Duke professors participate in world-wide land mine project

Leslie Collins usually works on hearing implants, a product aimed, in any context, solely at helping people. So as she learned about her latest project, she grew more and more shocked; an assistant professor of electrical and computer engineering, Collins is playing a part in the worldwide campaign against land mines as part of a government-funded study at the University.

Laying out a few photos of land mines on her desk, Collins points to a picture of an M-14 anti-personnel mine, about the size of a camera's lenscap. "They're not designed to kill you, they're designed to injure you," she said, noting that a maimed soldier would demoralize a squad much more than a dead one. "I couldn't believe people thought like this," she said.

Along with associate professor Lawrence Carin and professor Richard Fair, both in the department of electrical and computer engineering, Collins has worked for a year and a half to create better techniques and equipment to detect these hidden killers.

Aided by a $6-million multi-university research grant from the U.S. Army, the team of researchers is relying on a three-pronged approach: Increased basic research into the physical properties of detection, enhancement of the hardware used for detection and improved analysis of the signals that come from land mine detectors.

Carin, the principal investigator for the project-which includes researchers at the California Institute of Technology, the Georgia Institute of Technology and Ohio State University-is working on the basic research aspect of the study.

One of the technologies Carin is currently developing is called nuclear quadrupole resonance, a technique related to the magnetic resonance imaging used in medicine.

"Many explosives have a crystalline structure that aligns the atoms in a special way," he explained. "Whenever you bring electromagnetic radiation near them, the aligned atoms oscillate. If you can detect that [motion], you have a good chance at detecting the mines."

He added that the new technique detects the chemicals that give mines their explosive power rather than the easily disguised case. "With NQR we're not looking for the metal or plastic casing but for the explosive. That is why we're excited and that is why it's important."

This innovation represents an important improvement in the detection process because today's mine sensors usually rely on metal detectors that cannot distinguish between a harmless piece of metal and a land mine. "The false-alarm rate is prohibitively high," Carin said.

Decreasing this false alarm rate is a critical part of the project. On the hardware side, Fair is designing a detector that relies on chemicals such as TNT that leach into the soil after a mine is buried.

The detector uses ultra-sonic energy to lift chemical particles up into an airstream where they land on a microchip made of a bimetallic strip of gold and silicon. The particle is irradiated, causing it to decompose and give off energy, which in turn makes the strip bend and change much like a common thermostat. The change in the strip's shape sets off a signal to alert the user to the presence of a land mine.

Using small samples of TNT, PETN, and REDX-explosives common to land mine production-Fair and his colleagues have categorized the signals given off by different chemicals. "We're ready to look for the signatures of decomposing explosives with our sensors," he said.

Like Carin's work, this process helps detect non-metallic mines. For example, Fair said, "The ones made in China are made out of plastic, with no metal materials to detect."

But for metal mines, conventional detectors are still useful, and Collins is attempting to improve this aspect of the detection process.

Collins' research emphasizes the processing of signals from existing metal detectors as well as new models such as Fair's.

Currently, Collins said, metal detectors used to hunt land mines work by measuring the energy given off by metal buried in the ground. In the course of creating a signal to alert the user, the individual signals of the metals merge into one signal.

When the signal is processed, the unique characteristics of the mine are drowned out, making the mine indistinguishable from a nail.

"Previously, the information in the signal had been washed out by the process which transferred the signal to the soldier," Collins said.

The new technology will allow for faster searching of sites and fewer false alarms by reading the individual, tell-tale sign of land mines.

The University scientists working on this project seem very satisfied with their work, and the humanitarian nature of their efforts is a constant reminder of the work's importance.

"There's a huge international effort for banning land mines," Fair said. "They're insidious because they stay there and they maim people in peace time also."

"It's very important that the U.S. is being seen to care and putting research dollars into the problem," Collins added.

"As a nation of caring people that's important that we do that."

Discussion

Share and discuss “Duke professors participate in world-wide land mine project” on social media.