Reference
Auto-ignition Temperatures - Fire without Spark
By Mark C. Hughes, P.Eng.
© 2005 Sintra Engineering Inc.
The flashpoint and auto-ignition temperatures for flammable liquids have often been an area of limited understanding and research. I recently came across some really valuable information regarding the auto-ignition temperatures for most fluids present in automobiles and how auto-ignition works, so I thought I would share it with you.
In April of this year, I attended the World Congress of the Society of Automotive Engineers in Detroit and there were an extensive number of presentations related to vehicle fires. The research was conducted as part of a settlement between the National Highway Traffic Safety Administration (NHTSA) and General Motors (GM) related to the saddle fuel tanks in GM trucks. This settlement provided about $3 million (US) for research into crash induced fires. While most fire investigators know that crash related fires are relatively rare, the funding facilitated research into a number of areas that had not been very thoroughly explored in the past.
During this research, they discovered that very little was known about the fire related behaviour of many of the fluids present in vehicles. Most of these fluids had been extensively documented for mechanical properties but rarely for combustion properties. Moreover, while the flashpoint for these fluids was known, fire investigators are usually more interested in the auto-ignition temperature. To correct this situation, extensive testing was conducted on engine fluids to simulate auto-ignition temperature. Furthermore, the testing simulated several conditions in a vehicle environment to see how that would affect the auto-ignition temperature.
Auto-ignition temperature is, in theory, the temperature at which a fluid will ignite without a spark. There is an American Society for Testing and Materials (ASTM) test that lays out the method for determining the auto-ignition temperature. It is the results from this test that have been published in various books and standards as the auto-ignition temperature for various substances.
The problem with these tests is that while they are intended to be reproducible, they do not represent real world circumstances. The test requires that an insulated glass bulb container be heated up slowly to the test temperature with four rapid response thermocouples located within the container. The test is performed in a darkened room and a small quantity of the test fluid is added to the container. If a small flash is observed or a change is noted on the thermocouple outputs, then that temperature is assessed to be the auto-ignition temperature. The test is repeatable, meaning it reliably gives the same results for multiple tests, but unfortunately, this small flash does not really represent the temperature required to actually create a sustained ignition of a substance.
How does auto-ignition actually work? What causes the ignition without a spark? To cause the auto-ignition, the current theory is that the energy from breaking the molecular bonds (usually hydrogen/carbon or carbon/carbon for hydrocarbons) is sufficient to ignite the vapour.
What was readily apparent from the testing was that there were a number of parameters that affected the ability to auto-ignite. These parameters included air flow over the area, geometry of the surface and the method of adding the fluid. What they did discover was that the ASTM test for auto-ignition temperature does seem to provide a reliable minimum temperature for the auto-ignition of a substance, but that in the real world this temperature would be a little higher.
The results presented at the SAE congress were, unfortunately, not published as part of the congress; however, a recent article in Fire Technology (April 2005) published by the National Fire Prevention Association (NFPA) details the results from some of the testing. It is a useful reference for any fire investigators that are investigating vehicle fires and I highly recommend its addition to any investigatorís library.
Mark Hughes, P. Eng. is a principal with Sintra Engineering, a Forensic Engineering firm.
return to Reference articles...
