Chemistry is helping scientists in the US to detect explosives from a safer distance.

Detecting trace amounts of explosives from a distance quickly and accurately is a key aim for both the military and security sectors. Jennifer Gottfried and her team at the US Army Research Laboratory in Maryland have developed a detection system sensitive and selective enough to detect explosive residues at 20m. Their system uses laser induced breakdown spectroscopy (LIBS) to identify molecules.

LIBS uses a laser pulse to turn a small part of a sample into a plasma of excited atoms and ions. As the plasma cools the characteristic atomic spectra of its constituents can be detected. The idea is that energetic molecules (potential explosives) contain higher ratios of oxygen and nitrogen to carbon and so can be identified by looking at the ratios of these atoms in a sample. This is far from straightforward, however, not least because nitrogen and oxygen are the main components of air.

Gottfried and her colleagues think they have overcome the problems. Using an argon flow to displace air reduced the interference from air sufficiently, but is impractical for detecting explosives at a distance. For this, they discovered that using a second laser pulse instead of only a single pulse enabled sufficient separation of the N and O peaks due to air from those of the sample compounds. The team then developed suitable chemometric methods to enable interpretation of the data so that detection of energetic molecules was possible, even in the presence of interferences, such as dust.

Gottfried hopes these developments will fill an important gap in security systems. 'Currently there are no proven technologies that can accomplish residue explosives detection at a distance in a real-world scenario,' said Gottfried. She is optimistic that it will develop into a usable device, saying 'We expect that this technology will be available commercially very soon.'

Further development is still needed though, as Gottfried makes clear: 'This technology still needs to be verified and validated in real-world applications. We are moving in that direction.'

Edward Morgan