Automated Characterization of Compounds in Fire Debris Samples

Importance of the Topic

Accurate identification of hydrocarbon residues in fire debris is essential for forensic conclusions about the presence and type of accelerants. Traditional manual interpretation of complex gas chromatograms is time-consuming, subjective and can miss lower-abundance compounds. Automated software solutions improve throughput, reproducibility and objectivity, supporting more reliable fire investigations.

Objectives and Overview of the Study

The study evaluates Sherlock X, an automated analysis package designed to characterize C6–C26 hydrocarbons in fire debris extracts. The goals are to compare Sherlock X performance against established Ignitable Liquid Reference Collection (ILRC) standards, assess reproducibility across multiple fuel types and demonstrate its ability to deconvolute overlapping peaks and categorize compound classes.

Methodology

Samples of known ignitable liquids (charcoal lighter fluid, diesel, kerosene, and weathered gasoline) were prepared following ASTM E1412. Each batch included a calibration standard (NIST 2285) to adjust retention time expectations. Real-world debris extracts were collected by passive headspace onto activated charcoal and processed identically. Data acquisition used full-scan GC-MS, and a ChemStation macro invoked Sherlock X for peak identification, deconvolution and quantification.

Used Instrumentation

• Gas chromatograph: Agilent 7890 with HP5-MS column (30 m×0.25 mm, 0.25 µm)
  
• Mass spectrometer: Agilent 5977A MSD in full scan (36–350 amu)
• Software: Agilent MassHunter B.07.03, GCMS ChemStation F.01.01, Sherlock X macro integration

Main Results and Discussion

Sherlock X accurately reproduced ILRC gasoline profiles, naming all four major compounds in a 25 % weathered sample and detecting 21 peaks above 1 % abundance. Across four replicates of six ILRC fuels, run-to-run deviations were ≤1.23 %, confirming excellent reproducibility. The software successfully distinguished changes in compound distribution between 25 % and 75 % weathered gasoline and differentiated low-sulfur from regular diesel. In real-world samples, overlapping peaks (e.g., cymene and trimethylbenzene) were deconvolved, and common background pyrolysis products (styrene, limonene) were identified, enabling clear accelerant versus background assessment.

Benefits and Practical Applications of the Method

• Reduces analyst workload and subjective bias by automating peak naming (~30 min saved per sample)
• Expands compound coverage beyond manual interpretation to improve sensitivity
• Provides standardized, categorized output aligned with ASTM E1618 grouping (alkanes, aromatics, cycloalkanes)
• Generates annotated chromatograms and summary tables suitable for case reporting

Future Trends and Applications

Further integration of high-resolution MS and machine-learning libraries could enhance compound specificity and handle more complex matrices. Extending automated workflows to non-hydrocarbon targets (e.g., oxygenates, polar residues) and incorporating real-time instruments may broaden forensic applications in environmental monitoring and quality control.

Conclusion

Sherlock X delivers rapid, reproducible and comprehensive characterization of fire debris hydrocarbons. By combining retention time calibration, spectral matching and deconvolution, it offers an unbiased alternative to manual interpretation, improving confidence in accelerant detection and supporting efficient forensic workflows.

References

1. ASTM E1412 – 12, Standard Practice for Separation of Ignitable Liquid Residues by Passive Headspace.
   
2. P. Koussiates, Fire and Arson Investigator, 66, 30–40.
   
3. ASTM E1618 – 14, Standard Test Method for Ignitable Liquid Residues by GC-MS.
   
4. Ignitable Liquids Reference Collection (ILRC), University of Central Florida.
   
5. J. Hendriske et al., Identifying Ignitable Liquids in Fire Debris, Academic Press, 2015, 43–44.