The complex chemical composition of crude oils presents many challenges for rapid chemical characterization in the case of a spill. A number of approaches are currently used to “fingerprint” petroleum‐derived samples. Gas chromatography coupled with mass spectrometry (GC‐MS) is the most common, albeit not very rapid, technique; however, with GC‐MS alone, it is difficult to resolve the complex substances in crude oils. This study examined the potential application of ion mobility spectrometry–mass spectrometry (IMS‐MS), coupled with chem‐informatic analyses, as an alternative high‐throughput method for the chemical characterization of crude oils. We analyzed 19 crude oil samples from on‐ and off‐shore locations in the Gulf of Mexico region in the United States using both GC‐MS (biomarkers, gasoline range hydrocarbons, and n‐alkanes) and IMS‐MS (untargeted analysis). Hierarchical clustering, principal component analysis, and nearest‐neighbor‐based classification were used to examine sample similarity and geographical groupings. We found that direct injection IMS‐MS performed either equal or better than GC‐MS in the classification of the origins of crude oils. In addition, IMS‐MS greatly increased the sample analysis throughput (minutes versus hours per sample). Finally, a tabletop science‐to‐practice exercise, utilizing both the GC‐MS and IMS‐MS data, was conducted with emergency response experts from regulatory agencies and the oil industry. This activity showed that the stakeholders found the IMS‐MS data to be highly informative for rapid chemical fingerprinting of complex substances in general, and specifically advantageous for accurate and confident source‐grouping of crude oils. Collectively, this study shows the utility of IMS‐MS as a technique for rapid fingerprinting of complex samples and demonstrates its advantages over traditional GC‐MS based analyses when used for decision‐making in emergency situations.

Biomarker data for crude oils from the Gulf of Mexico and adjacent onshore regions. The biomarker data used in this study was originally reported in Kennicutt et al 1991. Approximately 100 mg of crude oil was spiked with a surrogate mixture (5β-cholane, d₁₀-phenanthrene, ad d₁₂-chrysene), and the mixture was then fractionated into its saturated and aromatic fractions using HPLC. GC-based petroleum biomarker analyses were carried out using the Hewlett-Packard 5890 GC coupled with Hewlett-Packard 5790A MSD.