Detection and classification of ignitable liquid residues using a fluorescence-based vapor-sensitive microsphere array

This paper describes the application of microsphere vapor sensing arrays to the detection of ignitable liquid (IL) vapors as both pure vapors and as residues (ILRs) on simulated fire debris samples. The temporal fluorescence response profile of the microsphere array generated a reproducible pattern unique to each analyte that could be used to classify subsequent sensor responses. This system, together with a support vector machine pattern recognition algorithm, was used to address several different IL and ILR classification scenarios. High classification accuracy (98%) was maintained over more than 200 vapor responses and the array was able to identify ILs when presented to the pattern classification algorithm within a dataset containing 11 other volatile compounds. Both burned and unburned IL treated samples were classified correctly greater than 97% of the time. These results indicate that microsphere vapor sensing arrays may be useful for the rapid identification of ILs and ILRs.