Shortly after terrorist bombs ripped through central London's transit system on July 7, Scotland Yard dispatched trained sniffer dog teams to search for explosives and to scent out clues at the blast sites. Meanwhile, less than an hour up the M11 highway in Cambridge, engineers Billy Boyle, Andrew Koehl and David Ruiz-Alonso were lamenting the fact that the antiterrorist technology they had worked on since just after 9/11--a sensitive but inexpensive electronic nose--had not been ready to help avert this tragedy. The Ph.D. engineers have developed a button-size chemical sensor prototype that is designed, among other things, to detect trace amounts of explosives before they detonate.

The prospect of a modern-day coal-mine canary for trains and buildings still lies in the future for the entrepreneurial brain trust of Owlstone Ltd., the University of Cambridge spin-off company the trio established two years ago. But backed by $2 million in venture-capital funding, the device should be ready for field tests this fall. The three are confident that the low-power device can quickly identify tiny concentrations of substances in parts per billion. "Our idea is to put one on the lapel of every soldier and in every Tube carriage," Boyle states.

The R&D effort started in late 2001, when Koehl, an electrical engineer, arrived at Cambridge from the California Institute of Technology. "From the beginning, I had the idea to create a small, cheap chemical detection system for the military and Homeland Security and then, later, for commercial markets," Koehl says. He soon met Boyle and Ruiz-Alonso, and during the next months, the engineers looked at a lot of sensor technologies, "trying to evaluate what we could take to the next level," Boyle reports.

A chemical sensor "is essentially a filter device," Boyle explains. "Each substance has its own signature smell or fingerprint. That's what we filter out." For Owlstone's sensor, the team chose to develop a relatively new and little known analytic technique called high-field asymmetric waveform ion mobility spectrometry (or FAIMS). The approach sorts compounds according to how their charged forms--chemical ions--move through a gas when subjected to electric fields.

As the ions are made to pass between charged metal plates, varying electric fields (that is, the asymmetric waveforms) drive them toward either side and back again successively, eventually causing most of the ions to hit the plates. But careful application of direct-current voltages to the plates keeps targeted molecules from hitting the sides until they reach a detector at the end. Each DC voltage correlates to an ion type, so the device can be "tuned" to detect specific substances.

Boyle likens the process to "trying to balance several marbles on a sheet of cardboard. Think of the different-size marbles as different ions." Only the right rocking movements--the DC voltages--will keep the largest marbles on the sheet. The separation of ions occurs within only tens of millimeters. Conventional ion mobility spectrometers, the current standard detection technology, are the size of a shoebox or larger because they typically use bigger sensors and a pump to move air between the plates.

Advanced FAIMS sensor technology, which is being pursued in an alternative form by a Massachusetts-based company called Scionex as well as large sensor makers, can also be more selective than conventional detectors. To many standard machines, the "smellprint" of the deadly nerve gas sarin resembles that of common diesel fuel. "With FAIMS, we can simply apply a higher voltage" to distinguish between the two, Boyle notes.

The Owlstone sensor also has many nonsecurity applications--say, as a household fire detector that can discern precombustion products formed before the flames start. It could also work as a sophisticated breath analyzer to identify volatile compounds generated by disease (for example, acetone can be used to monitor diabetes). Antiterrorism and military applications, though, remain the focus for the engineers. Says Boyle: "We hope that one day we'll be sitting on the Tube, look up to see one of our sensors, and know that everybody there is safe because of something we did."