ANALYSIS OF FIRE DEBRIS SAMPLES BY GAS CHROMATOGRAPHYMASS SPECTROMETRY AND CHEMOMETRICS

Importance of the topic

In modern forensic investigations, identifying accelerants in fire debris is essential for reconstructing arson events. The growing use of biodiesel blends complicates analysis, as these biofuels contain long-chain methyl esters alongside traditional hydrocarbons. Chemometric tools can enhance detection of subtle differences that are otherwise indistinguishable by eye.

Objectives and overview of the study

This study aimed to evaluate the ability of gas chromatography-mass spectrometry combined with chemometric processing to distinguish diesel from biodiesel in simulated fire debris samples. Samples included liquid diesel, liquid biodiesel, burnt cloth spiked with one of the fuels, and blank controls. The goal was to assess whether statistical analysis could reveal characteristic markers of each accelerant.

Methodology and instrumentation

The investigation employed both liquid injection and solid-phase microextraction (SPME) headspace sampling, analyzed on an Agilent 7890B GC with 5975B MSD. A Gerstel MPS XT autosampler equipped for SPME ensured reproducible sampling. Data deconvolution and compound detection were performed in Mass Hunter Unknowns Analysis, and results were exported in CEF format for chemometric interpretation using Agilent Mass Profiler Professional (MPP).

Key results and discussion

• Volcano plot analysis identified several fatty acid methyl esters (FAMEs) such as methyl palmitate, methyl stearate, and methyl margarate as statistically different between diesel and biodiesel.
• Principal component analysis achieved clear clustering of diesel and biodiesel replicates, confirming distinct chemical profiles.
• SPME-GC-MS detected target FAME markers in both liquid and fabric samples, albeit at lower intensity in headspace sampling. Signal-to-noise ratios remained acceptable for forensic purposes.

Benefits and practical application of the method

The combined GC-MS and chemometric workflow provides a robust approach for forensic laboratories to differentiate biofuel accelerants. It enhances selectivity and sensitivity, reduces subjective interpretation, and supports casework with statistical confidence.

Future trends and application possibilities

• Extension to complex fuel blends and emerging biofuels to broaden forensic databases.
• Integration of machine learning algorithms for automated pattern recognition.
• Miniaturized or portable GC-MS platforms coupled with real-time chemometric analysis for in-field screening.

Conclusion

This study demonstrates that chemometric analysis of GC-MS data enables reliable discrimination between diesel and biodiesel in fire debris. Key FAME markers and multivariate statistics together offer a powerful forensic tool to detect biofuel accelerants.

Reference

Jamie Minaeian, Eleanor Miller, Phine Banks, Camilla Liscio and Karen Robertson. Analysis of fire debris samples by gas chromatography-mass spectrometry and chemometrics. Anatune Ltd and SPA Forensic Services, 2023.