By chemically modifying a silicon nanotube, researchers have created a fluid-based transistor that can conduct either positively or negatively charged ions dissolved in solution. As in ordinary field-effect transistors, the voltage on a “gate” electrode controls the rate of ion flow through the water-filled tube. The degree of chemical modification determines whether the tube conducts positive or negative ions, or switches from one to the other depending on the gate voltage, the group reports in the 19 August PRL. The technology might be useful in highly sensitive biological sensors, and the result demonstrates the basic components needed to build fluid-based “computer chips” by this method, according to the team.
Controlling the flow of electrons with transistors is the basis of modern digital technology. Manipulating the flow of individual charged molecules may be instrumental to future machines designed to diagnose disease, for example, by sifting through femtoliters or less of blood or saliva for telltale molecules. To see whether today’s materials are sufficient to control the flow of charged atoms, Peidong Yang and his colleagues from the University of California, Berkeley, began investigating silicon nanotubes. Unlike water-repelling carbon nanotubes, silicon naturally forms a charged oxide surface on its outer layer–perfect for attracting ions in solution.
For each experiment, Yang’s group used a roughly 15-micron-long nanotube to connect two reservoirs containing a potassium chloride solution. Atop the nanotube, which lay on a silica surface, they deposited a gate electrode. As they had demonstrated in a previous paper  negative voltage applied to the gate electrode strips positively charged hydrogen ions from the nanotube’s interior oxide surface, increasing the surface’s negative charge. The tube then attracts positively charged ions, or cations, of potassium and repels negatively charged ions (anions) of chloride.
As in their previous work, when the team generated an electric current through the tube by applying voltages to the reservoirs, it conducted only cations. The more negative the gate, the more cation current flowed, and the more positive the gate, the less the current–exactly the characteristics of a so-called p-type field-effect transistor. This process occurred only because the inner width of the tube was 40 to 50 nanometers, narrow enough that the charge at its surface could be felt by ions in the center.
To make nanotubes conduct anions instead, they now report that they soaked the device in a nitrogen-silicon organic chemical. The soaking caused the tube to incorporate positively charged amine groups, which counteracted the oxide’s negative charges and increased the tube’s preference for anions. After two days of soaking, a tube could be made to conduct cations or anions, depending on the gate voltage. After four days, the tube only conducted anions.
“Just like the [charge-] carrier concentration being controlled by the doping level in semiconductors, here the concentration is controlled by the surface charge density of these tubes,” says Yang. The anion-conducting tubes are analogous to n-type transistors, he says, so the group can now begin assembling more complicated systems such as diodes–perhaps made of two nanotubes–and, in principle, integrated circuits for building miniature biochemical analysis devices.
This type of control is much simpler than mechanical pumping for moving small volumes of fluids, but until now has been nearly impossible in highly conductive fluids such as blood, urine and saliva, says engineer Chih-Ming Ho of the University of California, Los Angeles. “This concept will make it possible,” he says, thanks to the narrowness of the nanotubes.
JR Minkel is a freelance science writer in New York City.
- R. Karnik, R. Fan, M. Yue, D. Li, P. Yang, and A. Majumdar, “Electrostatic Control of Ions and Molecules in Nanofluidic Transistors,” Nano Lett. 5, 943 (2005).