The research could lead to simple detectors capable of differentiating between a wide variety of pathogens, including viruses, bacteria and toxic organic chemicals.
The experiment, an extension of earlier work in which similar devices were used to detect the mass of a single bacterium, is reported in a paper, "Virus detection using nanoelectromechanical devices," in the September 27, 2004, issue of Applied Physics Letters by Cornell research associate Rob Ilic of the Cornell NanoScale Facility (CNF), Yanou Yang, a Cornell graduate student in biomedical engineering, and Harold Craighead, Cornell professor of applied and engineering physics. The work was done with the assistance of Michael Shuler, Cornell professor of chemical and biological engineering, and microbiologist Gary Blissard of the Boyce Thompson Institute for Plant Research on the Cornell campus.
At CNF, the researchers created arrays of tiny silicon paddles from 6 to 10 micrometers (millionths of a meter) long, half a micrometer wide, and about 150 nanometers (billionths of a meter) thick, with a one-micrometer square pad at the end. Think of a tiny fly-swatter mounted by its handle like a diving board. A large array of paddles were mounted on a piezoelectric crystal that can be made to vibrate at frequencies on the order of 5 to 10 megaHertz (mHz). The experimenters then varied the frequency of vibration of the crystal. When it matched the paddles' resonant frequency, the paddles began to vibrate, as measured by focusing a laser on the paddles and noting the change in reflected light, a process called optical interferometry.
The natural resonant frequency at which something vibrates depends on, among other things, its mass. A thick, heavy guitar string, for example, vibrates at a lower tone than a thin, light one. A single one of these silicon paddle weighs about 1.2 picograms, and vibrates at frequencies in the neighborhood of 10 megaHertz. The virus used in the experiment weighs about 1.5 femtograms. (A picogram is 1/1,000,000,000,000th of a gram, and a femtogram is 1/1000th of a picogram.) Adding just a few virus particles to a paddle turns out to be enough to change its resonant frequency by about 10 kiloHertz (kHz), which is easily observable.
To trap viruses, the researchers coated the paddles with antibodies specific to Autographa californica nuclear polyhedrosis virus, a nonpathogenic insect baculovirus widely used in research. The paddle arrays were then bathed in a solution containing the virus, causing virus particles to adhere to the antibodies. Because air damps the vibration and greatly reduces the "Q," or selectivity, of the system, the treated paddles were placed in a vacuum for testing. From the frequency shift of about 10 kHz the researchers calculated that an average of about six virus particles had adhered to each paddle. It might be possible, the researchers say, to demonstrate detection of single particles by further diluting the virus solution. The system also can differentiate between various virus concentrations, they say.
As expected, the smallest paddles were the most sensitive. The researchers calculated that the minimum detectable mass for a six-micrometer paddle would be .41 attograms (an attogram is 1/1000th of a femtogram.) This opens the possibility that the method could be used to detect individual organic molecules, such as DNA or proteins.
Other members of the Craighead Research Group at Cornell have experimented with "nanofluidics," creating microscopic channels on silicon chips through which organic molecules can be transported, separated or even counted. Ilic speculates that a simple field detector for pathogens -- the much-heralded "laboratory on a chip" -- could be built by combining a paddle oscillator detector with a nanofluidic system that would bathe the paddles in a suspect sample, then automatically evacuate the chamber to a vacuum for testing. Arrays of paddles coated with various antibodies could allow testing for a wide variety of pathogens at the same time.