U.S. flag

An official website of the United States government, Department of Justice.

Development of Tools for Smoke Residue and Deposition Analysis

NCJ Number
Date Published
January 2010
256 pages
In developing methods and a scientific basis for the use of smoke deposition analysis as a forensic tool, this study examined the physics of smoke deposition from a hot layer to a wall, using a hood apparatus specifically designed to study the smoke deposition based on thermophoresis experimentally and analytically; for the first time, the optical density method was used to measure the amount of smoke deposited on the surface.
Although smoke deposition pattern analysis is included in standard fire-scene pattern analysis, the chemistry and physics of smoke deposition is poorly understood. This limits the interpretive value of smoke pattern analysis. The current work provides scientifically based tools for use by fire investigators and forensic chemists for general fire investigations and arson investigations in particular. By using both gravimetric and optical measurement methods, correlations between optical density and smoke deposition were determined for several fuels and test conditions. Solid phase mass specific extinction coefficient values were introduced for the first time for various fuels. Also, the optical properties of the smoke deposited on the surface were determined and compared to the smoke properties in the gas phase. An analytical thermophoretic smoke deposition model was developed using the measured smoke properties. This model is validated by using experimental results from this work. This model is suitable for predicting smoke deposition due to a fire. The effect of fire size and change in the flow regime due to the fire size was also studied in this research. Results from the smoke pattern predictions for the wall tests produced good agreement between the digital images from the smoke deposition on the walls and the processed data. 164 figures, 99 tables, 31 references, and appended supplementary data

Date Published: January 1, 2010