Printer Friendly

Evaluation of fire scene contamination by using positive-pressure ventilation fans.


In most laboratories measures are taken to prevent contamination and to verify the absence of contamination. This paper addresses the contamination of a fire scene before samples are selected. Specifically, this paper determines if using a gasoline-powered positive-pressure ventilation fan could contaminate a fire scene.

The purpose of a positive-pressure ventilation fan is to help reduce smoke, gases, and heat in a burning structure. This is accomplished by placing the fan outside an opening, usually a doorway, of the structure so that it forms a cone of air that effectively seals the opening. This causes a rapid increase in the pressure in the structure. A door or window on the other side of the structure is opened to allow the escape of smoke and gases.

Most of the articles in the literature consider positive-pressure ventilation fans from the perspective of the fire fighter. This includes safety, fire containment, exposure to hazardous materials, and search-and-rescue efforts (Bolstad-Johnson et al. 2000; Gustin 1997; Gustin 1998; Pressler 1997). Only one article was found that considered the possibility of scene contamination by the exhaust of a positive-pressure ventilation fan from a forensic perspective (Lang and Dixon 2000).

Materials and Methods

Sample Preparation

* 100 percent cotton terry towels (D.C. May Corporation, Durham, North Carolina) cut in 25-centimeter squares. This is a lightweight terrycloth towel.

* Activated carbon strip (Albrayco Laboratories, Incorporated, Cromwell, Connecticut) cut to 21 millimeters by 5 millimeters

Samples were tacked to a wood strip (121 centimeters long) to keep them from being blown by the fan.

Laboratory Instruments/Settings

* Gas chromatograph: ThermoQuest Trace Gas Chromatograph (Thermo Finnigan, Austin, Texas) with A200S autosampler

* Column: Rtx-5MS (Restek, Bellefonte, Pennsylvania) 15M x 0.25 mm ID x 0.25um film thickness

* Carrier gas: helium, 0.9 ml/minute

* Temperature program: 50[degrees]C for 1 minute, ramp at 20[degrees]C per minute to 300[degrees]C, hold for 2 minutes

* Detector: Finnigan Trace Mass Spectrometer (Thermo Finnigan, Austin, Texas)

* Scan range: Full Scan 40-400m/z

* Source temperature: 200[degrees]C

* Gas chromatograph interface temperature: 300[degrees]C

* Emission current: 350 [micro]V

* Software: Xaminer 1.0, Xcaliber 1.0

On-scene Instrument and Structure

* RamFan Turbo Ventilator (RAMFAN Corporation, Spring Valley, California) powered by a Honda GX160 5.5 horsepower gasoline engine

The residential structure is diagrammed in Figure 1. The sample locations are indicated by letters and correspond to Table 1 as follows:

* O = outside samples, 2 meters from positive-pressure ventilation fan and in the path of the engine exhaust of the fan

* L = living room samples, 2 meters from front door

* D = doorway samples, doorway from living room to kitchen

* K = kitchen samples, 1.5 meters from rear door leading to small back porch



The positive-pressure ventilation fan was set up according to the standard-operating procedures of the City of Marianna Fire Department, Marianna, Florida, with the fan output blowing air into the doorway. The fan's engine exhaust was vented to the left.

Samples for Trial 1 were positioned and allowed to equilibrate for 20 minutes. The first sample was taken from each of the sample locations as a background. The positive-pressure ventilation fan was then operated for 30 minutes, a common time span used at actual fire scenes. Samples were collected from each sample location at time = 0 (immediately after the fan was shut off), one hour, and two hours. Because investigators do not typically take samples from a scene until at least four hours after the fans have been shut off (Gunn 2002), this scenario should have allowed every opportunity for any contamination that was going to occur to still be present.

Trial 2 was similar to Trial 1 with a deviation. In order to simulate a spill a fire fighter might make while refueling the positive-pressure ventilation fan at a fire scene, 250 milliliters of gasoline was spilled on the side of the gas tank of the engine of the positive-pressure ventilation fan and on the ground. The fan was then operated for 30 minutes, and the rest of the experiment was repeated.

An exhaust sample was taken by holding an activated carbon strip approximately 30 centimeters from the exhaust of the gasoline engine of the positive-pressure ventilation fan for two minutes. Additional exhaust samples were taken by holding an activated carbon strip approximately 30 centimeters from the exhaust of two vehicles, a 3-cylinder 1994 Geo Metro and a 6-cylinder 1991 Chevrolet S-10 truck.

The towel samples were collected and placed in airtight one-quart cans commonly used for the collection of fire debris samples. The activated carbon strips were collected directly into two milliliter autosampler vials and sealed.

The towel samples were prepared by passive headspace extraction (ASTM 2000). Carbon disulfide (0.5 milliliters) was added to the activated carbon strip samples. All samples were analyzed by gas chromatography-mass spectral detection according to ASTM guidelines (ASTM 1994).


Table 1 shows the results. All towel and activated carbon strip samples from Trial 1 were determined to be negative for the presence of gasoline. All the towel samples from Trial 2 were determined to be negative for the presence of gasoline.

A trace quantity of some of the early components of gasoline was observed on the activated carbon strip samples taken from Trial 2, the trial in which gasoline was spilled before operating the fan and collecting the samples. This was not apparent from the total ion chromatogram (Figure 2) but was visible in the extracted ion chromatogram (Figure 3). This trace amount consisted of some early alkanes and low levels of toluene and xylenes. This is attributed to gasoline because of the known origin of the samples. In actual casework this would not be sufficient for a determination of gasoline or any other ignitable liquid and would be reported as negative.



The activated carbon strips were used to maximize the detection of contamination at fire scenes. Samples taken from fire scenes might contain matrices that strongly adsorb the organic compounds present in most ignitable liquids. However, it is unlikely that these materials would provide better adsorption than an activated carbon strip that is designed to adsorb organic compounds.

The trace levels of contamination found on the activated carbon strips compare favorably to the exhaust collected from the positive-pressure ventilation fan (Figure 4: A and B). The trace levels of contamination do not compare favorably to an extracted ion chromatogram of an actual case sample determined to be positive for the presence of gasoline (Figure 4: C).


Actual case samples tend to be depleted of the early, more volatile components of gasoline and possess a greater concentration of the later components of gasoline. Actual case samples compare more favorably to gasoline that has been weathered to 50 percent of its original volume, thus eliminating most of the early components (Figures 4: C and D). This is quite different than the exhaust and trace levels of contamination seen in this experiment.

The trace levels of contamination found on the activated carbon strips also compare favorably to the exhaust collected from two vehicles (Figure 5: A, B, and C). None of these compared favorably to gasoline that has been weathered to 50 percent of its original volume (Figure 5: A, B, C, and D). This would indicate that vehicles or other power equipment left running near the scene would not have a significant impact on the detection of gasoline at a fire scene.

The results from this experiment demonstrated that exhaust from a gasoline engine-powered positive-pressure ventilation fan is not sufficient to contaminate samples taken from a fire scene. Only when gasoline was spilled, providing a greater concentration of gasoline vapors in the air to be pulled through the fan, was any trace-level contamination observed. It is unlikely that one would be able to verify whether such contamination had taken place in actual casework. Further, the contamination that might occur from gasoline-powered engines of different types is not similar to what would be expected from gasoline recovered from an accelerated fire.

This work was designed to continue and complement that begun by Lang and Dixon in 2000. Whereas no fire scenes are identical and exact replication of any actual scene is impossible, by adding to the variety of structures and conditions tested, it is hoped that reasonable assumptions may be drawn regarding the potential contamination issues of positive-pressure ventilation fans.

The author acknowledges the assistance of the City of Marianna Fire Department, Marianna, Florida.


ASTM. Standard Guide for Ignitable Liquid Residues in Extracts from Fire Debris Samples by Gas Chromatography-Mass Spectrometry, ASTM-E1618-94, 1994.

ASTM. Standard Practice for Separation of Ignitable Liquid Residues from Fire Debris Samples by Passive Headspace Concentration with Activated Charcoal, ASTM-E412-00, 2000.

Bolstad-Johnson, D. M., Burgess, J. L., Crutchfield, C. D., Storment, S., Gerkin, R., and Wilson, J. R. Characterization of firefighter exposures during fire overhaul, Journal of the American Industrial Hygiene Association (2000) 61:636-641.

Gunn, J. Florida State Fire Marshal, Personal communication, April 2002.

Gustin, B. Fog streams and PPV: Their effects on two fires, Fire Engineering (1997) 150 (11):49-54.

Gustin, B. Search and rescue in single-family dwellings: Part 1, Fire Engineering (1998) 151(8):73-82.

Lang, T. and Dixon, B. M. The possible contamination of fire scenes by the use of positive pressure ventilation fans, Journal of the Canadian Society of Forensic Science (2000) 33 (2):55-60.

Pressler, B. More two-minute drills, Fire Engineering (1997) 150(5):26-29.

Perry Michael Koussiafes

Crime Laboratory Analyst

Fire and Arson Laboratory

State of Florida Fire Marshal

Havana, Florida
COPYRIGHT 2002 Federal Bureau of Investigation at
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2002 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Koussiafes, Perry Michael
Publication:Forensic Science Communications
Geographic Code:1USA
Date:Oct 1, 2002
Previous Article:Further discussion of the consistent treatment of length variants in the human mitochondrial DNA control region.
Next Article:Validating IND (1,2--indanedione).

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |