Profiling the volatile organic compound signature of illicit drugs

Understanding the intricate chemical profiles of illicit drugs is crucial for combating their unregulated distribution. While gas sensors have been proposed for their detection, dogs remain the primary tool used for law enforcement agencies. However, concerns about their reliability persist due to limited information on the specific odor they detect. To improve both sensor development and dogs training, robust chemical characterisation is needed. This study aims to characterize the volatile chemical profiles released by illicit drugs seized by the Belgian Police-including amphetamine, cannabis, opium, cocaine, heroin, MDMA, and ketamine-using headspace volatile sampling coupled with gas chromatography and mass spectrometry. In addition to identifying the most abundant compounds for each drug, a partial least square discriminant analysis (PLS-DA) was employed to define their chemical signature. The model achieved a 92% success rate in detecting and differentiating the tested drugs. These findings highlight the potential for field-deployable gas sensors targeting key signature compounds, and may significantly enhance the detection dog's performance.