14 February 2011—Flash memory, scanning tunneling microscopes...and a fly’s sense of smell. According to new research, the same strange phenomenon—quantum tunneling—makes all three possible. If confirmed, the discovery could pave the way for a new generation of artificial scents, from perfumes to pheromones—and, perhaps someday, artificial noses.
The conventional theory of smell holds that the nose’s chemical receptors—some 400 different kinds in a human nose—sense the presence of odorant molecules by a lock-and-key process that reads the odorant’s physical shape. That theory has some problems, though. For instance, ethanol (which smells like vodka) and ethanethiol (which smells like rotten eggs) have essentially the same shape, differing from each other by only a single atom. (Ethanol is C2H6O, and ethanethiol is C2H6S.)
Evidence has emerged over the past decade suggesting that at least part of a molecule’s scent comes from chemical receptors in the nose that pump current through the odorant molecule and cause it to vibrate in an identifiable way. Lacking a direct electrical hookup to the odorant, the nose’s receptors would likely transmit electrons via quantum tunneling, a well-studied process that allows electrons to hop through nonconducting regions if they are small enough. Tunneling is what allows charge to be stored in flash memory cells. It also forms the image in scanning tunneling electron microscopes and is a source of wasted power in microchips.
A group of four scientists from MIT and the Alexander Fleming Biomedical Sciences Research Center, in Vari, Greece, says it has proved the tunneling theory in fruit flies (Drosophila), a favorite lab specimen of geneticists. If the quantum ”molecular vibrational” theory of smell is correct, then the flies should be able to smell the difference between molecules that have regular hydrogen in them versus the same molecules that contain its heavier isotope, deuterium. (The nucleus of regular hydrogen is a proton; the nucleus of deuterium is a proton plus a neutron.)
While regular and ”deuterated” molecules have exactly the same shape, when set in motion by the receptors’ tiny tunneling current, the two would vibrate in a very different pattern of molecular wobbles. A tunneling-based sense of smell should be able to detect the different vibrations. Deuterated molecules would, in other words, smell different to the flies.
According to the new research—published in this week’s Proceedings of the National Academy of Sciences—the flies appear to be able to smell the difference. The researchers had set up a tiny maze with a T-junction. To the right, the molecule acetophenone (which has a sweet and flowery scent to human noses) filled the air. To the left, the air contained an isotope of acetophenone in which eight of the molecule’s hydrogens were swapped out for deuterium. Nearly 30 percent more flies went toward the regular acetophenone. Meanwhile, flies that had been genetically engineered to lack a sense of smell showed no preference.
The group repeated the experiment with a different molecule, octanol, and its deuterated cousin, and got the same result. In a third experiment, the researchers in Greece trained their flies to avoid a deuterated octanol. The vibrations between carbon and deuterium in octanol, says study coauthor Luca Turin, a visiting scientist in biomedical engineering at MIT, are very similar to those of carbon and nitrogen in chemicals called nitriles. So if the vibration theory holds, then some deuterated molecules might share similar odors to the nitriles, even though their shapes are worlds apart.
Indeed, when the trained flies were presented with a whiff of a nitrile, they avoided it. Conversely, flies conditioned to avoid nitriles also avoided the deuterated octanol. ”The only thing the deuterium and nitrile have in common is vibration [pattern],” Turin says.
Andrew Horsfield, senior lecturer in the department of materials at Imperial College London, has been working on early applications of the vibrational theory that Turin’s group tested. Horsfield and his colleagues have developed an indium-arsenide nanowire detector that might crudely ”smell” itself by tunneling electrons between special structures within the nanowire and reading off the vibrations produced. Horsfield’s group plans to extend this idea to smaller devices that can smell external molecules. The group has so far only been fine-tuning its nanowire setup, and Horsfield says its first ”sniff” test might be more than a year away.
Horsfield says that the research being done by Turin’s group justifies his group’s continued nanowire studies. ”It’s very clear to me that this is a very important paper,” he says. ”Time will prove that to be true.”
About the Author
Mark Anderson is an author and science writer based in Northampton, Mass. In the January 2011 issue he described an alternative way to measure the power of supercomputers.