Reference
Sampling for Dummies Part I
By Mark Hughes, P. Eng.
© 1999 Sintra Engineering Inc.
One of the black arts of fire investigation has always been sampling for accelerants. Often fire investigators will rely on the use of a dog trained to detect accelerants rather that take on the risk associated with the selection of locations to sample for accelerants.
Accelerants are commonly used in incendiary (set) fires to help speed the development of the fire. As such there is often a significant amount of pressure on investigators to confirm the presence of accelerants and to determine the type of accelerant. To sample correctly, samples of the debris must be taken, either based upon the preliminary assessment of an accelerant detection dog or based upon the fire investigator's training.
To get a better understanding of how to sample accelerants, it is necessary to understand what they are. Most types of accelerants are based upon some type of hydrocarbon chain. Unfortunately few accelerants are made up of a single molecule but are generally mixtures made up of a number of different compounds. To further complicate the issue, fire will weather the mixtures and alter their composition.
Typically accelerants are quite volatile (easily convert to a gas form) which is one of the features that make them especially flammable or combustible (as fire is the combustion of the gases of a compound). This volatility makes them attractive to someone trying to set a fire but also makes narrowing down where to sample difficult as the remaining fuel will tend to evaporate. If a long enough period passes after the fire, any chance of finding the fuel may be lost.
Other factors influencing the effectiveness of the samples are the extent to which the fire developed, what extinguished the fire, and whether the accelerants were consumed in the fire. Therefore, the trace remains of any accelerants are often particularly miniscule and difficult to detect.
Accelerant dogs are good at detecting certain types of accelerants but are not capable of detecting all types. If the type of accelerant used was one that the dog is not trained to detect, the dog will miss locations that should be sampled.
Complications aside, once a particular location has been selected, what should happen next? Proper analysis fire debris requires that samples be sampled and tested in a particular way to achieve the sensitivity required to confirm the presence of an accelerant.
Sample containers should be one of three types in descending order of preference: nylon bags, glass mason jars (sterilized) or metal containers. All the containers should be properly labeled and sealed. The sample containers should then be sent to a laboratory for analysis using a chain of custody form to ensure all the samples are maintained and not lost.
When the samples arrive in the lab, the containers are heated and a small portion of the vapours in the head space are collected and analyzed by gas chromatography/mass spectroscopy (GC/MS). This process produces a profile of the constituents of the vapour. A trained chemist in fire debris sampling will then compare the profiles to weathered known fire debris samples. This allows the chemist to verify if there are any traces of accelerants but the analysis will be limited to the types of weathered samples to which he/she can compare the sample to. Once a possible hit has been found, a further refinement of the sampling can be performed using a second iteration of the mass spectroscopy (GC/MS/MS).
There are a number of steps which can result in the loss of a good sample. You could have an old sample; a weak sample (poor readings overall); the accelerant could have been completely consumed; the volatile compound could be difficult to detect; the sample container could be improperly sealed; the sample container could be contaminated; the lab could attempt to sample the debris instead of the headspace; the lab could use an inappropriate type of test or the lab could use unweathered standard references.
The public sector has some access to specialized labs that can perform this work (e.g. the RCMP forensic labs), however, in the private sector, sample testing is particularly limited to a few specialized laboratories in Eastern Canada or the United States
Large tomes (a big book) can and have been written about sampling and while this article is by no means exhaustive it should give you a basic understanding of the factors at play when trying to sample a fire scene.
If you have any questions regarding sampling, please feel free to call, I may be able to point you in the right direction.
return to Reference articles...
