Purser Physiol Effects

PHYSIOLOGICAL EFFECTS OF COMBUSTION PRODUCTS
AND FIRE HAZARD ASSESSMENT
By Professor David A. Purser
Hartford Environmental Research, UK
Europacable Seminar “Safety during Fire”, Brussels 6th May 2009
3
ABSTRACT
A major design requirement for any occupied enclosure is to ensure that occupants are able to
escape safely in case of fire. The main cause of injury and death in fires is incapacitation
resulting from exposure to fire effluent (smoke and gases). During fires, physiological effects
dominate and determine time to incapacitation. The sequence of physiological hazards in
developing fires can be applied to fire safety engineering design of buildings and to the
evaluation of products for use in them, using appropriate fire data and calculation methods for
assessment of time to incapacitation
INTRODUCTION
A major design requirement for any occupied enclosure (such as a building or transport vehicle)
is to ensure that occupants are able to escape safely in case of fire. The main cause of injury
and death in fires is exposure to toxic fire effluent (smoke and gases), while the next most
important cause is exposure to heat. It is therefore necessary to ensure that the performance of
the building and it systems, including any combustible structural products or contents, is such
that occupants are able to escape before they are overcome by toxic smoke or heat. This is a
regulatory performance requirement in the UK Building Regulations. Requirement B1 of
Schedule 1 to the Building Regulations 20001 states that: "The building shall be designed and
constructed so that there are appropriate provisions for the early warning of fire, and appropriate
means of escape in case of fire from the building to a place of safety outside the building
capable of being safely and effectively used at all material times". A similar objective is stated in
the ISO fire engineering Technical Report (ISO/TR 13387-8 Fire safety engineering Part 8: Life
Safety - Occupant behaviour, location and condition)2. “Should a fire occur in which occupants
are exposed to fire effluent and/or heat, the objective of the fire safety engineering strategy is to
ensure that such exposure does not significantly impede or prevent the safe escape (if required)
of essentially all occupants, without their experiencing or developing serious health effects”.
Table 1: Fire causes - houses and flats Table 2: Causes – electrical fires houses and flats
Frequency Cause Frequency Cause Cause detail
25 Electrical 3 Electrical Washing machine
13 Chip pan 3 Electrical Tumble dryer
7 Grill/frying pan 3 Electrical Immersion heater
5 Gas 3 Electrical Electric blanket
5 Arson 2 Electrical Wiring
4 Misuse of heating 2 Electrical TV
4 Ignition by child 2 Electrical Mains
3 Chimney 1 Electrical Toaster
3 Unknown 1 Electrical Stereo
2 Cigarette 1 Electrical Iron
2 Cooking 1 Electrical Hi Fi
1 Coal falling from fire 1 Electrical Fridge freezer
1 Overheating of wax 1 Electrical Fridge
1 Paint stripping 1 Electrical Tumble dryer lead
1 Candle 25 SUM
1 Cooker, misuse of
78 SUM
In a detailed study of 100 domestic fires in the United Kingdom3, by far the greatest single cause
4
of fire was electrical (32% in houses and flats), mostly arising from faulty appliances. The next
biggest cause was related to cooking (22 fires = 28% for dwellings). If cooking fires and
deliberate fires are excluded, leaving 51 fires due to other causes, then half of all accidental
fires resulted from electrical faults. After electrical and cooking fires the next most important
causes were gas appliances and deliberate ignition. Cigarettes and smokers materials generally
were a very minor cause at 2.6%.
Electrical fires are detailed in Table 5. Half due were due to faults in domestic appliances and
one third due to faults in supply or building services. These were of particular concern to
occupants because they were completely unexpected, largely beyond their control and the
cause was often never fully established. Occupants felt they needed to know why these fires
had started and what they could have done to prevent them. They were some of the most
serious and damaging incidents. In a number of major incidents in larger building electrical fires
have presented major hazards. In some cases a key problem has been the size of the source
fire, such as large fires in plant rooms in locations such as hotel basements, which have
involved large volumes of toxic effluents spreading through a building via vents and stairwells.
In other incidents involving fires in cable installations, both fire and fire effluents have spread
between floors via ducts and plenum spaces. Particular issues with most fires involving
electrical installations are the large amounts of smoke and toxic gases evolved in the effluent
plumes, so that even small fires can lead to serious contamination of large buildings.
For these reasons it is particularly important that the potential physiological and toxic hazards of
fires involving electrical installations are considered in parallel with the potential hazards from
other combustible products in building and transport design
Since fire hazards are essentially time-based phenomena. The objective is to ensure as far as
possible that should a fire occur, the occupants receive a timely warning and have suitable
escape routes available for a sufficient time to enable escape before conditions deteriorate to
such an extent that they can no longer use them. In fire safety engineering terms it is necessary
to ensure that available safe escape time (ASET) is greater than required safe escape time
(RSET) by an acceptable margin of safety[2].
The main tenability limits for ASET are conditions that cause incapacitation of occupants such
that they are unable to escape.
Toxicity resulting from exposure to any substance arises from a combination of physiological
effects on the function of body systems and pathological effects on body tissues. For the toxic
effects of exposure to combustion products in fire effluent, particularly in relation to the survival,
injury or death of fire victims, it is mainly the more immediate physiological effects that dominate
at the fire scene. This is because physiological effects occur very rapidly, often within a few
seconds, while pathological changes tend to occur over time scales of from hours to years.
Physiological effects also occur in response to heat and smoke exposure in fires, interacting to
some extent with the effects of toxic combustion products.
The acute physiological hazards in fires affecting escape capability are as listed below4,5.
ACUTE SURVIVAL HAZARDS DURING FIRES

• Impaired vision from smoke obscuration
• Impaired vision, pain and breathing difficulties from effects of smoke irritants on eyes
and respiratory tract.
• Asphyxiation from toxic gases leading to confusion and loss consciousness
• Pain to exposed skin and respiratory tract followed by burns from exposure to radiant
and convected heat leading to collapse
These tend to be encountered more or less in the order shown, with exposure first to smoke,
which is likely to be irritant, followed by asphyxia or burns, depending upon the type of fire
scenario and the proximity of the person to the seat of the fire. Once a victim has become
5
trapped or incapacitated in a fire, then conditions usually become lethal within a further few
seconds or minutes. This is because flaming fires grow exponentially, so that concentrations of
smoke and toxic gases and the heat intensity increase rapidly, resulting in death either from
asphyxiation or heat exposure depending upon the fire scenario. For this reason the key
determinant of survival is incapacitation, while the lethal potency of fire effluent is of limited
relevance.
Figure 1 shows an example of a typical set of time-concentration curves from a full-scale fire
test of an item of upholstered furniture in the open lounge of a house6. In order to estimate the
time when a room occupant would be incapacitated it is necessary to consider the developing
hazards from each toxic gas, the smoke and heat and their interactions. For some toxic
products (such as irritants and smoke particulates) the effects depend mainly upon the
immediate concentration to which the subject is exposed, while for others (asphyxiant gases
and heat) the total exposure dose is the most important feature (expressed in terms of varying
functions of concentration x exposure time), so that effects are delayed until a sufficient dose
has built up in the body to produce a given physiological effect5,7
Other important considerations are the nature of the physiological effects and the relationships
between exposure concentration (or exposure dose) and the type and severity of effect. The
intensity of physiological effects of stimulation of sense organs such as hearing, vision or pain
reception tend to show a progressive increase, logarithmically related to the intensity or
concentration of the stimulus. Thus walking speed in smoke is inversely proportional to the
smoke optical density8, while the painful effects of irritant gases lie on a continuum from mild
eye and nasal irritation at low concentrations to intense pain, involuntary eye closure and
breathing difficulties at high concentrations5,7,9
Asphyxiant gases act by limiting the supply or use of oxygen in the body tissues (tissue
hypoxia). Since the body systems are designed to tolerate a degree of hypoxia (for example
during exercise), the effects of low doses or asphyxiants are minor, with little change as the
body dose increases other than some degree of reduction in exercise tolerance, and subtle
sensory or cognitive effects, until a critical threshold level is reached above which a sudden
severe and dramatic loss of functionality occurs. At this point subjects pass rapidly from near
normal behaviour and function to collapse and loss of consciousness5,10,11,12,13
35 12000
Test CDT18
Lounge
30
CO2%, O2%, Temperature degC/10 and
10000
25
8000
CO, HCN and HCl .
smoke OD/m
20
6000
15
Landing Landing
ionization optical 4000
10
Lounge Lounge
ionization optical
2000
5
0 0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Time (min)
CO2 % O2 % Smoke OD/m Temperature /10 CO ppm HCN ppmx10 HCl ppm
Copyright HER
6
Figure 1: Example of time-concentration curves for smoke, toxic gases and temperature
at head height in the domestic lounge of a house during an armchair fire. The doorway to
the hall is open but the house is otherwise enclosed. The time for triggering of smoke
detectors is shown
BASIC REQUIREMENTS FOR LIFE SAFETY HAZARD ANALYSIS

Life safety hazard from fires in buildings (or other enclosures) therefore depends upon the
performance of a dynamic system involving interactions between the building, the fire and the
occupants.
In order to determine the ASET time for any system it is necessary to determine
• The time-concentration (or time-intensity) curves for the major toxic products, smoke and
heat in the fire at the breathing zone of the occupants, which in turn depend upon:
o Fire growth curve (mass loss rate of the burning fuel [kg/s] and its dispersal
volume [kg/m3] with time )
o The yields of the major toxic products (kg/kg) and heat (kj/kg) (for example kg
CO per kg of material burned).
These terms can be measured directly in full-scale tests or calculated using appropriate fire
dynamics computations, with appropriate input data including reaction-to-fire properties and
data on product yields under a range of fire conditions.
• The concentration/time/physiological effect relationships of these products in terms of the

physiological/toxic potency of the products and heat (the exposure concentration [kg/m3]), or
exposure dose (kg·m-3·min or ppm·min) required to cause toxic effects (and the equivalent
effects for heat and smoke obscuration).
o Concentrations, doses (or heat intensity) likely to impair escape efficiency due to
behavioural and/or physiological effects
o exposure concentrations or doses likely to cause incapacitation or prevent
escape due to behavioural and/or physiological effects
o Lethal exposure concentrations or doses
These terms can be calculated by the application of appropriate physiological methods
(Fractional Effective Dose methods) outlined in this paper and presented in detail in the SFPE
Handbook of Fire Protection Engineering5 and other publications14,15,16. A simplified method is
present in ISO 1537117.
In order to evaluate the fire safety of any built environment, or any combustible material or
product used in a built environment it is therefore necessary to consider the overall fire
performance of the system, either by performing a full-scale fire test (or series of tests) on the
end-use system or by modelling the full-scale fire conditions, using suitable fire dynamics
models run with a input data on a range of performance parameters obtained from a set of small
or large scale fire tests.
Problems in representing “toxicity” of a burning material or product as a
single test result constant
It should be evident from this that for systems containing combustible materials, no single small-
scale test can possibly supply all the data necessary to evaluate the full-scale fire performance.
It should be equally evident that representing “toxicity” as a single parameter for which data can
be obtained directly from any small-scale test represents a gross oversimplification of a more
complex reality. Even if the time-varying nature of both fire conditions and physiological effects
are ignored, it is not possible to represent the “toxicity” or “toxic potency” of the combustion
products from a material or commercial product by a single number because the yields of
7
individual toxic gases responsible for the overall toxic effects vary considerably between
different combustion conditions. Combustion products consist of a mixture of individual toxic
gases and particulates that vary in concentration throughout the fire exposure. The yields of
each toxic species from any one burning material vary considerably (by up to several orders of
magnitude) depending upon the combustion conditions, which in turn depend upon the varying
fuel/air ratios and temperatures in the fire. Taking PMMA as an example, Figure 2 shows how
the yield of carbon monoxide varies by a factor of approximately 50 as combustion conditions
vary from well-ventilated (phi<1) to vitiated (phi >1) in two bench-scale apparatus (the ASTM
E2058 flammability apparatus and the ISO TS19000 tube furnace18,19). Similar variations in
toxic product yields with phi have also been shown to occur with other toxic combustion
products from different materials including smoke particulates, total organics, hydrogen cyanide
and oxides of nitrogen20, 21. These two test methods have both been developed specifically to
measure the yields of toxic species from materials and products over a range of combustion
conditions occurring in full-scale fires. The data obtained can then be used as part of the input
needed for toxic hazard assessment or full-scale fire scenarios.
Although it is considered to have limited relevance to full-scale toxic hazards, small-scale tests
such as the tube furnace test have been used to calculate figures for the overall toxic potency of
the combustion product mixture obtained under any specific test condition. Toxic potencies
estimated from small-scale test methods are usually quoted in terms of lethal toxic potency or a
toxicity index number. The lethal toxic potency expressed as the LC50 is the mass loss
concentration of a material (g/m3) capable of causing death in 50% of a group of exposed rats
after a 30-minute exposure period, with 14-day post-exposure observation period. Historically
this was measured directly using animal exposures but is now usually estimated from the
measured concentrations of toxic gases in a test using a calculation method. Toxicity indices
are usually based upon published IDLH (immediately dangerous to life or health) values for
specific toxic gases. For both types of potency estimates a simple additive model is used for the
contributions from each individual toxic gas component, thereby ignoring the different time-
related physiological effects and the interactions occurring between the different components.
Figure 3 illustrates how the calculated LC50 concentrations for different materials decomposed in
the tube furnace vary with different combustion conditions, depending upon the varying yields of
the main toxic species22. In practice the lethal toxic potencies for common materials over
different combustion conditions vary by a factor of around 100, while one group of materials can
have a toxic potency approximately 1000 times greater than average under certain thermal
decomposition conditions5,23.
There are therefore two major problems in attempting to derive toxicity performance from
existing small-scale toxicity test methods.
• Firstly most test methods reproduce only a single combustion condition, for which the
relationship to combustion conditions existing in any particular type of stage of full-scale
fires has often not been established
• Secondly the concept of an overall toxicity value for the mixed combustion products from
any material or product has little meaning in relation to full-scale fire hazards, because it
ignores the time-based nature of both fire atmospheres and the effects of the different toxic
components.
In order to evaluate the fire safety of any built environment, or any combustible material or
product used in a built environment it is therefore necessary to consider the overall fire
performance of the system, either by performing a full-scale fire test (or series of tests) on the
end-use system or by modelling the full-scale fire conditions using suitable fire dynamics models
run with input data on a range of performance parameters.. It is then necessary to apply
suitable algorithms to calculate the time at which different physiological ASET endpoints occur.
8
PMMA Phi vs. LC50 concentrations for some materials
100 100 PMMA

PMMA @12 and 10% O2
90 wood
wood @12 and 10% O2
polyamide
Rat 30-min LC50 concentration (g/m3)

80 polyamide @12 and 10% O2
Velour fabric
70 PU foam
CO yield vc/CO yield w v
boucle fabric
Boucle-FR
60 PVC
MDF
10 50 MDF@12and 10%O2
LD PE
40
Factory m utual
apparatus 30
BRE tube
furnace 20
10
1
0
0.1 1 10
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Equivalence ratio Phi
Copyright HER
Figure 2: Variation of CO yield with Figure 3: Variations in lethal toxic
combustion conditions in the potency (LC50) concentrations with
ASTME2058 and ISO TS1900 test combustion conditions in the ISO
methods
Methods for estimating time to incapacitation for the main toxic gases, smoke and heat in fires
(Fractional Effective Dose models [FED models]) have been in existence for somewhat longer
than suitable fire dynamics models have been available for their application, but the increasing
sophistication of fire modelling methods now makes the wider application of such methods
feasible in addition to their direct application to measured results of full-scale fire tests
This paper outlines the physiological basis of the FED models for which a detailed description is
provided in the SFPE Handbook5 and other references, and a more simplified form in ISO
1357117. Some options are then presented on suitable approaches to product evaluation in
relation to potential toxic hazards with specific comments on cable fires
METHODS OF FIRE HAZARD ANALYSIS BASED UPON PHYSIOLOGICAL

ALGORITHMS
Fire hazard assessment is based upon a “step through” approach whereby the extent of the
hazard is calculated for each successive minute (or other appropriate time interval) during the
fire, until the point is reached where different hazard endpoints are predicted
The behavioural and physiological effects of exposure to toxic smoke and heat in fires combine
to cause varying effects on escape capability, which can lead to physical incapacitation and
permanent injury or death.
For a basic fire engineering design there are two main considerations:
• The time at which the concentration of effluents reaches a level such that safe escape is
compromised due to:
a) behavioral and physiological effects of exposure to heat and toxic smoke on escape
capability;
b) in the absence of direct exposure, behavioural effects caused by seeing fire effluents
on escape behaviour.
9
The effects of exposure to smoke result from impaired vision due to the optical opacity of
smoke and from the painful effects of irritant smoke products and on the eyes and
respiratory tract. Occupant movement speed and wayfinding ability is impaired.
Behavioural effects of seeing or being immersed in smoke result in a proportion of
occupants being unwilling to approach smoke or heat-logged areas or escape routes.
• The time at which the exposure dose of asphyxiant toxic gases or heat reaches a level
at which occupants is are likely to become incapacitated - such that they cannot save
themselves and are likely to die within a minute or so unless rescued.
In a regulatory or design evaluation context several different considerations exist. In flaming
fires the effluent is buoyant, so tends initially to fill enclosures from the ceiling downwards. This
presents a potential opportunity for occupants to escape before the smoke layer descends to
near head height, but occupants are still exposed to heat radiation from the hot fire effluent in
the upper layer. Larger building such as shopping malls are often designed with smoke venting
or extraction to maintain the upper layer above a minimum clear layer height of 2.5 m with a
maximum upper layer temperature of 200°C. Occupants are considered to be willing and able
to escape in clear air under such a layer and the downward heat radiation is considered
tolerable. Above this temperature the downward radiation is capable of causing pain and burns
within a short period. While such layering may be maintained for some minutes in large open
structures, in most fires some degree of smoke mixing to near floor level occurs as effluent
moves through building spaces and as the fire grows.
EFFECTS OF EXPOSURE TO NON-IRRITANT AND IRRITANT SMOKE
Smoke contains a mixture of gases, liquid droplets and solid particles capable of exerting a
range of adverse physiological effects on escaping occupants. The size and composition of
soot particles varies with the nature of the burning fuel and the combustion conditions. The main
solid component under well-ventilated combustion conditions is soot particles comprised mainly
of carbon. These cover a range of sizes but a considerable proportion is small and highly
respirable. As combustion becomes less efficient these particles become mixed with a wide
variety of organic compounds many of which are irritant to the eyes and respiratory tract. These
compounds adhere to the particles and may also be present in vapour or liquid droplet form. If
the burning fuel contains hetero-elements such as halogens, nitrogen or sulphur, then irritant
acid gases are also present5.
The first physiological effect: impairment of visibility, occurs even before a subject is immersed
in smoke. Seeing smoke-logged areas, for example in escape routes, may impair escape
behaviour and wayfinding.
In situations where smoke is mixed down to near floor level, some building occupants may move
through dense smoke in some situations, but in other situations people may be unwilling to enter
smoke logged escape routes, turn back or be unable to find an exit. Where heat is not an issue,
the immediate effects of smoke depend upon the visibility distance, and the sensory irritancy of
the smoke if people are exposed directly. For such situations it is necessary to set tenability
criteria for design purposes, depending upon the level of adverse effects on occupants
considered acceptable or non-acceptable
In a number of studies of fires in buildings, a proportion of people (approximately 30 %) were found
to turn back rather than continue through smoke-logged areas5,24, 25,26. The average density at
which people turned back was at a “visibility” distance of three metres. This represents an optical
density (OD·m-1) of 0.33, (extinction coefficient 0.76) and women were more likely to turn back
than men. A difficulty with this kind of statistic is that, in many fires in buildings, there is a choice
between passing through smoke to an exit or turning back to take refuge in a place of relative
safety such as a closed room. In some situations, people have moved through very dense smoke
when the fire was behind them, while in other cases people have failed to move at all.
10
Behaviour might also depend on whether layering permits occupants to crouch down to levels
where the smoke density is lower and whether low-level lighting is used to improve visibility.
Based upon considerations such as these in relation to parameters such as the size and
complexity of the building, it is possible to set design limits for optical density of smoke. As an
approximate guide it might be assumed that occupants will not use an escape route if the
visibility in that route is less than three metres (OD/m = 0.33, extinction coefficient 0.76).
However, if they enter an escape route contaminated to this optical density and become
exposed to the smoke, then their ability to progress depends upon both the optical density and
the irritancy of the smoke.
Figure 4 shows the effects of exposure to non-irritant smoke and irritant wood smoke on walking
speed, and walking speed in darkness, derived from the work of Jin8. For Jin’s study, volunteers
walked along a corridor filled either with non-irritant theatrical smoke or irritant smoke generated
using a bee smoker. The fuel consisted of wood chips, and although the exact decomposition
conditions are unknown, they most likely involved a mixture of smouldering and non-flaming
oxidative decomposition. This would produce a smoke containing a range of irritant organic
species at significant concentrations. For non-irritant smoke there is a more or less linear
relationship between walking speed and OD/m, and at around 0.55 (representing about 2
meters visibility), peoples’ behaviour was observed to change from walking as if in the light to
walking in darkness. Under these conditions people stopped walking directly along the corridor
and started to feel their way along the walls using their hands. In an unfamiliar building subjects
move very slowly for fear of tripping over obstacles or falling down a stair.
An important aspect of Jin’s experiments is the difference between walking speed in non-irritant
and irritant smoke. As Figure 4 shows, for a given optical density, walking speed was much
slower in irritant smoke, with people walking as if in darkness at an optical density of around 0.2.
A further problem in smoke is not just the visibility through the smoke, but the reduction in
overall illumination. Lighting, including warning signs, are placed mainly at high level, so that
when a dense smoke layer forms under the ceiling, the conditions in a building can become
quite dark.
Walking speed against smoke optical density

1.2
non irritant smoke
1 irritant wood smoke
Walking speed (m/s)
0.8
30% of people
turn back rather
0.6 than enter
0.4
wallking speed in
darkness
0.2
Visibility (m): 10 5 4 2.8 2.3
0
0 0.1 0.2 0.3 0.4 0.5 0.6
Smoke OD/m
Figure 4 — Walking speeds in non-irritant and irritant smoke

Equations 1 and 2 for the relationship between walking speed and smoke optical density (OD·m-
11
1
), (together with all other physiological equations) are presented in the last section of the paper
.
Based upon the finding that people move as if in darkness at a visibility of five metres in irritant
smoke, and that smoke from most fires contains a variety of irritant chemical species, the
generic design tenability limits shown in Table 3 are proposed, on the basis that concentrations
exceeding these levels could impair or even prevent occupants’ safe escape (for situations
where the concentrations of acid gases are considered unlikely to be significant due to the fuel
composition):
Another advantage of using these smoke concentration limits as tenability criteria is that the
concentrations of asphyxiant gases present are most unlikely to exceed levels capable of causing
incapacitation (loss of consciousness) in exposed occupants within 30-60 minutes.
Table 3 — Smoke tenability limits
Smoke density and Irritancy Approximate visibility

Reported effects
OD·m-1 (extinction coefficient) diffuse illumination
None Unaffected Walking speed 1.2 m/s
0.5 (1.15) non-irritant 2m Walking speed 0.3 m/s
0.2 (0.5) irritant reduced Walking speed 0.3 m/s
0.33 (0.76) mixed 3 m approx. 30 % people turn back rather than
enter
Suggested tenability limits for buildings with: OD·m-1 = 0.2 αk 0.5 (visibility 5 m)
– small enclosures and travel distances; OD·m-1 = 0.08 αk 0.18 (visibility 10
– large enclosures and travel distances. m)
(where OD/m is log10 (Io/I), the logarithm of the ratio of the intensities of light transmitted over a
pathlength of 1 m from a light source to a receiver in the absence and presence of smoke,
respectively. The light extinction coefficient αk is ln (Io/I)).
For situations where smoke is expressed in terms of particulate mass concentration, these
equate to approximately 0.7 and 0.3 g particulates/m3 respectively (where particulates g/m3 ~
0.356 x OD/m).
Although these generic values are useful, a problem with the prediction of the effects of irritant
smoke on walking speed is that the irritant composition, and hence the level of irritancy, is very
dependent upon the composition of the fuel and the decomposition conditions. Above certain
concentrations, it is considered that exposure to irritant gases in smoke will severely impair and
even prevent escape. For the majority of flaming fires, it is considered that the concentrations of
mixed smoke irritants will be below this level provided the smoke optical density does not
exceed OD·m-1 = 0.2. Exceptions could be smouldering fires, for which the organic irritant yields
tend to be high, and fires involving fuels giving off significant yields of inorganic acid gases (HCl,
HBr, HF, SO2, NOx). Between zero and the concentration causing incapacitation there will be a
relationship between the irritancy of the smoke and walking speed (as demonstrated by Jin). In
order to provide some indication of possible effects on walking speed between these limits
expressions have been developed for any irritant based upon the concentration estimated to be
very painfully irritant. The model is based upon a concept that at low concentrations an
increase in irritancy will have a relatively minor effect as does smoke (for example walking
speed in 10 m visibility smoke should be the same as in 100 m visibility smoke). There is then a
middle range, over which an increase in irritancy is likely to have large effect on walking speed,
and then a point where walking is slow and further increases in irritant concentration have less
incremental effect. This concept is illustrated in Figure 5. This shows a general case for the
12
effect of exposure to any irritant gas or mixture or irritant gases on fractional walking speed.
The x-axis show the fractional irritant concentration (FIC) where FIC = 1 represents
incapacitation (e.g. 1000 ppm HCl) (see equations 3 and 4).
Unlike the situation for non-irritant smoke, the curve reaches a fractional speed of zero when
FIC is 1.0. This is because painful effects on vision and breathing are predicted to be of
sufficient severity to cause incapacitation and cessation of effective escape movements.
1
0.9
0.8
Fractional Walking Speed
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 0.2 0.4 0.6 0.8 1
FIC
Figure 5: Estimated relationship between fractional walking speed and FIC for a sensory
irritant
PREDICTING TIME TO INCAPACITATION BY IRRITANTS, ASPHYXIANT

GASES AND/OR HEAT
Although the above smoke density limits may be used as design limits, they do not
represent conditions under which escape is necessarily prevented. A design calculation should
also evaluate predicted time to incapacitation due to the effects of heat or toxic gases. It might
also be important to know the time between the visibility criteria being exceeded and that when
incapacitation is predicted, especially for probabilistic design. The methods presented here are
designed to predict time to incapacitation for an average person, after which a factor may be
applied to allow for more susceptible members of the occupant population.
Time to loss of tenability due to smoke, irritants, asphyxiant gases and heat in fires is calculated
using Fractional Effective Dose (FED) methodology as described in the following sections.
Fractional Effective Dose methodology for hazard analysis

As stated, some toxic or physical effects of exposure to combustion products occur almost
immediately on exposure and the severity of the effect is proportional to the concentration of the
substance and its potency. This applies to visual smoke obscuration and to the painful effects of
exposure to irritants. For example, irritant smoke in the eyes or nose immediately causes pain
(sensory irritation), reflex closure of the eyes and breathing difficulties.
For other substances, such as asphyxiant gases, the effect depends upon the dose inhaled.
The effects therefore take some time to develop and depend upon the concentration inhaled
and the time over which it is inhaled. The effects tend to more persistent that those of sensory
irritation, since it take some time for the toxic material inhaled to be detoxified (for example
13
hydrogen cyanide) or expelled (for example carbon monoxide) An example of dose-dependent
effects is collapse from asphyxia resulting from exposure to carbon monoxide.
In practice, for asphyxiant substances (and to some extent also for heat) a distinct threshold
concentration or exposure dose can be identified at which serious effects are predicted. This is
known as the effective concentration or exposure dose for a given toxic or physiological
endpoint (for example, the exposure dose of CO required to cause loss of consciousness or the
exposure dose of heat required to cause skin pain). For application to toxic hazard calculations
the concept of Fractional Effective Concentration, or Dose, is used whereby the exposure
concentration or dose at any point during a fire is expressed as a fraction of the exposure
concentration or dose predicted to produce a given effect. For example the concentration of
smoke present at any time during a fire can be expressed as a fraction of the concentration
required to seriously impair escape capability.
Thus FECsmoke (Fractional Effective Concentration of smoke) = 1 represents a smoke

concentration considered capable of seriously limiting escape capability
where FECsmoke = Concentration (of smoke) present in a fire at any time divided by the
concentration considered to significantly affect escape efficiency
The exposure dose of CO can be expressed as a fraction of the exposure dose predicted to
cause incapacitation .
Thus FIco (fraction of an incapacitating dose of carbon monoxide) can be expressed as:
conc. gas present × time 1000 ppm CO × 20 min

FIco = e.g. FIco = = 0.57
conc. × time for incapacitation 35,000 ppm.min
Although the FED calculations for different asphyxiant gases are based upon this simple
concept, they are in practice more complex for a variety of physiological reasons, but the user is
required to know only the concentration and exposure duration to perform the FED exposure
dose calculation.
Another reason for expressing the concentration or dose as a fraction of an effective

concentration or dose for each toxic product is that fire atmosphere contains a mixture of toxic
products of differing potencies. In order to sum the effects of the different effluent components
it is necessary to normalize them in terms of the effective dose. Furthermore, the
concentrations of toxic products change with time during a fire, so that for constituents whose
effects are dose-related, it is necessary to calculate the effective doses received, based upon
the concentrations averaged over short periods of time, and then integrate these over
successive periods. The aim is to calculate the time at which the summed effective doses
reach unity, at which point the endpoint (such as incapacitation) is predicted to occur (see
equation 5).
Effective concentrations and exposure doses for defined endpoints for smoke, heat and toxic
effluent mixtures are presented in the following sections.
Effects of irritant fire effluents
Irritant chemical products in smoke constitute one of the parameters know to affect people in
fires in several adverse ways5,7; affecting behaviour and use of escape routes, reducing travel
speeds, causing painful incapacitation, and threatening post-exposure survival by effects on the
lungs (bronchopneumonia, bronchiolitis, inflammation and oedema). It is therefore important to
consider all these effects in a hazard analysis, taking into account that effects delaying or
14
preventing escape, while they may not in themselves be fatal, may lead to death from exposure
to heat or asphyxiant gases. It is also necessary to consider the time-concentration curves for
irritant and other gases during the course of a fire. If the material properties or combustion
conditions are such that certain irritant compounds are not formed, or are rapidly cleared, or not
released at a stage when occupants may be exposed, then a potential irritant hazard may not
develop into a real one or may subside.
A number of attempts have been made to provide guidance on effects and acceptable limit
exposure levels to humans for irritants both in the context of industrial hygiene and emergency
planning, and in the context of fire safety. The industrial context differs somewhat to that for
fires, in that industrial concerns tend to relate to exposures to individual irritant gases, and
where the specific gas constitutes the hazard. In the fire context a mixture of irritant gases and
smoke particulates is present and the hazard consists of the combination of smoke and irritant
gases with asphyxiant gases and heat. A detailed description of the effects of irritants and
tenability limits is presented in Purser5 in the SFPE Handbook. Industrial and emergency limits
are usually expressed in terms of concentration over a given exposure period (US ERPG -
Emergency Response Planning Guidelines27, AEGL – US EPA Acute Exposure Guidance
Levels for Hazardous Substances)28. These take into account both immediate irritant effects
and lung damage. In the fire safety context, guidance has been provided on separate
phenomena for irritant fire effluent mixtures by Purser: setting ASET tenability limits for
impairment of escape capability and incapacitation, estimating the relationship between irritant
smoke concentration and walking speed (RSET) and estimation of lethal exposure doses for
irritant smoke5,7. Tenability limits and calculation methods have also been set in British
Standard BS 7899-2 (1999)29, with guidance on estimation of levels of irritancy likely to inhibit
escape or cause incapacitation, lethal exposure doses for smoke irritant mixtures and effects of
irritant smoke on walking speeds. ISO 13571 addresses only one aspect – concentrations of
mixed irritants estimated to cause incapacitation (prevent escape).
Setting Tenability Limits For Painful Sensory Irritant Effects

The basic effects and mechanisms of acute sensory irritation are similar in mammals30,31,32,33. In
humans, irritation is characterised by an irritant or painful burning sensation in the upper
respiratory tract (nose, mouth throat and chest), and also the eyes and skin of the face. There
is a reflex decrease in breathing rate, with pauses during which the breath is held in, and there
may be some bronchoconstriction. Irritation of the cornea can result in blepharospasm (reflex
eye closure). Setting tenability limits depends upon what degree of pain and distress is
considered to constitute acceptable limits. In the firesafety context Purser 20025 and BSI29
recommend two possible ASET limits for consideration for any specific irritant:
• A concentration predicted to impair escape of an average person
• A concentration predicted to cause incapacitation of the average person.
ISO13571 specifies concentrations of each irritant “expected to seriously compromise the ability
of most occupants to escape”. Allowing for the range of susceptibility in the human population
a safety factor of 0.3 x limit concentration is recommended for normal populations.
In order to estimate limiting concentrations for any specific irritant it is necessary to consider the
relationship between exposure concentration and the severity of the effects. In order to achieve
this, consideration has been given to published data on sensory irritant effects in humans and
animal models (rodents and primates) for a variety of irritant substances. A difficulty with
achieving precise predictions of effects is that it is not possible for ethical reasons to carry out
experimental studies on humans. The same stricture applies to a lesser extent to animal
studies, especially those using non-human primates, and with animal studies there is the added
difficulty of extrapolating from animal data to predicted effects in humans. The resulting data-
base consists of limited human data from experimental and accidental exposures to individual
irritant gases and irritant fire effluent, some very limited experimental data from experiments in
non-human primates, and more extensive data on experiments using rats and mice. Very few of
15
these experiments are related to the endpoint of interest – compromise of escape capability.
One helpful consideration is that the range of effects of most sensory irritants are very similar,
so that detailed understanding of the effects of any one irritant can be used to predict those of
another within certain limitations. The main difference between different sensory irritants is
their potency, so that for a given degree of painful stimulation, different concentrations are
required for different individual substances. If the differences in potency of different substances
are measured, then the relationship between concentration and severity of effects can be
estimated for any specific irritant.
Mouse RD50 model
A good physiological model for measuring both the potency of any specific irritant and the
severity of irritant stimulation is the mouse Respiratory Rate depression (RD50) test, which
makes use of the reflex decrease in breathing rate when exposed to an irritant atmosphere,
which is common to humans and rodents. This reflex is particularly well developed and stable in
rodents, especially mice, such that the percentage decrease in breathing rate is proportional to
the log of the inhaled concentration.At this concentration the effects on breathing rate are almost
instantaneous, the rate decreasing to around 30% of the pre-exposure level. Similar effects in
primates breathing thermal decomposition products from wood, polypropylene and flexible
polyurethane foam are shown in Purser15. At concentrations of most irritants producing a high
intensity sensory irritant response, the effects occur almost instantaneously and are
concentration related as anyone who has experienced the following will be able to confirm
(eating a hot chilli or rubbing your eyes with chilli on your hands, bonfire or barbeque smoke
blowing into the face, exposure to CS riot control gas)5,7.
For most irritant substances, human exposure at the mouse RD50 concentration is reported as
being very painful and disruptive. Table 4 shows the wide range of irritant potencies for
different substances and the similarity with the mouse RD50. A particularly good example is one
sensory irritant for which the effects on symptoms and behaviour have been studied
experimentally in humans – CS riot control gas. For CS, the RD50 concentration is 0.52 ppm
and human exposure to 0.5 ppm is reported as very painful .This relationship is generally
applicable, so that at concentrations for most irritants around the mouse RD50, severe effects on
behaviour and escape capability are likely to occur. CS is know to inhibit behaviours such at
those important for escaping from a building during a fire incident, and there have been a
number of incidents where CS released inside occupied buildings, resulted in severe effects on
occupants, leading to deaths from crowd crush injuries during escape attempts. This is
therefore not to say that the symptoms and behavioural effects in mice are identical to those in
humans, but that the RD50 model provides a good analytical tool for the assessment of irritant
potency in humans.
In estimating concentrations for different irritants likely to cause severe disruption of escape
capability other data have been taken into account in addition to the mouse RD50. Table 3
illustrates the range of data available for HCl. Based upon these data it is considered that
between 100 and 1000 ppm the effects on escape capability from an occupied building are likely
to be severe (remembering that severity is proportional to log concentration). However for HCl
and acrolein experiments have been performed in which baboons were capable of escaping
from an exposure chamber using a simple door lever after 5 minutes exposure to concentrations
of HCl as high as 15,000 ppm. Taking into account differences in sensitivity between humans
and other primates, and the more complex scenario of escaping form a crowded building in
irritant smoke, it is suggested that around 1000 ppm HCL would be likely to seriously impair
escape capability and likely to cause incapacitation due to severe pain and breathing difficulties
in most people, while around 300 ppm most people would suffer significant effects on escape
capability and behaviour such as entering or turning back from contaminated escape routes.
16
Table 4: (for details see reference 4)
Sensory and Pulmonary Irritancy of Combustion Products
Irritant RD50 (ppm) Severe sensory 30-minute LC50 LC50/RD50
mouse irritancy in humans (ppm) mammal
(ppm)
0.1-1.0
Toluene diisocyanate 0.20 1.0 100 500
ο-chlorobenzylidene
– malonitrile (CS)* 0.52 0.5 150-400 529
α-chloroacetophenone (CN)* 0.96 6-50 300-400 365
1.0-10
Acrolein 1.7 1-5.5 140-170 91
Formaldehyde 3.1 5-10 700-800 242
Chlorine 9.3 9-20 100 11
10-100
Crotonaldehyde 4-45 200-1500
Acrylonitrile >20 4,000-4,600
Penteneone 1,000
Phenol >50 400-700
100-1,000
SO2 117 50-100 300-500 3
NH3 303 700-1700 1,400-8,000 16
HF 120 900-3,600
HCl 309 100 1,600-6,000 12
HBr 100 1,600-6,000
NO2 349 80 60-250 0.4
Styrene 980 >700 10,000-80,000 46
Acetaldehyde 1,000-10,000
4946 >1,500 20,000-128,000 15
10000-100000
Ethanol 27,314 >5,000 400,000 15
Acetone 77,516 >12,000 128000-250000 3
The potential for causing sensory irritation spans six orders of magnitude, while that for causing death
spans approximately three orders of magnitude. For substances down to NO2 death is likely to be
due to lung irritation, while for the remainder from styrene to acetone death is likely to be due to
asphyxiation
*Substances not detected in combustion atmospheres. RD50 concentrations from Alarie
LC50 concentrations have been normalised to a 30-minute exposure time according to Haber’s rule
Copyright HER
17
Table 5: Irritants effects of exposure to hydrogen chloride
ppm Effect
1.8 AEGL-1 34
3 ERPG-1 Maximum concentration human exposure up to one hour without experiencing
other than mild transient health effects of a clearly defined objectionable odour35
<5 Minor nasal irritation can be detected below 5 ppm (the OEL)
20 ERPG-2 Maximum concentration human exposure up to one hour without impairing
ability to take protective action or develop serious health effects
43
50 AEGL-2 for 30 minute exposure
10 - 50 IDLH (immediately dangerous to life and health)36
Perceived as irritant, but work is possible at up to approximately 50 ppm.
50 - 100 Strongly irritant, and some people report exposure to 100 ppm as being
excruciatingly painful to the eyes and respiratory tract37.
100
ERGP-3 Maximum concentration human exposure up to one hour exposure not
140 life threatening. Also AEGL-2 for 10 minute exposure and AEGL-3 for 1 hour
190 Incapacitation in guinea pigs after 16.5 minutes
200 No obvious signs in baboons during 5 minute exposure
210
300
Purser SFPE Handbook – predicted to impair escape in average person5
309
500 AEGL-3 (lethal) human 30 minutes
ISO-13571 maximum concentration to prevent human incapacitation (allowing for
620 sensitivity variations in the population)17
700 Mouse RD5034
Hyperventilation (lung irritation) in sedated baboons and post exposure chronic lung
900 injury after 15 minute exposure
AEGL-3 (lethal) human 10 minutes

Severe upper respiratory tract and eye cornea damage following 15 minute exposure -
rats38,
Purser SFPE Handbook – predicted to cause incapacitation average person5

800-1000 Signs of severe irritancy in baboons during 5 minute exposure including eye rubbing,
profuse salivation and blinking. Able to perform shuttle box escape
1000
1000-2000 ISO-13571 maximum concentration to prevent human incapacitation in average
1095 person
1293 Brief exposure regarded as being dangerous to lethal to humans
30-minute LC50 tracheal cannulated mice
30-minute rat exposure (nose or mouth breathing) severe necrotizing rhinitis, turbinate
2810 necrosis, psudomembrane formation, ulcerative tracheitis, necrosis, polymorhonuclear
3800 leucocyte infiltration down to alveoli
5000 60-minute rat LC50
15,000 30-minute rat LC50 (representing an exposure dose of 114,000 ppm.min)
40% decrease in arterial blood oxygen in sedated baboons

5 minute lethal exposure concentration in rats and baboons is around 15,000 ppm
but baboons able to perform shuttle box escape from a chamber after 5 minute
exposure
18
Thresholds and variations in sensitivity in humans
Based upon all this information, what can be developed to predict effects in humans during
fires? It is important in this respect to recognise that all people are not the same, and that a
considerable range of sensitivities can be expected to occur human subjects. Also, although the
severity of irritant effects lies on a continuum related to the exposure concentration, it might be
possible to recognise to important thresholds for effects related to fires:
• A concentration capable of seriously impairing escape capability and movement speed
• A concentration capable of causing incapacitation, such that the subject effectively
cannot move.
The threshold concentrations shown in Table 4 are proposed for common fire irritants likely to
severely affect escape capability in most humans. In order to allow for more sensitive
individuals it is recommended that design limit threshold of 0.3 x these levels might be used.
Since fire effluents contain a mixture of irritants it is necessary to consider how they work in
combination. It is currently recommended that they should be considered additive (see equation
6).
Post-exposure lung inflammation and survival
In addition to incapacitating effects of sensory irritation, affecting escape capability, a proportion
of inhaled irritants penetrates to the deep lung, and when a sufficient exposure dose is
accumulated this can lead to inflammatory processes which can be fatal over periods of several
hours to several days after exposure. A guide to concentrations and exposure doses of
common irritant gases likely to cause incapacitation at the scene or death following exposure is
given in Table 6 (see equation 7).
Table 6: Exposure concentrations and exposure doses for incapacitation and lethal lung damage
Gas Concentration concentration Exposure doses
predicted predicted to cause predicted to be lethal
to impair escape incapacitation to half the population
(ppm) (ppm) (ppm.min)
HCl 200 1000 114,000
HBr 200 1000 114,000
HF 200 500 87,000
SO2 24 150 12,000
NO2 70 250 1,900
NO - >1000 ~30,000
CH2CHO (acrolein)* 4 30 4,500
HCHO 6 250 22,500
(formaldehyde)*
* where the concentrations of acrolein and formaldehyde (or other important irritants) are
unknown, a term derived from smoke density a term derived from smoke density 0.5
OD/metre may be used as an indication of irritancy likely to impair escape efficiency.and 90
OD/metre.min may be used as an indication of lethal organic irritant exposure dose
Tenability limits and hazard calculations for asphyxiant gases

The main cause of incapacitation and death during and immediately after fires is exposure to
asphyxiant gases. Incapacitation results from loss of consciousness due to the combined
hypoxic effects of carbon monoxide, hydrogen cyanide and carbon dioxide, with some additional
effects from low oxygen hypoxia, nitrogen oxides and inhaled irritants. Loss of consciousness
prevents escape and further uptake of asphyxiants while comatose is likely to result in death
within a further minute or so unless the occupant is rescued. The most useful tenability
endpoint to work to is therefore considered to be loss of consciousness (FEDIN = 1), with design
limits set to prevent this occurring. Since individual susceptibility varies in the population, this is
19
predicted to represent the median of the distribution of exposure doses resulting in collapse.
Approximately 11.3% of the population is considered likely to be susceptible below an FED of
0.3. (see ISO 1357117). It will be necessary for the designer or regulatory authority to select an
FED level suitable to protect vulnerable sub-populations in the chosen application (for example
0.3 or some other value depending upon the application).
The effects of combinations of asphyxiant gases causing incapacitation in fires are considered
to be approximately additive, but a number of interactions need to be considered5:
• The FED for CO and HCN are considered directly additive as has been demonstrated
experimentally
• NO and NO2 (designated as NOx in mixtures) also act as asphyxiants, reducing oxygen
carriage in the blood due to the conversion of haemoglobin to methaemoglobin. To this
extent their asphyxiant effects can be considered additive with those of HCN and CO.
However, methaemoglobin combines with HCN in the blood, thereby reducing its
asphyxiant effect. NO2 is also a potent lung irritant.
• The effects of irritants on lung function also cause some hypoxia and so an additive term
is included consisting of the FLDirr
• The main effect of carbon dioxide is to increase the breathing rate and thus the rate of
uptake of CO and HCN. A multiplicatory term VCO2 is used to calculate this effect.
• Low oxygen hypoxia will be additive with the overall effects, but is not increased by
VCO2 (in fact it is improved)
• The direct intoxicating effects of CO2 are considered unlikely to occur before other
effects so are normally ignored.
The overall equation for these relationships is Equation 8. For each of these gases it is
necessary to obtain an expression for the fraction of a dose required to cause incapacitation.
The derivation of these expressions is detailed in the section on asphyxiation by fire gases in
the SFPE Handbook5. Basically, the exposure dose acquired over any period of time during a
fire is expressed as a fraction of the dose required to cause incapacitation for each asphyxiant
component. These are then summed and corrected for VCO2 to provide an overall FIN for each
time period.
A complication is the extent to which the dose response effects of each gas follow Haber’s rule
and where it is necessary to allow for deviations from this oversimplification used by all small-
scale toxicity test methods. Haber’s rule is that any given physiological or toxic endpoint (for
example loss of consciousness due to asphyxiation), occurs at a constant c x t exposure dose,
irrespective of whether the subject has been exposed to a high concentration for a short period
or an equivalent lower concentration exposure for a longer time. This effect is illustrated in
Figure 6, which shows time to incapacitation for primates exposed to constant concentrations of
CO or HCN for different exposure periods of up to 30 minutes. For exposures to CO, the
animals were largely unaffected until they achieved and exposure dose of approximately 27,000
ppm.minutes, at which point they showed signs of confusion and lethargy followed by loss of
consciousness within less than a minute. The exposure dose for incapacitation was constant for
exposure concentrations of 1000-8000 ppm, providing the smooth curve shown. This
contrasts strongly with the effects of exposure to HCN, whereby exposure to concentrations of
around 200 ppm and above resulted in loss of consciousness within two minutes (c x t = 380
ppm.minutes), while a much large exposure dose over a long period was required at lower
concentrations (c xt = 2610 ppm.minutes at 87 ppm). Cyanide is also much more potent that
CO by a factor of around 20 or so.
20
8000 800
CO HCN
7000 c t ct c t ct 700
900 30.00 27000 87 30.00 2610
1000 26.60 26600 98 19.1 1872
6000 600
2000 14.05 28097 151 8.5 1284
4000 6.72 26868 200 1.9 380
5000 8000 3.26 26086 300 0.9 270 500
HCN (ppm)
CO (ppm)
4000 400
Carbon monoxide
3000 300
Hydrogen cyanide
2000 200
1000 100
0 0
0 5 10 15 20 25 30
Time to loss of consciousness (minutes)
Figure 6: Time to incapacitation for cynomolgus monkeys exposed to different

concentrations of CO or HCN. CO exposures were for active animals exercising in a
chamber, HCN exposures in resting animals
This behaviour of HCN is important because it means that a short exposure to a high
concentration (above around 150 ppm) will cause rapid collapse of a fire victim, who then
remains incapacitated, inhaling more HCN and CO which may then lead to death. This
variation of exposure dose required for a given endpoint is yet another reason why use of small-
scale toxicity test methods or toxicity indices based upon a constant exposure period is likely to
lead to error, while the time-based FED methodology allows for these effects in calculating time
to incapacitation. Similar dose-response effects deviating from Haber’s rule occur for low
oxygen hypoxia and heat5
Another important aspect of time to incapacitation for fire victims, especially with regard to CO
intoxication, is the level of physical activity. A person at rest breathes only around 8.5 litres of
air each minute, while a person engaged in light activity such as walking breathes around 25
l/min, so the rate of CO uptake is much more rapid in active subjects. Also, due to the energy
and oxygen demands of physical activity, an active subject is likely to collapse at a lower blood
%COHb level than a sedentary subject.
However, the effects are based on data for healthy young adult animals or humans. The
exposure dose or concentration causing incapacitation therefore represents the maximum in a
statistical distribution of subjects’ responses surrounding that exposure dose or concentration,
that is, the mode, or most frequently expected exposure dose for an exposed population.
Individual exposure doses or concentrations for the response would, in practice, be statistically
distributed around the mode in a probability curve. The overall human population contains a
number of subpopulations, which exhibit greater sensitivity to various fire effluent toxicants,
principally due to compromised cardiovascular and pulmonary systems.
21
35
Fire deaths Non-fire deaths

30
25
% of all deaths
20
15
10
0
<30 30-40 40-50 50-60 60-70 70-80 80-90 90-100
% COHb range ©HER
Figure 7: Distribution of fatal %COHb in non-burned fire victims and non-fire CO

poisoning cases (After Nelson)
This variability of susceptibility in the population is illustrated in Figure 7, which shows the
distribution of %COHb in non-burned fire victims and non-fire CO poisoning cases in the United
States (data compiled by Gordon Nelson). The Figure illustrates the range of sensitivity in the
exposed human population. The mode is between 70-80% COHb. Note that the fatal levels in
fire victims are somewhat lower than for CO poisoning cases. This may reflect the effects of
other toxic fire gases such as hydrogen cyanide.
Two of the largest such subpopulations are the elderly and the approximately 15 percent of
children and 5 percent of adults who are asthmatic. The elderly, and particularly those with
impaired cardiac perfusion, are especially susceptible to asphyxiant gases. Thus the average
lethal carboxyhemoglobin (COHb) concentration in adults dying in fires or from accidental CO
exposure is lower in the elderly. The results in Figure 7 show that some individuals died at
COHb concentrations below 30 percent while others survived long enough to obtain blood
concentrations above 90 percent COHb. Many fire fatalities occur at lower COHb concentrations
than for cases of CO poisoning alone. This may partly reflect the influence of other toxic gases
in addition to CO in fire atmospheres. Also, it has been shown in experimental studies that as
little as 2 percent COHb significantly reduces the time to the onset of pain in an exercise test of
angina sufferers. This could be very important when attempting to escape from a fire. A further
complication with the CO lethality data is that in practice few people survive an exposure of
more than 50 percent COHb if rescued and treated, even though much higher levels are found
in the bodies of decedents. This is because uptake continues in comatose people until the point
where respiration ceases. This is illustrated in data from a study by Pach39 (Table 5), which
shows the proportions of survivors and fatalities from a sample of 260 CO poisoning cases. The
data show that survival is rare above 50%COHb, around 0.67 in the 40-50%COHb range, and
that most people survive below 40%COHb. Equations for calculating the FEDs for CO and each
of the other gases are numbers 9-12.
22
1
0.9
0.8
0.7
Proportion surviving
0.6
0.5
0.4
0.3
0.2
0.1
0
0 10 20 30 40 50 60 70 80 90 100
%carboxyhamoglobin ©HER
Figure 8: Proportions of survivors and fatalities in different COHb ranges from a sample
of 260 CO poisoning cases (after Pach 1978)39
Tenability and hazard calculations for development of pain, incapacitation,

injury and death from exposure to heat and burns
There are three basic mechanisms whereby heat exposure may lead to incapacitation and
death in fires:
• Heat stroke (hyperthermia)
• Skin pain followed by body surface burns
• Respiratory tract burns
Hyperthermia involves prolonged exposure (approximately 15 minutes or more) to heated
environments at ambient temperatures too low to cause burns. Under such conditions, where
the air temperature is less than approximately 121°C for dry air or 80°C for saturated air, the
main effect is a gradual increase in body core temperature. Hyperthermia leads to confusion
and collapse. Exposure to air temperatures above approximately 121°C (or to radiant heat
fluxes above 2.5 kW/m2) leads to pain to exposed skin followed by body surface burns and
hyperthermia if exposure is prolonged. Respiratory tract burns can also occur if exposure to
heated air is sufficient to cause facial burns. For a basic engineering design it is proposed that
a tenability endpoint for exposure to heat should be taken as the limits of heat tolerance due
either to the onset of hyperthermia or to severe pain to exposed areas of skin (such as the head
and hands). It is proposed that under such conditions exposed building occupants may be
unable to escape (if affected by hyperthermia) or unwilling to move through painfully hot
environments. It is known that in some fires people have moved through (and escaped from)
scenarios under conditions hot enough to cause severe burns. It is suggested that severe
incapacitation is likely to occur under conditions likely to cause second degree burns, and death
under conditions likely to cause third degree burns.
23
There are also three basic heat exposure scenarios for building occupants during a fire:
1. Exposure to convected heat in a hot air environment.
2. Exposure to radiant heat direct from a fire or from a hot upper smoke layer
3. Exposure of a subject immersed in hot smoke – subjected to both radiant heat from hot
smoke particles and convected heat from contact with hot gases in the hot smoke
environment
Experimental human exposure data are available for the first two cases, from which time-
tolerance curves have been obtained for exposure to convected or radiant heat. For direct
exposure to radiant heat (for example from a fire or heater), empirical relationships between
exposure time and effect have been measured by a number of authors. For these experiments
heat radiation is expressed in terms of received heat flux (kW/m2). See equation 13.
The tenability limit for exposure of skin to radiant heat is approximately 2.5 kW/m2, below which
exposure can be tolerated for at least several minutes. Radiant heat at this level and above
causes skin pain followed by burns within a few seconds, but lower fluxes can be tolerated for
more than 5 minutes. Above this threshold, time (minutes) to incapacitation due to radiant heat
tIrad, at a radiant flux of q kW/m2 is given by Equation 135. The effects of heat on an occupant
response may depend upon the situation. The threshold for pain occurs at a value between
approximately 1.33 and 1.67 (kW.m-2)4/3.min. Second degree burns occur at 4.0-12.2 (kW.m-
2 4/3
) .min and third degree (full thickness) burns at appro16.7 (kW.m-2)4/3.min.
A figure of 1.33 (kW.m-2)4/3 is used to represent a tolerance threshold and 10 (kW.m-2)1.33 a
threshold for incapacitation and serious injury. For infrared radiation it is also proposed that 10
(kW.m-2)1.33 min represents a fatal level for a vulnerable population (over 65 years of age) or a
1% fatality level for the average population, while 16.7 (kW.m-2)1.33 .min represents a 50%
probability lethal level for the average population.
Calculating effects of exposure to convected heat only
Some experimental data are available for exposure of unclothed or lightly clothed subject to hot
air environments in terms of tolerance time to pain or hyperthermia. The Blockley40 curve shown
in Figure 9 is for dry air and humid air (saturated at normal room temperature), which is then
heated. This is very different from the effects of air saturated with water at higher temperatures.
Water is a serious potential problem in fires due to its high latent heat. If fire effluent contains
water at above 100°C (i.e. steam) it releases considerable heat if it comes in contact with the
skin or is inhaled. Although the hazards of contact with steam are well known, it may be less
obvious that air saturated with water vapour at lower temperatures can be dangerous. The
highest temperature at which saturated air can be breathed for more than a few minutes is
60°C. As an approximate guide, the volume concentration of water vapour in fire effluent is
similar to the CO2 concentration, which might reach as much as 10%. This is less than the
concentration in saturated air at the breathable limiting temperature of 60°C. For this reason it
is considered that the fuel-derived water vapour content of normal fire effluent is not likely to
present a serious hazard.
An expression has been derived for exposures of up to two hours to convected heat from air
containing less than 10% by volume of water vapour (see equation 14) .
As with toxic gases, the body of a fire victim may be regarded as acquiring a “dose” of heat over
a period of time during exposure, with short exposure to a high radiant flux or temperature being
more incapacitating than a longer exposure to a lower temperature or flux. The same fractional
incapacitating dose model as with the toxic gases may be applied and, providing that the
temperature in the fire is stable or increasing, the fractional dose of heat acquired during
exposure can be calculated (see equations 14-16)..
24
300
Dry
250 Humid (saturated at room temperature)
data points for clothed subjects from other authors for comparison
Temperature ( C) 200
o
150
Above 121oC tolerance limited by skin pain
Below 121oC tolerance limited by hyperthermia
100
50
0
0 5 10 15 20 25 30 35 40 45 50 55 60
Exposure time (minutes) for men at rest, unclothed: air movement low (< 30 m/min)
Copyright HER
Figue 9 Thermal tolerance to convected heat exposure for humans at rest, naked skin
exposed. Adapted from Blockley40.
OVERALL HAZARD ANALYSIS FOR A FIRE

Figure 10 shows an example of an overall hazard analysis for the furniture house fire
illustrated in Figure 1. The fractional effective concentration or dose of each physiological effect
is summed with time during the fire, with the endpoint for each effect reached at an FIC or FED
of 1. For the example shown the first hazard to exceed the tenability limit is impairment of
escape capability due to the effects of irritant organics and acid gases in the smoke products at
1.5 minutes. This is followed one minute later by the limit for optical obscuration by smoke
particulates. The combined effect of these two hazards in to seriously impair or prevent escape.
After five minutes it is predicted that an occupant would lose consciousness, primarily due to
the asphyxiant effects of hydrogen cyanide, with death occurring within approximately a further
30 seconds. For this fire escape impairment due to the effects of heat exposure is predicted
after six minutes, although several minutes further exposure would be required for burns to
occur. Also shown for comparison is the predicted time to incapacitation if there had been no
hydrogen cyanide present (after 7 minutes), this would be due to mainly to the asphyxiant
effects of carbon monoxide and carbon dioxide. The exposure dose of irritants capable of
causing potentially fatal deep lung inflammation some hours after exposure is only around 0.2
after 10 minutes, so this effect is predicted not to occur, although elderly subjects have been
found to be particularly sensitive to potentially fatal bronchopneumonia following exposures of
this kind in fires.
Since hazards from toxic effluent are the main causes of incapacitations and death in fires it
may seem sensible to regulate for toxic hazard or toxicity, with respect to both the overall design
and performance of the built environment (such as buildings and vehicles) and the individual
products used in their structure and contents. This may be achieved by means of prescriptive
requirements supported by pass-fail performance or ranking of specific products in standard
small or large-scale tests, or through performance-based (fire engineering-based) design using
engineering calculations with appropriate test data inputs to assess the overall performance or
built systems or products used in them.
25
Time to incapacitation
5
3
FED or FIC
0
0 1 2 3 4 5 6 7 8 9 10
Time (min)
Smoke Sensory irritancy Lung irritation Asphyxia Asphyxia no HCN Heat
Figure10: Fractional effective dose analysis in term of times to incapacitation for an

occupant of a domestic house lounge during an armchair fire.
METHODS FOR REGULATING TOXICITY AND TOXIC HAZARD

Historically, prescriptive guidance for fire safety has been based upon design requirements
for aspects such as detection, warnings and provision for means of escape, combined with test-
based fire resistance and reaction-to-fire performance of products and materials. Toxic hazard
has been controlled indirectly, mainly by controlling reaction-to-fire properties such as
ignitability, flame-spread and heat release rate using a battery of small-scale and medium-scale
test methods. This approach works reasonably well because these tests either address
fundamental properties, or fire properties for which small-scale behaviour is predictive of end-
use performance. For example, materials incapable of supporting combustion due to a very low
organic content, or products failing to support spreading flaming combustion under high radiant
heat fluxes in small-scale tests, are unable or unlikely to support formation of a large, growing,
hazardous fire in the end-use application. Because of this, they are unlikely to make a
significant contribution to a rapidly developing fire hazard from either toxic smoke or heat.
The same cannot be said of small-scale tests for toxicity or toxic hazard, since both are in
practice more complex outcomes of the performance of fire scenarios in the full-scale end-use
system, and specifically:
• Small-scale toxicity tests cannot possibly replicate the complex, dynamic, full-scale fire
hazard scenario
• Toxicity tests use very primitive fire models that generally have not been shown to replicate
the decomposition or combustion conditions in different types and stages of full-scale fires,
so they do not represent the range of yields and toxic product mixes occurring in such full-
scale fires.
• Toxicity tests employ very simplistic toxic potency or toxicity index models, which in no way
reflect the physiological effects and interactions of fire effluent mixtures
• The simplistic direct application of such index-based toxicity criteria is likely to result in
sanctioning the use of products or materials that may be hazardous in end-use applications,
while proscribing the use of other products or materials that might is some situations have a
26
high toxic potency, but in end-use have been shown to provide a good level of overall safety
with respect to fire performance and toxic hazard.
In contrast to this, performance-based approaches to toxic-hazard assessment, as used in fire
engineering-based design, have the potential for a realistic evaluation of end-use hazards both
of entire systems (buildings and transport) and specific products in end-use applications.
Performance-based design for buildings, trains and airplanes, whereby available safe escape
time is compared with required safe escape time for relevant fire scenarios, is increasing used
for fire-safety design compliance. For ASET evaluations these make use of fire dynamics data
and models, or full-scale tests, to obtain the time-concentration and intensity curves for design
fires, in combination with physiological FED models (such at those described here and detailed
in ISO 1357117 or in the SFPE Handbook of Fire Safety Engineering)5 ,in order to calculate times
to loss of tenability
Although such approaches are not generally applied to the application or regulation of products
used in building (or transport) structure or contents, it is proposed that standard methods could
be developed for applying physiological FED models for this purpose. The main advantages to
this approach would be that all products could be evaluated on a level-playing field with respect
to their ultimate end-use performance, taking into account specified combustion conditions, and
that the standards could be tailored for different potential applications.
An example might be a regulatory performance requirement for the use of a wall covering or an
item of upholstered furniture in a domestic dwelling. The regulatory performance requirement
could be expressed in terms of time available for escape for a specified fire scenario (or set of
fire scenarios) in an enclosed room: for example a period of 5 or 10 minutes following
application a defined ignition source.
A protocol might developed be along the lines of the following:
1. Reaction-to-fire tests performed using a specified small or medium – scale test such as
the Cone Calorimeter41, furniture calorimeter or Single Burning Item (SBI) apparatus42.
These are used to obtain the heat release rate/mass loss rate curves for the fire.
2. Small-scale combustion tests are performed to measure the yields of a standard set of
toxic fire effluents (kg/kg product mass loss) under a range of thermal decomposition
and combustion conditions representing those occurring at in different stages and types
of full-scale fires. Suitable test methods might include the ISO 19700 tube furnace
method18 or the ASTM E2508 flammability apparatus21.
3. These test data are then input into a specified fire calculation model (such as the NIST
CFAST43 zone model) using a specified standard domestic room (for example a 50 m3
room with specified linear dimensions).
4. From this, a set of time-concentration and time-temperature curves are calculated for the
smoke, toxic gases and heat in the defined fire scenario.
5. The standard ISO 13571 physiological FED model is applied to calculate the predicted
time to loss of tenability.
6. Depending upon the tenability time obtained, the product could be specified for use in
different applications. For example a longer time might be require for a care home than
for student accommodation or general domestic use.
The advantages of this type of approach is that once the basic test data are input, it is simple to
run a range or variety of cases, either for different applications of for different scenarios in any
one application. For the domestic case it would be possible to specify a case involving an open
fire room door with effluent spread into the greater volume of a specified standard apartment or
house, with assessment of tenability for occupants in locations remote from the fire room. Or
tenability in the room of origin could be examined for several different ventilation scenarios
(such as enclosed room, open door, open window). In this way the applicability of a product
could be evaluated quickly for a range of common dwelling fire scenarios. The same input data
might also be used to test applicability to a hotel room case, with different spaces and
27
dimensions specified. Reference scenarios for different cases could involve full-scale fire tests
or higher-level computational methods such as computational fluid dynamics models.
For the basic approach envisaged, the tests required to obtain data would consist of some
methods commonly in current use and the calculation/modelling could be standardised so as to
be performed within an hour or so.
As with any standard test method some degree of generalisation is necessary to represent the
diversity of actual built environments by a small set of standard enclosures (or enclosure sets)
and the diversity of possible fire scenarios by a small set of standard scenarios, but it is
considered that such an approach could provide a far more realistic and balanced evaluation of
actual end-use performance than currently used methods based upon unproven and overly
simplistic small-scale toxicity tests and indices, which bear no real relationship to either the end-
use fire conditions or their effects on occupants. It would also be possible as part of a
performance-based assessment for a product producer to tailor the modelled scenarios to
specific end-use applications where their product might show particular performance
advantages.
PARTICULAR ISSUES IN RELATION TO CABLE FIRES
The time concentration curves for toxic gases and smoke, and the time intensity curves for
radiant and convected heat in any fire depends upon the full-scale fire scenario, in terms of
the disposition and rate of combustion of the fuels and their interaction with the fire
enclosure, and it ventilation. When evaluating any product from a fire hazard perspective it
is important that the testing and hazard assessment strategy is relevant to the end use fire
hazard scenarios of interest.
In some installations, such as in plant rooms, cable runs may be open in the enclosure, in
which case the basic fire scenarios are somewhat similar to those for other combustible
fuels such as structural products and contents. The rate of combustion depends upon the
reaction to fire properties of the cables (as installed) and the enclosure ventilation. In large
enclosures, particularly if vents such as windows, doors or smoke extractors are open, then
combustion tends to be relatively efficient, leading to rapid fire growth combined with
relatively low toxic product yields for non-fire retarded fuels, and generally slower fire growth
rates from fire-retarded fuels, but accompanied by higher yields of toxic products. In
smaller enclosures, or where vents are mostly closed, then fires soon become vitiated, with
the fire size controlled by the available ventilation. If vents are opened, then flashover may
occur, but the fire is still vitiated and ventilation controlled. Under these conditions the yields
of smoke and toxic gases are greatly increased and even small fires can rapidly fill open
spaces within a building with lethal concentrations of smoke and toxic gases.
For cables in enclosed vertical or horizontal ducts, or in ceiling or floor plenum spaces, the
potential fire scenarios are somewhat different form those in room-sized enclosure fires.
Vertical cable runs are often in small enclosed spaces or ducts, often running up many floors
in a building. The rate of spread for fire and the rates of production, and yields, of toxic
products are likely to be very dependent upon the ventilation within the duct, the height of
the run, and the efficiency of fire stopping between compartments. From a hazard
perspective the extent of sealing of the cable runs within the ducts and the efficiency of fire
stopping is of crucial importance, since in a well-sealed environment the potential for fire
growth is limited and fire effluents are largely contained within the space of origin. Where
containment is less efficient, a developing fire is likely to be extremely vitiated, with very high
yields of smoke and toxic products and large volumes of toxic smoke spreading between
floors and into occupied spaces at different locations. In situations where vertical runs have
air flowing through them, the potential for more rapid fire growth and spread exists, and the
extent of vitiation and toxic product formation depends upon the fuel/air ratio in the
combustion zone. The preEN50399 test apparatus would appear to be a reasonable
28
approximation of an end-use application for vertical cable run fire scenarios44, although for
enclosed vertical runs the enclosure depth would often be much less than 2 metre depth in
the test apparatus, perhaps providing a greater “chimney” effect. Also it might be relevant to
consider a range of different ventilation conditions to evaluate different end-use applications.
Other important full-scale fire scenarios involve cable runs in horizontal unventilated or
ventilated plenum spaces. As with vertical runs, the potential for fire growth and spread
depends partly on the reaction-to-fire properties of the cable runs, especially the propensity
for fire spread along horizontal cables, and partly on the size and extent of enclosure of the
plenum space and its ventilation. As with vertical runs, an important aspect of potential
hazards is the extent of containment within separate enclosures and fire compartments, the
separation between the floor or ceiling spaces and the adjacent occupied rooms (including
the probability of tiles being removed), and the efficiency of fire stopping or shutter
activation. A consideration with small, vitiated fires is the extent of toxic effluent production
before operation of heat-activated shutters. If significant fire spread and propagation occurs
in plenum cable fires then the resultant fires are likely to be relatively small and slow growing
if ventilation is very restricted, but producing highly toxic fire effluent, while more ventilated
spaces may lead to larger fires, but in the restricted spaces within plenums, especially those
with high fuel loads, then vitiated combustion with high toxic product yields is also likely to
occur. The general position is that since most cable fires tend to occur in confined spaces
with limited ventilation, they are likely to involve vitiated combustion with high toxic product
yields, while the rate of fire development and ultimate fire size will depend upon the reaction-
to-fire properties of the cables as installed and the available ventilation.
Physiological Calculation Equations
An equation for the relationship between walking speed and smoke optical density (OD·m-1) is
given by:
Walking speed in non-irritant smoke (m/s) = 1.36 – 1.9 × smoke optical density (OD·m-
1
) [1]
Walking speed in irritant smoke (m/s) = 2.27 – 9 × smoke optical density (OD·m-1)
[2]
An equation for the curve in Figure 5, for walking speed during exposure any individual irritant
compound, or to the fractional irritant concentrations for a mixture of compounds, is given by:
2
Fwvirr = 1 − ((1 − e − ( x / b ) ) + ( − 0 .2 x + 0 .2 ) / 1 .2 ) [3]
Where
Fwvirr = fractional walking speed (1 = normal walking speed 1.2 m/s)
b = 160
x = FIC
The overall effect of exposure to an irritant smoke on walking speed (Fwv) would then be given
by:
Fwv = 1 – (1-Fwvsmoke) – (1-Fwvirr)
[4]
Where:
Fwv = Overall fractional walking speed
Fwvsmoke = Fractional walking speed due to smoke effects on visibility
Fwvirr = Fractional walking speed due to irritant effects for irritants compounds 1 to n
29
The general FED equation for summing exposure doses for each toxic product during each time
period and then integrating with time is:
t2 n
Ci
FED = ∫ ∑
i =1 (Ct ) i
Δt [5]
t1
where:
Ci is the average concentration, of an dose related toxicant such as an asphyxiant gas “i” over
the chosen time increment;
∆t is the chosen time increment, expressed in minutes (min)
(Ct)i is the specific exposure dose expressed as concentration × minutes, that would constitute
an effective dose (i.e. an exposure dose producing the a defined endpoint such as preventing
an occupant’s safe escape)
The overall FIC for an irritant mixture is as follows:
FIC = FICHCl +FICHBr + FICHF +FICSO2 +FICNO2 + FICCH2CHO + FICCH2O + ΣFICx [6]
Where ΣFICx = FICs for any other irritants present.

The FED equation for the lethal effects of inhaled irritants (i.e FLD or Fractional Lethal Dose) is:
FLDirr= FLDHCl + FLDHBr + FLDHF + FLDSO2 + FLDNO2 + FLDCH2CHO+FLDHCHO+Σ FLDx [7]
The overall FED equation for asphyxiant gases is :

FIN = ( FICO + FICN + FINOx + FLDirr ) × VCO 2 + FEDIO [8]
Where:
FIN = Fractional effective dose for incapacitation (loss of consciousness) due to
asphyxiants
FIco = Fractional effective dose for incapacitation by CO
FIcn = Fractional effective dose for incapacitation by HCN
Note: If necessary this can be corrected for the presence of other nitriles besides
HCN and for the protective effect on cyanide poisoning of NO and NO2. [CN] can
then be calculated as: [CN]=[HCN]+[Total organics nitriles]- 0.66 x [NO + NO2]
FINOx = Fraction of an incapacitating dose of NO + NO2 (= [NOx ppm x tmin]/15000)
FLDirr = Fractional lethal dose for irritants
Note: where significant concentrations of acid gases are present this term is
calculated according to equation 22. Otherwise it may be expressed in terms of
smoke optical density as OD/m x t(min)/90
VCO2 = Multiplicatory effect of inhaled CO2
FIo =Fractional effective dose for incapacitation by low oxygen hypoxia
For a simple analysis the direct asphyxiant effects of NOx and those of NOx on HCN asphyxia
may be ignored without significant error.
The expressions used to calculate the FEDs for each individual component are as follows:
For the effects of CO the FED is expressed in terms of %COHb. The denominator is the
%COHb predicted to cause loss of consciousness in an active (escaping) person = 30% COHb
30
(or 40% COHb for a resting person). The numerator is the Stewart equation45, by which the
%COHb in the subject is calculated from the inhaled CO concentration in the fire, the exposure
time and the volume of air breathed each minute:
FIco = 3.317 x10 −5 [CO ]1.036 )(V )(t ) / D [9]
Where:
[CO] = carbon monoxide concentration (ppm v/v 20°C)
V = volume of air breathed per minute (liters/min)
t = exposure time in minutes
D = exposure dose (%COHb) for incapacitation
The following values are taken for V and D:

Activity level of subject V (l/min) D %COHb
Resting or sleeping 8.5 40
Light work – walking to escape 25 30
Heavy work – slow running, walking up stairs 50 20
Note 1: This expression (the Stewart equation) was obtained from young adult male human
volunteers. It is suitable for adults in situations where the CO concentration is high in relation to
the blood COHb concentration (as for most flaming fires and short exposure durations). Where
long exposures may lead to near equilibrium conditions, the Coburn Forster Kane46,47 equation
should be used since significant deviations from the Stewart equation may then occur. The
Stewart equation somewhat underestimates uptake rates for children. Where more precise CO
uptake calculations are required, differences in body size (including children) can be
accommodated using the CFK equation. For a basic design, use of the Stewart equation is
recommended. Differences in body size and other susceptibilities can be considered as
allowed for in the overall safety margin chosen to accommodate more susceptible sub-
populations (for example an FED 0.3 or other endpoint).
Note 2: As an alternative to using this expression the FEDIco may be expressed as a CO

exposure dose ratio. This method is used for ISO 1357117. For this method the FEDico is
expressed as CO ppm x t/35,000. This is approximately equivalent to a “light work” case for a
subject breathing approximately 20 l air/minute.
Since occupants must at least walk in order to escape from a fire, the default case suggested is
that for light work. However this could be varied according to the case. For example a sleeping
person escaping from a basement might start by being at rest, then awaken and walk to a stair
(light work) then climb the stair (heavy work).
For the effects of hydrogen cyanide and low oxygen hypoxia, the expressions are more
complex, because the denominators are not constants. Exponential expressions have been
developed to fit the time to incapacitation versus exposure concentration curve from
experimental exposures in non-human primates (HCN) and humans (hypoxia), so that the
Fractional Incapacitating Doses for HCN (FICN) and hypoxia (FIO) as follows:
⎡⎛ exp([CN ] 43) ⎞ ⎤ [10]

FICN = ⎢⎜ ⎟ − 0.0045⎥t
⎣⎝ 220 ⎠ ⎦
Where CN = HCN concentration (ppm v/v at 20°C)

31
FIo = t (exp[8.13 − 0.54(20.9 − [%O 2 ])]) [11]
Where [%O2] = oxygen concentration (% v/v at 20°C)

I = exposure dose for incapacitation
Similarly, a curve has been fitted to the effect of CO2 on ventilation (breathing volume per
minute) based upon human experimental data.
Ventilatory stimulation by CO2: VCO2 = exp ([CO2]/5) [12]
Where [CO2] = carbon dioxide concentration (% v/v at 20°C)
FEDs are calculated for successive short periods during the fire and then integrated with time in
order to calculate the time when incapacitation is predicted (FED = 1). Due to the rapid (t2) rate
of increase of asphyxiant gas concentrations in most flaming fires, variations in individual
susceptibility and uncertainties in prediction of incapacitating doses tend to have relatively minor
effects on predicted times to incapacitation.
The tenability limit for exposure of skin to radiant heat is approximately 2.5 kW/m2, below which
exposure can be tolerated for at least several minutes. Radiant heat at this level and above
causes skin pain followed by burns within a few seconds, but lower fluxes can be tolerated for
more than 5 minutes. Above this threshold, time (minutes) to incapacitation due to radiant heat
tIrad, at a radiant flux of q kW/m2 is given by Equation 135.
1.33
t Irad = 1.33
(8) (25)13 [13]
q
Where:
tIrad = time to endpoint (pain in this case) in minutes
q = heat flux kW/m2.
The effects of heat on an occupant response may depend upon the situation. The threshold for
pain occurs at a value between approximately 1.33 and 1.67 (kW.m-2)4/3.min. Second degree
burns occur at 4.0-12.2 (kW.m-2)4/3.min and third degree (full thickness) burns at appro16.7
(kW.m-2)4/3.min.
A figure of 1.33 (kW.m-2)4/3 is used to represent a tolerance threshold and 10 (kW.m-2)1.33 a
threshold for incapacitation and serious injury. For infrared radiation it is also proposed that 10
(kW.m-2)1.33 min represents a fatal level for a vulnerable population (over 65 years of age) or a
1% fatality level for the average population, while 16.7 (kW.m-2)1.33 .min represents a 50%
probability lethal level for the average population.
The following expression has been developed for the mid-humidity case from Blockley40 and
other data. Tolerance time ttol (minutes) is then given by:
ttol = 2 x 1031 x T-16.963 + 4 x 108 x T-3.7561 [14]
Where T = room temperature, degrees centigrade

For Equations 13 and 14, the “Fractional Effective Dose” of heat received each minute is given
by the reciprocal of the expression.
For situations when a subject is exposed to both radiant heat (for example from a heated upper
layer) and covected heat from exposure to heater air, the overall heat dose received may be
estimated by summing the radiant and convected fractions using Equation 34:
32
t2 1 1
FED = ∫ ( + ) Δt (10) ( [15]
t1
t Irad t Iconv
The doses acquired each unit of time are then integrated. The tenability limit is predicted when
the FED for heat = 1.
A possible method for estimating tolerance time for the case of a subject immersed in hot
smoke could be to calculate the total heat flux from the radiant and convected components
of the smoke. On this basis it is proposed that the total incident flux to the skin of a person
immersed in hot smoke is given by:
q =εσ (Ti4 – Tm4) + hc (Ti – Tm)/1000 [16]

Where:
q = heat flux kW/m2
Ti = heat source temperature °K
Tm= material surface temperature °K
ε = emissivity (0.05 for a gas to 1 for a black body, perhaps 0.5 for smoke)
σ = Stefan Bolzmann constant (5.67 x 10-8 Wm-2K-4)
hc = convective heat transfer factor. For air this depends upon the flow rate past the object.
It will be approximately 5-8 for slow moving air.
The first term in the equation represents the radiant component of heat flux and the second
term the convected component of heat flux. Using this equation it is therefore possible to
calculate total heat flux from the room temperature at body height. The radiant component
is relatively small at low temperatures, and is negligible for hot air due to its low emissivity.
However, for smoke the emissivity is likely to be must higher (around 0.5) so that at higher
temperatures both components should be considered in order to calculate the total heat flux
to the skin. In addition to the heat flux from the fire effluent enveloping a subject there is
likely to be additional heat radiation from hot upper layers and/or directly from the fire. If the
subject is in air (with a low emissivity), below a hot smoke layer, the only significant radiative
heat flux sources are likely to be the upper layer, the fire or hot surfaces.
Time to the different heat effect endpoints for total heat fluxes in excess of 2.5 kW/m2 is then
given by equation 16 using the appropriate exposure dose endpoints as the numerator:
1.33 (kW.m-2)1.33 .min (tolerance limit/pain/first degree burns)
10 (kW.m-2)1.33 .min (severe incapacitation and second degree burns)
16.7 (kW.m-2)1.33 min (fatal exposure with third degree burns).
or all the expressions in this section time to effect is estimated based upon measured effects on
exposed skin of subjects for given exposures to hot air or different levels of incident radiant heat
flux.
REFERENCES
1
The Building Regulations 2000: Approved Document B 2006 edition Communities and
Local Government 2006 edition
2
ISO/TR 13387-8 Fire safety engineering Part 8: Life Safety - Occupant behaviour, location
and condition.
3
Purser D.A. and Kuipers, M. (2004) Interactions between buildings, fire and occupant
behaviour using a relational database created from incident investigations and interviews.
3rd International Symposium on Human Behaviour in Fire. Europa Hotel, Belfast, 1-3rd
September 2004. Proceedings pp. 443-456 Interscience Communications, London UK.
4
Purser, D.A. Human Tenability. The Technical Basis for Performance Based Fire
Regulations. United Engineering Foundation Conference. San Diego CA January 7-11
33
5
2001 Proceedings pp.88-129.
Purser D.A. Assessment of Hazards to Occupants from Smoke, Toxic Gases and Heat. The
SFPE Handbook of Fire Protection Engineering 4th ed),.DiNenno P.J (ed.), National Fire
Protection Association, Quincy, MA 02269, 2009, pp. 2/96 – 2/193.
6
Purser, D. A Toxic product yield and hazard assessment for fully enclosed design fires
involving fire retarded materials. Polymer International 49: 1232-1255, (2000).
7
Purser, D.A, The application of exposure concentration and dose to evaluation of the
effects of irritants as components of fire hazard. Interflam 2007. Royal Holloway College,
University of London, UK 3-5th September 2007 Proceedings pp 1033-1046.
8
Jin, T., Visibility through fire smoke, Part 5. Allowable smoke density for escape from fire.
Report of Fire Research Institute of Japan, No. 42, 12 (1975).
9
Alarie Y. Sensory irritation by airborne chemicals. CRC Crit. Rev. Toxicol. 1973; 2: 299-362.
10
Purser, D.A. and Berrill, K.R. (1983) Effects of carbon monoxide on behaviour in monkeys
in relation to human fire hazard. Arch. Environ. Hlth. 39, 308-315.
11
Purser, D.A. and Woolley, W.D. (1982) Biological effects of combustion atmospheres. J.
Fire Sci. 1, 118-144.
12
Purser, D.A.and Grimshaw, P. (1984) The incapacitative effects of exposure to the thermal
decomposition products of polyurethane foams. Fire and Materials. 8, 10-16.
13
Purser, D.A. Grimshaw, P and Berrill, K.R. (1984) Intoxication by cyanide in fires: a study
in monkeys using polyacrylonitrile. Arch. Environ. Hlth. 39, 394-400.
14
Purser, D.A. Modelling time to incapacitation and death from toxic and physical hazards in
aircraft fires. NATO-AGARD Conference Proceedings No.467: Aircraft Fire Safety. pp. 41-
1 to 31-12. Sintra, Portugal, 22-26 May 1989.
15
Purser, D.A. (1996) Behavioural impairment in smoke environments. Toxicology, 115, 25-
40.
16
Purser, D.A. (2008) Physiological effects of combustion products. In: Hazards of
combustion products: Toxicity, Opacity, Corrosivity and Heat Release Conference. The
Royal Society, London, 10-11 November 2008. Proceedings pp 97-122.Interscience
Communications Limited.
17
Life threatening components of fire – Guidelines for the estimation of time available for escape
using fire data. ISO TS13571 (2007).
18
Controlled Equivalence Ratio Method for the Determination of Hazardous Components of
Fire Effluents,” ISO/TS 19700, International Organization for Standardization, Geneva
(2007).
19
Purser, D.A. (2002) ASET and RSET: addressing some issues in relation to occupant
behaviour and tenability. 7th International Symposium On Fire Safety Science, Worcester
Polytechnic Institute – Worcester Massachusettes, USA 16 – 21 June 2002. FIRE
SAFETY SCIENCE – Proceedings of the seventh international symposium. International
Association for Fire Safety Science, 2003 pp. 91-102.
20
Purser, D.A. and Purser, J.A. HCN yields and fate of fuel nitrogen for materials under
different combustion conditions in the ISO 19700 tube furnace and large-scale fires. 9th
International Symposium on Fire Safety Science, University of Karlsruhe, Germany. 21-26th
September 2008
21
Tewarson A. Generation of Heat and Chemical Compounds in Fires. In SFPE Handbook
of Fire Protection Engineering, 3rd Ed. NFPA Quincy MA, pp 3-82 to 3-16
22
Purser, D.A. Influence of Fire Retardants on Toxic and Environmental Hazards from
Fires. FRPM07 – 11th European Meeting on Fire Retardant Polymers.The Albert Halls,
Bolton, Manchester. 2nd-6th July 2007.
23
Purser, D.A. (1992) Recent developments in understanding the toxicity of PTFE thermal
decomposition products. Fire and Materials 16, 67-75.
24
PD 7974-6, The application of fire safety engineering principles to fire safety design of
buildings — Part 6: Human Factors: Life safety strategies — Occupant evacuation,
behaviour and condition. British Standards Institution 2004.
34
25
Bryan, J.L. Smoke as a determinant of human behaviour in fire situations (project people).
National Bureau of Standards, Washington D.C. USA Report NBS GCR-77-94.
26
Bryan, J. L. Behavioural response to fire and smoke, in SFPE Handbook of Fire Protection
Engineering, Third Edition, Ed. By P.J. DiNenno et al.,Society of Fire Protection Engineers,
Boston, MA and National Fire Protection Association, Quincy, MA, USA, (2002). Section 3,
Chapter 12, pp. 3-315 - 3-341.
27
Emergency Response Planning Guidelines, American Industrial Hygiene Association, Akron,
OH (1989).
28
National Advisory Committee for Acute Exposure Guideline Levels for Hazardous Substances
(NAC/AEGL).
29
Assessment of hazard to life and health from fire – Part 2: Guidance on methods for the
quantification of hazards to life and health and estimation of time to incapacitation and death in
fires. BS 7899-2 (1999)
30
Alarie Y. Sensory irritation by airborne chemicals. CRC Crit. Rev. Toxicol. 1973; 2: 299-362.
31
Alarie Y. Bioassay for evaluating the potency of airborne sensory irritants and predicting
acceptable levels of exposure in man. Food. Cosmet. Toxicol. 1981; 19: 623-26.
32
Beswick FW., Holland P. and Kemp, KH. Acute effects of exposure to orthochlorobenzylidene
malonitrile (CS) and the development of tolerance. Br. J. Industr. Med. 1972; 29: 298-306.
33
Nielsen GD. Mechanisms of activation of the sensory irritant receptor by airborne chemicals.
Toxicology 1991; 31: 183-208.
34
Acute Exposure Guideline Levels for Selected Airborne Chemicals Volume 4 (2004)
Chapter 2 Hydrogen chloride 77-122. National Advisory Committee for Acute Exposure
Guideline Levels for Hazardous Substances, Subcommittee on Acute Exposure Guideline
Levels, Committee on Toxicology, Board on Environmental Studies and Toxicology.
National Academic Press, Washington.
35
Emergency Response Planning Guidelines, American Industrial Hygiene Association,
Akron, OH (1989).
36
NIOSH (National Institute for Occupational Safety and Health). (1994). Documentation for
Immeditaely Dangerous to Life or Health Concentrations (IDLHs). U.S. Department of
Health and Human Sevices, National Institute for Occupational Safety and Health,
Cincinatti, Ohio, USA.
37
Personal communication from two individuals exposed to HCl.
38
Barrow, C.S, Lucia, H., and Alarie (1979). A comparison of the acute inhalation toxicity of
hydrogen chloride versus the thermal decomposition products of polyvinylchloride.
J.Combust. Toxicol. 6, 3-12.
39
.Pach J. et al., Analysis of predictive factors in acute carbon monoxide poisonings. Folia
Medica Cracoviensia, 20, 159-168, 1978.
40
Blockley, W.V. 86. Temperature Tolerance: Man: Part 1. Heat and cold tolerance with and
without protective clothing. In Biology Data Book, 2nd edition, Volu,me II, p.781 Federation
of American Societies of Experimental Biology. Bethesda, MD.(1973).
41
“Reaction-to-Fire Tests—Heat Release, Smoke Production, and Mass Loss Rate—Part 1:
Heat Release (cone calorimeter method),” ISO 5660-1, ISO, Geneva, Switzerland (2002).
42
EN13823:2002 Reaction to fire tests for building products. Building products excluding
flooring exposed to thermal attack by a single burning item. Euopeqan Committee for
Standarization (CEN). Brussels.
43
Jones, W.W., Peacock, R.D., Forney, G.P. and Rneke, P.A. CFAST – Consolidated Model
of Fire Growth and Smoke Transport (Version 6) NIST Special Publication 1026, 2005.
44
Journeaux, T.I.. (2008) Hazards of combustion products – industry sector report – electric
cables. In: Hazards of combustion products: Toxicity, Opacity, Corrosivity and Heat
Release Conference. The Royal Society, London, 10-11 November 2008. Proceedings pp
331-320.Interscience Communications Limited.
45
Stewart, R.D. The Effects of Carbon Monoxide on Man, J. Combustion Toxicology 1, 167-
35
176 (1974)
46
Coburn, R.F. Forster, R.E and P.B. Kane. Consideration of physiological variables that
determine the blood carboxyhaemoglobin concentrations in man. Journal of Clinical
Investigation. 44, 1899-1910 (1965).
47
Peterson, J.E and. Stewart R.D. Predicting the caboxyhaemoglobin levels resulting from
carbon monoxide exposures. Journal of Applied Physiology 39, 633-638 (1975).
36

Purser Physiol Effects

Uploaded by

Copyright:

Available Formats

Purser Physiol Effects

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Purser Physiol Effects

Uploaded by

Copyright:

Available Formats

PHYSIOLOGICAL EFFECTS OF COMBUSTION PRODUCTS

AND FIRE HAZARD ASSESSMENT

By Professor David A. Purser

Hartford Environmental Research, UK

Europacable Seminar “Safety during Fire”, Brussels 6th May 2009

ACUTE SURVIVAL HAZARDS DURING FIRES

BASIC REQUIREMENTS FOR LIFE SAFETY HAZARD ANALYSIS

• The concentration/time/physiological effect relationships of these products in terms of the

100 100 PMMA

Rat 30-min LC50 concentration (g/m3)

METHODS OF FIRE HAZARD ANALYSIS BASED UPON PHYSIOLOGICAL

Walking speed against smoke optical density

Figure 4 — Walking speeds in non-irritant and irritant smoke

Table 3 — Smoke tenability limits

Smoke density and Irritancy Approximate visibility

PREDICTING TIME TO INCAPACITATION BY IRRITANTS, ASPHYXIANT

Fractional Effective Dose methodology for hazard analysis

Thus FECsmoke (Fractional Effective Concentration of smoke) = 1 represents a smoke

conc. gas present × time 1000 ppm CO × 20 min

Another reason for expressing the concentration or dose as a fraction of an effective

Setting Tenability Limits For Painful Sensory Irritant Effects

AEGL-3 (lethal) human 10 minutes

Purser SFPE Handbook – predicted to cause incapacitation average person5

40% decrease in arterial blood oxygen in sedated baboons

Tenability limits and hazard calculations for asphyxiant gases

Figure 6: Time to incapacitation for cynomolgus monkeys exposed to different

Fire deaths Non-fire deaths

Figure 7: Distribution of fatal %COHb in non-burned fire victims and non-fire CO

Tenability and hazard calculations for development of pain, incapacitation,

OVERALL HAZARD ANALYSIS FOR A FIRE

Smoke Sensory irritancy Lung irritation Asphyxia Asphyxia no HCN Heat

Figure10: Fractional effective dose analysis in term of times to incapacitation for an

METHODS FOR REGULATING TOXICITY AND TOXIC HAZARD

Physiological Calculation Equations

Where ΣFICx = FICs for any other irritants present.

FLDirr= FLDHCl + FLDHBr + FLDHF + FLDSO2 + FLDNO2 + FLDCH2CHO+FLDHCHO+Σ FLDx [7]

The overall FED equation for asphyxiant gases is :

FIco = 3.317 x10 −5 [CO ]1.036 )(V )(t ) / D [9]

The following values are taken for V and D:

Note 2: As an alternative to using this expression the FEDIco may be expressed as a CO

⎡⎛ exp([CN ] 43) ⎞ ⎤ [10]

Where CN = HCN concentration (ppm v/v at 20°C)

Where [%O2] = oxygen concentration (% v/v at 20°C)

ttol = 2 x 1031 x T-16.963 + 4 x 108 x T-3.7561 [14]

Where T = room temperature, degrees centigrade

q =εσ (Ti4 – Tm4) + hc (Ti – Tm)/1000 [16]

You might also like