Vapour sensing of explosives, chemical agents and drugs

There is significant civil and military interest in the accurate detection of the presence of hazardous and illicit materials. The current state of the art for detection of explosives and narcotics is the olfactory sense of a trained sniffer dog, however there are significant benefits that could be added with augmentation with sufficiently sensitive vapour detection equipment. There is clearly a need for automated ultra-sensitive detectors: this requirement currently being met in part by sensor technology based on ion mobility spectroscopy. The drawback with these detectors is that they are large, expensive, and delicate. These attributes make them suitable for use at fixed monitoring points, but they cannot be readily adapted into a lightweight, mobile device.

The key requirement is to have access to explosives sensing technology that can be readily miniaturised and will thus allow molecular detection devices to be incorporated into small instruments such as watches or clothing. A further requirement is that the sensor is cheap and does not require protracted training for the user. The creation of such a device requires an entirely new type of gas detection technology, which is currently being developed in Professor Castell's research group. The work on a novel ultra-sensitive electrochemical sensor was supported through an EPSRC grant (EP/I015973/1).

The main advantages of electrochemical sensors are that they are light, cheap, and robust. However, they have the drawback that they tend to lack the necessary sensitivity requirements for explosives detection. This can be a particularly acute problem if the explosive to be detected is buried and has a low volatility. For example, the vapour pressure for 2,4,6-trinitrotoluene (TNT) at 25 Celsius is 7.7 x 10-6 mbar, which is less than 10 parts per billion (ppb) in air, so for a sensor to be of any practical use it will need to be able to operate at a better than 1 ppb level. For the sensor to be effective it also needs to be highly selective and only respond to a prescribed analyte.

The new Oxford sensor design is based around an electrical percolation network of conducting polymers. With few conducting regions no charge can flow through the network because there is no continuous conducting pathway from one side of the network to the other. As the proportion of conducting elements is increased, the probability increases that a conducting pathway is formed. There is a critical point, known as the percolation threshold, when the number of conducting elements is sufficient to provide a conducting pathway in half the random configurations. The critical point of the new sensor design is to create a detection method where target molecule adsorption onto the percolation network affects it around the percolation threshold. As a result of molecular adsorption the electrical conductivity of the network will change by orders of magnitude. This allows very low concentrations of molecules to be detected by measuring the change in electrical resistance of the percolation network.

In this DPhil project percolation sensors will be developed for the detection of explosives, chemical agents, and drugs. The student will be involved in a broad range of interdisciplinary activities including design, building, and testing of the sensor. The student will have regular contact with the Dstl Explosives Detection Group, based near Sevenoaks in Kent and Porton Down in Wiltshire. Some extended visits to Dstl of a few weeks per year may also be required. A key element of the work will be the characterisation of the sensor, particularly the selectivity and sensitivity for target explosives and incidence of false positives. Under the CTW programme, a significant amount of work has been undertaken over the past two years to create variable, quantifiable vapour sources of a range of explosives, which can be used for this work.

The Project is aligned with the EPSRC Global uncertainties theme.