Since 9/11, we are increasingly threatened by terrorist plots — the release of noxious substances into crowded public places or on airplanes, in crowded buildings, and in sports arenas. Anxieties about these rare events are mostly unrealized, but such events have occurred frequently enough — such as in Madrid, London, or Delhi — that protective measures that will reassure the public are needed. To be effective, sensors and screening devices must be deployed widely, be inexpensive, support high-volume throughput (respond rapidly), and be available for a wide range of threats (i.e., flexible and adaptable to different or new scenarios).

Given these scenarios, devices are needed that identify defined chemical targets, yet at the same time, remain flexible enough to respond to rapid shifts in terrorist tactics employing new explosive and toxic materials. Selectivity based on sensitivity to single chemical attributes such as differential retention on a chromatographic column, allows sorting in one dimension, whereas selectivity based on multiple chemical attributes can enable devices to identify more "components" or compounds with fewer sensors.

The use of sensors that can examine multiple chemical attributes, thereby allowing discrimination across a number of different dimensions, can be engineered in various ways. For example, in the case of gas chromatography (GC), this can be implemented by using multiple, different columns; in ion mobility spectroscopy (IMS), this can be implemented by testing for both positive and negative ions; and in electronic noses, this approach can be exploited by incorporating overlapping, broadly selective sensors. Detection and identification may be augmented by using parameters that include the time course and amplitude of the response. The more flexibility in the dimensionality of detection space, the greater the ability of an analytical device to detect and identify multiple disparate compounds.

(A) Portable configuration of the ScenTraK™ a Handheld Optoelectronic Platform that models the way biological noses work to detect, identify, and discriminate airborne compounds. This production prototype accepts replaceable cartridges, with space available for mounting networking transceivers or other devices near the snout at the left. (B) Exploded, schematic view of the ScenTraK™ sensor chamber and airflow path.

Electronic noses that sniff the environment can be tailored to respond to many different compounds. The identification of appropriate sensors can require extensive empirical searching or synthesis of new materials that can take many months. A new, recently patented system utilizes short oligomers of unique DNA sequences coupled to a fluorescent dye to detect vapor phase molecules. Libraries of billions of DNA oligomers can be subjected to high-throughput screening to identify sensitive, rapidly responding sensors to new targets. The identified sequences can be synthesized rapidly in large volumes using standard molecular techniques and deployed on sensing cartridges put into devices currently being developed for the security arena. Optoelectronic circuits are used to measure fluorescence changes in DNA polymers or other dye-coupled polymers interacting with vapor phase compounds (see figure).

One way to develop devices for odor detection is to mimic multiple attributes of biological olfactory systems, taking advantage of the millions of years of evolutionary optimization. In addition to the cross-reactive sensor arrays that replicate biological odorant receptors, such devices can respond rapidly without prior sample preparation, can implement real-time dynamic modulation of sampling and detection parameters, and can utilize sophisticated analyte detection algorithms. These devices exhibit high sensitivity, discrimination, and detection capability for target analytes, in some cases at sub-parts-per-billion levels.

Both DNA-based sensors and other fluorescent dye-coupled sensor materials may be used to generate spatio-temporal optical response patterns to volatile compounds that are identified through use of algorithms consisting of template matching, neural networks, delay line neural networks, and statistical analysis. Interferents are handled in a variety of ways by judicious choice of sensors unresponsive to ambient vapors, or by subtracting responses identified through use of other sensors relatively specific for defined interferents.

Research supported by the Homeland Security Advanced Research Projects Agency (HSARPA) has demonstrated that optoelectronic noses using these principles can detect a variety of toxic industrial compounds and chemical warfare agents in the presence of a number of interferent compounds. Ongoing tests are examining the ability to detect these compounds at low concentrations that may endanger life or health with an eye to providing such a detection system to the first responder community.

This article was written by John S. Kauer, Ph.D., Chief Scientific Officer and Chairman of the Board; and Barbara R. Talamo, Ph.D., Vice President for Business Development, at CogniScent, Inc. For more information, phone 508-839-7973, or visit .