ArticlePDF Available

The Propensity of Lit Cigarettes to Ignite Gasoline Vapors

Authors:

Abstract and Figures

Fire investigators regularly evaluate available fuels and potential ignition sources to determine the cause of a fire. This work examined the propensity of lit cigarettes to ignite gasoline vapors, expanding on previous work to include a large number of trials and a wide range of test conditions. Experiments were conducted exposing lit cigarettes, both at idle and under draw, to gasoline vapors in various configurations including pools/pans of gasoline, gasoline on textile substrates (clothing), and sprays of gasoline. Five major brands of commercially-manufactured tobacco cigarettes were tested. The experiments conducted for this study consisted of 70 distinct tests involving a total of 723 cigarettes and over 4,500 instances of exposure of a lit cigarette to ignitable concentrations of gasoline vapor in air. There were no instances of the ignition of gasoline vapors from the exposure of those vapors to a lit tobacco cigarette during any of the experiments.
Content may be subject to copyright.
The Propensity of Lit Cigarettes to Ignite
Gasoline Vapors
Howard A. Marcus and Justin A. Geiman*, Bureau of Alcohol, Tobacco,
Firearms, and Explosives (ATF), U.S. Department of Justice,
6000 Ammendale Rd., Beltsville, MD 20705, USA
Received: 8 August 2013/Accepted: 7 December 2013
Abstract. Fire investigators regularly evaluate available fuels and potential ignition
sources to determine the cause of a fire. This work examined the propensity of lit cig-
arettes to ignite gasoline vapors, expanding on previous work to include a large num-
ber of trials and a wide range of test conditions. Experiments were conducted
exposing lit cigarettes, both at idle and under draw, to gasoline vapors in various
configurations including pools/pans of gasoline, gasoline on textile substrates (cloth-
ing), and sprays of gasoline. Five major brands of commercially-manufactured
tobacco cigarettes were tested. The experiments conducted for this study consisted of
70 distinct tests involving a total of 723 cigarettes and over 4,500 instances of expo-
sure of a lit cigarette to ignitable concentrations of gasoline vapor in air. There were
no instances of the ignition of gasoline vapors from the exposure of those vapors to a
lit tobacco cigarette during any of the experiments.
Keywords:Fire investigation,Ignition,Cigarettes,Gasoline
1. Introduction
Fire cause determination involves determining how a fuel, an ignition source and
an oxidizer combine to cause a fire [1]. NFPA 921 cautions investigators not to
prematurely conclude the cause of a fire based solely on the presence of a readily
available fuel and a potential ignition source, but rather to understand the igni-
tion sequence and determine whether a competent ignition source exists for the
first material ignited (see §18.4.2 and §18.4.4 of NFPA 921 [1]).
Lit cigarettes are a potential ignition source that may need to be considered by
fire investigators at many scenes. There were an estimated 90,800 reported smok-
ing-material fires in the US in 2010 [2]. An additional 155,000 home cigarette fires
go unreported each year, according to a study by the US Consumer Product
Safety Commission [3]. Despite the prevalence of smoking material fires overall,
cigarettes are generally considered to be poor ignition sources for ignitable liquids.
This study examines whether a lit tobacco cigarette, commercially manufactured
in the US, is a competent ignition source for gasoline vapors.
Gasoline is one of the most prevalent flammable liquids [4]. It has also been
referred to as the ‘‘most dangerous substance routinely handled by untrained
* Correspondence should be addressed to: Justin A. Geiman, E-mail: justin.geiman@atf.gov
Fire Technology, 50, 1391–1412, 2014
2013 Springer Science+Business Media New York (Outside the USA). Manufactured in The United States
DOI: 10.1007/s10694-013-0380-3
12
individuals’’ [5]. Gasoline is an engineered mixture of volatile, low boiling point,
midrange hydrocarbons blended with a variety of additives that readily produces
flammable vapors at ambient temperatures [6]. Recent US fire statistics show that
there are an estimated 2,400 residential structure fires annually involving gasoline,
with approximately half of these fires categorized as intentional [7].
The objective of this study was to expand the previous work on the propensity
of lit cigarettes to ignite gasoline vapors to a wide range of test conditions in con-
figurations of practical interest to fire investigators. Experiments were conducted
exposing lit tobacco cigarettes, both at idle and under draw, to gasoline vapors in
various configurations. Tested configurations included gasoline in a pan, on a hor-
izontal surface, on a variety of textile substrates (clothing), and as a mist.
1.1. Literature Review
Cigarette ignition of ignitable liquids and gases has been studied previously in the
literature [814]. Babrauskas [5] and Holleyhead [12] provide comprehensive
reviews of the subject. The consensus in the literature is that cigarettes make poor
ignition sources for most ignitable liquids and gases, but that it is possible to
ignite a limited set of ignitable liquids and gases with a cigarette. The eight sub-
stances that have been ignited by lit cigarettes in laboratory experiments are
shown in Table 1.
In contrast, 45 different ignitable liquids and gases have been shown not to
ignite when exposed to cigarettes in laboratory experiments (see Geiman and Fuss
[15] for a detailed list). Many common ignitable liquids and gases do not ignite in
the presence of lit cigarettes including fuels such as butane, propane, heptane,
methane; alcohols such as methanol, ethanol, isopropanol; and solvents such as
acetone, methyl ethyl ketone, and xylene.
Prior to examining the previous experimental studies involving ignition of gaso-
line with cigarettes, it is instructive to examine the properties of both cigarettes
and gasoline. Commercial cigarettes are typically 60 mm to 100 mm long, with a
diameter of 7.8 mm to 8.1 mm, and a weight of 0.6 g to 1.1 g [5]. Holleyhead [12]
provided a detailed explanation of the cigarette combustion process, including the
Table 1
Ignitable Liquids and Gases Ignited by Cigarettes
Fuels References
Acetylene [9]
Carbon disulphide [8,9]
Ethylene oxide [9]
Diethyl ether [9,11]
Hydrogen [9,10]
Hydrogen sulphide [9]
Phosphine [9]
Toluene
a
[8]
a
a
One test only, no ignition in subsequent tests
1392 Fire Technology 2014
temperature, velocity, and gas species concentrations (oxygen, carbon monoxide,
and carbon dioxide) produced. Table 2summarizes the gas- and solid-phase tem-
peratures associated with lit cigarettes.
Gasoline has a flash point of -38Cto-43C and an auto-ignition tempera-
ture (AIT) of 280C to 456C[16]. A range of values for flashpoint and AIT are
listed because they vary depending on the grade (octane-rating) of the gasoline.
Modern automotive gasoline typically has an AIT of 440C to 450C[5]. Based
solely on the temperatures listed in Table 2and the combustion properties of gas-
oline, fire investigators often assume that ignition of gasoline by a cigarette is not
only possible, but likely. However, evaluating the ignition process for this scenario
is more complicated than simply comparing these temperature measurements.
Research into the ignition of gasoline on a hot surface, a problem with similari-
ties to cigarette ignition of gasoline, has shown that, in general, hot surface tem-
peratures need to significantly exceed the AIT for ignition to occur. Hot surface
ignition of ignitable liquids is probabilistic in nature, and occurs over a range of
temperatures over which ignition becomes more likely [1719]. Colwell and Reza
[17] developed a logistical regression curve based on series of hot surface ignition
experiments with unleaded 87 octane gasoline, which showed a 10 % probability
of ignition at approximately 610C and a 90 % probability of ignition at approxi-
mately 660C. Davis et al. [18] reported hot surface ignition at temperatures as
low as 667, 670, and 735C, respectively for 87-, 89-, and 93-octane gasoline. They
showed that the hot surface ignition temperature generally increases with increas-
ing octane number for standard gasoline blends. Recent work by Shaw et al. [19]
found minimum hot surface ignition temperatures of 530C for gasoline using a
different experimental approach.
While the hot surface ignition temperatures for gasoline may seem to support
the possibility that cigarettes can ignite gasoline, it is important to remember that
hot surface ignition is a complex phenomenon, dependent on numerous factors
including surface properties, environmental conditions, and properties of the ignit-
able liquid [18]. Of particular relevance to cigarette ignition of gasoline, it has
been shown that hot surface ignition temperature is dependent on the size of the
hot surface [5,20]. As the area of the hot surface decreases, the hot surface igni-
tion temperature increases. The previously mentioned studies on the hot surface
ignition of gasoline involved hot surfaces several orders of magnitude larger than
the size of the hot surface of a lit cigarette. Since hot surface ignition tempera-
tures are so strongly dependent on the test conditions, the hot surface ignition
Table 2
Temperatures Associated with Lit Cigarettes [12]
Peak temperature Idle cigarette (C) Puffed cigarette (C)
Gas phase 700–850 850
Solid phase 700–850 900–950
a
a
Maximum transient temperatures of 1,200C were measured at the paper burn line in one study, at the start of a
puff only
The Propensity of Lit Cigarettes 1393
temperatures for gasoline listed above may not be relevant to cigarette ignition of
gasoline, but they do suggest that the surface temperature of the cigarette would
likely need to be significantly greater than the AIT to cause ignition.
Unlike hot surface ignition, cigarette ignition of gasoline has the additional
complication of having conditions in the combustion region of the cigarette that
are unfavorable to ignition. Holleyhead [12] reported oxygen-deficient conditions
in the combustion region of the cigarette, with high concentrations of carbon
dioxide, and air velocities through the cigarette of up to 400 cm/s. At the center
of the cigarette coal, where temperatures are highest, oxygen is almost entirely
depleted, with other areas of the coal having oxygen concentrations of less than
10 % by volume [12]. Concentrations of carbon dioxide in the cigarette coal are
reported to be as high as 17.5 % by volume [12]. With velocities of up to 400 cm/s
in the cigarette while being puffed, the residence time of gases or vapors in the
high temperature region of the cigarette coal are typically less than 1 ms [12]. As a
result, gases or vapors entering the cigarette coal when the temperatures are at a
maximum (during puffing), have a limited exposure period during which to heat
to their ignition temperature. All these conditions create an environment unfavor-
able to ignition, and result in an ignition process that is more complex than a sim-
ple temperature comparison.
There is not a clear consensus on the factors that explain the ignition, or lack
of ignition, for substances by cigarettes in laboratory experiments. Various factors
have been proposed, including flammability limits, surface temperature, residence
time and ignition delay, burning velocity, minimum ignition energy (MIE),
quenching distance, and insulating effect of the ash [1]. No single combustion
parameter explains the ignition, or lack thereof, for ignitable liquids or gases by
cigarettes [12]. However, in general, substances that have been ignited via a ciga-
rette have smaller quenching distances, wider flammable ranges, lower MIE, and
slightly lower AITs than substances that do not ignite [12]. Table 3compares gas-
oline to substances that have been ignited by a cigarette.
Table 3
Comparison of Gasoline to Substances that Have Been Ignited by
Cigarettes [5,20]
Fuels
Quenching
distance (mm)
Flammable
range (%) MIE (mJ) AIT (C)
Ignited
by cigarettes
Gasoline 2–3
a
1–7 0.8 440 No
Acetylene 0.65 2.5–100 0.03 305 Yes
Carbon disulphide 0.55 1.2–50 0.039 90 Yes
Ethylene oxide 1.18 3.6–100 0.105 429 Yes
Diethyl ether 1.85 1.9–36 0.79 195 Yes
Hydrogen 0.64 4.0–75 0.03 400 Yes
Hydrogen sulphide 4.0–44 0.077 260 Yes
Phosphine 1–100 – 100 Yes
a
Value estimated from correlation of MIE and quenching distance [21]
1394 Fire Technology 2014
Despite the lack of a theory in the literature as to why cigarettes ignite some
substances, and not others, the general trends in Table 3are consistent with the
hypothesis that gasoline is unlikely to be ignited by a cigarette. The quenching
distance may be relevant to cigarette ignition in that it is possible that the tobacco
fibers of the cigarette may act similar to a flame arrestor, and prevent any ignition
that occurs within the cigarette to propagate to fuel vapors outside the cigarette
[12]. This effect has not been shown experimentally, so it is unclear whether the
larger quenching distance of gasoline is indeed significant.
The flammable range for gasoline is narrower, with a significantly lower upper
flammable limit (UFL), than the substances that have been ignited by cigarettes.
The narrow flammable range means that the volume of the fuel–air mixture over
which ignition may occur is significantly less than the substances ignited by ciga-
rettes. MIE is a metric for the amount of energy transfer required for ignition to
occur. Stresse [9] observed that fuels with an MIE of less than about 0.08 mJ were
ignited by cigarettes, while those with higher MIE values generally did not ignite.
The MIE of gasoline is an order of magnitude greater than the value suggested by
Stresse, indicating that it is unlikely to ignite. However, diethyl ether has an MIE
similar to gasoline and it is ignited by cigarettes. Similarly, the AIT of gasoline
was higher than most substances ignited by cigarettes. However, hydrogen and
ethylene oxide were ignited by cigarettes despite AITs similar to gasoline. When
taken as a whole, the data in Table 3indicate substances ignited by cigarettes are,
in general, more reactive and more easily ignited than gasoline. However, there
are some inconsistencies in the data, and therefore experimental investigation is
necessary to fully address the issue.
Existing experimental evidence supports the hypothesis that cigarettes will not
readily ignite gasoline. The Ignition Handbook [5] notes that several papers report
that gasoline vapors are not ignited in cigarette experiments [8,10,12]. DeHaan
[6] noted anecdotally that throwing or dropping lit cigarettes into the vapors
above a gasoline pool had never resulted in ignition, however no details on the
experiments were provided.
Yockers and Segal [8] conducted a series of experiments involving various flam-
mable liquids and gases. In the experiments, they saturated a 7.6 cm (3 in.) square
of asbestos cloth with the liquid and exposed it to three different cigarette condi-
tions—a lit cigarette, a lit cigarette under draw, and glowing coals from a ciga-
rette. Each liquid was tested three times, and none of the experiments with
gasoline resulted in ignition.
Schuh and Sanderson [13] found that no ignitions occurred in 100 attempts to
ignite gasoline with a lit cigarette, in various configurations. Specifically, none of
the 20 cigarettes they dropped into liquid gasoline caused ignition. Likewise, no
ignitions occurred when cigarettes under draw were placed above the surface of
the liquid [up to 0.3 m (1 ft) away]. Even when gasoline was misted on lit ciga-
rettes no ignitions occurred.
A recent study by Jewell et al. [14] examined lit commercial cigarettes, hand-rol-
led cigarettes, and cannabis resin joints exposed to gasoline vapors. A total of 30
ignition attempts were conducted by placing lit cigarettes/joints 20 cm to 50 cm
(8 in. to 20 in.) above a pool of liquid gasoline in a pan. An additional nine
The Propensity of Lit Cigarettes 1395
ignition attempts involved a mannequin dressed in cotton clothing that was spla-
shed with various amount of gasoline ranging from 20 mL to 500 mL and
exposed to lit cigarettes/joints under draw from a pump. No ignition of gasoline
vapors was observed in the 39 ignition attempts.
The primary criticisms, or caveats, related to the previous experimental work on
the ignition of gasoline by a lit cigarette were the limited number of experiments
and limited range of conditions under which tests were conducted [5]. It has even
been suggested that an ‘‘infinite number of experiments’’ would be required to
completely disprove cigarettes as a possible ignition source for ignitable liquids
and gases [5,11]. While this may be true, in a strict scientific sense, the existing
data has shown that the ignition of gasoline vapors by a cigarette is an unlikely
event requiring a unique set of circumstances that has yet to be identified. From a
practical standpoint for fire investigators, a cigarette has not been shown to be a
competent ignition source for gasoline vapors in laboratory testing. The present
study was designed to address the limitations of the previous experimental work
by conducting a large number of cigarette exposures to gasoline and considering a
wide range of conditions to explore the possibility of the ignition of gasoline by a
cigarette in as empirically-sound manner as possible.
1.2. Case Studies
Whether cigarettes are a competent ignition source for gasoline vapors is more
than just an academic argument, the scenario is encountered by fire investigators
on a regular basis. A common scenario where the ignition of gasoline vapors by a
cigarette is in question involves two people, often a man and a woman, where one
or both are smokers. The scenario often unfolds where the victim accidentally
came in contact with liquid gasoline, according to the witness, and the gasoline
vapors were subsequently ignited by a cigarette. The witness, who was in close
proximity to the victim, sustained no injuries or only minor injuries. Several case
studies are presented in this section to illustrate the prevalence of this scenario.
On 23 June 2002 a Grand Rapids, Michigan woman was severely burned, and
later died, after gasoline on her body was ignited while she was sitting in the pas-
senger seat of a Ford Bronco. Her boyfriend, who was in the driver’s seat, sus-
tained burns to the inside of his arms and hands, and flash burns to his face.
Initially he reported that an unsealed five gallon plastic gasoline container con-
taining one gallon of gasoline was located on the floor of the vehicle between the
victim’s feet. His account of the event was that the victim was smoking a cigarette
and she suddenly ignited due to gasoline spilled on her hands from filling the con-
tainer. Later, he changed his story and reported that they were arguing and she
poured gasoline on herself and committed suicide. Evidence showed that the vic-
tim and her boyfriend were arguing violently prior to the incident with the gaso-
line container on the center console prior to the fire and that the boyfriend
poured gasoline on the victim and ignited it with a lighter. The boyfriend was
convicted of second degree murder after a jury trial in Kent County (Michigan)
Circuit Court and was sentenced to life in prison.
1396 Fire Technology 2014
On 28 March 2003, a South Holland, Illinois woman was burned after gasoline
was spilled or poured on her torso and ignited. A friend claimed that his burning
cigarette accidentally ignited the spilled gasoline, which caused serious burns to
her head, face, arms, neck, upper back and upper chest. He received treatment for
burns on his right hand. The man was later convicted of heinous battery, a Class
X felony in Illinois, and was sentenced to 12 years in prison.
On 4 October 2004, two police officers were burned in Harlingen, Texas as they
responded to a domestic dispute. The officers followed a man carrying a gasoline
container into his residence. The man claimed he dropped the gasoline container,
splashing gasoline onto one of the officers, and dropped a cigarette accidentally,
which ignited the gasoline. The first officer received third degree burns over 75 %
of his body and the second officer, a few feet behind him, received flash burn inju-
ries. Evidence showed that the man had poured a trail of gasoline behind him
while running through the house. He then ignited the opening of the gasoline con-
tainer with his lighter and threw the container at the officers. The man was found
guilty of two counts of aggravated assault on a peace officer and one count of
arson of a residence, and was sentenced to consecutive sentences of 50 and
25 years, respectively.
2. Experimental Setup
Experiments were conducted exposing lit, commercially-manufactured tobacco cig-
arettes, both at idle and under draw, to gasoline vapors in various configurations.
Testing configurations included gasoline in a pan, on a horizontal surface, on a
variety of textile substrates (clothing), and as a spray. These configurations were
selected so as to consider realistic scenarios that provided favorable conditions for
the ignition of the gasoline vapors.
All experiments used regular 87 octane automotive gasoline purchased at a
commercial gasoline station in January 2005. The exact composition of the gaso-
line was not determined, however it is likely that the gasoline contained up to
10 % ethanol. Prior to each test, the desired amount of liquid gasoline was
poured from a portable fuel storage container into a graduated cylinder. The gas-
oline was then splashed, poured, or sprayed during the test according to the spe-
cific test configuration. Some tests involved multiple cigarettes per test, with
multiple probes of the gasoline vapors per cigarette. At the conclusion of each
test, an open flame ignited the gasoline vapors. This process verified that the ciga-
rettes were exposed to a gasoline vapor–air mixture that was within the flamma-
bility limits during the experiment.
2.1. Cigarette Exposures
The cigarettes utilized in these experiments were commercially manufactured
tobacco cigarettes purchased from retail outlets. The cigarette brands selected for
this study were Marlboro (M), Marlboro Light (ML), Newport (N), Camel (C),
and C without filter (CNF). M, N, and C represent the three most popular ciga-
rette brands purchased by consumers in the US, representing approximately 56 %
The Propensity of Lit Cigarettes 1397
of overall sales [22]. The physical parameters and burning behavior of commercial
tobacco cigarettes, including their competence as an ignition source, were consis-
tent across cigarette brands [5,23].
None of the cigarettes used in this study were Fire Standards Compliant (FSC)
cigarettes. FSC cigarettes, often referred to as ‘‘fire safe cigarettes’’, are designed
to self-extinguish more quickly than standard cigarettes. At the time these tests
were conducted (2005), FSC cigarettes were not prevalent. Any differences in igni-
tion performance between current FSC cigarettes and the non-FSC cigarettes used
in this study are expected to be negligible with regards to the ignition of gasoline.
Cigarettes were tested under two basic conditions—idle and under draw. An
idling cigarette refers to a lit and glowing cigarette not actively being smoked.
Holleyhead [12] referred to this condition as natural smoldering, the burning that
occurs between draws or puffs of a cigarette by a smoker. A cigarette tested under
draw refers to a suction or draw being applied to the unlit end of a cigarette to
provide a severe exposure at the lit end of the cigarette. This condition is also
referred to as induced smoldering or puffing of the cigarette.
For experiments conducted with cigarettes under draw, a vacuum draw appara-
tus was constructed to produce a steady draw condition on the lit cigarette to pro-
vide a severe exposure at the lit end of the cigarette [24]. Guidance on the design
and operation of this apparatus was based on a mock-up ignition test method
procedure developed by the National Institute of Standards and Technology for
cigarette ignition of upholstered furniture [24]. Figure 1shows the vacuum draw
apparatus. The test apparatus consisted of a vacuum pump connected via flexible
tubing to a flowmeter with a needle valve, which was connected via flexible tubing
to a cigarette holder. The holder contained a flexible diaphragm with a hole sized
to seal around the filter end of the cigarette and a fibrous particulate filter. This
apparatus provided a means to pull air through the lit cigarette at a rate of 1 L/min ±
200 mL/min. The holder was attached to a 0.9 m (3 ft) rod so that a lit cigarette,
under draw, could be used to probe the gasoline vapors from a safe distance.
Figure 1. Cigarette vacuum draw apparatus.
1398 Fire Technology 2014
Several methods were employed to bring the cigarettes and gasoline into con-
tact—gasoline introduced near lit cigarettes, tossing cigarettes into gasoline liquid/
vapors, and probing gasoline liquid/vapors with a cigarette. The first method
involved gasoline poured, splashed, or sprayed in the presence of a lit cigarette,
with no movement of the cigarette. Cigarettes were also tossed into liquid gasoline
and gasoline vapors to consider movement of the cigarette through the vapor
space.
For cigarette exposures involving pans of gasoline and gasoline on textiles,
probes with the cigarette were used. A probe refers to moving the cigarette into,
through, and possibly out of the gasoline vapor space. Probing was accomplished
with the cigarette placed in the holder of the vacuum draw apparatus. Probing at
idle was accomplished with the vacuum pump off and no forced draw through the
cigarette. Probing was conducted at various locations in the area of the gasoline
exposure. For the pan experiments, probing was performed at multiple locations
within the pan. For the experiments with gasoline on clothing, probing was per-
formed at various locations on and around the spill/soak location. For example,
probes were conducted in the center of the spill, at the spill edges, and above and
below the spill location. The exact probing locations varied for each test, and
multiple probes at the same location were typically conducted during a test.
Another cigarette condition that was explored for a limited number of experi-
ments was a shower of burning tobacco fragments produced when a lit cigarette
strikes a solid surface. Babrauskas [5] suggested that this scenario provided favor-
able conditions for ignition. To create a shower of burning tobacco fragments, a
lit cigarette was tossed against a vertical surface, as shown in Fig. 2. Note the
black marks on the vertical surface, just above the burning tobacco fragments in
Fig. 2. These marks are a result of the cigarettes striking the surface. The vertical
surface was located such that the cigarette and resulting shower of burning
tobacco fragments fell through gasoline vapors into a pan.
Figure 2. Burning tobacco fragments created by striking a solid surface.
The Propensity of Lit Cigarettes 1399
2.2. Gasoline Configurations
Three gasoline configurations were considered in this study—pools/pans of gaso-
line, gasoline on textile substrates (clothing), and sprayed gasoline. Tests con-
ducted with pools/pans of gasoline were intended to simulate spills of gasoline
onto solid surfaces, or open containers of liquid gasoline. Table 4summarizes the
five pool/pan fire scenarios tested.
A 0.6 m (2 ft) diameter steel pan, 4 cm (1.5 in.) deep, containing 500 mL
(17 oz) of gasoline was used for the majority of the pan experiments. In one
experiment, the pan was turned upside down and only 150 mL (5 oz) of gasoline
was placed on the backside of the pan. This allowed the cigarettes to be placed
adjacent to the liquid gasoline without the influence of the lip of the pan.
Tests were conducted with the gasoline poured in the pan prior to, and after,
the burning cigarettes were in place. In some tests, screens were placed into the
pan to allow the cigarettes to burn in the vapor space above the liquid surface,
without making contact with the liquid. Due to the vapor density of gasoline, the
vapor space above a pool of gasoline could be fuel rich, or even above the UFL
just above the liquid surface in a quiescent atmosphere [25]. The concentration of
gasoline vapors will decrease with increasing height above the surface of the liquid
gasoline due to diffusion and normal air circulation. To account for the potential
of variable concentrations of gasoline vapors at different heights, screens were
used to offset cigarettes at different heights in the gasoline vapor space. Four
screens, each at a different height, were located within the pan. The height of each
screen above the bottom of the pan was not recorded. Figure 3shows the height
and arrangement of the screens.
Although a wire mesh screen is often used as a flame arrestor to prevent fire
spread through the screen, the screens used in the pan experiments did not pre-
vent ignition. Since the fuel (gasoline vapors), oxygen (from the ambient air), and
ignition source (cigarette or open flame) were all present on the same side of the
screen, ignition still occurred. This was verified when an open flame was
Table 4
Summary of Pool/Pan Fire Scenarios
Scenarios
Gasoline
quantity (mL) Description
P1 500 Cigarettes inserted into the vapor space above a pan of
gasoline
P2 500 Cigarettes placed on a screen above a pan of gasoline.
Gasoline was poured into pan past cigarettes
P3 500 Cigarettes placed on screens at four different heights
within the vapor space above a pan of gasoline. Gaso-
line was poured into pan past cigarettes
P4 150 Gasoline poured on the back of a pan. Cigarettes
placed in/around gasoline on pan
P5 500 Cigarettes and glowing tobacco fragments fall into a
pan of gasoline
1400 Fire Technology 2014
introduced above the screen, at the same location as the cigarettes, and consis-
tently resulted in sustained ignition of the gasoline vapors in the pan.
This procedure of introducing an open flame ignition source at the conclusion
of each test was performed for each pan experiment in order to demonstrate that
a flammable concentration of gasoline vapors existed in or above the pan in the
area of cigarette exposure. After ignition, the fire was then extinguished, the pan
was cleared of any cigarettes, and the pan was rinsed with water prior to the next
test.
In tests involving gasoline on clothing, gasoline was either soaked or splashed
onto a portion of the garments. The area of the garment to which gasoline was
applied varied in each test based on the quantity of gasoline used, method of
applying the gasoline, and the clothing material. In general, liquid gasoline was
present only on a limited portion of the garment. Figure 4shows the two methods
used to apply the gasoline to the clothing. Soaking was accomplished by slowly
pouring 150 mL (5 oz) of gasoline onto the garment to absorb the maximum
amount of fuel into the material. Conversely, splashing the gasoline onto the
clothing refers to the application of the entire volume of gasoline to the clothing
surface in one abrupt motion. A smaller quantity of gasoline, 50 mL (1.7 oz), was
used when the gasoline was splashed onto the clothing. In both cases, a pan was
Figure 3. Screens were used to offset cigarettes at different heights
above the pan.
Figure 4. Gasoline was applied to the clothing via (a) splashing and
(b) pouring.
The Propensity of Lit Cigarettes 1401
located below the target garment to collect any liquid gasoline falling from the
material. Table 5summarizes the fire scenarios tested involving gasoline on cloth-
ing.
By using varied textile materials, a variety of absorption and wicking scenarios
were considered. The use of garments in particular, rather than fabric swatches,
was selected so that the tests were similar to fire scene situations that investigators
encounter. Secondly, much of the previously conducted research into the ignition
propensity of gasoline vapors by a cigarette involved the pool/pan of gasoline sce-
nario. By using a textile substrate in the vertical orientation, the gasoline vapors
fall from the garment due to the vapor density of gasoline. This potentially results
in the mixture of the gasoline vapors and the ambient air in concentrations that
may be different than those achieved in the more tightly packed vapor column
present above a pool of gasoline.
The clothing utilized in these experiments was obtained both new and used
from retail outlets and resale shops. Clothing description data was obtained from
the labels attached to the garments; garments lacking labels were not used.
Table 6summarizes the clothing used in the experiments. Figure 5contains pho-
tographs of selected clothing items used in the experiments.
Two methods were used to support the clothing. In some tests, mannequins
were dressed with the clothing tested. Other tests were conducted with the gar-
ment supported on a wire clothing hanger.
At the conclusion of each test involving gasoline on clothing, an open flame
ignited the remaining gasoline to demonstrate the presence of a flammable concen-
tration of gasoline vapors.
Table 5
Summary of Gasoline on Clothing Fire Scenarios
Scenarios Description Gasoline
Cigarette
exposure Wind
T1 Cigarette probed clothing soaked with gasoline Soak Idle Probe No
T2 Cigarette under draw probed clothing soaked with
gasoline
Soak Draw Probe No
T3 Cigarette under draw probed clothing splashed
with gasoline
Splash Draw Probe No
T4 Gasoline splashed on clothing while mannequin is
holding a cigarette
Splash Idle Holding No
T5 Gasoline splashed on clothing while mannequin is
holding a cigarette and clothing probed with a
cigarette
Splash Idle Both Yes
T6 Gasoline splashed on clothing while mannequin is
holding a cigarette and clothing probed with a
cigarette under draw
Splash Both Both Yes
T7 Gasoline soaked on clothing while mannequin is
holding a cigarette and clothing probed with a
cigarette under draw
Soak Both Both Yes
1402 Fire Technology 2014
The final gasoline configuration considered in this study was a spray of gaso-
line. Spray from a failed hose, mist liberated by a malfunctioning pump, or splat-
ter from dropped or spilled gasoline could account for the presence of a spray,
mist, or fine droplets of the fuel in the presence of a cigarette. Liquids, in the
form of a spray, are known to ignite readily, even at temperatures below their
flashpoint [20,26]. In addition, Babrauskas [5] notes that the MIE of gasoline
decreases from 5 mJ at its flashpoint to 0.6 mJ at 10C when in the form of a
spray. Given the ease of ignition and low energy levels required to ignite a spray
of gasoline at ambient temperatures, exposing a lit cigarette to a spray of gasoline
would seem to be a scenario highly favorable to ignition.
For this study, cigarettes were introduced to a spray of gasoline while in the
draw apparatus previously described. Tests were also conducted where an open
flame was introduced into the spray of gasoline. The spray was produced with a
newly purchased small spray bottle, similar to those used to spray cologne. The
spray bottle was mounted on an apparatus that could actuate the pumping mech-
anism of the sprayer multiple times from a safe distance. Figure 6shows the gaso-
line sprayer apparatus used in this study. The size distribution of the spray
produced by this apparatus was not quantified.
2.3. Environmental Conditions
Environmental conditions in the laboratory during the testing were monitored and
recorded. Ambient temperatures and gasoline temperature were measured using
Type K thermocouples. Ambient air temperatures monitored in the laboratory
during testing ranged from 12Cto19C (54F to 66.0F). The temperature of the
Table 6
Clothing Summary
IDs Materials Description Colors
S1 100 % Cotton T-shirt White
S2 100 % Polyester ‘‘Dri Wick’’ fabric T-shirt Orange
B1 100 % Polyester Blouse Black with flowers
B2 100 % Polyester Blouse Blue/purple
B3 100 % Polyester velvet Blouse Maroon
B4 100 % Rayon Blouse Multi-colored
B5 55 % Ramie/45 % cotton Blouse Multi-colored
J1 100 % Cotton Jeans Blue
SS1 80 % Cotton/20 % polyester Sweatshirt Blue
SW1 65 % Polyester/35 % cotton Sweater Red
SW2 55 % Ramie/45 % cotton Sweater Multi-colored
D1 100 % Silk Dress Red
C1 100 % Nylon Coat Black
C2 Exterior: 100 % leather
Liner: 50 % nylon/50 % acetate
Coat Black
C3 Exterior: 52 % polyester/48 % acetate
Liner: 100 % acetate
Coat Black
C4 Exterior: 100 % wool
Liner: 100 % nylon sateen
Coat Gray
The Propensity of Lit Cigarettes 1403
gasoline used in testing ranged from 8Cto17C (47Fto63F). Tests were con-
ducted in both a quiescent environment, and with wind provided by an electric
fan. Wind speeds of up to 1 m/s (2.2 miles/h) and 2 m/s (4.5 miles/h) were used
Figure 5. Selected clothing used in the experiments.
Figure 6. Gasoline sprayer apparatus.
1404 Fire Technology 2014
during testing. Wind speed was measured using a Kestrel Pocket Thermo Wind
Meter approximately 15 cm (6 in.) above the upwind lip of the pan containing
gasoline or at the target in tests involving garments. The relative humidity of the
ambient air during testing ranged from 29 % to 82 %. Barometric pressure during
the testing was nominally one atmosphere.
3. Results and Discussion
A total of 70 experiments were conducted over a period of 7 days from 6 January
2005 to 14 January 2005. Of the 70 experiments conducted, 64 involved exposing lit
cigarettes to gasoline. Each experiment typically involved multiple exposures of ciga-
rettes to gasoline. A total of 723 cigarettes were used during the test series. The
remaining six tests (Tests 1.0, 1.1, 1.2, 9.5, 9.7, and 20.0) did not include cigarettes;
gasoline was ignited with an open flame in these tests. The identifiers for the experi-
ments are not numbered sequentially; they were retained from an unpublished report
by one of the authors which has been admitted to court on multiple occasions.
Thirty experiments were conducted with lit cigarettes exposed to pools or pans
of gasoline. In these 30 experiments, 591 cigarettes were exposed to gasoline
vapors, often multiple times, in various conditions. Tables 7,8, and 9detail the
experiment conditions and results of the cigarette exposures to pans of gasoline.
In these tables, the brands of cigarette used in the experiments are referred to
using the following nomenclature: M, ML, N, C, and CNF. The naming conven-
tion for the pan scenarios is given in Table 4.
Table 10 and Fig. 7summarize the overall results from these experiments. No
ignition of gasoline vapors was observed in any of the 734 ignition attempts in
this configuration.
Table 7
Summary of Pan Experiments for Scenario P1
Test Cigarette brand Wind speed (m/s)
Number of ignitions/number of attempts
Probe (idle) Probe (draw) Open flame
1.0 1/1
1.1 1/1
1.2 1/1
2.0 M 0/1 0/0
2.1 M 0/7 0/0
2.2 M 0/7 1/1
2.3 M 0/16 1/1
3.0 M 0/5 0/0
3.1 M 0/3 1/1
4.0 M 1.1 0/15 1/1
4.1 M 2.0 0/12 1/1
5.0 M 1.1 0/9 1/1
5.1 M 2.0 0/15 1/1
Total 0/58 0/32
The Propensity of Lit Cigarettes 1405
Thirty-seven experiments were conducted with lit cigarettes exposed to gasoline
on textile substrates (clothing). In these 37 experiments, 129 cigarettes were
exposed to gasoline vapors, often multiple times, in various conditions. Tables 11
Table 9
Pan Experiments for Scenario P3
Test Scenarios Cigarette brands
Number of ignitions/number of attempts
Idling Probe (draw) Open flame
22.0 P3 M, ML, N, C 0/16 1/1
22.1 P3 M, ML, N, C 0/16 1/1
22.2 P3 M, ML, N, C 0/16 1/1
23.0 P3 M, N, C, ML 0/16 1/1
23.1 P3 M, N, C, ML 0/16 1/1
23.2 P3 M, N, C, ML 0/16 1/1
23.3 P3 M, N, C, ML 0/16 1/1
23.4 P3 M, N, C, ML, CNF 0/16 0/17 1/1
23.5 P3 M, N, C, ML, CNF 0/16 0/52 1/1
23.6 P3 M, N, C, ML, CNF 0/20 1/1
23.7 P3 M, N, C, ML, CNF 0/20 1/1
Total 0/184 0/69
Table 10
Summary of Pan Experiments
Description Number of ignitions/number of attempts
Idling cigarettes 0/405
Thrown cigarettes 0/170
Probes with idle cigarette 0/58
Probes with cigarette under draw 0/101
Total 0/734
Table 8
Summary of Pan Experiments for Scenarios P2, P4, and P5
Test Scenarios Cigarette brands
Number of ignitions/number of attempts
Idling Thrown Open flame
17.0 P4 M, ML, N, C 0/51 0/31 1/1
18.0 P2 M, C, N, ML, CNF 0/50 1/1
18.1 P2 M, C, N, ML, CNF 0/48 1/1
19.0 P2 and P5 M, ML, N, C 0/24 0/55 1/1
19.1 P2 and P5 M, ML, N, C 0/24 0/56 1/1
19.2 P2 and P5 M, ML, N, C 0/24 0/28 1/1
Total 0/221 0/170
1406 Fire Technology 2014
and 12 detail the experiment conditions and results of the cigarette exposures to
gasoline on clothing. The same abbreviations are used for the cigarette brands as
in the previous tables. The scenario and clothing identifiers listed in Tables 11 and
12 refer to those listed in Tables 5and 6, respectively.
Table 13 and Fig. 8summarize the overall results from the clothing experi-
ments. No ignition of gasoline vapors was observed in any of the 3,533 ignition
attempts in the clothing experiments.
Three experiments were conducted with sprays of gasoline. Table 14 details the
results of the spray experiments. The same abbreviations are used for the cigarette
brands as in the previous tables. When the gasoline spray was applied to an open
flame, the gasoline ignited and a fireball immediately ensued. No ignitions caused
by cigarettes occurred during these experiments.
One observation noted in both the pan experiments and spray experiments was
the ability of the liquid gasoline to extinguish combustion in the cigarette. This
phenomenon was observed in ten of the pan experiments. For example, when a lit
cigarette was dropped into a pan of liquid gasoline, white vapors were visible
when the cigarette first entered the liquid. Once in contact with the liquid gaso-
line, the cigarette absorbed the gasoline and wicked gasoline through the remain-
der of the cigarette, ceasing combustion. A similar process was observed with
sprays of gasoline following multiple sprays on the same cigarette.
Figure 7. Summary of pan fire experiments by scenario.
The Propensity of Lit Cigarettes 1407
Table 15 summarizes all the cigarette ignition attempts of gasoline conducted in
this study. In total, over 4,500 attempts to ignite gasoline were conducted. No
ignition of gasoline by a cigarette was observed in any configuration tested.
In contrast, an open flame introduced into the same gasoline configurations to
which the cigarettes were tested ignited the gasoline vapors in 87 of the 88 ignition
attempts. This confirmed that the cigarettes were exposed to flammable concentra-
tions of gasoline vapor, and indicated that the cigarettes were not competent igni-
tion sources.
The results of these tests are limited to commercially-manufactured tobacco cig-
arettes. This study did not address hand-rolled cigarettes, cigars, marijuana ciga-
rettes, or other tobacco products. See Babrauskas [5] and Jewell et al. [14] for
discussion of these topics.
One question that often arises as part of a study on ignition propensity of ciga-
rettes is whether the results show that it is ‘‘impossible’’ to ignite gasoline with
cigarettes. Some researchers have even gone as far as saying that an ‘‘infinite num-
ber of experiments’’ would be required to completely disprove cigarettes as a pos-
sible ignition source [5,11]. The scientific method dictates that hypotheses should
be testable and falsifiable, meaning that a test of the hypothesis can be conducted
that would disprove the hypothesis. In addition, Popper [27] introduced the con-
cept of degree of corroboration of a hypothesis, meaning that ‘‘it is not so much
Table 11
Clothing Experiments for Scenarios T1–3
Test Scenarios Clothing IDs Cigarette brands
Number of ignitions/number of attempts
Probe (idle) Probe (draw) Open flame
6.0 T1 S1 M, N, C 0/215 1/1
6.1 T1 B4 CNF 0/164 1/1
7.0 T2 S1 M, N, C, ML 0/108 1/1
7.1 T2 B1 ML, N, C, M 0/314 1/1
7.2 T2 J1 N, ML, M, C 0/484 1/1
7.3 T2 B4 CNF 0/45 1/1
7.4 T2 B5 CNF 0/29 1/1
7.5 T2 B2 N 0/32 1/1
7.6 T2 SS1 ML 0/38 1/1
7.7 T2 SW2 M 0/39 1/1
7.8 T2 S2 C 0/20 1/1
9.0 T3 SW1 C, M, ML, N 0/863 1/1
9.1 T3 B3 ML, M, C, N 0/922 1/1
9.2 T3 C1 N 0/33 1/1
9.3 T3 C1 C 0/40 1/1
9.4 T3 C2 ML 0/9 1/1
9.5 T3 C2 1/1
9.6 T3 C2 C 0/12 0/1
9.7 T3 C2 1/1
Total 0/379 0/2,988
1408 Fire Technology 2014
the number of corroborating instancesas the severity of the various tests to
which the hypothesis in question can be, and has been, subjected’’.
The hypothesis that lit cigarettes do not ignite gasoline is highly testable, falsifi-
able, and has a high degree of corroboration. As the number of studies examining
this hypothesis [6,8,10,1214] indicates, it is trivial to construct a test that would
falsify the hypothesis. A single ignition of gasoline by a lit cigarette would invali-
date the hypothesis, and given the availability of both gasoline and cigarettes,
designing such a test of the hypothesis is straightforward. One of the goals of the
present study was to address the degree of corroboration of this hypothesis by
exposing it to severe tests including cigarettes under draw, which offer a severe
cigarette exposure, and gasoline in various configurations including sprays of
Table 12
Clothing Experiments for Scenarios T4–7
Test Scenarios
Clothing
IDs
Cigarette
brands
Wind speed
(m/s)
Number of ignitions/number of
attempts
Idling
Probe
(idle)
Probe
(draw)
Open
flame
10.0 T4 C3 C 0/3 1/1
11.0 T5 C3 N 1.0 0/1 0/5 1/1
11.1 T5 C3 ML 1.0 0/1 0/5 1/1
11.2 T5 C3 C 2.0 0/1 0/5 1/1
11.3 T5 C3 M 2.0 0/1 0/5 1/1
13.0 T6 C3 M 1.0 0/1 0/5 1/1
13.1 T6 C3 CNF 1.0 0/1 0/5 1/1
13.2 T6 C3 N 2.0 0/1 0/9 1/1
13.3 T6 C3 ML 2.0 0/1 0/5 1/1
13.4 T6 C4 C 0/3 0/6 1/1
13.5 T6 C4 M 0/3 0/5 1/1
13.6 T6 C4 CNF 1.0 0/3 0/7 1/1
15.0 T7 C4 N 1.0 0/3 0/8 1/1
15.1 T7 C4 C 1.0 0/3 0/6 1/1
15.2 T7 C4 ML 2.0 0/3 0/5 1/1
15.3 T7 C4 M 0/3 0/10 1/1
15.4 T7 D1 M 0/3 0/20 1/1
15.5 T7 D1 N 0/3 0/17 1/1
Total 0/38 0/20 0/108
Table 13
Summary of Clothing Experiments
Description Number of ignitions/number of attempts
Idling cigarettes 0/38
Probes with idle cigarette 0/399
Probes with cigarette under draw 0/3,096
Total 0/3,533
The Propensity of Lit Cigarettes 1409
0
500
1000
1500
2000
2500
3000
3500
Idling Cigarettes Probes with
Idle Ci
g
arette
Probes with Cigarette
Under Draw
Ignition Attempts
T1 T2 T3 T4 T5 T6 T7
Figure 8. Summary of clothing fire experiments by scenario.
Table 14
Spray Experiments
Test Cigarette brand
Number of ignitions/number of attempts
Idling Probe (draw) Open flame
20.0 16/16
21.0 M 0/106 3/3
21.1 M 0/145 5/5
Total 0/106 0/145
Table 15
Summary of Cigarettes Ignition Attempts
Description
Number of ignitions/number of attempts
Pans Clothing Sprays Total
Idling cigarettes 0/405 0/38 0/106 0/549
Thrown cigarettes 0/170 0/170
Probes with idle cigarette 0/58 0/399 0/457
Probes with cigarette under draw 0/101 0/3,096 0/145 0/3,342
Total 0/734 0/3,533 0/251 0/4,518
1410 Fire Technology 2014
gasoline, which present the fuel in a form highly favorable to ignition. Popper [27]
notes that a well corroborated hypothesis is one which ‘‘has been severely tested
[and] has stood up well to the severest tests we were able to design so far’’. The
hypothesis that lit cigarettes do not ignite gasoline has withstood a significant
number of challenges, by various investigators, under a variety of severe tests. In
this way, the hypothesis is well corroborated.
From a practical standpoint, fire investigators want to know whether cigarettes
are a competent ignition source for gasoline. Based on the currently available
experimental data, derived under controlled laboratory conditions from this and
other studies, there is no evidence that a cigarette is a competent ignition source
for gasoline vapors.
4. Conclusions
Fire investigators regularly evaluate available fuels and potential ignition sources
to determine the cause of a fire. This work examined the propensity of lit ciga-
rettes to ignite gasoline vapors, expanding on previous work to include a large
number of trials and a wide range of test conditions. Experiments were conducted
exposing lit cigarettes, both at idle and under draw, to gasoline vapors in various
configurations including pools/pans of gasoline, gasoline on textile substrates
(clothing), and sprays of gasoline.
The experiments conducted for this study consisted of 70 distinct tests involving
a total of 723 cigarettes and over 4,500 instances of exposure of a lit cigarette to
ignitable concentrations of gasoline vapors in air. There were no instances of the
ignition of gasoline vapors from the exposure of those vapors to a lit, major
brand, tobacco cigarette, commercially manufactured in the US, during any of the
experiments. In contrast, an open flame ignition source exposed to the same gaso-
line configurations resulted in ignition in all but one attempt.
While it is impossible to conduct an infinite number of experiments to provide
absolute proof that ignition of gasoline could never happen with a lit cigarette,
based on the currently available experimental data, derived under controlled labo-
ratory conditions from this and other studies, there is no evidence that a cigarette
is a competent ignition source for gasoline vapors.
References
1. NFPA 921 (2011) Guide for fire and explosion investigations, 2011rd edn. National
Fire Protection Association (NFPA), Quincy
2. Hall JR Jr (2012) The smoking-material fire problem. National Fire Protection Associa-
tion (NFPA), Quincy
3. Greene MA, Andres C (2009) 2004–2005 National sample survey of unreported residen-
tial fires. Table 6-6. U.S. Consumer Product Safety Commission
4. Slye OM Jr (2008) Flammable and combustible liquids. In: Fire protection handbook,
20th edn, Section 6, Chapter 12. National Fire Protection Association (NFPA), Quincy,
pp 6–198
5. Babrauskas V (2003) Ignition handbook. Fire Science Publishers, Issaquah
The Propensity of Lit Cigarettes 1411
6. DeHaan JD (2007) Kirk’s fire investigation, 6th edn. Pearson Education, Inc., Upper
Saddle River
7. NFPA (2011) Gasoline at home fact sheet. National Fire Protection Association
(NFPA). http://www.nfpa.org/categoryList.asp?categoryID=302. Accessed 4 April 2011
8. Yockers JR, Segal LS (1956) Cigarette fire mechanisms. NFPA Q 49:213–222
9. Stresse G (1970) Zundmoglichkeit von brennbaren Gasen und Dampfen durch glim-
menden Tabak. Sonderdruck aus Bundesarbeitsblatt—Fachteil Arbeitsschutz 3:66–70
10. Hagimoto Y, Kinoshita K (1981) Ignition possibility of inflammable mixtures with
burning cigarettes. J Jpn Soc Saf Eng 20:197–202
11. Hards DL (1983) Examination of the effect of lighted cigarettes on flammable vapour–
air mixtures. Section Paper: IR/L/IN/83/1. Health and Safety Executive, Harpur Hill
12. Holleyhead R (1996) Ignition of flammable gases and liquids by cigarettes : a review.
Sci Justice 36:257–266
13. Schuh DA, Sanderson JL (2008) Gasoline vapor testing: what makes a competent igni-
tion source?. Fire Find 16(1):1–3
14. Jewell RS, Thomas JD, Docids RA (2011) Attempted ignition of petrol vapour by lit
cigarettes and lit cannabis resin joints. Sci Justice 51:72–76
15. Geiman JA, Fuss SP (2013) Investigation of cigarettes as an ignition source for Cole-
man fuel. In: Proceedings of the fire and materials 2013 conference. Interscience Com-
munications Ltd., London, pp 759–768
16. Colonna GR (2010) Fire protection guide to hazardous materials, 14th edn. National
Fire Protection Association (NFPA), Quincy
17. Colwell JD, Reza A (2005) Hot surface ignition of automotive and aviation fluids. Fire
Technol 41(2):105–123. doi:10.1007/s10694-005-6388-6
18. Davis S, Kelly S, Somandepalli V (2010) Hot surface ignition of performance fuels. Fire
Technol 46(2):363–374. doi:10.1007/s10694-009-0082-z
19. Shaw A, Epling W, McKenna C, Weckman B (2010) Evaluation of the ignition of die-
sel fuels on hot surfaces. Fire Technol 46(2):407–423. doi:10.1007/s10694-009-0098-4
20. Drysdale D (2011) An introduction to fire dynamics, 3rd edn. Wiley, West Sussex
21. Glassman I (1997) Combustion, 3rd edn. Academic Press, San Diego, p. 345
22. Maxwell JC (2010) The Maxwell report: year end and fourth quarter 2009 sales esti-
mates for the cigarette industry. John C. Maxwell, Jr., Richmond
23. Ohlemiller TJ, Villa KM, Braun E, Eberhardt KR, Harris RH, Lawson JR, Gann RG
(1995) Quantifying the ignition propensity of cigarettes. Fire Mater 19:155–169
24. Ohlemiller TJ, Villa KM, Braun E, Eberhardt KR, Harris RH Jr, Lawson JR, Gann
RG (1993) Test methods for quantifying the propensity of cigarettes to ignite soft fur-
nishings, NIST Special Publication 851. National Institute of Standards and Technol-
ogy, Gaithersburg
25. Friedman R (1998) Principles of fire protection and chemistry, 3rd edn. National Fire
Protection Association (NFPA), Quincy
26. Crowl DA (2003) Understanding explosions. Center for Chemical Process Safety of the
American Institute of Chemical Engineers, New York
27. Popper K (2010) The logic of scientific discovery. Routledge Classics, New York, pp
266, 375
1412 Fire Technology 2014
Research
Full-text available
This represents one of several sections of "A Bibliography Related to Crime Scene Interpretation with Emphases in Geotaphonomic and Forensic Archaeological Field Techniques, Nineteenth Edition" (The complete bibliography is also included at ResearchGate.net.). This is the most recent edition of a bibliography containing resources for multiple areas of crime scene, and particularly outdoor crime scene, investigations. It replaces the prior edition and contains approximately 10,000 additional citations. As an ongoing project, additional references, as encountered, will be added to future editions. The newest category of evidence added to this bibliography includes that related to fire and explosives evidence. Previously a few references about these topics could be found in General Crime Scene and Death Scene Investigation. (641citations)
Book
Full-text available
This is the only authoritative treatise to encompass the entire field of ignition. The 1116-page handbook was published in cooperation with the Society of Fire Protection Engineers, under whose auspices the peer review was performed.
Article
Full-text available
Establishing a competent ignition source is a critical part of any fire investigation. While it is generally recognized that cigarettes are unlikely ignition sources for common ignitable liquids, the experimental data on specific fuels is limited. This study examined the propensity of cigarettes to ignite Coleman camp fuel in two configurations - clothing soaked with Coleman fuel and pools of Coleman fuel. The first configuration involved a clothed mannequin splashed with Coleman fuel and subjected to a variety of cigarette exposures, including thrown cigarettes, direct contact with lit cigarettes, and direct contact with lit cigarettes under draw. In the experiments that exposed cigarettes to pools of Coleman fuel, cigarettes were placed on a screen above the liquid surface and were allowed to burn their entire length while the lit cigarette was under draw. A total of 16 experiments were conducted, with each involving multiple cigarette exposures. No ignitions of Coleman fuel vapor were observed in approximately 530 cigarette exposures.
Article
The cause of a fire has often been attributed to lighted cigarettes when, in fact, the combustible materials available in many cases cannot be set on fire by such an ignition source. Experiments with flammable gases and vapours have shown that many of the most commonly-encountered substances, including methane and petrol vapour, were not ignited by a lighted cigarette. This review discusses the physical and chemical parameters which govern the ignition of flammable gases and vapours. Temperature and gas concentration measurements inside a burning cigarette tip have been reviewed and data from such work have been used in conjunction with combustion parameters to explain the experimental findings.
Article
Research funded under the Fire Safe Cigarette Act of 1990 (United States Public Law 101–352) has led to the development of two test methods for measuring the ignition propensity of cigarettes. The Mock-Up Ignition Test Method uses substrates physically similar to upholstered furniture and mattresses: a layer of fabric over padding. The measure of cigarette performance is ignition or non-ignition of the substrate. The Cigarette Extinction Test Method replaces the fabric/padding assembly with multiple layers of common filter paper. The measure of perfomance is full-length burning or self-extinguishment of the cigarette. Routine measurement of the relative ignition propensity of cigarettes is feasible using either of the two methods. Improved cigarette performance under both methods has been linked with reduced real-world ignition behavior; and it is reasonable to assume that this, in turn, implies a significant real-would benefit. Both methods have been subjected to interlaboratory study. The resulting reproducibilities were comparable to each other and comparable to those in other fire test methods currently being used to regulate materials which may be involved in unwanted fires. Using the two methods, some current commercial cigarettes are shown to have reduced ignition propensities relative to the current best-selling cigarettes.