Simulation of Fire Debris for the Training of Chemometric Models for the Identification of Ignitable Liquids

• Author / Creator

  Lee, Xiao Qin

• Arson is one of the most challenging crimes for forensic scientists to investigate. The variability in the composition of ignitable liquids, including changes in chemical composition during and after the fire, and the presence of pyrolysis products generated from burning substrates yields a very complex mixture of volatile compounds in samples of fire debris. Headspace extraction of debris samples followed by gas chromatography-mass spectrometry (GC-MS) is the most common approach for fire investigation. For many laboratories, data interpretation is the bottleneck in the workflow, consuming an inordinate amount of analyst time. It is also a process that is highly dependent on the experience and skill of analysts which gives rise to subjective results.
  Chemometrics offers an alternative to manual data interpretation. However, for this work to be applicable in real-world fire investigations, the chemometric model must be able to classify all major classes of ignitable liquids that can be possibly found in a fire. Construction of a chemometric model requires abundant casework data. This is this not a problem for gasoline, which is the most commonly used ignitable liquid, but it is a challenge for other ILs. The lengthy time needed for the collection of casework debris containing other ILs for the model construction limits the practical use of this work. Therefore, it would be a great benefit if models applicable to casework samples could be generated based on simulated debris profiles.
  An established debris simulation protocol has been shown to be effective in generating realistic debris for training human analysts. This thesis evaluates the applicability of this simulation protocol for generating debris that are chemometrically identical to casework debris. It was discovered that models trained on the simulated debris were not applicable to casework samples without a significant loss in the accuracy of the model. It was established that the reason for the inadequacy of the simulated debris was that it did not contain sufficient C2-alkyl benzenes and non-aromatic hydrocarbons. Consequently these features which are not characteristic of gasoline were selected by the chemometric model and model quality degraded for real samples.
  Thus research turned to a study of the effects of temperature on the pyrolysis of household materials, mainly flooring and roofing materials, at temperatures above 400 °C. I was particularly interested in finding conditions that will generate additional BTEX and aliphatic hydrocarbons, which were generally lacking in debris pyrolyzed at 400 °C with the established simulation method.

• Subjects / Keywords

  • Chemometrics
  • Gas chromatography-mass spectrometry
  • Forensic
  • Simulated Fire Debris

• Graduation date

  Spring 2015

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/R3V11VT89

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.