Electronic Nose Is Designed to Sniff Out Hazards

Harry Tuller copies the physiology of natural noses to design an electronic nose.

Electronic noses sense smells in a way similar to human noses.

If an aroma were a color, it might be printed as a dappled mixture of cyan, magenta and yellow. Wine tasters use vague terms like "floral and vegetal nuances" to describe bouquet because scents are complex overlays of volatile chemicals that excite different sensors in the human nose.

"The way we distinguish between coffee odor and fish odor is not that we have one sensor designed to detect coffee and one designed to detect fish, but our nose contains arrays of sensors sensitive to various chemicals," says Harry Tuller, an electrical engineering professor at MIT. "Over time, we train ourselves to know that a certain distribution of vapors corresponds to coffee."

Tuller and his lab members are borrowing from the physiology of natural noses to design an electronic one that picks up scents in much the same way. Using a programmable ink-jet printer head, the researchers have devised a way to print tiny droplets of gas-sensitive barium carbonate rapidly onto a microchip. Each drop is a hollow microsphere with walls only billionths of a meter thick that contain countless docking sites for scent molecules.

The device could be used to sniff out hazards such as carbon monoxide, industrial solvents and even explosives. But for a sensor to be useful, it must be able to distinguish between gases. A sensor at an airport, for example, would need to know the difference between a toxic chemical and perfume, Tuller said. Sensors should have an array of "droplets" that each respond differently to different gases. This is similar to the way the human sense of smell works, Tuller explained.

The inkjet print head, like that in a regular inkjet printer, can deposit patterns of different gas-sensitive inks quickly and controllably, just as in a color printer. The process allows the researchers to rapidly produce many small, identical chips and could eventually allow for mass production of highly sensitive gas detectors.

"Mass production would be an enormous breakthrough for this kind of gas sensing technology," said Tuller, who presented the research this week at the Composites at Lake Louise Conference in Alberta, Canada.

—By Leslie Fink, NSF

This material is provided by the National Science Foundation, an independent federal agency that supports fundamental research and education across all fields of science and engineering, in partnership with U.S. News & World Report.