people who can access the research, Johnson said the ethos of this kind of publishing is to increase the importance the public in determining what research is most notable.

“There is a collection of social-media tools available on the journal’s Web site, which allows the readership to decide the importance of the various papers,” Johnson said. “The idea is to build multiple communities of people that include both researchers and the public, and then to have them tell each other which papers have the most impact, rather than have a small group of reviewers or editors make that determination.”