Issue Date: April 4, 2011
Electrochemists have been using carbon electrodes for decades. And they’ve done their best with what nature provided or with glassy carbon. But these materials, which are hard to fabricate into the increasingly tiny devices required in modern electronics and electrochemistry, are nearing the limits of their capabilities.
Scientists are developing methods to push past these limits. They are adopting materials from the semiconductor industry that allow the scientists to harness the characteristics of carbon in complex shapes and tiny devices. Examples of this work were described in a symposium at Pittcon 2011, held last month in Atlanta.
“Carbon is amazing stuff,” said Richard L. McCreery, symposium co-organizer and senior research officer at the National Institute for Nanotechnology in Edmonton, Alberta. “It’s not a metal, even though it conducts. It’s the basis of organic chemistry, so it’s got all kinds of variety. It’s a very versatile material.”
But it’s hard to handle. “Carbon by itself is difficult to machine” into desired shapes, Marc J. Madou, a mechanical engineering and bioengineering professor at the University of California, Irvine, pointed out. In contrast, polymers are easily manipulated into a variety of structures.
Materials scientists and chemists are now turning to materials that give them a way to “machine” carbon while circumventing carbon’s drawbacks and harnessing its strengths. The materials making such advances possible are photoresists—the same ones used to fabricate masks for patterning silicon-based devices in the semiconductor industry.
These flexible polymeric materials, usually phenolic resins, are easy to manipulate and deposit as films on substrates. Heating the films to more than 1,000 °C pyrolyzes them, driving off heteroatoms and additives, leaving nothing but carbon. Pyrolysis causes the structures to shrink in all directions, maintaining the shape in a much smaller package.
Pyrolyzed photoresist films, or PPFs, can be used to make unusually shaped devices. For example, Madou is using them to construct “washline” nanosensors, which consist of nanoscale carbon wires suspended between microscale carbon posts. The posts are first fabricated on a surface via photolithography, and then the wires are generated by electrospinning a polymer onto them. The posts and wires can be made of the same or different starting materials. The structures are then pyrolyzed together, which forges carbon-carbon bonds between the wires and their supports.
Such wires are “almost all surface,” Madou said. Any electronic change induced at their surface “leads to a large resistance change in the wire and better limits of detection.” Madou plans to add specificity to such sensors by hanging molecular laundry—DNA or proteins—that can serve as recognition elements on the washlines.
“Electrospinning aligns the polymer fibers,” Madou said. “Then, when you do the pyrolysis, you get graphitic material.” Graphite formation usually requires much higher temperatures, but the fiber prealignment results in graphite wires being formed at temperatures as low as 900 °C.
Carbon films made with PPFs are much flatter than conventional glassy carbon films, McCreery said. Despite its smooth appearance, commercial glassy carbon has surface roughness on the order of 4 to 40 nm when measured by atomic force microscopy, whereas PPFs have roughness less than 0.5 nm. “When you’re making devices smaller than 5 nm, flatness is very important,” McCreery said.
McCreery is using PPFs to impart molecular function to otherwise conventional electronics. For example, he attaches aromatic molecules to carbon in the PPFs to make memory devices or chemical sensors. In addition, he has shown that he can deposit metals such as copper and gold on the PPF surface, creating electronic junctions between carbon-based molecules and metals that allow the devices to be integrated with semiconductor systems.
Different methods can be used to pattern the photoresist. Exposing the photoresist film to ultraviolet radiation prior to pyrolysis is the most common approach.
But Mark T. McDermott, a chemistry professor at the University of Alberta and a research officer at the National Institute for Nanotechnology, instead etches the PPF layer after it’s been pyrolyzed. He also uses electron-beam lithography to pattern the film. Unlike light-based methods, electron-beam lithography is limited not by diffraction but by electron scattering. Because an electron beam, or e-beam, is easily steered, McDermott can make extremely small features with any shape, including curved lines.
“Our record so far is a 10-nm-wide line,” McDermott said. “With normal lithography, you couldn’t even approach that size.”
With other nanoscale technologies, such as carbon nanotubes, connecting nanoscale devices to the outside world is a particular challenge, McDermott said. E-beam fabrication offers a way around such problems.
“We can write a nanoscale line and also write a micron or bigger pad to connect it to,” McDermott said. “Everything can be written from the nanoscale to the micron scale with the same e-beam device.”
McDermott demonstrated that he can use e-beam lithography to program device characteristics such as resistance. Because the resistivity of pyrolyzed films remains constant at different size scales, the overall resistance of a device depends on its dimensions, so “by changing those parameters you get different resistors,” he said.
These carbon microfabrication techniques make it easy to fabricate arrays of electrodes. R. Mark Wightman of the University of North Carolina, Chapel Hill, and Gregory S. McCarty of North Carolina State University are collaborating on PPFs to make electrode arrays that Wightman can use for neurochemical measurements.
Microelectrode arrays can provide more information than individual carbon-fiber microelectrodes in neurochemistry studies, Wightman said. If the electrodes are closely spaced, they can provide a picture of the workings of a single brain area. Broader spacing in an array can reveal how multiple brain areas work together.
For example, Wightman used PPF-based microelectrode arrays to measure the neurotransmitter dopamine in rats’ brains. The response to dopamine was different at each electrode in the array, demonstrating heterogeneity of the response to dopamine within the striatum region of the brain. Adding cocaine, which blocks dopamine uptake, also resulted in a spatially heterogeneous dopamine response.
Despite the applications of PPF-based microdevices that have already been demonstrated, widespread adoption faces challenges. “Microelectronic fabrication often involves high temperatures,” McCreery told C&EN, “but 1,000 °C for more than an hour would be difficult in a commercial manufacturing line.”
Nonetheless, he remains optimistic about the future for carbon devices made with PPFs. “If the microelectronics industry can get something out of the technology, they’ll make it work,” McCreery said. Devices made this way have a better chance of being practical than ones made with carbon nanotubes, he noted. “You don’t make a billion devices one at a time. The wonderful thing about PPFs is not just that the devices are small but that they’re massively parallel.”
He believes carbon devices will augment, not replace, existing technology by adding functionality on top of silicon. “Look at the 1950s,” McCreery said. “There were tubes, and next to the tubes were transistors. They coexisted. Maybe someday we will make microelectronic devices out of just [carbon-based] molecules, but not in my lifetime.”
- Chemical & Engineering News
- ISSN 0009-2347
- Copyright © American Chemical Society