|
Performance Abstract on the EST-4200 Ultimate Vapor Tracer |
|
Ultra-high speed gas chromatography is a powerful analytical method for analysis of odors, fragrances, and chemical vapors produced by explosives, chemical and biological weapons, contraband, and hazardous industrial materials. A portable chemical profiling system incorporating an ultra-high speed chromatography column, a solid-state sensor, a programmable gate array microprocessor, and an integrated vapor preconcentrator is described.
Using ultra-high speed chromatography, chemical vapors within containers can be speciated and their concentration measured in less than 10 seconds with picogram sensitivity using a SAW (EST-4200) sensor with electronically variable sensitivity. Odor concentration and intensity are measured directly with an integrating GC sensor. The solid-state sensor produces high resolution 2-dimensional olfactory images unique to many complex odors. Examples involving odors from explosives, contraband drugs of abuse, hazardous chemicals, and even biological life forms are presented.
An important requirement for a chemical profiling system is that it must recognize odors and fragrances based upon their full chemical signature. Unlike a trace detector, it must see everything and miss nothing. A library of retention time indices for chemicals allows for the creation of hundreds of specific virtual chemical sensors. Virtual chemical sensors combined with odor profiling can be an effective method for recognizing the presence of hazardous materials. Chemical libraries and electronic odor profiles allows users to quickly distribute signatures of hazardous materials or new threat vapors of any kind.
Chemical sensor arrays have interested developers of neural networks and artificial intelligence algorithms for some time, yet physical sensors have limited performance because of overlapping responses and physical instability. Using ultra-high speed gas chromatography, arrays of virtual chemical sensors with non-overlapping response are possible. Long term stability coupled with picogram sensitivity may lead to the possibility of applying artificial intelligence and neural networks to detect and recognize an unlimited number of threat vapors.
The U.S. now inspects 4 percent of the 6 million shipments that arrive at more than 100 ports, twice the 2 percent before the Sept. 11 attacks in 2001. About 20 percent of that cargo passes through overseas ports such as Hong Kong, where U.S. inspectors are being stationed. Cargo worth $1.2 trillion, or half of U.S. imports, arrives by sea. The rest comes from Canada and Mexico. There is a clear and present danger yet the problem is daunting.
Figure 1- Nearly 7 million cargo containers come into U.S. ports from overseas every year. Officials say it would be impossible to examine every one of them.
Current sensor capabilities are fairly limited; in many cases, the best “technology” for practical use continues to be trained dogs. Manufactured sensors are often designed for use in specific environments and to be selective for only one or two chemicals. Yet because there is a spectrum of possible threats, sensor systems are needed that can detect a large number of possible chemicals. In addition, sensor systems need a number of different subsystems, including sample collection and processing, presentation of the chemicals to the sensor, and sensor arrays with molecular recognition.
In this paper, an electronic nose using a single solid-state sensor is able to create an unlimited number of specific virtual chemical sensors for chemically profiling odors in cargo containers. Virtual sensor arrays and recognizable olfactory images for explosives, hazardous substances, drugs of abuse, and even the cargo itself provides a cost effective screening tool for shippers and inspectors alike. In support of container security protocols, odor profiles can also be attached to an electronic manifest file and forwarded to authorities at the country of destination for comparison purposes. Next Page >>>
<<< Back to Ultimate Vapor Tracer Main Page
|
