Agricultural and Biosystems Engineering Publications

Campus Units

Agricultural and Biosystems Engineering, Civil, Construction and Environmental Engineering, Food Science and Human Nutrition, Transportation

Document Type


Publication Version

Accepted Manuscript

Publication Date


Journal or Book Title

Forensic Science International



First Page


Last Page


Research Focus Area(s)

Biological and Process Engineering and Technology




This report highlights the importance of an individual chemical's odor impact in the olfactory identification of marijuana, cocaine, and heroin. There are small amounts of highly odorous compounds present in headspace of these drugs, with very low odor detection thresholds, that are more likely responsible for contributing to the overall odor of these drugs. Previous reports of the most abundant compounds in headspace can mislead researchers when dealing with whole odor of these drugs. Surrogate scent formulations, therefore, must match the odor impact of key compounds and not just the chemical abundance of compounds. The objective of this study was to compare odorous volatile organic compounds (VOCs) emitted from illicit drug samples of marijuana, cocaine, and heroin to surrogate smell formulations using simultaneous sensory (via human olfaction) and chemical analyses. Use of solid phase microextraction (SPME) allowed VOCs in drug headspace to be extracted and pre-concentrated on site, and analyzed by multidimensional gas chromatography–mass spectrometry–olfactometry (MDGC–MS-O). Use of MDGC–MS-O allowed for further separation of odorous compounds and simultaneous detection by the human nose of the separate odor parts that make up the total aroma of these drugs. The compounds most abundant in headspace were not the most odor impactful when ranked by odor activity values (OAVs) (defined as ratio of concentration to odor detection threshold, ODT). There were no apparent correlations between concentrations and OAVs. A 1 g marijuana surrogate lacked in odor active acids, aldehydes, ethers, hydrocarbons, N-containing, and S-containing VOCs and was overabundant in odor active alcohols and aromatics compared with real marijuana. A 1 g cocaine surrogate was overabundant in odor active alcohols, aldehydes, aromatics, esters, ethers, halogenates, hydrocarbons, ketones and N-containing compounds compared with real. A 1 g heroin surrogate should contain less odor active acids, alcohols, aromatics, esters, ketones, and N-containing compounds. Drug quantity, age and adulterants can affect VOC emissions and their odor impact. The concept of odor activity value, then, is useful to researchers without access to more sophisticated instrumentation. Odor activity values can be calculated from published odor detection thresholds. More research is warranted to expand the database, and determine odor detection thresholds for compounds of interest. Additional information could be obtained from establishing ODTs of key odorants for canines.


This is a manuscript of an article published as Rice, Somchai, and Jacek A. Koziel. "The relationship between chemical concentration and odor activity value explains the inconsistency in making a comprehensive surrogate scent training tool representative of illicit drugs." Forensic Science International 257 (2015): 257-270. DOI: 10.1016/j.forsciint.2015.08.027. Posted with permission.

Copyright Owner

Elsevier Ireland Ltd.



File Format


Published Version