Recursive Least Squares algorithm dedicated to early recognition of explosive compounds thanks to multi-technology sensors
This paper describes a system for the detection of nitroaromatic explosives consisting of a portable detector based on a specific fluorescent material. The developed sensor was able to perform an ultra trace detection of explosives, such as trinitrotoluene (TNT) or its derivate 2,4-dinitrotoluene (DNT), in ambient air or on objects tainted with explosives. In the presence of nitroaromatic vapors, the fluorescence of the material was found to decrease due to the adsorption of nitroaromatic molecules on its specific adsorption sites. The sensor exhibited a large sensitivity to TNT or DNT at their vapor pressures (respectively 6 and 148 ppbv) and the detection threshold was evaluated on a laboratory test setup and was found to be 0.75 ppbv for TNT. Moreover, the detector demonstrated no loss of performance in the presence of humidity or interfering compounds. All the tests led to the conclusion that the sensor fulfills the main requirements for the identification of suspect luggage, forensic analyses or battlefields clearing.