Uncommon Sensors

By Katherine McIntire Peters

July 1, 2002

Forget about ships, tanks and airplanes. The future of national security rests with tiny devices that detect danger before humans can.

while defense analysts and pundits talk of a revolution in military affairs, James Gole holds one in his hand: a glowing wafer about the size of a dime. Gole, a professor of physics at Georgia Tech, along with colleague Peter Hesketh, a professor in the university's mechanical engineering school, have discovered a way to detect minute amounts of lethal gases nearly instantaneously, using specially treated porous silicon. Although silicon sensors already existed, they were expensive to manufacture, required a lot of energy to operate, and could detect gases only in relatively high concentrations. The highly sensitive gas sensor developed by Gole and Hesketh consumes a fraction of the power of a typical wristwatch, is cheap to manufacture and is reusable-traits that would be impressive enough by themselves, but in combination represent a remarkable leap forward in gas-detection technology.

How did they do it? By tapping the power inherent in silicon's unique properties, Gole says. When ultraviolet light passes through porous silicon, it emits a brilliant orange color. Scientists have long debated the reason for silicon's photoluminescence. Gole and Hesketh haven't found a definitive answer, but they have found a way to stabilize this light emission and use the energy it produces to plate the silicon with metal ions. The metalization process renders the silicon vastly less resistant, meaning that it will react almost instantaneously to the presence of ammonia, hydrochloric acid and nitrogen oxides, even in minute concentrations of 10 parts per million and lower.

This might not mean much to the average person, but the implications for security, and perhaps even medicine, are enormous, says Gole. Such sensors could be easily-and cheaply-dispersed across a battlefield or military staging area to detect an enemy's use of poison gas, for instance. "Right now the device we have built is basically a centimeter in diameter. You can hold this in your hand. A soldier could take it into the field. The other thing is the response time, which is seconds. That's a very big breakthrough." The next step, which is admittedly a big one, Gole says, is to make the sensor more selective in detecting other gases and agents. Eventually, the researchers believe they will be able to build networks of such sensors that could monitor air and water quality, analyze blood for abnormalities and test for pathogens with a level of precision and economy heretofore unimaginable.

Gole and Hesketh aren't the only scientists promising to revolutionize the way we look out for trouble. In laboratories across the country, researchers are making significant breakthroughs in sensor technology, a catchall description for devices that detect changes in temperature, light, sound, motion, biological or chemical makeup, and a host of other properties. Sensors aren't new-smoke detectors, car alarms and even refrigerators rely on them to function properly-but advances in manufacturing, microelectronics and wireless technology will dramatically improve their usefulness over the next few years, scientists say.


Mike Horton, founder and chief executive of Crossbow Technology Inc., a sensor technology firm based in San Jose, Calif., believes sensors will be virtually everywhere in a few years. "Ubiquitous sensing," he calls it, "because [micro-electromechanical technology] makes the devices so much cheaper and also smaller-you can use a lot more of them than heretofore was possible."

For years, industry officials have been interested in how sensors could improve productivity in manufacturing, but more recently military and now homeland security requirements are driving growth. "The push is really coming from all over," Horton says.

"The notion of what kinds of things you can sense is still fairly traditional-you can sense physical things, motion, chemicals, people-but the ability to make them inexpensive, much smaller and connected to the digital world makes it possible to have a lot more of them in a lot more places collecting a lot more data," Horton says.

Many of the advances in sensor technology are the direct result of investment in military space and missile defense programs. The Defense Department has placed a premium on the development of remote sensing capability, a term that generally refers to sensing done from satellites. Once the purview of the military and intelligence agencies, high-resolution images from space are now widely available over the Internet. The commercial market for images generated from satellites has exploded in recent years as urban planners, cartographers, meteorologists and others regularly depend on remote sensing technology to do their jobs.

Developments in new imaging technologies will only fuel the trend as images become increasingly informative. For instance, with funding from the Missile Defense Agency, the Kestrel Corp., based in Albuquerque, N.M., is developing a hyperspectral imager, a sensor so sensitive that it can distinguish a maple leaf from an oak leaf from thousands of miles above the earth's surface.

Other researchers are demonstrating equally impressive leaps in imaging technology. Xi-Cheng Zhang, a physics professor at Rensselaer Polytechnic Institute in Troy, N.Y., has developed a sensor that collects data from pulses of terahertz radiation, called T-rays, as they pass through human tissue. The images produced from the data are far more detailed than X-rays or ultrasound. Zhang also has demonstrated that T-rays, which exist in the infrared region of the electromagnetic spectrum, can be used to detect underground toxins and anthrax bacteria concealed in packages. Zhang's pioneering work has peaked the interest of federal agencies; he has received $7 million in grants from the National Science Foundation, the Army and the Energy Department.

"This is really a scientifically rich area," Zhang says. "And technically, it's unexplored." The potential applications for T-ray imaging in environmental testing and security are great, but the payoff is likely to be in biomedical applications, he says. Eventually, Zhang believes T-ray sensors will be widely used to detect medical problems ranging from cavities to cancers and other diseases far earlier and far more safely than they are detected now, since T-rays are safer than X-rays, he adds.


Even comparatively simple sensor technology is finding dramatic new applications, thanks to advances in wireless technology. Horton's Crossbow Technology recently demonstrated for the Air Force a wireless microradar sensor network that monitors activity over a broad area. Each sensor, really a collection of sensors bundled together and housed in a small box the size of two packs of cigarettes, continuously sends out radio waves. When waves bounce off an object, they produce a "signature," basically a set of data. Each box then relays that signature back to a base station, which compiles data reported from multiple sensors. The data are calibrated to distinguish between objects-an animal will look different from a truck, for example-and are used to produce an image of the area being monitored. Each box, which can run 24 hours a day for several months on two AA batteries, can "see" about 100 feet. The more boxes that are deployed, the broader the area that can be monitored. The sensors don't provide a perfect picture, Horton says, but they still can provide vital information. "If you're trying to protect an area, it's a very useful device," Horton says.

For the military, the benefits of having such a network are obvious. When troops deploy to unfamiliar territory, the sensors could add significantly to the capability of scouts and security personnel. But the advantages of these artificial eyes and ears at home are not lost on government officials. At a time when nearly everything seems a feasible target for terrorists, networks of relatively inexpensive sensors are beginning to look attractive, especially in rural areas where a single sheriff, and maybe a deputy or two, have to cover a lot of ground.

Martin Harmless, president of Clarion Sensing Systems Inc. in Indianapolis, says sensors should be the first line of defense in protecting the nation's water supply. Multiple sensors capable of detecting chemical and biological agents could be deployed throughout the water distribution system and linked by a wireless network. Any security breaches would immediately be relayed to a regional monitoring station, which would then notify authorities and close off the affected area using remote controls, before contaminated water could reach the public.


In a Feb. 22 interview with the Massachusetts Institute of Technology's online publication Technologyreview.com, Gerald Yonas, vice president and principal scientist at Sandia National Laboratories in Albuquerque, N.M., said sensor networks will provide a critical edge in the war on terrorism. Precision weapons are of little use without real-time target information. Such information increasingly will be gleaned from sensor networks, he said.

"The whole point is to be able to deploy a network of sensors in a system that can locate, observe, and then give us the knowledge to neutralize the enemy. I'd say it's doable in three to five years," Yonas said. When asked about the next step, Yonas replied, "In addition to us going into another region and blanketing it with an information web of ubiquitous sensors, we build our own infrastructure that way. We build our own buildings and cities with instruments that are interconnected and can share information with each other and with decision-makers, and adapt and respond."

Yonas' vision for the future may not be far off. More fiber-optic strain sensors are being built into bridges and buildings, Horton says. "There's a lot of work being done in structural monitoring-sensing buildings, bridges, railway tracks and things like that for structural changes, to see if the structure is deteriorating or if there is seismic activity or potential terrorist activity. The building itself can tell you if it's healthy or not."

The explosion in sensor technology and capability will have benefits that go beyond the military and homeland security. For now, though, sensors that boost security are likely to receive the most attention and funding.

Tom Bevan, a principal scientist at the Georgia Tech Research Institute in Atlanta, was concerned about homeland security long before Sept. 11 brought the issue to the public's attention. About three years ago, soon after he moved to Atlanta to take the job at Georgia Tech, he learned one of his neighbors had received a letter in the mail that purportedly contained anthrax. The letter turned out to be a hoax, but it made an impression on Bevan.

"That was a personal epiphany," Bevan says. He had spent the previous two years living in the former Soviet Union, where he worried about his personal safety. He never dreamed he'd have to worry about terrorism when he returned home. As a result of his proximity to the threat of anthrax, and the growing concern among public officials, Bevan founded the Center for Emergency Response Technology, Instruction and Policy, a public-private partnership between the Marine Corps and Georgia Tech. The center's mission is to put better technologies in the hands of those who would be most likely to respond to an act of terrorism.

"Microelectronics have come a long way. That underlies a lot of the sensor technology that we and others are developing," Bevan says. A less obvious factor influencing the center's research is chicken processing, he adds. Because the poultry industry is the biggest industry in Georgia, Georgia Tech has done a tremendous amount of research in developing sensors to detect salmonella and E. coli bacteria and other diseases that affect chickens and make humans sick.

"What we're doing is adapting a lot of that research to chemical, biological or commercial agents that might be involved in [a terrorist] incident," Bevan says. One of the primary pieces of equipment the center is developing is an integrated optical sensor. The sensor, which is about the size of a shoebox, contains multiple sensors capable of detecting numerous chemical and biological agents.

"You need at least one sensor for each thing you want to look for. On a 1-by-2 centimeter chip, you can get about 70 to 75 sensors looking for different kinds of chemicals or biological agents," Bevan says. Built using parts largely available at Radio Shack-a laser like the one in your CD-ROM drive, a microchip, and components cannibalized from a Web camera-a chief virtue of the system is its low cost. "It compares favorably to what firemen have to pay for the chemistry sets they use to identify agents," Bevan says. "And the sensors are cheaper, easier to use, and they show quicker results."

"It's simple," he adds. "It doesn't take a Ph.D. to operate." One day, he says, such devices may become as prevalent as smoke alarms.

By Katherine McIntire Peters

July 1, 2002