Air Quality

Hdb Env Chem Vol.
4, Part H (2005): 324

DOI 10.1007/b107234
Springer-Verlag Berlin Heidelberg 2005
Published online: 8 August 2005
Aircraft Environmental Control Systems

M. Dechow C.A.H. Nurcombe (u)
Airbus Deutschland GmbH, Kreetslag 10, 21129 Hamburg, Germany
martin.dechow@airbus.com, claire.nurcombe@airbus.com
1
Introduction/Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Regulatory Requirements and Guidelines . . . . . . . . . . . . . . . . . .
Environmental and Occupant-Related Constraints . . . . . . . . . . . . .
4
4.1
4.2
Cabin Pressurisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Absolute Cabin Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Cabin Pressure Rate of Change . . . . . . . . . . . . . . . . . . . . . . . .
7
8
9
5
5.1
5.2
Thermal Comfort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Flow Patterns in the Cabin . . . . . . . . . . . . . . . . . . . . . . . . . . .
Temperature and Air Velocity . . . . . . . . . . . . . . . . . . . . . . . . .
9
9
12
Humidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
13
Ventilation Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
15
8
8.1
8.2
8.2.1
8.2.2
8.2.3
8.2.4
8.2.5
8.2.6
Contaminant Control . . . . . . . . . . . . . . . . . . . . . . . . . .
Cabin Operations Contamination Sources . . . . . . . . . . . . . . .
External Contamination Sources . . . . . . . . . . . . . . . . . . . .
Carbon Dioxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Carbon Monoxide . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Volatile Organic Compounds and Semi-volatile Organic Compounds
Ozone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Bacteria, Viruses, Fungi, Moulds . . . . . . . . . . . . . . . . . . . .
Particulates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
16
16
16
17
17
18
19
20
21
Interdependency of Factors . . . . . . . . . . . . . . . . . . . . . . . . . .
22
10
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
23
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
23
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Abstract The external environment at 41 000 ft (12 500 m), a typical cruise altitude for
modern civil aircraft, is hostile to human life. Aircraft environmental control systems
are designed to ensure the survival of the aircraft occupants as well as providing them
with a comfortable atmosphere. Major design drivers for the environmental control system are thermal comfort, pressurisation and cabin air quality. However, these parameters
cannot be considered independently. They interact between themselves and with other
parameters, which may or may not be controllable by the system designer. These interactions occur in a highly complex manner. Research has led to a good understanding
M. Dechow C.A.H. Nurcombe
of the basic functions to allow safe and comfortable aircraft environmental conditions.
Future research efforts will be increasingly focussed on identifying and elaborating the
interdependency of factors in order to further enhance the aircraft cabin environment.
Keywords Environmental control system Thermal comfort Cabin air quality
Pressurisation Humidity control
Abbreviations
APU
Auxiliary power unit
ASHRAE American Society of Heating, Refrigerating and Air Conditioning Engineers
ASICA Air management simulation for aircraft cabins
CDC
Centers for Disease Control
CFD
Computational uid dynamics
cfm
Cubic feet per minute
CPCS
Cabin pressurisation control system
DIN
Deutsches Institut fr Normung (German Standardisation Institute)
ECS
Environmental control system
FAA
Federal Aviation Administration
FACE
Friendly aircraft cabin environment
FAR
Federal Aviation Regulations
FL
Flight level
HEACE Health effects in aircraft cabin environment
HEPA
High-efciency particulate arrestor
IFE
In-ight entertainment
JAA
Joint Aviation Authorities
JAR
Joint Aviation Requirements
LF
Load factor
MAK
Maximale Arbeitsplatz-Konzentration (German Maximum Workplace Concentrations)
MIL
Military standard
ppm
Parts per million
RH
Relative humidity
SARS
Severe acute respiratory syndrome
SVOC
Semi-volatile organic compounds
VOC
Volatile organic compound
WHO
World Health Organisation
1
Introduction/Summary
The necessity to provide aircraft occupants with not only survivable but
also comfortable conditions for work and relaxation drives ECS design. Due
to the hostile environment outside the aircraft during ight conditions this
aim requires the control of several interdependent factors. The analysis of
cabin environment factors is mainly limited in this discussion to the major design requirements for air conditioning systems, e.g. air contaminants,
thermal comfort and cabin pressure, however, a short discussion of the interdependency of factors will be revisited at the end of this paper. Where
possible, measurement results as well as calculations are shown and related
to aircraft system design methods; specically the Airbus design philosophy.
Gaps in knowledge surrounding the perception of the cabin environment by
unhealthy, very young or elderly aircraft occupants are identied. Where appropriate, reference to current and future technology developments is made
to show how the design process is evolving as these knowledge gaps are
closed.
2
Regulatory Requirements and Guidelines
The conditions outside the aircraft during ight are hostile for humans. An
articial climate must be established within the cabin to support life under
these conditions. Additionally, the cabin environment is an important inuence on crew performance and passenger comfort.
Although there are certication requirements for at least some of the
cabin environment factors [1, 2], air quality in particular is the subject of investigation by governmental organisations [35], as well as standardisation
committees established by ASHRAE (SPC161), DIN (DIN6032) and CabinAir,
an EU funded research programme. The reason for this interest is the increasing sensitivity of the public and press to potential health threats and the
recognition that a set of new information generated through recent research
may require inclusion in the certication requirements. Existing standards
also do not address the specic environment of the aircraft cabin in detail,
if at all. The aircraft cabin environment is unique when compared to other
indoor spaces due to the combination of elevated cabin altitude, low humidity, high passenger density, the long sedentary position of the passengers and
ights across time zones.
3
Environmental and Occupant-Related Constraints
Due to the specic external environment, the primary function of the ECS is
to preserve the lives of the occupants of the aircraft. At a cruise altitude of
41 000 ft ambient pressure may be as low as 200 hPa, the temperature lower
than 60 C and the water content of the air almost zero. Without life support
systems humans would not be able to survive under these conditions. The
ECS encompasses the air conditioning packs, consisting of heat exchangers,
compressor, water extraction and turbine, and the air distribution, recircu-
Fig. 1 The environmental control system Airbus Library
lation and pressurisation systems, including the associated fans, valves and
ducting. The bleed system delivers hot air from the engines, APU or external sources to the ECS, and a trim system taps some of this hot air off before
it passes through the air conditioning pack to be added in the distribution
ducting for temperature control reasons. These systems, and how they are
interconnected are shown in Fig. 1.
The ECS designer must also ensure that the rates of change of pressure
and the minimum pressure within the cabin are controlled in such a way as
to prevent physiological damage to the occupants. Once the basic life preservation functions have been fullled the ECS designer must then consider the
system performance for heating and cooling as well as comfort control systems for the occupants. Comfort control design is much more difcult than
design of the life preservation functions since individuals have varying ideas
of what acceptable comfort is. Additionally, the requirements of both passengers and cabin crew must be fullled within the same cabin conditions. Flight
crew comfort, with its own attendant requirements must also be carefully
considered.
The comfort requirements for the cabin crew and passengers are not generally analogous. While passengers are mainly sedentary, cabin crew may
combine periods of activity with periods of inactivity, which may be within
the galley, cabin, or special crew rest areas completely separated from the
cabin. Equipment in these specic areas may also have an effect on comfort,
such as the temperature effect of ovens or chillers in the galleys. The cabin
crew may also have specic uniform requirements regarding the clothing they
have to wear for each activity, whereas passengers are free to remove or add
clothing or blankets to improve their personal thermal comfort.
The ight crew comfort requirements may be considered to be similar to
passengers, although temperatures may generally be controlled to lower levels during periods of high workload, such as take-off. There are however
some additional design constraints that must be considered carefully when
designing for cockpit occupant comfort. One consideration is the amount
of heat-generating electrical equipment that is installed in the cockpit. This
signicant heat load requires high air exchange rates to ensure equipment
cooling and prevent occupants overheating. Additionally, the large expanse of
windows can be a signicant source of either heating or cooling, depending
on the outside conditions. Due to the cockpits small volume, high heat loads
and the effect of radiant heat loads it is a signicant design challenge to prevent temperature stratication and drafts and ensure a good thermal comfort
level.
4
Cabin Pressurisation
The pressure outside the aircraft is hostile to human life at cruise levels of
modern aircraft. To assure a habitable environment for the occupants the
fuselage has to be pressurised during ight. As the cabin pressure is slightly
reduced from ground level pressure during ight conditions, appropriate
pressure gradients have to be considered for the CPCS design.
Fig. 2 Typical pressure schedule SAE ARP1270
4.1
Absolute Cabin Pressure
The current certication requirement is to keep the cabin altitude lower or
equal to 8000 ft, equivalent to 2440 m [1]. This is seen as the best compromise
between the occupant health and comfort on one hand, and the aircraft structure weight, which would increase with a higher pressure difference between
cabin and the outside, on the other. However, the maximum cabin altitude is
only seen at the highest certied ight altitude of Airbus aircraft. Many ights
are operated substantially below this altitude with cabin altitudes controlled
lower than the 8000 ft maximum. The maximum cabin altitude for the Airbus
long-range aircraft (A330/A340) is set to 2240 m (7350 ft) for longer ights,
providing an additional margin (see Fig. 2 for a typical pressure schedule
used by the pressurisation control system).
The percentage of oxygen in the atmosphere remains constant at around
21% for the altitudes at which modern aircraft y. What is of greater concern
for the denition of the cabin pressure requirement is how much of that oxygen the body is able to absorb. Therefore the major driver dening the cabin
pressure requirement is the required oxygen saturation of the blood to keep
crew performance high and prevent passenger health problems.
The blood oxygen saturation is dependent on the oxygen partial pressure
of the cabin air, which is dependent on the cabin pressure itself. As the cabin
pressure reduces (with increasing cabin altitude) the oxygen partial pressure
decreases. The oxygen partial pressure is however only one factor for the
actual oxygen saturation of the blood. The pH-value, occurrence of carbon

monoxide and personal constitution also play roles, the exact details of which
have not been fully investigated. The effects of elevated cabin altitude on children, elderly or sick people could be derived from chamber studies with their
population groups but ethical concerns have up until today prevented such
studies. However, study programmes are increasingly highlighting this area
as a high priority and research programmes are starting to be set up which
could address some of these issues.
4.2
Cabin Pressure Rate of Change
Furthermore, since the maximum cabin altitude increases in comparison to
airport altitudes (with some exceptions, such as take-off or landing at Mexico City at 2237 m (7341 ft), Quito at 2808 m (9213 ft) and Lhasa at 3570 m
(11 712 ft)) the cabin pressure must be reduced during aircraft climb to cruise
conditions and raised during aircraft descent to the external conditions at
the landing eld. The rate of pressure change is noticed by many passengers
through natural physiological phenomena, such as pressure discomfort at the
eardrum, frontal sinuses or in the intestines. The eardrum is especially sensitive to pressure changes. These effects are amplied if illness is pending or,
for instance, nasal cavities are blocked or cavities in the teeth are present. The
pressure adaptation is easier with decreasing pressure (equivalent to aircraft
climb).
The cabin pressure rate of change is therefore limited by the CPCS for
these physiological reasons; the cabin altitude should not increase at more
than 150 m/min (500 ft/min) and as the adaptation is more difcult during
the aircraft descent (re-pressurisation of the cabin), the cabin altitude should
not decrease at more than 90 m/min (300 ft/min). These gradients are appropriate for healthy people, but may not prevent problems for occupants with
deteriorated health, such as ear problems or a severe cold.
5
Thermal Comfort
Thermal comfort is often seen as a matter of temperature alone. However,
thermal comfort is actually an agreeable combination of temperature, air
velocity, rate of velocity uctuations and humidity. These four control parameters are further related to the ow pattern achieved within the cabin. The
ow pattern is a critical parameter to ensure that the ventilation air supply is
correctly distributed throughout all areas of the cabin.
10
Fig. 3 Typical circular ow pattern within cabin Airbus Library
5.1
Flow Patterns in the Cabin
The ventilation system is designed such that the air is adequately distributed
throughout the length of the cabin. It is just as important to distribute the air
appropriately in each temperature control zone (the cabin is divided into temperature control zones to allow temperature control with respect to the cabin
section layout). Thus the allocation and design of the cabin outlets are the
main tasks to be carried out during an air conditioning system design process. There are several different philosophies regarding how best to achieve
optimal ow patterns in the cabin. Large commercial aircraft tend to have
a circular ow pattern within the cabin where the ventilation air enters at the
top of the cabin and circulates as two counter-rotating advection rolls before
being exhausted at oor level (see Fig. 3 for a typical example).
The number of air outlets per side may be optimised, depending on the
specic requirements of the aircraft cabin layout. Airbus designs the air outlet positions so as to achieve the necessary air exchange rate of air within
the cabin. Lateral outlets signicantly improve air movement at the window
seats and in co-ow with upper ceiling outlets establish a stable, quasi twodimensional, advection ow.
The circular advection ow pattern which develops as a result of this
design is seen as being an optimal solution to ensure homogenous air distribution through the cabin for the majority of operating conditions. It is
important, however, to note that the design of the air outlets and their blowing characteristics is very much dependent on cabin lining. A smooth lining
contour can allow only one air outlet installation per side, which creates two
counter-rotating uid ows in each semi-section of the cabin. However, as the
Fig. 4 Typical CFD calculation result for temperature Airbus Library
Fig. 5 Typical CFD calculation result for air velocity Airbus Library
11
12
trend to exibility and customisation of the cabin interior increases, the challenge for the ECS engineer is to ensure that regardless of the cabin interior
layout an optimal ow pattern remains. For the A380, for instance, several air
outlet congurations have been developed, and would be applied depending
on the individual airlines cabin interior layout choice.
Systems providing a ow pattern from bottom to top (air inlets at oor
level, outlets at head level) have been suggested from time to time but major
disadvantages of this type of ow pattern include the downward convection
ow direction, with the subsequent difculty of achieving the cooling demand, and the negative inuence of high momentum air ow in the vicinity
of seated passengers. Additionally, contamination on the oor will be carried
up into the faces of the passengers. Another major factor in the difculty of
designing bottom up ow designs is the effect that the cabin furnishings may
have on the ow patterns. In top down ow designs the outlet is positioned
to leave the ow free from disruption by the overhead storage compartments.
The seats and seat-back tables only inuence the ow once it is distributed
and therefore low momentum. With a oor up ow, the air ow will be
disturbed before reaching the heads of the sitting or walking passengers and
crew. This would therefore make it much more difcult for the designer to
develop stable, quasi two-dimensional ow patterns within the cabin.
CFD simulations are used for new aircraft programs to optimise the ow
pattern within the cabin in the early phases of the design process. This has
an added benet of easing the validation of the ventilation and distribution
systems during rig and aircraft tests. Both the temperature distribution and
air velocities can be predicted by CFD (see Figs. 4 and 5 for example output).
Good ow patterns within the cabin will ensure good temperature distribution, preventing hot spots occurring and ensuring a uniform temperature
throughout the particular temperature control zone.
5.2
Temperature and Air Velocity
Just as temperature control is linked to correctly calibrated ow patterns,
so is temperature perception interdependent on air velocity. One difculty
with designing temperature control systems is that temperature perception
depends on individual preferences. Every individual has a particular sensitivity to temperature. This may create difculties for airlines in meeting the
differing comfort expectations of the passengers.
Additionally, working ight attendants have different temperature requirements than the seated passengers, typically requiring cooler temperatures
while they are working and more elevated temperatures during their rest
breaks. Temperature control is typically limited to the occupied cabin areas,
with the galleys drawing air from the cabin, or being incorporated into the
adjacent cabin temperature zone. There are however increasing moves to-
13
wards introducing temperature control possibilities into the galleys with the
installation of heated oor panels, dedicated heaters and individual air outlets
being studied industry wide.
The temperature perceived by the individual (the resultant surface temperature) is inuenced by the direct air temperature, the wall temperature
(radiation) and the air velocity (both the mean value and the uctuation
level). Additionally, humans like the head to be in a slightly cooler environment than the feet. While compensation for individual variability can be
made by using blankets and adding or removing clothing, the aim of the designer is to ensure that an acceptable temperature is provided globally when
considering the total cabin. Two main requirements have to be considered for
cabin air temperature to ensure this desire is met: the temperature selection
must be highly exible and the temperature must be as comparable as possible throughout a cabin zone in all three directions (longitudinal, vertical
and horizontal). As discussed in the previous section, temperature stratication and drafts are avoided by ensuring that ow patterns within the cabin are
optimised.
6
Humidity
The main source of humidity in the cabin is the occupants. The release of
humidity through metabolic processes is fairly predictable and can be calculated depending on the passenger load and in relation to the outside air
Fig. 6 Calculated decrease of relative humidity Airbus Library
14
ow per occupant provided to the cabin. Figure 6 indicates that fairly low outside air ows are necessary to achieve a humidity level usually perceived as
comfortable.
During ight, the air entering the cabin from outside is extremely dry (below 1% when the cabin temperature is taken into consideration), even when
the aircraft ies through clouds. The reason is the low saturation concentration of water in the cold air outside ( 40 to 70 C). As the occupant-released
humidity does not increase the level substantially, RH levels between 5 and
20% [6, 7] are usually found in aircraft cabins during cruise. Higher humidity
levels may be seen on the ground depending on the climate at the airport.
Comfort standards usually dene the lower RH limit at 30% [8, 9]. Low humidity is often perceived as the main comfort degradation for airline passengers from the environment control point of view. However, expert evidence
has not conrmed health risks associated with the low level of humidity [10].
Millions of people live in low-humidity environments, either in deserts, e.g.
Arizona, or in cold areas with indoor heating during the winter, e.g. Scandinavia. It must be taken into account however that those people are adapted to
low humidity levels, which is not necessarily the case for aircraft passengers
and crew.
Active humidication systems may be used by airlines to increase humidity levels in the cabin and thus improve thermal comfort. However, due to
weight constraints for the equipment and water required for current systems,
the RH can usually only be raised to around 20% in specic cabin compartments (crew rest areas, for example). The generally accepted comfort zone for
humidity is above 30% which means that the increases possible with todays
systems still fall short of optimal comfort.
Another area of consideration, aside from system design and capability, is
condensation. During ight the aircraft skin cools below the dewpoint temperature of the cabin air and what humidity there is in the cabin air may
condense onto the inner surface of the aircraft skin. During ight this water freezes, but during descent and ground phases this ice defrosts leading
to phenomenon such as rain in the plane where drops of water may fall
into the cabin through gaps around the ceiling panels. As well as having an
impact on the airlines image, this free water contributes to electrical faults,
particularly as more electronic systems are introduced into the cabin for IFE
and other cabin comfort systems. This effect may be seen on aircraft with
high density seating layouts combined with high load factors and short turn
around times (giving the aircraft less time to dry out). Drying systems, which
blow dry air into the ceiling area, are becoming increasingly available and
their use to combat the condensation effects of high density layouts and active
humidication systems is likely to expand.
A further challenge with respect to design for humidity control in the cabin
is during the ground phase in hot and humid environments. To cool the cabin
the air conditioning pack air outow must be cooler than the outside air,
15
which in hot and humid external conditions may lead to free water in the
distribution ducting. To prevent condensation, or even icing, in the distribution network, water is removed from the air stream before being cooled
in the pack. This is achieved by the introduction of a water extractor in the
air conditioning packs before the air passes through the nal cooling loop.
An energy saving measure may also be obtained by re-injecting this water
in to the ram air ow (ram air is used to provide the heat sink for the heat
exchangers), which through evaporative cooling further reduces the ram air
temperature, leading to increased heat exchanger efciency. The humidied
ram air does not enter the air conditioning pack air stream and is exhausted
overboard.
7
Ventilation Rates
As discussed in the section on thermal comfort, a major requirement for
the ECS is to provide a well-mixed, uniform temperature environment in
the cabin. This must be done without introducing drafts or temperature
stratication within the individual temperature control zones. To full this
requirement the ventilation ow must be sufcient to remove the heat load
in the cabin generated by the occupants, IFE and cabin operations such as
food preparation. Heat load dissipation is the main driver for the setting of
the ventilation ow rate. A minimum ventilation ow is required to maintain a sufcient level of pressurisation of the cabin at ight altitudes as well
as for contaminant dilution. However gaseous contaminant removal can be
achieved with relatively low outside air ow rates [11, 12] of around 5 cfm.
Flow rates required for heat load dissipation are generally higher than the
minimum required to full pressurisation and air exchange requirements.
With current development of equipment to remove gaseous contaminants
from recirculation air, complementing the standard installation of particulate
lters, it is probable that future aircraft designs will not require outside air
ow to manage internally generated contaminants, although some outside air
ow may be needed for pressurisation and temperature control reasons.
Recirculation of the extracted cabin air, after appropriate ltering to remove particulate contamination, helps to prevent temperature stratication
within the cabin. When the air leaves the cabin at oor level it is well mixed
with a stable temperature. A proportion of this air is remixed with cool air
delivered by the air conditioning packs. The recirculated air increases the
temperature of the pack air towards the lowest temperature demand among
the cabin temperature control zones. Where warmer air is required, trim air,
drawn from the bleed supply upstream of the air conditioning packs, is added
in the individual distribution ducts. An additional benet of adding recirculation air is a reduced requirement for bleed air, therefore less bleed air is drawn
16
off from the engines leading to better engine efciency and reduced fuel burn.
This helps reduce engine emissions in the atmosphere and therefore offers an
environmental benet.
8
Contaminant Control
Knowledge of air composition and contaminants in the cabin is developed
from either measurements or simulations or both. Some of the contaminants
of concern are relatively easy to measure, and due to their having a homogenous, unique source are easy to predict. Examples are carbon dioxide and the
oxygen content. Others are emitted by several sources or in different quantities, such as VOCs.
8.1
Cabin Operations Contamination Sources
Most of the contaminants can only be measured properly during normal service ights, as either the occupants themselves or the cabin operations are the
major emission sources. Cabin occupants are a source of gaseous and biological contaminants through normal metabolic processes. The cabin occupants
also introduce particulate contamination with their movement around the
cabin (levels of particulates have been found to be signicantly higher during
boarding than during cruise [13]). Animals carried on board also introduce
additional contaminants, as may items brought on board as carry-on luggage.
With respect to the cabin operations themselves, food and beverage preparation specically introduce particulate and gaseous contamination into the
cabin, while cleaning procedures in the cabin may introduce other contaminants. Currently gaseous contaminant removal is carried out through high
exchange rates of the cabin air. On those aircraft that recirculate air, particles
and biological matter are removed by lters in the recirculation system. HEPA
lters are recognised as being the current best practise and they are currently
installed as standard or optional equipment on all Airbus aircraft.
8.2
External Contamination Sources
Contaminant entry from outside the cabin during specic ground operations
is possible. In this case gaseous contaminants from the exhaust of surrounding aircraft and ground servicing vehicles may cause odour in the cabin.
These odours are caused by VOCs, present as combustion products. Techniques have been recently developed to remove these unpleasant odours from
the bleed air in the form of catalytic converters which use oxygen to break-
17
down VOCs into non-odorous compounds such as water vapour and small
amounts of carbon dioxide [14].
8.2.1
Carbon Dioxide
Carbon dioxide (CO2 ) is present in the external atmosphere at levels around
0.035%, equivalent to about 350 ppm. It is also produced in the cabin through
human metabolic processes the major emission source of CO2 on aircraft. Another source, related to specic airline cabin operations, is sublimation of dry
ice used for cooling of galley storage areas. Many major airlines have installed
dedicated galley cooling devices (air chillers) to cool the galleys storage areas,
especially for the long range aircraft eets, however some airlines do still use
dry ice for local cooling within the galley.
Measurements have shown CO2 levels in the cabin are generally between
400 and 3000 ppm, depending on passenger density and ight/ground phase,
with mean values around 1000 ppm [7, 16]. The certication requirements of
5000 ppm (0.5%) set by the FAA and JAA [1, 2] are not approached within the
breathing zones of passengers and crew during normal operations.
CO2 is frequently used, in general indoor air quality applications, as a surrogate for actual air quality [9, 17]. This is especially true for visitors entering a room with a certain occupancy level. Many people have experienced
this when coming into a room that has already been occupied for a certain
time without any air exchange occurring; the air is obviously odorous to the
new entrant. However, occupants who have been able to slowly adapt to the
air CO2 concentration will not detect CO2 related odours. This is precisely the
scenario in the aircraft cabin. There are no visitors to the cabin during ight
and thus elevated concentrations of CO2 cannot lead to a perception of bad
air quality.
8.2.2
Carbon Monoxide
Carbon monoxide (CO) is a toxic, non-coloured, non-odorous gas which prevents inhaled oxygen being taken up by haemoglobin when present in high
concentrations. It is a product of incomplete combustion. Since there are no
combustion sources within the ECS, CO is not usually present in the cabin
under normal operating conditions during ight. During ground operations
CO contained in the exhaust gases of surrounding ground servicing vehicles
or taxiing aircraft may enter the cabin under certain conditions. This cannot
be controlled and is limited to the ground phases.
Various indoor air quality standards bodies, depending on the environment and application, have set threshold values for CO between 25 and
50 ppm [18, 19]. The FAA and JAA certication requirement is 50 ppm [1, 2].
18
Values measured in aircraft cabins are usually below one or two ppm, with
peak values under 10 ppm [7, 8], these occurring usually during ground
phases. At these low contamination levels the air quality within the cabin is
not degraded.
8.2.3
Volatile Organic Compounds and Semi-volatile Organic Compounds
VOCs and SVOCs are generic terms for thousands of compounds containing
carbon and hydrogen, often oxygen and nitrogen, and sometimes elements
such as phosphorus and sulphur. SVOCs have a higher boiling point (above
250 C) compared to VOCs (50260 C). Emission sources range from vehicle exhaust gases on the ground, cabin occupants through human metabolism
(so-called bio-efuents), out-gassing of cabin furnishings and cabin operation (meals, beverages, servicing). As there are many compounds and several
possible emission sources, an assessment of repercussions on occupant health
or general cabin comfort is difcult.
Several measurement campaigns have been carried out to measure VOCs
during ight, [7, 13, 20]. The use of various adsorbent tubes, e.g. Tenax, and
different analysis methods allow as many compounds as possible to be detected with the maximum possible accuracy.
8.2.3.1
Volatile Organic Compounds
The largest fraction of the total VOCs present in the cabin air, between 70%
and 90% of the detected organic compounds [13], consists of ethanol. Its origin is simply the alcoholic beverage served during the ight. In fact, most of
the substances found in the cabin air during these measurement campaigns
are also present in the normal indoor air of homes and thus present no unusual exposure situation. Additionally, wherever the measured values could
be compared to existing or proposed indoor guide values, these values were
not exceeded in a single case. This was also the case for health and safety at
work exposure guidelines, such as the German MAK, in most cases even with
the application of an appropriate safety factor of 100.
The notable exception to the above in this study [13] was nicotine, which
was identied in the air of the smoking section. Concentrations approaching MAK-limit concentrations were determined in these areas, but did not
approach this in non-smoking sections of the cabin.
The general results described above correspond to the results of other
measurement campaigns and reviews, most performed in the USA [7, 15, 20].
The measurement campaigns cover most aircraft types from the main aircraft
manufacturers. With such results it can be assumed that VOC concentrations
19
are not heavily dependant on aircraft type but rather are connected to passenger loads and other sources of contaminants.
8.2.3.2
Semi-volatile Organic Compounds
SVOCs may be present as the breakdown products of aircraft uids such
as engine lubricants and hydraulic oils. These contaminants do not enter
the cabin air under normal operating and through specic design measures
ingress following failure conditions is also minimised. Precautions are taken
to prevent hydraulic liquid and fuel entering the air system; for instance ensuring air conditioning ducts are placed above the hydraulic uid lines in the
aircraft. APU oil seal leakages are extremely rare and engine oil seal leakages even more rare. Oil and lubricant ingestion into the bleed system from
spillages during servicing is also a rare event and recommendations for correct servicing practices are provided in aircraft maintenance manuals. As
SVOCs are conjectured to be present only after a very unlikely incident, and
highly infrequently, no reliable data has been collected on the possible types
and concentrations that could be expected.
8.2.4
Ozone
Ozone (O3 ) is produced in the upper layers of the atmosphere. It is usually present in relatively high concentrations above 65 000 ft (20 km) altitude.
Depending on season and latitude, pockets of ozone may occur at lower
altitudes, including altitudes at which commercial aircraft y. The ozone concentration is generally higher at northern latitudes during the winter and
spring months [21]. Ozone is highly toxic for humans and the regulatory authorities have set the maximum allowable concentrations within the cabin
and cockpit as follows [1, 2]:
0.1 ppm for any three-hour period when the aircraft is above FL 270
(27 000 ft (8230 m))
0.25 ppm as maximum concentration at any time when the aircraft is
above FL 320 (32 000 ft (9755 m))
Measurement of ozone during in-ight surveys have returned results showing
ozone remains within the certied limits set down above [6], although lack of
data has been highlighted as a concern [3].
The outside ozone concentration may reach 1 ppm for a short time at
cruise altitudes. Ozone is unstable and its decomposition is accelerated by
heat (for instance, in the bleed system) and contact with metallic surfaces and
components within the cabin itself. However, this natural breakdown process
is not sufcient for such elevated outside concentrations. Ozone converters
20
are therefore required for aircraft ying routes that pass through latitudes
where high levels of ozone are expected. These routes cover almost all city
pairs being bounded through northern ight paths (north Atlantic, north
Pacic) and therefore most long-range aircraft are equipped with ozone converters. Ozone converters are also offered as optional equipment by Airbus
for their shorter range aircraft types.
8.2.5
Bacteria, Viruses, Fungi, Moulds
The presence of microbiological matter in the cabin has been the subject of
debate for some years, with the SARS outbreak providing additional data for
the understanding of the mechanisms at work. There is a general misconception that bacteria and viruses are propagated through the ECS. Several
studies have now shown that the mechanism of transmission is proximity
and person-to-person contact. Both the WHO tuberculosis study [22] and
the CDC led study of possible airline transmission of SARS [23] indicate that
proximity, specically in the few rows in front of the index case, is a major factor in the transmission of these diseases. In the case of the tuberculosis study,
exposure time was also found to have played a signicant part, with no transmission occurring on ights with duration under eight hours. These studies
conrm the position of the aircraft manufacturers that micro-organisms are
not spread through the ECS.
Bacteria and fungi can be measured with handheld devices collecting onto
agar plates. The analysis of these plates is carried out in a laboratory where
two complementary analyses are performed; an overall identication of the
amount of bacteria and fungal spores in the air in addition to an analysis
regarding the species captured. Viruses however cannot be measured with
handheld devices and virus concentrations are therefore not measured directly in the cabin during in-service ights. Some studies [24] have taken
a sideways look at the question of infection rates although they do not address
transmission methods and thus do not allow conclusions to be drawn with respect to the amount of viral material in an aircraft cabin at any given time.
They also fail to take into account other mechanisms, such as the number of
people from different backgrounds met by the cabin crew, environment differences between departure and arrival airports and physiological factors such
as jet lag.
In [13] it was found that very low concentrations of mainly non-pathogenic
bacteria and fungi were detected in the different cabin sections. It was also determined that there was a proportional increase in the germ concentration as
the passenger density increased. Thus the highest concentrations were found
in the economy class section of the aircraft measured. On the other hand,
the germ concentrations in the in-coming ventilation air was so low that it
would full requirements for operating theatres (Airbus A340) or intensive
21
care wards (Airbus A310). When peak concentrations of bacteria occurred,

for instance following a sneeze or a cough in the vicinity of the measurement equipment, the concentrations reduced rapidly to the background level.
This indicated the efcacy of the aircraft ventilation system. The Institute for
Hygiene and Environmental Medicine of the Medical University of Lbeck
provided expert evidence that bacteria of the type and concentration found
are irrelevant for health considerations and conrmed that in their view the
only actual health risk is in person-to-person contact. The infection is transmitted over short distances as droplets after a sneeze or cough.
Other studies have come to the same general conclusions. Another
study, [15], conrmed a low bacteria concentration in aircraft cabins when
compared with other modes of transportation. Considering these results one
can summarise that concentrations of bacteria and fungal spores are generally low in aircraft cabins and are not spread throughout the cabin by the air
recirculation system, with the biological contaminant survival rate generally
decreasing at low RH.
8.2.6
Particulates
Measurements of the dust load (weight per volume air) have shown relatively
low concentrations of particulates within aircraft cabins [6, 7, 16]. This is particularly so in aircraft cabins when smoking is prohibited. However, most
studies have not measured the particle load in the outside air or the recirculation air. Particle count and size distribution have also generally not been
measured.
In order to better clarify total particulate contaminant loads, cabin interior particulate loads, the dust loads within the recirculation air and a control
measurement outside the aircraft at departure and arrival airports have been
studied [13]. The study involved xed installed particle counters in the cabins of two aircraft, an Airbus A310 and an Airbus A340. The particle counters
were installed in the rst or business class, the economy class and in the
smoking section (if available on the aircraft). Particle counters were also
installed in the recirculation system with handheld measurements being conducted for the outside measurements.
The two aircraft represent two different cabin distribution and recirculation philosophies, with the A310 having local mixing in the dedicated cabin
zones and the A340 having a central mixing unit with distribution to the cabin
zones. Additionally, the A310 recirculation air is ltered with EU9 class lters (90% efciency at 0.5 micron) while the A340 has HEPA-lters installed
(lters as dened by EN 1822-1 [25]) approximately equivalent to the EU13
classication.
The study results conrmed the hypothesis regarding the effect of lter efciency. As was expected, the mean particle concentration within the
22
recirculation air was found to be lower than, or equal to, the outside air
concentrations for all ground and ight cases for both aircraft types. Additionally, on those aircraft with HEPA lters installed, the outside air is up to
2800 times more highly contaminated with particles compared to the recirculation air. Even during cruise, when the outside air has a very low particle
concentration, the outside air is still up to 250 times more contaminated with
particles compared to the recirculation air.
Additionally, the occupants, cabin furnishings such as carpets and cabin
operations such as meal services could be conrmed as the main emission
sources for particles, since substantially higher concentrations were measured in the cabin than in the supply air. As might be anticipated, the study
also found that particle concentrations were much higher in the smoking section when compared to the non-smoking section, especially during ight.
9
Interdependency of Factors
When considering aircraft design parameters many factors from single contaminants to individual aspects of the cabin environment are taken into
account. Some of these factors are shown in Fig. 7. However, the human perception of comfort is not just a sum of several variables. Most of the factors
already discussed inuence others and are themselves inuenced by others.
To make matters even more complicated, the perception of the cabin environment is not only inuenced by these main factors but also by a lot of
other variables that are not inuenced or controlled by aircraft systems. There
are person-related aspects such as demographics and personal constitution,
cabin operations and work schedule related items for the cabin crew, and the
specic physiological aspects such as the long sedentary position for the pas-
Fig. 7 Interdependency of factors Airbus Library
23
sengers and ight crew. Other factors such as the perception of space, fear
of ying, the colour and light perception also have a signicant inuence
on the overall comfort perception, however, communication of these feelings
and perceptions may be difcult for the occupants themselves. For instance,
there is some evidence that low cabin pressure combined with some alcohol
consumption increases the complaint rate substantially.
However there is a considerable knowledge gap within both the engineering and medical communities regarding these interdependencies. Work has
been carried out within the EU-funded research programmes ASICA, FACE,
CabinAir and HEACE to further identify and understand interdependencies,
however, it is certain that such a complex subject will continue to draw research efforts. In order to full recommendations from the NRC report, [3],
a centre of excellence is being put in place in the USA. An ASHRAE sponsored
research project into multiple factors combined with an FAA sponsored monitoring package installation has also started, and these results will hopefully
ll in some currently existing knowledge gaps. Knowledge of how the interdependencies function is ultimately seen as the key to decreasing complaint
rates from passengers and crew, since none are individually critical for human health and most do not substantially reduce the comfort perception by
themselves.
10
Summary
The main challenge for aircraft ventilation systems is to ensure that the predicted air requirement for a special zone is in fact achieved. This means that
they realise the objectives of an equal distribution throughout the length of
the cabin as well as an appropriate ow pattern within the cabin.
The analysis of the mentioned parameters shows that none of them are
critical for the health of the cabin occupants and do not adversely affect comfort. This is especially true for healthy individuals. However, there is a lack of
know-how surrounding single parameters, such as the cabin pressure for unhealthy, very young and elderly persons, as well as for the interdependencies
of the different factors, which are not limited to the cabin environment but
also include personal and operational aspects.
References
1. JAR 25.831/841, Amendment 16
2. FAR 25.831/841, Amendment 105
3. National Research Council (2002) The Airliner Cabin Environment and the Health of
Passengers and Crew. National Academy Press, Washington DC
24
4. House of Lords (15/11/00) Air Travel and Health. Her Majestys Stationery Ofce, UK
5. Parliament of the Commonwealth of Australia (2000) Air Safety and Cabin Air Quality
in the BAe 146 Aircraft. Senate Printing Unit, Canberra, Australia
6. ASHRAE Research Project 957-RP (1999) Relate Air Quality and Other Factors to
Symptoms Reported by Passengers and Crew on Commercial Transport Category
Aircraft 7.
7. Dumyahn TS, Spengler JD, Burge HA, Muilenburg M (2000) Comparison of the Environments of Two Transportation Vehicles: Results of Two Surveys. In: Nagda NL (ed)
ASTM 1393. American Society for Testing and Materials, West Conshocken, PA
8. DIN 1946, part 2 (1994) Ventilation and Air Conditioning. Beuth Verlag, Berlin
9. ANSI ASHRAE Standard 55 (1992) Thermal Environment Conditions for Human Occupancy
10. Thibeault C (1997) Special Committee Report: Cabin Air Quality. Aviat Space Envir
Md 68:8082
11. ANSI ASHRAE 62 (2001) Addendum n, Ventilation for Acceptable Indoor Air Quality
12. Cain WS, Leaderer BP, Isseroff R, Berglund LG, Huey RJ, Lipsitt ED, Perlman D (1982)
Ventilation Requirements in Buildings Control of Occupancy Odour and Tobacco
Smoke Odour. Atmos Environ 17(6):11831197
13. Dechow M, Sohn H, Steinhanses J (1997) Chemosphere 35(1):2133
14. Nurcombe C (2004) Treatment of Aircraft Ground Air Sources (VOC Removal). Presentation to ASHRAE Summer Meeting, Nashville
15. Spengler JD, Burge H, Dumyahn T, Muilenberg M, Forester D (1997) Environmental
Survey on Aircraft and Ground-based Commercial Transportation Vehicles. Harvard
School of Public Health, Harvard University, Cambridge, MA
16. Lindgren T, Norbck D, Andersson K, Dammstrm BG (2000) Aviat Space Envir Md
71:774782
17. ANSI/ASHRAE Standard 62-89 (1989) Ventilation for Acceptable Air Quality
18. American Conference of Governmental Industrial Hygienists (ACGIH) Threshold
Limit Value (TLV): 25 ppm, 29 mg/m3 TWA
19. World Health Organisation (WHO) Threshold Limit Value (TLV): 10 ppm, 11.6 mg/m3
8h TWA
20. Space DR, Johnson RA, Rankin WL, Nagda NL (2000) The Airplane Cabin Environment: Past, Present and Future Research. In: Nagda NL (ed) ASTM 1393. American
Society for Testing and Materials, West Conshocken, PA
21. FAA (1980) Transport Category Airplanes Cabin Ozone Concentrations AC 120-38.
FAA, Washington, DC
22. Valway S, Watson J, Bisgard C, Scudeller L, Espinal M, Raviglione M (1998) Tuberculosis and Air Travel: Guidelines for Prevention and Control. WHO, Geneva
23. Olsen SJ, Chang HL, Cheung TY, Tang AF, Fisk TL, Ooi SP, Kuo HW, Jiang DD,
Chen KT, Lando J, Hsu KH, Chen TJ, Dowell SF (2003) New Eng J Med 349:24162422
24. Whelan EA, Lawson CC, Grajewski B, Petersen MR, Pinkerton LE, Ward EM, Schnorr
TM (2003) Occup Environ Med 62:929934
25. European Standard (1998) EN 1822-1 High Efciency Particulate Air Filters (HEPA
and ULPA)
Hdb Env Chem Vol. 4, Part H (2005): 2551

DOI 10.1007/b107235
Occupational and Public Health Risks

Judith Murawski
AFL-CIO, Association of Flight Attendants-CWA, K Street NW, 1275, #500,
Washington, DC 20005, USA
jmurawski@afanet.org
1
1.1
1.2
1.3
Introduction . . . . . . . . . . . . . . . . . . .
Regulatory Environment . . . . . . . . . . . .
Sources of Publicly Available Data . . . . . . .
Health Risks Attributed to Aircraft Air Quality
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
26
26
27
31
2
2.1
Elevated Levels of Bioefuents . . . . . . . . . . . . . . . . . . . . . . . . .

Reported Symptoms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
32
34
3
3.1
Infectious Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
35
36
4
4.1
Extreme Temperatures (See also Chap. 3) . . . . . . . . . . . . . . . . . . .

38
38
5
5.1
Exhaust Fumes, Deicing Fluid, Fuel Fumes, and Cleaning Products . . . .

39
40
6
6.1
Heated Engine Oil and Hydraulic Fluid (See also Chap. 10, and 12) . . . .
40
41
7
7.1
Reduced Oxygen Supply (See also Chap. 3, and 4) . . . . . . . . . . . . . .

44
44
8
8.1
Ozone Gas (See also Chap. 3) . . . . . . . . . . . . . . . . . . . . . . . . .

45
46
9
9.1
Insecticides (See also Chap. 9) . . . . . . . . . . . . . . . . . . . . . . . . .

47
47
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
48
Abstract Symptoms reported by passengers and crewmembers on commercial aircraft are

described according to individual air quality-related sources, including: (1) elevated levels
of bioefuents; (2) infectious agents; (3) extreme temperatures; (4) exhaust fumes, deicing uid, fuel fumes, and cleaning products; (5) heated engine oil and hydraulic uid;
(6) reduced oxygen supply; (7) ozone gas; and (8) insecticides. A brief overview of the aircraft regulatory environment and available sources of data on the hazards and associated
health effects is also provided.
Keywords Carbon monoxide Insecticides Ozone Tricresylphosphates Ventilation
26
J. Murawski
Abbreviations
APU
Auxiliary power unit
BLS
US Bureau of Labor Statistics
CAMI US Civil Aeromedical Institute
CAR
Canadian Aviation Regulation
CFM
Cubic feet of air per minute
CO
Carbon monoxide
CS
European Certication Specication
EASA European Aviation Safety Agency
ECS
Environmental control systems
ETS
Environmental tobacco smoke
FAA
US Federal Aviation Administration
FAR
US Federal Aviation Regulation
ITF
International Transport Workers Federation
JAA
Joint Aviation Authorities (Europe)
L/s
Liters per second
NIOSH US National Institute for Occupational Safety and Health
SARS Severe Acute Respiratory Syndrome
TB
Tuberculosis
TC
Transport Canada
TCPs Tricresylphosphates
TWA Time-weighted average
WHO World Health Organization
1
Introduction
Passenger and crew reporting of symptoms attributed to substandard aircraft air quality is a controversial subject, and has been a source of debate
among regulators, airlines, aircraft/component manufacturers, passengers,
and crewmembers. As background, it is helpful to understand: (1) the regulatory environment, (2) the sources of available data on aircraft air quality
hazards, and (3) the potential sources of symptoms reported by passengers
and crew.
1.1
Regulatory Environment
Air quality standards intended to protect airline worker safety and health are
under the jurisdiction of each countrys aviation authority. In the US, the Federal Aviation Administration (FAA) asserted its exclusive claim of jurisdiction
over airline worker safety and health on registered civil aircraft in operation [1]. Basic protections established and enforced by the US Occupational
Safety and Health Administration do not apply to crewmembers, and the FAA
has not published occupational safety and health protections for air quality-
27
related hazards. Aviation authorities in most other countries have the same
arrangement. They argue that air quality standards may inuence equipment
selection and maintenance, which may in turn require modications to the
aircraft structure, which could compromise the safety of ight.
Aviation authorities worldwide issue two basic types of aircraft air quality
regulations: design standards that manufacturers must comply with during
the aircraft certication process, and operating standards that airlines must
comply with when operating an aircraft. There are very few design standards
for aircraft environmental control systems (ECS) (Table 1), and even fewer
operating standards for such systems (Table 2). There is no requirement to install and operate air quality monitoring equipment on aircraft, so compliance
with operating standards that dene chemical exposure limits is not ensured.
Aviation authorities occasionally require airlines to implement specic
maintenance procedures, sometimes in response to recommendations from
aircraft or component manufacturers. For example, British Aerospace issued
maintenance recommendations to airlines that operate their BAe146 aircraft
in response to incidents that involved impaired performance of ight crew
and circumstantial evidence that such incidents were caused by oil leaking into the air supply systems [2]. These maintenance procedures have since
been incorporated into regulations by several aviation authorities.
Similarly, in response to reports of smoke and odor in the passenger cabin
and cockpit due to hydraulic uid leaking into the auxiliary power unit inlet, and subsequently, into the air conditioning system, the FAA now requires
that airlines implement specic maintenance procedures to increase the robustness of specic hydraulic uid lines shown to be prone to failure in the
auxiliary power unit (APU) of certain aircraft types [3]. This requirement will
address one source of contamination on one series of aircraft.
Fleet-wide regulations allow airlines to dispatch aircraft that have an inoperative APU for a limited period of time, typically 10 days. Even if the APU
is inoperative due to contamination with oil or hydraulic uid, the aircraft is
still considered airworthy.
No aviation authority requires airlines or aircraft/component manufacturers to provide affected crew or passengers who are exposed to airborne toxins
during a ight with either aircraft maintenance or mechanical records (when
relevant), or product information. Also, there are no requirements for routine
in-ight air quality monitoring. For these reasons, it is often difcult for affected crew and passengers to prove an association between their symptoms
and deciencies in the quality of the air onboard.
1.2
Sources of Publicly Available Data
It is challenging to dene the extent of occupational and public health risks
attributed to aircraft air quality because there is no large-scale and indepen-
Provide a sufcient amount of

uncontaminated air to enable
the crewmembers to perform their
duties without undue discomfort
or fatigue, and to provide reasonable
passenger comfort (FAR 25.831(a)).
Effective 5 July 1996, systems on new
aircraft types must also be designed
to provide a minimum 0.55 lb
of outside air per min to each
occupant
Carbon dioxide 5000 ppm in compartments normally occupied by passengers
or crewmembers (FAR 25.831(b)(2)).
Carbon monoxide 50 ppm
(FAR 25.831(b)(1)).
Ozone 0.25 ppm (ceiling) when operating above 32 000 feet, and 0.1 ppm
TWA during any 3-h interval
above 27 000 feet (FAR 25.832(a))
Note: All concentrations provided as sea level equivalent
Carbon
monoxide
Ozone
Carbon
dioxide
Ventilation
US (FAA)
Carbon dioxide < 30 000 ppm

for crewmembers
(CS 25.831(b)(2))
Same as US regulation;
see CS 25.831(b)(1).
None required
Each passenger and crew compartment must be ventilated, and each

crew compartment must have enough
fresh air (but no less than 10 cubic feet
per min per crew member)
to enable crew members to perform
their duties without undue discomfort
or fatigue (CS 25.831(a));
see also AMC 25.831(a)
Europe (EASA)
see CAR 525.831(b)(1).
see CAR 525.832(a)
see CAR 525.831(b)(2)
see CAR 525.831(a).
Canada (TC)
Table 1 Aircraft system design standards published by the US Federal Aviation Administration (FAA), the European Aviation Safety Agency
(EASA), and Transport Canada (TC)
28
J. Murawski
None required
Combustion
heating
Note: All concentrations provided as sea level equivalent
None required
System operation
at low
temperatures
Provide an effective altitude (pressure)

in the cabin of not more than 8000 feet
operating altitude of the airplane
under normal operating conditions
at the maximum (FAR 25.841(a))
US (FAA)
Cabin
pressure
Table 1 (continued)
None required
None required
see CS 25.841(a)
Europe (EASA)
Combustion heaters must be

approved (CAR 525.833)
Must demonstrate satisfactory

airplane operation after it has
experienced prolonged exposure
to ground ambient temperatures
equal to or less than 35 C
(CAR 525.1301-1)
see CAR 525.841(a)
Canada (TC)

29
30
J. Murawski
Table 2 Aircraft air quality operating standards published by the European Joint Aviation Authorities (JAA), US Federal Aviation Administration (FAA), and Transport Canada
(TC)
Ozone
Ventilation
Carbon
monoxide
US (FAA)
Europe (JAA)
Canada (TC)
0.25 ppm (ceiling)

when operating above
32 000 feet,
and 0.1 ppm TWA
during any 4-h
interval above
27 000 feet
(FAR 121.578(b));
also see AC 120-38.
Each passenger or
crew compartment
must be suitably
ventilated
(FAR 121.219)
Carbon monoxide
50 ppm and fuel
fumes may not be
present
(FAR 121.219)
None apply
Same as
US regulation
None apply
Same as
US regulation
None apply
Same as
US regulation
dent incident reporting system for either crewmembers or passengers. In the

US, only incidents that render a pilot unable to perform his or her duties must
be reported to the National Transportation Safety Board [4].
Aviation authorities do not collect work-related illness reports from either airlines or crewmembers, although US airlines must log crewmemberreported occupational illnesses and recordable injuries (i.e., serious enough
to require lost work time or medical follow up). Each year, a sample of airlines is enlisted to participate in the US Bureau of Labor Statistics (BLS)
annual survey of occupational injuries and illnesses. The BLS has extensive
documentation of crewmembers work-related injuries and illnesses, and its
analysts provide extensive information upon request. A major union representing cabin crew collected copies of these injury and illness logs from 13
airlines, and calculated that 10% of the 31 422 employed cabin crew had reported a work-related illness or a recordable injury [5]. The national average
across industries for the same year was 3.1% [6]. Although the data that the
BLS collects are useful, there is evidence that the lists of air quality-related illnesses currently maintained by the airlines underestimate the true rate, even
of reported incidents, highlighting the need for independent oversight of any
data collection.
31
Some major government-funded reports have been published on the subject of aircraft air quality [712]. The majority of published data is funded by
industry sources, which has generated concerns about potential conicts of
interest.
1.3
Health Risks Attributed to Aircraft Air Quality
Symptoms reported by passengers and crew are described in Sects. 29, according to the eight potential exposure sources listed in Table 3. Aircraft
occupants may be subject to any combination of these, or none at all, during
a given ight. Physical hazards such as high noise levels, cosmic/solar radiation, and electromagnetic elds have been documented in the aircraft cabin
and cockpit, but are beyond the scope of this chapter.
An historical aircraft air quality hazard, at least in most parts of the world,
is environmental tobacco smoke (ETS). Key historical developments in the
smoking ban on commercial ights are listed in Table 4 [1316]. Smoking is
prohibited on domestic ights in most countries, and is restricted on most international ights; for example, airlines in the UK, Germany, Italy, and Russia
have imposed a voluntary smoking ban on international ights, while airlines
in France and the Ukraine allow smoking in sections of the aircraft, depending on the destination [17]. Even in countries that have enacted smoking bans
on all commercial ights, the majority of currently employed crewmembers
have been exposed to ETS in the aircraft cabin during their careers. ETS exposure for a full-time cabin crewmember on smoking ights has been described
as equivalent to living with a one-pack-per-day smoker [18]. Research into
tobacco-related diseases among crewmembers is currently being conducted
in the US by the Flight Attendant Medical Research Institute, funded by a settlement from a 1991 lawsuit led against tobacco companies.
Table 3 Potential sources of air quality related symptoms reported by passengers and
crewmembers during commercial airline ights
Section
2
3
4
5
6
7
8
9
Potential source of air quality related symptom

Elevated levels of bioefuents
Infectious agents (bacteria and viruses)
Extreme temperatures
Exhaust fumes, deicing uid, fuel fumes, and cleaning products
Heated engine oil and hydraulic uid
Reduced oxygen supply
Ozone gas
Insecticides
32
J. Murawski
Table 4 Key historical developments in the smoking ban on commercial aircraft

Year
Development
1988
The US congress passed legislation that banned smoking on domestic

ights lasting 2 h or less [13]
The US smoking ban was expanded to include domestic ights
lasting 6 h or less [14]
The International Civil Aviation Organisation issued a non-binding resolution
calling for smoke-free ights by 1 July, 1996. Only 13% of the more than
300 member countries had implemented the ban by then [15]
The US congress expanded their smoking ban to include all international ights
traveling to and from the US [16].
1990
1992
2000
It has been suggested that some symptoms reported by crew and passengers may be explained, not by problems with aircraft air quality, but with
multiple factors such as jet lag, dehydration, fatigue, or simply hysteria.
Certainly being on duty for long ights, crossing time zones, and attending
to the public are stressors in and of themselves. However, documented symptoms are often correlated with documented air quality problems on aircraft.
In some cases, incident type or symptoms are more prevalent on certain aircraft models, aircraft, or ight paths, making air quality related problems
easier to dene.
2
Elevated Levels of Bioeffluents
There is no operating standard for a minimum per person ventilation rate in
the passenger cabin. Airlines need only maintain cabin pressure, which requires a per person outside air ow rate of approximately 3 cubic feet per
minute (CFM) or 1.4 L/s). Reducing outside air supply conserves fuel. On
most commercial jet aircraft, the air supply is approximately 50% outside air
and 50% recirculated; this will vary somewhat between aircraft manufacturers [7]. Some regional aircraft operate with 100% outside air.
A discussion of ventilation and bioefuents (including infectious agents)
on aircraft often relies on comparisons to building data. There is a shortage of
reliable aircraft data compared to the volumes of documented ground-based
investigations into non-specic symptoms that are reported by a proportion of building occupants and typically diminish upon leaving the building.
Aside from some obvious differences between aircraft and buildings, there
33
are ve key distinctions that inuence exposure to bioefuents in these environments:

Low per person outside airow: Published data show that levels of gaseous
bioefuents (such as carbon dioxide) are considerably higher in aircraft
than typical building environments, indicative of low per person ventilation rates (Table 5) [1923]. Although carbon dioxide itself is not considered to be hazardous at the concentrations reported in these surveys, it
is an indicator of outside air supply, and elevated levels have been associated with an increased prevalence of non-specic symptoms reported by
occupants in ground based environments [24].
Low per person recirculated airow: The per person volume of ltered, recirculated supply air is considerably lower on aircraft (710 CFM;
3.34.7 L/s) than in buildings (70 CFM; 33 L/s), raising concerns about exposure to increased airborne levels of particulate, including viruses and
bacteria. It is possible that the quality of recirculated air on aircraft may
be higher than in buildings, although the trade off between volume and
quality of ltered, recirculated air has not been quantied.
Airborne contaminants from off-gassing surfaces: The per person building surface area is considerably smaller on aircraft than in building environments. If surface off-gassing is the primary cause of reported symptoms, then fewer aircraft occupants should report symptoms, all other
things being equal. However, it may not be possible to draw a direct comparison because the key sources of off-gassing surfaces are different on
aircraft (e.g., lavatory, galley kitchen) than in buildings (e.g., photocopier).
Airborne contaminants from aircraft occupants: The occupant-generated
contaminant load per unit volume of space is considerably greater on
aircraft than in building environment. If elevated levels of bioefuents explain reported symptoms then one would expect a higher prevalence of
symptoms reported by aircraft occupants, all other things being equal.
Small volume of air assigned to each person to dilute contaminants:
The per person volume of air space that is effectively provided to each
aircraft occupant for dilution of airborne contaminants is approximately
one-tenth that provided to building occupants [25, 26], such that equilibrium concentration of airborne contaminants generated by occupants is
reached more quickly. This high occupant density on aircraft compared
to buildings also has implications for the overlap of occupants breathing zones, particularly on full ights between neighboring passengers, and
between cabin crew and passengers during beverage and meal services.
34
J. Murawski
Table 5 Results of carbon dioxide monitoring on commercial aircraft

Waters et al., 2002 [19]
Pierce et al., 1999 [20]
Spengler et al., 1997 [21]
Jurgiel et al., 1994 [22]
Nagda et al., 1992 [23]
Funded by the US National Institute for Occupational

Safety & Health to monitor air quality
on 36 ights. The average ightlong concentration
of carbon dioxide was 1387 ppm
Funded by the American Society of Heating,
Refrigerating, and Air Conditioning Engineers
to monitor air quality on eight B777 ights.
Conditions during ground operations when carbon
dioxide levels are expected to peak were not
monitored. Average in-ight levels in the cabin
were 1509 ppm. Average levels in the aft galley
were 2480 ppm, not including a measurement
of 4915 ppm that was attributed to a nearby
bin of dry ice
Funded by the Boeing Company to collect air quality
measurements on aircraft and other modes of
transportation. Average CO2 levels on four B777 ights
were reported as 12001800 ppm during cruise,
and 10002300 ppm during boarding
Funded by Trans World Airlines to collect air quality
monitoring data on two B747-100 non-smoking ights
according to the location in the cabin and phase of
ight. Carbon dioxide levels during ground operations
averaged 2480 ppm. Cabin class also inuenced
ventilation, with 10471510 ppm measured
during cruise in the economy section, compared to
884950 ppm in the rst class, and 720 ppm in the
upper deck. Cockpit levels averaged 740 ppm
Funded by the US Departement of Transportation to
conduct air quality monitoring on 23 domestic,
non-smoking ights. The average carbon dioxide
concentration was 1756 ppm with 87%
of the data exceeded the 1000 ppm upper limit
recommended by ASHRAE Standard 62
2.1
Reported Symptoms
In ground-based environments, the following symptoms have been associated
with low outside air ventilation rates: digestive problems; dizziness; dry or
burning mucous membranes in nose, eyes, or throat; fatigue or lethargy; forgetfulness; headaches; inability to concentrate; irritability; nausea; sneezing;
and stuffy or runny nose [24]. Few published studies have investigated either
the prevalence of these symptoms reported by crew or passengers on com-
35
mercial aircraft, or whether there is an association with outside air ventilation

rates.
A survey conducted by the US National Institute for Occupational Safety &
Health (NIOSH) compared the prevalence of self-reported respiratory symptoms and illnesses between cabin crewmembers, teachers, and an external
population of women blue collar workers with no known occupational exposures [27]. Cabin crew were four to six times more likely to report workrelated eye, nose, and throat symptoms than the referent working women.
Cabin crew were also less likely than teachers or the referent working women
to report ever having been diagnosed with asthma (8.2%, 13%, and 12%, respectively).
A survey commissioned by Scandinavian Airlines Systems compared the
perceptions of workplace air quality and reported symptoms of crewmembers (n = 1513) and ofce workers (n = 168) employed by SAS [28, 29]. At the
time, smoking was permitted on intercontinental ights. Crewmembers were
less satised with their work environment than ofce workers, and reported
more nasal and throat symptoms, as well as dermal symptoms on the hands
and face. It would be worth repeating this survey now that smoking is not
permitted on aircraft.
A survey commissioned by Cathay Pacic Airways evaluated the health
and comfort of cabin crew on 16 international ights over an eight-month
period [30]. Most of the surveyed ights were smoke-free except for a few
short-haul ights that were divided into smoking and non-smoking sections.
On each ight, researchers distributed health surveys to cabin crew (N = 185)
and collected basic air monitoring data. The average carbon dioxide level
was 934 ppm (6831557 ppm). Half of the respondents classied air quality
as poor or adequate, both less than acceptable according to the rating
scale.
3
Infectious Agents
Anecdotally, passengers and crew report an association between infectious
disease transmission and air travel. These reports are consistent with the
close proximity of cabin occupants and the low ventilation rates on aircraft;
however, it is difcult to substantiate these claims because of the latency
period between infection and symptoms, and the challenge of contacting passengers and crew after any given ight.
Aircraft occupants can be infected by two routes of exposure: (1) airborne,
and (2) surface contamination (i.e., touching an infected surface such as a cup
or lavatory door handle, and then touching ones mouth or eyes).
Risk factors for airborne exposure include low total ventilation rate per
person, inadequately ltered recirculated air, and close proximity of occu-
36
J. Murawski
pants. Seating conguration and activity will also affect the degree of overlap
between occupants breathing zones, and therefore the risk of infection. Risk
factors for transmitting an infection via contact with a contaminated surface
include inadequate hand washing and touching ones face.
The relative contributions from airborne and surface sources within the
aircraft has not yet been dened, although the exposure potential, and therefore, risk of infection, are expected to increase relative to the duration of the
ight. Other variables include pathogen type (i.e., clinically relevant dose)
and individual susceptibility to infection.
For the airborne route, maximizing the airow through the cabin will reduce the residency time of airborne infectious agents. Recirculated air that is
ltered through properly maintained and installed true high efciency particulate (HEPA) lters should provide similar protection to outside air, in terms
of infectious disease control. Bacteria can easily be trapped by a true HEPA
lter and, although viruses are smaller than the pores of a HEPA lter, they
should be removed from the air stream if they travel in clusters or on droplets
of water or mucous. Even the best HEPA ltered air still contains gaseous
contaminants, and therefore does not have the dilution capacity provided by
outside air.
The benet of recirculated air is that it is more humid than the outside air
supplied by the engines; one downside is that the source of humidication is
aircraft occupants breath. Presently, although some of the major airlines report that they have done so, there is no minimum requirement to install or
properly maintain HEPA lters on aircraft.
3.1
Reported Symptoms
A survey conducted by NIOSH and introduced in the previous section reported that ight attendants were more likely to report ve or more episodes
of colds or u in the past year than either teachers or working women in
a comparison population (10.2%, 8.2%, and 2.3%, respectively) [27]. Flight attendants were also signicantly more likely than teachers and referent working women to report chest illness (33%, 19%, and 7.2%, respectively).
A study of self-reported colds and u tracked 1100 passengers that traveled on one of 250 2-h ights between the same two US cities, half on aircraft
with a 50%50% mix of recirculated and outside air, and half on similarly
congured aircraft with 100% outside air [31]. There was no signicant difference in the frequency of self-reported colds and u between groups, leading
the researchers to report no evidence that aircraft cabin air recirculation
increases the risk for (upper respiratory tract) symptoms in passengers traveling aboard commercial jets. A subsequent letter to the editor noted that
the rate of upper respiratory infection reported by the cohort of airline passengers was four times the national average, suggesting an increased risk of
37
disease transmission on commercial ights [32]. This increase may be explained by the overlap of occupants breathing zones, low per person total
ventilation rates compared to other environments, contact with infected surfaces, or some combination thereof.
In its 2003 investigation of Severe Acute Respiratory Syndrome (SARS), the
World Health Organization (WHO) dened contacts as passengers within
two seats rows of an infected person and all on-board cabin crewmembers [33]. Presumably, the WHO recognizes the potential for ambient air to
drift between seat rows before returning to the air supply system or being dumped overboard. On one ight, passengers sitting seven rows in front
and ve rows behind a person with symptomatic SARS developed the disease;
however, in this case, the route of transmission (i.e., airborne versus contact)
was not conrmed. If a cabin crewmember is the suspected or probable SARS
case, then all the passengers are considered contacts [33].
The implications of contracting SARS during a ight are especially serious
given the overall case fatality rate of 15% [34], and evidence that a surface can
stay infected for up to 4 days [35]. To date, there have been 27 reported cases
of SARS transmission on aircraft, involving four cabin crewmembers and 23
passengers [36].
There has been considerable interest in the risk of transmitting tuberculosis (TB) on aircraft. One of the more conclusive investigations was conducted
by the US Centers for Disease Control and Prevention (CDC) and involved
802 (87%) of passengers and crew who had traveled on one of four ights
with a person who had infectious TB [37]. There were two outbound ights
and two return ights a month later, by which time the patients condition
was reported to have worsened. On the rst three ights, a total of 14 contacts had positive tuberculin skin prick tests, although of these, 13 had other
risk factors for TB. However, on the last ight (lasting 8.75 h), 15 contacts
had positive tuberculin skin tests and, of these, six had no other risk factors for TB and were seated in the same cabin section as the index case,
four within two rows of her. The observed pattern of infection within the
cabin suggests the potential for drift of infected air between rows. The
absence of reported skin-test conversions in other cabin sections implies
that bacteria were not transmitted through the aircrafts air recirculation
system.
A less conclusive investigation into the risk of TB transmission on aircraft
involved 225 (73%) passengers and crew on a 14-h ight with one person who
was highly infectious [38]. Of these, 184 had positive tuberculin skin prick
tests for TB, although only nine had skin conversions. Of those nine, the possibility of transmission from the index patient could not be ruled out in three
cases, although all three were sitting between 15 and 23 rows from the index
patient, not a compelling nding. The authors concluded that the risk of TB
transmission on aircraft was no greater than that in other conned settings,
noting that TB outbreaks often occur as a result of overcrowded conditions
38
J. Murawski
in poorly-ventilated facilities when there is prolonged close exposure to an

infectious person.
An investigation into a pilot with active TB who had own with 48 other
pilots over a 6-month period found no risk of transmission [39]. It is possible
that this reduced risk is explained by the approximate 20-fold increase in the
supply rate of outside air in the cockpit, compared to the economy section of
the cabin.
The potential for transmitting measles and meningococcal disease on
aircraft has also been described. From February 1999 through May 2001,
the CDC received reports of 21 suspected cases of air-travel associated
meningococcal disease from local health departments, an average of one
every 6 weeks [40]. In April 2004, the CDC issued a report describing nine
young children with serologically conrmed or suspected cases of measles
that had traveled by air, three of whom would have been infectious on the
aircraft [41]. State and local health departments were concerned enough to
attempt to identify and evaluate potential contacts, and provide prophylaxis
when indicated.
4
Extreme Temperatures (See also Chap. 3)
Extreme temperatures are primarily a concern during ground operations if
there is insufcient capacity for the cooling or heating necessary to effectively manage extreme ambient temperatures. The temperature and humidity
of the air supplied to the cabin and cockpit will vary according to destination,
season, and air supply equipment. The fuselage while sitting on hot tarmac,
especially in the sun, can absorb heat and open doors introduce unconditioned ambient air into sections of the cabin and cockpit. Temperatures as
high as 110 F (43 C) have been documented in the cabin during ground
operations in the northeastern US during the summer [42].
During ground operations, occupants metabolism is generally elevated
as they move through the cabin and stow or retrieve their luggage. The
metabolic rate of cabin crew is elevated relative to passengers and cockpit
crew throughout much of a given ight. Relative humidity also inuences
thermal comfort, and will typically change considerably during a ight as
a function of ight phase and ambient environmental conditions.
In-ight, cabin crewmembers report: (1) cold drafts at ankle level when
working in galleys and sitting in jumpseats adjacent to one or more doors
with sweeping door seals; (2) exposure to stagnant, warm air in the upper
section of galleys, especially if ovens are operating; and (3) highly variable
temperatures between zones on some aircraft types.
39
4.1
Reported Symptoms
Although there are no published studies dedicated exclusively to temperature on commercial ights, either during ground operations or in-ight, a few
surveys have measured temperature and surveyed crewmembers, passengers,
or both about perceived comfort. One such study noted that complaints of
draftiness, and of temperatures that were too high or too variable, were more
common among surveyed crewmembers than ofce workers, and that female
crew were more likely to describe the temperature as too low [29]. Another
study found that 23% of surveyed crew rated temperature as cool or too
cold, and that too-cool temperature was one of the most common air quality
complaints among passengers [30].
Temperature-related incident reports have been submitted to the International Transport Workers Federation (ITF), an international labor organization that operates a safety and health incident reporting system for crewmember unions [42]. Reported symptoms include heat exhaustion (i.e., clammy
skin, dizziness, extreme fatigue, headache, nausea) during excessively hot
ground operations, and aching legs and feet from working in cold galleys.
5
Exhaust Fumes, Deicing Fluid, Fuel Fumes, and Cleaning Products
The poor quality and insufcient quantity of air supplied to the cabin while
the aircraft is occupied and sitting at or near the gate generate complaints
from passengers and crewmembers. The source of the cabin air supply during ground operations will vary according to airport equipment and aircraft
type. The most common is a conditioned air intake attached to the base of the
passenger boarding bridge. Other options include the auxiliary power unit
(APU) located in the aircraft tail, the interior of the airport terminal, the airport terminal supply air ducts, and ground carts. With the possible exception
of airport terminal air, these sources supply the cabin with ambient air that
can be polluted by the following sources:
Exhaust fumes: Fumes from diesel-powered ground service vehicles and
other aircraft can be ingested into the supply air, especially if the air intake is located near to the vehicles. Engine exhaust can contain ozone and
nitrous oxides, both respiratory irritants.
Deicing uid: Deicing uid that contains propylene glycol, diethylene
glycol, or methylene glycol can be ingested into the aircraft engines or
auxiliary power unit when the aircraft is being deiced, contaminating the
air supply systems [43] and creating a safety hazard on certain aircraft
types [44].
40
J. Murawski
Fuel vapor: Fueling operations at the airport, airplane tank venting during
lling, and failed engine relights are sources of fuel vapor that may enter
the cabin and cockpit air supply systems. Although there have been no formal exposure studies in the aircraft cabin and cockpit, an animal study
that reported adverse effects on learning abilities suggests that regular,
high-level exposure to jet fuel fumes may carry implications for groundbased fueling workers [45].
In addition, noxious disinfectants and deodorizers can be applied in the
cabin, primarily by airline cleaning staff. The labels for these products indicate that they are intended for use in a well-ventilated environment, not the
small, enclosed lavatories in the aircraft cabin. The chemical components of
cleaning agents are described elsewhere [7].
5.1
Reported Symptoms
Formal investigations into symptoms attributed by crew and passengers to
chemical exposures associated with ground operations have not been conducted. Symptoms reported to the ITF include reports of dizziness, eye and
throat irritation, headaches, and nausea that crewmembers attribute to exposure to cleaning products, deodorant sprays, lubricants applied to beverage
carts when stationed in the galley, and exhaust fumes [42].
6
Heated Engine Oil and Hydraulic Fluid (See also Chap. 10, and 12)
The potential for heated engine oil and hydraulic uid to contaminate the
aircraft air supply due to maintenance, operation, and design failures or
deciencies has been documented in the aviation industry for more than
35 years [46, 47]. Passengers and crew can be exposed to toxic components of these products, such as tricresylphosphates (TCPs), and their heated
byproducts, such as carbon monoxide (CO). Despite this history, the frequency and the severity of these contamination incidents are under debate,
particularly when not all crew and passengers are affected during a given
incident.
Carbon monoxide generation will depend on the temperature at the source
of contamination, whether the APU or aircraft engines. Ground-based exposure limits for CO are not applicable in-ight because the reduced oxygen
environment will exacerbate the toxic effects. For example, breathing air with
50 ppm CO at a cabin altitude of 6000 feet has been dened as physiologically
equivalent to a cabin altitude of 12 000 feet [48]. Similarly, breathing air with
150 ppm CO at a cabin altitude of 8000 feet effectively raises the cabin altitude
41
to 19 000 feet [49]. Alcohol consumption, elevated physical activity, cardiopulmonary disease, and cigarette smoking will magnify this effect. Smoking
1.5 packs per day can raise a persons carboxyhemoglobin to 10% [50]. This
altitude effect has not been tested for other airborne contaminants or for mixtures of contaminants.
Pilot incapacitation is one long-recognized hazard of air supply contamination. In response to a recommendation from the US National Transportation Safety Board, the US Civil Aeromedical Institute (CAMI) exposed rats to
heated aircraft engine oils and measured their response time, following several unexplained crashes of turboprop aircraft that might possibly be due to
pilot incapacitation from toxic fumes [51]. The CAMI rat study claimed to
rule out CO as a causative agent, but did not eliminate the possible presence of an additional (chemical) component with signicant animal toxicity.
More recently, the UK aviation authority reported incidents of pilot impairment caused by cockpit air supply contamination, emphasizing the need for
pilots to be regularly trained in incapacitation procedures [52].
The TCP content of these oils and hydraulic uids is typically reported
as 3% by volume and is likely a complex mixture of up to ten isomers,
six of which contain mono- or di-ortho isomers that are ve to ten times
more toxic than even the tri-ortho isomer of TCP [53], even though the
tri-ortho isomer is regularly mistaken as the most toxic isomer [54]. One
engine oil manufacturer assessed and ruled out the physical manifestations
of organophosphate-induced delayed neuropathy among test animals made
to ingest these oils [55], but this research does not mimic the exposure pattern of crews and passengers on commercial aircraft, and is therefore of little
relevance. Investigations into the causes of Gulf War Syndrome have found
evidence of altered brain chemistry and immunosuppression in test animals
following the inhalation of sub-clinical concentrations of an organophosphate
nerve gas [56, 57]. There is now interest in testing whether these ndings apply to other organophosphates, such as TCPs.
6.1
Reported Symptoms
A recent health survey of pilots operating B757, B737, and A320 aircraft at
a major airline in the UK identied symptoms associated with 1674 reported
incidents of cockpit air supply contamination, all but seven of them on the
B757 [58]. Although the response rate was low (106 of 600 pilots), the reported symptoms are cause for concern, given the implications for ight
safety. The most commonly reported symptoms were: eye, nose, and throat
irritation (63%); headaches, lightheadedness, and dizziness (55%); fatigue,
weakness, decreased performance (56%); concentration difculties, confusion (32%); nausea, vomiting, gastrointestinal problems (22%); skin irritation
(21%); short-term memory impairment (19%); numbness (17%); joint pain,
42
J. Murawski
muscle weakness (16%); intolerance to chemicals or odors (16%); and disorientation (12%).
The ITF has been informed of smoke in the cabin incidents by ight attendants safety representatives based in Australia, Canada, Denmark, France,
Sweden, the US, and the UK [59]. The reported symptoms are sometimes
consistent with exposure to carbon monoxide and neurotoxic agents, and include blurred or tunnel vision, confusion, disorientation, dizziness, giddiness,
headache, lack of motor control, loss of cognitive function, nausea, tremors,
unusual fatigue, and vomiting.
In 2000, a bipartisan senate committee in Australia released a report summarizing its two-year inquiry into reports of air supply contamination on the
BAe146 aircraft [9]. The report concluded that exposure of air crew and,
potentially, passengers to cabin air which may be...even minutely affected,
by fumes originating in an aircrafts engines raises the potential of occupational illness and, for certain individuals, an incapacity to continue work.
Although the committee had been charged with investigating complaints on
the BAe146, their report identied similar problems on other aircraft, including the A320 and MD90, and recommended that the Australian Civil Aviation
Safety Administration introduce regulations that dictate specic preventive
maintenance procedures, a national standard for checking and monitoring
the engine seals and air quality in all passenger commercial jet aircraft, and
a national incident reporting system.
In November 1999, on the rst two of three ights on a BAe146 aircraft,
cabin crew reported symptoms that included fainting, odd pressure in the
head, and the feeling of moonwalk. On the third ight, both the pilot
and copilot were affected and had to don their oxygen masks. Prior to the
ights, the airline had found a minor external oil leak in one of the engines.
A government investigation found no technical fault that could explain the
incident, but attributed it to exposure to probably polluted air [60].
In an effort to quantify the extent of air supply contamination at one US
airline, the cabin crew union initiated a review of all sources of reported
information related to these incidents [61]. The union identied 760 incidents involving 900 crewmembers and passengers over a nine-year period
(19891998) that involved either a visible aerosol in the cabin and symptoms
reported by crew or passengers, or mechanical records that indicated aircraft
air supply contamination, or both. This translated into approximately seven
incidents per month. In many cases, the documented symptoms reported by
crew and passengers were consistent with exposure to TCP additives in the
jet engine oils and hydraulic uids (e.g., abnormal gait, balance problems,
chills, delayed peripheral neuropathy, muscle aches, seizures, stomach cramping, and tremors) and/or asphyxiants (e.g., dizziness, metallic taste, severe
headaches, and tunnel vision).
The cabin crew union at the same airline also asked NIOSH to investigate the reports of neurological illness associated with smoke in the cabin
43
incidents [62]. Once NIOSH staff had arranged an acceptable sampling schedule with the airline, they monitored CO, carbon dioxide, relative humidity,
temperature, and volatile organic compounds on three test ights. The CO
data was unusable because the equipment had been improperly calibrated,
peak ozone concentrations ranged from 0.0580.093 ppm, and maximum
cabin altitudes were 80648218 feet (24592505 m), relative to ambient pressure on the ground. Cabin crew reported that their performance was not
affected during those three ights, although on one of the ights, three of
the four cabin crew reported a headache, and one reported lightheadedness
and dizziness. NIOSH measured CO on another 13 ights with direct-reading,
continuous electrochemical sensors. Peak CO concentrations were reported
as one minute averages that ranged from < 125 ppm, but there was no indication as to whether or not cabin crew had reported symptoms during those
ights. Also, NIOSH did not identify whether the readings were explained by
a CO source or chemical interference. Cabin crewmembers continue to report symptoms that are consistent with exposure to neurotoxic or asphyxiant
chemicals on an infrequent but persistent basis.
Similar health complaints from crewmembers, some of which required admission to the emergency department, prompted an investigation into the
BAe146 aircraft operated by a Canadian airline [63]. The most common reported symptoms were burning eyes and throat, disorientation, headache,
and nausea. Oil contamination was identied in the air supply system on
a test aircraft and TCPs were identied in the bulk oil sample. However, due
to sampling limitations, TCPs could not be demonstrated in the cabin air.
There is evidence of a possible connection between these exposure incidents and neurological decits that resemble multiple sclerosis (MS) and
Parkinsons Disease, as described by cabin crew in the US, as well as pilots in
Australia and the UK. Occupation-specic clusters of symptoms that resemble MS have been documented in another industry, with specic references
to exposure to hydraulic and machining uids that contain TCPs, just like
commercially-available aircraft engine oils and many hydraulic uids [64].
Signicant excess in mortality and morbidity from motor neuron disease has
been reported among pilots [65, 66], although such associations have never
been formally investigated among cabin crew.
Some explanations for differences in individual susceptibilities to the effects of exposure to particular organophosphates include evidence that genetics, levels of particular hormones, health status, and exposures to mixtures of
particular chemicals can inuence the efcacy of enzymes involved in their
metabolism, and could thereby inuence the degree of toxic effect [67, 69].
44
J. Murawski
7
Reduced Oxygen Supply (See also Chap. 3, and 4)
The aircraft cabin is pressurized because the oxygen content in unpressurized
air during ight is not adequate to sustain life. The introduction of compressed air into the aircraft cabin ensures that the internal cabin pressure
(and the corresponding partial pressure of oxygen) is higher than the outside
air pressure at the ight altitude. The cabin pressure is usually referred to in
terms of its corresponding altitude (cabin altitude). Aviation standards require that aircraft systems be designed to maintain a cabin altitude not higher
than 8000 feet (2440 m) at the maximum ight altitude during normal conditions (Table 1). This corresponds to a supply of approximately 75% of the
oxygen available at sea level.
No regulatory authority has issued an explicit operating standard for cabin
altitude, except that when the cabin altitude reaches 10 000 feet (3050 m), essentially an emergency condition, the pilots must don oxygen masks, and at
14 000 feet (4270 m), oxygen masks are automatically provided to the cabin
occupants. A recent sampling study reported that the maximum in-ight
cabin altitudes on 36 ights ranged from 30007500 feet (9152290 m) [19],
although cabin altitudes in the 60008000 feet (18302440 m) range are probably more typical, and readings as high as 8915 feet (2717 m) have been
reported [70].
There is no apparent health-based rationale for the 8000 feet design standard, probably because the FAA was not required to provide substantiating
material when it recodied the US Civil Aeronautical board regulations in
1964. Regulators must now thoroughly justify any new standards but the
pressurization standard has not been revisited, and an operating standard
has never been proposed. Based on the results of pressurization studies conducted in the 1940s, the 8000 feet design standard has been described as
a compromise between the aircraft design and operation requirements and
the human performance impairments, and when the standard was rst published, 50006000 feet (15201830 m) was recommended for routine operating conditions [7]. Occupants oxygen needs vary according to activity level,
health status, smoking status, and age.
7.1
Reported Symptoms
Symptoms associated with insufcient blood oxygen saturation include: decient color discrimination, dizziness, fainting, fatigue, headaches, nausea,
slowed reaction time, and weakness for novel tasks. There is little published
data on symptoms reported by crew or passenger and blood oxygen saturation. One study on a group of 42 airline pilots on 22 ights measured an
average blood oxygen saturation on the ground of 97% (9599%) compared
45
to the reduced and more variable saturation of 89% (8091%) at altitude,

although symptoms were not reported [71].
The reduced partial pressure of oxygen and the changes in pressure during a ight have been described as the deciding factors for tness to travel
due to the proportion of in-ight deaths attributable to coronary heart disease [72]. A test on the effects of simulated air travel on 17 patients with
chronic obstructive pulmonary disease reported that, for most patients, blood
oxygen levels dropped below the recommended levels and their hypoxia was
aggravated by mild exercise [73]. A review of medical diversions at one major
US airline concluded that neurologic symptoms, including dizziness/vertigo,
seizures, headaches, pain, and cerebrovascular complaints, were the largest
category of in-ight medical incidents prompting air-to-ground calls [74]. Reports of loss of consciousness/syncope were the most likely to result in an
actual diversion of an aircraft. Reduced cabin pressure has been suggested as
a risk factor for triggering or exacerbating lymphedema [75], although the
question of whether reduced cabin pressure contributes to the risk of deep
vein thrombosis appears unresolved [76, 77].
8
Ozone Gas (See also Chap. 3)
At ground level, ozone is unnatural a component of smog and a public
health menace. Exhaust fumes are a source of ozone that can be entrained
into the aircraft supply during ground operations, as stated earlier. Generally
though, ozone exposure is problematic during ight.
At altitude, ozone occurs naturally and is generally classied as protective of public health because it lters some of the ultraviolet light that can
otherwise burn skin and initiate cancer. Commercial aircraft prove the exception to this rule because they operate within the ozone layer, such that the
naturally occurring ozone gas not intended for human consumption can be
captured and concentrated in the air supply systems. Ozone levels start to increase in the troposphere (approximately 26 000 feet altitude at the poles and
50 000 feet at the equator) and generally continue to increase with altitude
up to 90 000 feet. The troposphere drops to lower altitudes in the late winter
and early spring, such that ozone concentrations increase at ight altitudes,
accordingly.
Some ozone will be removed from the air supply when it reacts with the
inside surface of the air supply ducts; some ozone will be converted into oxygen if a catalytic converter is installed and operating; some ozone is delivered
to the cabin and cockpit.
Ozone exposure limits are many and varied. The average 8-h workplace
limits include an enforceable limit of 0.1 ppm [78] and a recommended
limit of 0.050.1 ppm, depending on the level of exertion [79]. For the
46
J. Murawski
public, a maximum 8-h average of 0.056 ppm has been recommended to

protect public health, and a maximum 1-h average of 0.092 ppm recommended for population information [80]. Ambient ozone levels deemed
acceptable on aircraft are considerably higher (Table 1, Table 2), and vary
with operating altitude, suggesting that they are not health-based. Airlines need not monitor the air to demonstrate compliance with ozone limits; they need only show by analysis that ambient ozone levels are unlikely to exceed the said limits. The accompanying degree of statistical
uncertainty that must be demonstrated for these analyses is substantially
higher (p < 0.16) than what is generally considered sound scientic practice
(p < 0.05) [81].
Recent sampling conducted on 36 ights, half of them polar (i.e., some portion of the ight operating above 50 latitude), and all on aircraft equipped
with catalytic converters, reported gate-to-gate average levels of ozone ranging from < 0.05 to 0.24 ppm on ights with maximum cabin altitudes on the
low end of operation (30007500 feet or 9152290 m) [19].
Documented ozone concentrations during ight are variable, largely due
to ight path, altitude, and season, but the location of monitoring equipment
can also inuence results because ozone gas is so reactive. One survey found
that 40% of the ozone present at ceiling height in the economy class section
had disappeared when measured at a height of 4 feet above the oor [82],
presumably because it had already contacted and reacted with surfaces in the
cabin, including the occupants.
8.1
Reported Symptoms
The adverse effects of ozone on the respiratory system even at very low concentrations are well documented. One comprehensive literature review [83]
reported that a single ozone exposure to healthy, non-smoking young adults
in the range of 0.080.12 ppm produces a complex array of pulmonary responses. The observed association between long-term ozone exposure at
0.25 ppm and progressive and persistent lung function and structural abnormalities in test animals raises serious concerns about the effects of chronic
exposure on people. Cited animal studies also support the hypothesis that
chronic ozone exposure accelerates the aging of the human lung.
Respiratory symptoms associated with exposure include: aggravated
asthma; chest tightness; cough; inammation of the lung tissue; painful, labored, or rapid and shallow breathing; pulmonary edema; and temporary
decrease in lung capacity. There is also evidence that ozone gas can induce
immune system changes [84] and increase susceptibility to infection. Children, asthmatics, and people with existing respiratory disease are most at
risk. Both physical exertion and heat stress have been shown to exacerbate
the effects of exposure to ozone. The reduced supply of oxygen at altitude
47
may magnify the effects of exertion because of the attendant increase in the
breathing rate.
A series of articles on ozone-related symptoms reported by crewmembers
was published in the early 1980s in response to hundreds of complaints received by commercial airlines after the B-747-SP high altitude aircraft was
introduced into passenger service in 1976 [81]. Sampling data collected by
the US National Aerospace Administration in 1977 indicated signicantly elevated ozone levels on the B-747-SP aircraft compared to the standard B-747,
and the cabin ozone levels increased as the ights progressed. In one survey,
self-reported symptoms consistent with ozone exposure were reported three
to four times more often by cabin crew employed by airlines operating high
altitude, long-distance ights, than by those employed by airlines operating
low altitude, short duration ights [85]. Symptoms included burning sensations in the throat and eyes, chest pain, coughing, shortness of breath, and
wheezing. The study was limited by a relatively low response rate (61% of
active crewmembers) and the airlines refusal to allow in-cabin monitoring.
Another survey reported a signicant association with self-reported ozonerelated symptoms and 747-SP ights, although the survey response rate was
again low [86].
9
Insecticides (See also Chap. 9)
Insecticides are applied on aircraft for three key reasons: routine control of
domestic insects, particularly in the aircraft galleys; response to on-board
insect sightings; and compliance with foreign quarantine regulations. Sixty
countries publish and enforce foreign quarantine regulations that require
insecticide spraying on incoming aircraft to kill any insects that may be
on board and may carry disease or damage the environment [87]. Insecticides (typically pyrethroids) are applied in occupied or soon-to-be-occupied
aircraft cabin, and neither passengers nor crew are warned in advance. Historically, DDT and Sevin have been applied in the aircraft cabin [88]. Current
efforts focus on a possible mechanical means of disinsection as an alternative
to the current chemical spraying [89].
9.1
Reported Symptoms
There are few epidemiological studies into either the acute or chronic effects
of exposure to insecticides applied on aircraft specically. The most common symptoms recently reported by crewmembers exposed to pyrethroids
were cardiovascular, dermatological, gastrointestinal, neurological, ocular,
and respiratory [90]. In addition to the potential for acute illness, the po-
48
J. Murawski
tential for cumulative or chronic health problems associated with pyrethroid

exposure has been acknowledged both on [90] and off [91] aircraft. A moderate association between self-reported application of DDT on aircraft and
elevated risk of breast cancer among female ight attendants has also been
reported [92]. See also Sect. 4.
Acknowledgements The author thanks Dr. Martin Hocking at the University of Victoria,
Canada for his invaluable editorial work.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
US Federal Register (1975) 40:29114

British Aerospace (2002) Inspection Service Bulletin 21150
US Federal Aviation Administration (2000) Airworthiness directive 2000-15-17
US National Transportation Safety Board regulation (1995) 49 CFR 830.5
Murawski J (2002) Association of Flight Attendants AFL-CIO presentation to US General Accounting Ofce, Washington, DC
US Bureau of Labor Statistics (1999) Workplace injuries and illnesses in 1999. USDL
00-357, news release. Washington, DC
US National Research Council Committee on Air Quality in Passenger Cabins of Commercial Aircraft (2002) The airliner cabin environment and the health of passengers
and crew. National Academy Press, Washington, DC
UK Department of Transport, Local Government, and the Regions (2001) Technical
report on health in aircraft cabins prepared by the Building Research Establishment,
London, England
Parliament of the Commonwealth of Australia Senate Rural and Regional Affairs and
Transport Legislation Committee (2000) Technical report on air safety and cabin air
quality in the BAe146 aircraft. Senate Printing Unit, Canberra, Australia
UK House of Lords Select Committee on Science and Technology (2000) Air travel
and health HL paper 121-I and HL paper 121-II. The Stationery Ofce, London,
England
Nagda NL, Fortmann MD, Koontz MD et al. (1989) Technical report prepared for the
US Department of Transportation DOT-P-15-89-5 NTIS/PB91-159384
US National Research Committee on Airliner Cabin Air Quality (1986) The airliner
cabin environment: air quality and safety. National Academy Press, Washington, DC
US Congress (1987) Public Law 100-202 (22 December 1987)
US Congress (1989) Public Law 101-164 (21 November 1989)
Lautenberg F (1996) Testimony of US Senator FR Lautenberg on HR 969/S1524, the
Airliner Cabin Air Quality Act of 1995
US Congress (2000) Public Law 106-181 (5 April 2000)
Corrao MA, Guindon GE, Sharma M, Shokoohi DF (eds) (2000) Tobacco control country proles, 11th world conference on tobacco or health. American Cancer Society,
Atlanta, Georgia
Fiore M (1994) Effect of smoking withdrawal on ight performance: a 1994 update.
US Centers for Disease Control and Prevention, Atlanta, Georgia
Waters MA, Bloom TF, Grajewski B et al. (2002) Measurements of indoor air quality
on commercial transport aircraft. In: Levin H (ed) Indoor air 2002: proceedings of the
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
49
9th international conference on indoor air quality and climate, Santa Cruz, California,
pp 782787
Pierce WM, Janczewski JN, Roethlisberger B et al. (1999) ASHRAE J 9:2634
Spengler JD, Burge H, Dumyahn T et al. (1997) Environmental survey prepared for
ABC News by Harvard School of Public Health, Harvard University, Cambridge, Massachusetts
Jurgiel J (1994) Cabin air quality investigation. Consulting report by JA Jurgiel & Associates, St. Louis, Missouri
Nagda NL, Koontz MD, Konheim AK, Hammond SK (1992) Atmosph Environ
26A:22032210
Seppanen OA, Fisk WJ, Mendell MJ (1999) Indoor Air 9:226252
Hocking MB (2002) Rev Environ Health 17:149
Hocking MB (2000) Chemosphere 41:603615
Whelan EA, Lawson CC, Grajewski B et al. (2003) Occup Environ Med 62:929934
Lindgren T, Andersson K, Dammstrom BG et al. (2002) Int Arch Occup Environ
Health 75:47583
Lindgren T, Norback D, Andersson K et al. (2000) Aviat Space Environ Med 71:774
782
Lee SC, Poon CS, Li XD et al. (2000) Questionnaire survey to evaluate the health and
comfort of cabin crew. In: Nagda NL (ed) Air quality and comfort in airliner cabins
ASTM STP 1393. American Society for Testing and Materials, West Conshohocken,
Pennsylvania, pp 259268
Nutik-Zitter J, Mazonson PD, Miller DP et al. (2002) JAMA 28:483486
Hocking MB, Foster HD (2002) JAMA 288:2972
World Health Organization (2003) Weekly epidemiological record 78:97120
World Health Organization (2003) WHO update 49 (7 May 2003)
World Health Organization Communicable Disease Surveillance and Response (2003)
Technical report on stability and resistance of SARS coronavirus (4 May 2003)
World Health Organization (2003) WHO update 62 (22 May 2003)
Kenyon TA, Valway SE, Ihle WW et al. (1996) NEJM 334:933938
Wang PD (2000) Am J Infect Control 28:233238
Parmet AJ (1999) Aviat Space Environ Med 70:817818
US Centers for Disease Control and Prevention (2001) Morbidity and Mortality
Weekly Report 50:4859 (15 June 2001)
US Centers for Disease Control and Prevention (2004) Morbidity and Mortality
Weekly Report 53:12
International Transport Workers Federation Civil Aviation Section (2004) Aviation
Safety & Health Database International, London, England
Society of Automotive Engineers (1997) Aerospace Information Report 1539 Rev A
US National Transportation Safety Board (2002) Safety recommendation A-02-05
Ritchie GD, Rossi J, Nordholm AF et al. (2001) J Toxicol Environ Health 64:385415
Robbins CS (1969) Technical report for United Airlines prepared by the Boeing Company
van Netten C (2000) Analysis of two jet engine lubricating oils and a hydraulic uid:
their pyrolytic breakdown products and their implication on aircraft air quality. In:
Nagda NL (ed) Air quality and comfort in airliner cabins ASTM STP 1393. American
Society for Testing and Materials, West Conshohocken, Pennsylvania
US Air Force (1992) Guide specication, environmental control airborne MIL-E-87145
McFarland RA (1971) Aerospace Med 12:13031318
50
J. Murawski
50. US Army Headquarters (2000) Field manual 3-04.301, aeromedical training for ight
personnel
51. Crane CR, Sanders DC, Endecott BR, Abbott JK (1983) Aviation Medicine Report FAAAM-83-12, US Federal Aviation Administration
52. UK Civil Aviation Administration Safety Regulation Group (2002) Flight Operations
Department Communications (FODCOM) 17/2000, 14/2001, and 21/2002. Gatwick,
England
53. Henschler D (1958) Klinische Wochenschrift 36:663674
54. Spengler JD, Wilson DG (2003) Proc Instn Mech Engrs 217:323335
55. Mackerer CR, Barth ML, Krueger AJ et al. (1999) J Tox Environ Health 56A:293328
56. Henderson RF, Barr EB, Blackwell WB et al. (2002) Toxicol Appl Pharmacol 184:6776
57. Kalra R, Singh SP, Razani-Boroujerdi S et al. (2002) Toxicol Appl Pharmacol 184:8287
58. Michaelis S (2003) J Occup Health Safety Austr NZ 19:253261
59. ITF (2003) Minutes of biannual ITF international task group on aircraft air quality
(19992003), London, England
60. Statens Haverikommission Board of Accident Investigation (2001) Report RL
2001:41e. Accident investigation into incident onboard aircraft SE-DRE during ight
between Stockholm and Malm M county, Sweden
61. Witkowski CJ (1999) Remarks on airliner air quality. Presentation at ASHRAE conference, Chicago, Illinois
62. US National Institute for Occupational Safety and Health (1993) Health hazard evaluation report HETA 90-226-2281
63. van Netten C (1998) Appl Occup Environ Hyg 13:733739
64. Krebs JM, Park RM, Boal WL (1995) Arch Environ Health 50:1905
65. Nicholas JS, Butler GC, Lackland DT et al. (2001) Aviat Space Environ Med: 72:8216
66. Nicholas JS, Lackland DT, Dosemeci M et al. (1998) J Occup Environ Med 40:9805
67. Haley RW, Billecke S, LaDu BN (1999) Toxicol Appl Pharmacol 157:22733
68. Howard JK, East NJ, Chaney JL (1978) Arch Environ Health 277279
69. Davis ME, Yu EA, Fugo NW (1948) JCE 666673
70. Cottrell JJ (1988) Chest 92:8184
71. Cottrell JJ, Lebovitz BL, Fennell RG, Kohn GM (1995) Aviat Space Environ Med
66:126130
72. Shand D (2000) Occup Med (Lond) 50:56671
73. Christensen CC, Ryg M, Refvem OK et al. (2000) Eur Respir J 15:6359
74. Sirven JI, Claypool DW, Sahs KL et al. (2002) Neurology 58:173944
75. Casley-Smith JR, Casley-Smith JR (1996) Aviat Space Environ Med 67:5256
76. Crosby A, Talbot NP, Harrison P et al. (2003) Lancet 361:22078
77. Schobersberger W, Hauer B, Sumann G et al. (2002) Wien Klin Wochenschr 114:1420
78. US Occupational Safety and Health Administration (1971) Permissible exposure limit
79. American Conference of Governmental Industrial Hygienists (1998) Threshold limit
values
80. European Economic Community (1992) Council Directive 92/72/EEC Ofcial Journal
L 297:17
81. US Federal Aviation Administration (1980) Advisory circular 12038
82. van Heudsen S, Mans LGJ (1978) Aviat Space Environ Med 49:10561061
83. Lippmann M (1993) J Exposure Analysis Environ Epi 3:103129
84. Foster WM, Wills-Karp M, Tankersley CG et al. (1996) J Appl Physiol 81:794800
85. Reed D, Glaser S, Kaldor J (1980) Am J Ind Med 1:4354
86. Tashkin DP, Coulson AH, Simmons MS et al. (1983) Int Arch Occup Environ Health
52:117137
51
87. World Health Organization (2001) Working paper 12 presented at the 3rd meeting of
the Facilitation Panel (FALP/3-WP/12), 1216 February 2001, Montreal
88. Aviation Consumer Action Project et al. v. United States Department of Agriculture
(1977) Complaint in civil court, civil action no. 77-1941, 10 November 1977, US District Court for the District of Columbia
89. International Civil Aviation Organization (2004) FAL/12-WP/117 Facilitation Division 12th Session, 22 March2 April 2004, Cairo, Egypt
90. California Department of Health Services Occupational Health Branch (2003) Occupational illness among ight attendants due to aircraft disinsection, Oakland, CA
91. Muller-Mohnssen M (1999) Toxicol Letters 107:161175
92. Wartenberg D, Stapleton CP (1997) Abstract presented at the 9th annual conference of
the International Society of Environmental Epidemiology, 1720 August 1997, Taipei,
Taiwan

DOI 10.1007/b107236
Aircraft Cabin Indoor Air Environment Requirements

Nadia Boschi1 (u) Fariborz Haghighat2
1 Department
of Building Construction, Virginia Polytechnic Institute and State

University, USA
nboschi@vt.edu
2 Department of Building, Civil and Environmental Engineering, Concordia University,
Montreal, Canada
haghi@cbs-engr.bcee.concordia.ca
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
54
A New Health Metric . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
55
Occupational Safety and Health Responsibilities . . . . . . . . . . . . . . .
56
4
4.1
Global Thermal Comfort . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Thermal Comfort Requirements for Elderly Persons . . . . . . . . . . . . .
60
63
Health Considerations Relative to Environmental Parameters . . . . . . .
66
6
6.1
6.2
6.3
6.4
Indoor Environment Parameters .

Pressure (and Altitude Equivalent)
Ventilation Requirements . . . . .
Relative Humidity . . . . . . . . .
Ozone . . . . . . . . . . . . . . . .
.
.
.
.
.
70
71
72
76
77
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
80
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
80
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Abstract The purpose of heating or cooling systems is to provide an acceptable microclimate and maintain suitable conditions for the intended use of the space. Airliner cabins,
however, present different design and operation challenges because of the extreme external environmental conditions, the complexity of the operational systems and the diverse
authorities that govern such environments.
The environmental quality of a space is determined by the occupants response to various environmental stimuli and his integration of these inputs into a comfort and health
response. Most thermal comfort studies have dealt with a homogeneous population with
similar physical characteristics (neither ill nor old). The concept of comfort to meet the
requirements of the elderly and the new health metric called the disability-adjusted life
year (DALY) are introduced.
This contribution reviews environmental requirements such as ventilation, relative
humidity, carbon dioxide concentrations, ozone and pressure as expressed in relevant
guidelines and standards that are applicable to air quality in air cabins. Further, the
health effects associated with environmental exposures organized by level of concern are
discussed.
54
N. Boschi F. Haghighat
Based on existing data, it is concluded that at cruise altitude the pollutant of primary
concern is ozone (O3 ). Ozone standards are not regularly met. Carbon monoxide and
particulate matter concentrations appear to be lower than health-based standards for ambient air, while VOC and SVOC appear to be present in similar concentrations as in other
transportation vehicles.
Keywords Comfort Health Metric Environmental parameters
Abbreviations
ACGIH
American Conference of Governmental Industrial Hygienists
ASHRAE American Society of Heating, Refrigerating and Air Conditioning Engineers
cfm
Cubic feet per minute
CFR
Code of Federal Regulations
clo
Clothing value
CO2
Carbon dioxide
CO
Carbon monoxide
DALY
Disability-adjusted life year
DOT
US Department of Transportation
EPA
US Environmental Protection Agency
ETS
FAA
Federal Aviation Administration
FARs
FR
Federal Regulation
HEPA
High efciency particulate lter
JAAA
Joint Aviation Airworthiness Authority
NAS
National Academies of Science
NIOSH
National Institute of Occupational Safety and Health
NRC
National Research Council
Ozone
O3
OSHA
Occupational Safety and Health Administration
ppm
Parts per million
TLVs
Threshold limit values
WHO
World Health Organization
NAAQSs US National Ambient Air-Quality Standards
1
Introduction
People spend more than 90% of their time in an enclosed environment (e.g.,
dwelling, workplace, or transport vehicle). When energy costs soared during the energy crisis of the early 1970s, the outdoor air supply was sharply
reduced in many mechanically ventilated buildings and commercial aircraft
[1, 2]. Since then there has been growing attribution of adverse effects on
comfort, health, and productivity from this change [3]. The purpose of heating or cooling systems is to provide an acceptable microclimate and maintain
55
suitable conditions for the intended use of the space. The thermal environment must be considered in the design of a ventilation system, whether it is
for a building or an aircraft, as it is fundamental to the comfort and wellbeing of the human occupants.
The environmental quality of a space is determined by the occupants response to various environmental stimuli and the integration of these inputs
into a comfort response. If one assumes that sufcient heating or cooling
capacity is available to maintain the desired average temperature within
a space, then a comfortable thermal environment will be completely dependent upon the distribution of conditioned air in the space. From a thermal
standpoint, it is possible to have an average temperature, which may exist at
some point in the space, which satises overall criteria for thermal balance.
At the same time, there may well be conditions which cause the local temperatures throughout the space to vary from this average or mean value. The
objective of a good air distribution system is to produce within the occupied space the proper combination of temperature, air motion, and relative
humidity to keep the occupants comfortable. Designers and operators of ventilation systems should be familiar with the comfort and indoor air quality
requirements necessary to achieve an acceptable indoor climate.
2
A New Health Metric
Although the primary purpose of the built environment is to provide for the
needs of the occupants, history is replete with evidence of building-associated
illnesses. Studies in the USA and Europe indicate that occupants are exposed
to environmental conditions that result in high rates of complaints or illness
in 2030% of existing commercial buildings [4].
Since the time of Hippocrates (460377 BC) it has been known that polluted air is detrimental to health and that certain spaces are less healthy than
others. Since the time of Vitruvius (1st century BC) we have codied directions on urban planning, design, operation and maintenance of buildings,
materials and equipment in buildings, and appropriate energy saving measures to assure the good quality of the air we breathe, as well as our comfort
and well being [5]. The target is, without doubt, our health in the indoor environment. Health is the rst target we should establish when we design, build,
maintain, and operate our indoor air environment, including airliner cabins.
In 1948, the World Health Organization (WHO) dened health as: a state
of complete physical, mental and social well-being and not merely the absence
of disease or inrmity [6]. This denition provided the blueprint for dening future health measures. It identied two critical aspects: rstly, health has
more than one dimension; secondly, it is important to measure the full range
of health states.
56
In the last 50 years the context of health has been broadened and has
started to be inclusive of aspects that characterize the way people live. In
1948, the Universal Declaration of Human Rights established the links among
health, way of living, and cultural context by stating: Everyone has the right
to a standard of living adequate for the health and well-being of himself and
of his family, including food, clothing, housing and medical care and necessary social services ... [7].
The elimination of illnesses or accident is an ambitious plan. Instead, the
achievement of better living environments is more realistic. Sometimes, contingent situations take over. In the late twentieth century, for example, during
the years of the energy crisis, indoor air quality was not a priority. On the
contrary, outside air was treated as a luxury to be minimized. The energy crisis had a huge impact on ways of designing and building the structure and
overall building systems as well as on the health and lifestyle of people. It is
important to look back to the basic of our health and comfort targets so that
mistakes such as the Sick Building Syndrome can be prevented.
WHO estimates that 3040% of respiratory diseases are caused by air pollution. Mortality gures do not capture the huge toll of sickness and disability
caused by diseases that keep workers off the job, children out of school, and
generally slow both economic and social development. In addition, the few
gures that do exist tend to be biased because wealthier people seek medical care much more often than the poor. Over the years various investigators
have attempted to overcome these limitations by developing new metrics that
factor in disability or quality of life along with mortality. One of the most
recent measures, although still controversial, is the Disability-Adjusted Life
Year, (DALY) [8]. Such assessment combines various health outcomes using
explicit weighting schemes.
As shown in Table 1, using this metric, by comparing the ten leading causes
of death with the ten leading causes of DALYs, communicable diseases are the
single most important cause of ill health globally, accounting for 44% of the
total. This increase in the relative importance of infectious diseases reects in
large part the early age at which they strike. Globally, of the top ten causes
of DALYs, communicable diseases account for seven, with lower respiratory
infections and diarrhea diseases heading the list.
3
Occupational Safety and Health Responsibilities
In modern micro-built environments, control is achieved by designing and
operating the mechanical systems in compliance with prevailing standards
such as those developed by the American Society of Heating, Refrigerating and Air Conditioning Engineers (ASHRAE) [9]. Airliner cabins, however,
present different design and operation challenges because of the extreme
57
Table 1 Comparing causes of death worldwide with disability-adjusted life year

Top 10 causes of disabilityadjusted life years, 1990
Top 10 causes of death, 1990
Ischemic heart disease

Cerebrovascular disease
Lower respiratory infections
4.
5.
Lower respiratory infections

Diarrhea diseases
Conditions arising during
the perinatal period
Unipolar major depression
Ischemic heart disease
6.
Cerebrovascular disease
1.
2.
3.
7.
8.
9.
10.
Tuberculosis
Measles
Road trafc accidents
Congenital anomalies
Diarrhea diseases
Conditions arising
during the perinatal period
Chronic obstructive
pulmonary disease
Tuberculosis
Measles
Road trafc accidents
Trachea, bronchus, and lung cancers
Communicable diseases. (Source: Adapted from Murray JL and Lopez AD [8])
environmental conditions, complexity of the operational systems, and the authorities that govern such environments.
In 1995, Hunt et al. describe how the environmental control in jet powered
aircraft is achieved by integrating four subsystems: (1) a bleed air system,
which diverts high-pressure air from the pre-combustion compressor stage
of engines, (2) an air-conditioning pack, which regulates thermal treatment
of the air, (3) an air distribution system, which distributes and recirculates
the treated air throughout the cabin, and (4) a cabin pressure control system
continuously monitoring the pressure [10]. This system controls the opening
and closing of the cabin pressure outow valve in the fuselage for continuous air escapes from the airplane. According to Space et al. (2000) this airow
corresponds to a total air change rate of 2030 air changes per hour [11].
In 1958, the US congressional Federal Aviation Act (Public Law 85-726)
established the Federal Aviation Administration (FAA), which has the regulatory authority over the operation of civil aircraft. In 1970, all federal agencies
were granted the right to exercise jurisdiction over their own workers. In
1975, FAA dened their jurisdiction over the safety and health of cockpit
cabin crew (40 FR 29114, DOT 1975). Specically, FAA stated in 40 FR 29114:
Every factor affecting the safety and healthy working conditions of aircraft
crew members involves matters inseparably related to FAAs occupational
safety and health responsibilities under the Federal Aviation Act.
This authority applies to aircraft in operation, which is the time from
when the aircraft is rst boarded by a crew member to when the last crew
member leaves the aircraft, including the time when the engines are shut
down during stops but at least one crew member is on board (40 FR 29114,
58
July, 1975). Further, the FAA, as dened in 49 USC 40101D and 49 USC
44701A, is authorized to protect the health and safety of passengers. Because
of this regulatory authority over health and safety, FAA has promulgated
specications for air quality in commercial aircraft in Federal Aviation Regulations (FARs). Similarly to this US regulatory situation, the European Joint
Aviation Airworthiness Authority (JAA) applies the Joint Aviation Regulations to European aircraft.
Safety has always been the primary factor of concern. Air quality in airplanes is a growing concern. In 1986, the National Research Council (NRC) of
the National Academies published The Airline Cabin Environment: Air Quality and Safety. The scope of the task force working on this report was to dene
the health impact of the air cabin characteristics on crew and passengers
health problems. The task force concluded empirical evidence is lacking in
quality and quantity for a scientic evaluation of the quality of airliner cabin
air or of the probability of health effects of short or long exposure to it [12].
But the task force did develop several conclusions and recommendations in
regard to air quality issues including environmental tobacco smoke (ETS),
carbon dioxide (CO2 ), ozone (O3 ), and ventilation.
FAA adopted several of the recommendations listed in the NRC 1986 report [12]. In 1988, the US Congress passed Public Law 100-202 banning
smoking on ights lasting less than 2 h. In 1989, in Law 101-164, the duration of the ights was raised to 6 h. In 1996, FAA lowered the permitted
level of CO2 from 30 000 to 5000 ppm (61 FR 63952). Because of a continued
concern about air quality in aircraft, Congress directed FAA in the Wendell
H. Ford Aviation Investment and Reform Act of the 21st century, enacted in
2000, to request NRC to perform another independent study to examine cabin
air. This report, which focused on aircraft environmental control systems,
sources, toxicity and health effects associated with possible air contaminants,
was published in 2001 with the title The Airliner Cabin Environment and the
Health of Passengers and Crew [13].
Currently, FARs address O3 , carbon monoxide (CO), CO2 , ventilation and
cabin pressure. These air quality regulations are dened in 14 CFR 25 and are
intended as design specications for aircraft that are subject to certication
under 14 CFR 21 (i.e., it certies that the manufacturer and air carrier are in
compliance with the regulations during design, construction and operation
of the aircraft). 14 CFR 21 is the operational standard and applies to domestic, foreign, and supplemental air carriers. This standard species appropriate
O3 concentrations in the cabin at particular altitudes, and ventilation requirements. Specically, Section 121.219 states that each occupant must be
suitably ventilated, CO concentrations may not be more than 50 ppm and
fuel fumes must not be present.
The design ventilation standard, as specied in 14 CFR 25, Section 831,
indicates that the air in the cockpit and cabin must be free of harmful
or hazardous concentrations of gases and vapors. Starting in 1996 (Section
59
25.831), this standard is modied to include airow provisions not only for
crew members but for all passengers as well. Specically, in normal operating conditions, the ventilation system must be designed to provide each
occupant with an airow containing at least 0.55 lb of fresh air per minute
(or 10 ft3 /per min (cfm) at 8000 ft cabin altitude). CO concentrations higher
than 50 ppm are considered hazardous, and CO2 concentrations during ight
should not exceed 5000 ppm (or 0.5% by volume at 25 C and 760 mm Hg, sea
level equivalent) in compartments occupied by passengers or crew members
(FAR 25, Section 832).
The FAR (Section 25.832) indicated that O3 concentrations during ight
must not exceed 0.25 ppm at sea level equivalent at any time above ight
level 320 (i.e., 10 700 m) or to exceed 0.1 ppm at sea level equivalent for
a time weighted average during any 3-h interval above ight level 270 (i.e.,
9000 m). Cabin-pressure altitude is covered in Section 25.841 as being lower
than 8000 ft (2500 m) at the maximal operating altitude of the aircraft under
normal conditions.
Many guidelines and standards are applicable to air quality in air cabins
and the exposures are legally established by various organizations. Table 2
provides a summary of existing guidelines and standards that might be applicable to airplane cabins. Among these: FAA, which has regulatory authority;
ASHRAE which provides guidelines and standards on exposure in indoor environment, including those in aircraft [9]; the Occupational Safety and Health
Administration (OSHA) which establishes permissible occupational exposure
limits (PELs); the American Conference of Governmental Industrial Hygien-
Table 2 Air quality standards and guidelines

Guidelines/standards
Key environmental factors
ACGIH TLVs, 1998 [14] Threshold limits values for

exposure to chemical substances,
physical and biological agents
ASHRAE 62-89 [15]
Ventilation for acceptable
indoor air quality
FAA-Airworthiness
Acceptable exposures:
1998 [16]
CO, CO2 , O3 , cabin pressure
OSHA PELs, 1998 [17] Permissible exposure limits
for toxic chemicals
US EPA, 1998 [18]
National Air Quality

Standards for criteria
pollutants: CO, O3 , NOx ,
SOx , particulate matter and lead
Population/environment
Occupational
Industrial workplaces
General population
Buildings
General public and crew
Aircraft cabin
Occupational
Industrial workplaces
General population
Ambient outdoor
60
ists (ACGIH) which recommends threshold limit values (TLVs) to protect

worker health; and the Environmental Protection Agency (EPA) which promulgates the National Ambient Air-Quality Standards (NAAQSs) for outdoor
air. OSHA and ACGIH limit values focus on adults of working age while children and elderly are better covered under the EPAs NAAQSs.
The results of a number of scientic studies in Europe and in the US
will soon be published, and are likely to inuence current regulations on
air quality in aircrafts. Further, in 1995, ASHRAE established an Aviation
Standard Committee, SPC 161P, to develop a standard that will become the
point of reference for design and operation of aircraft ventilation, ventilation
and pressurization systems. The proposed ASHRAE standard titled Cabin Air
Quality Standard (161P) is still under development. The purpose of this standard is to dene requirements for air quality in passenger aircraft and specify
methods for measurement and testing in order to establish compliance with
the standard. A draft of the standard may be released in 2005.
4
Global Thermal Comfort
The environmental quality of a space is determined by the occupants response to various environmental stimuli and the integration of these inputs
into a comfort response. If one assumes that sufcient heating or cooling capacity is available to maintain the desired average temperature within a space,
then a comfortable thermal environment will be completely dependent upon
the distribution of treated air in the space. From a thermal standpoint, it is
possible to have an average temperature at a given location in the space which
satises overall criteria for thermal balance. At the same time, there may be
conditions which cause the local temperature throughout the space to vary
from this average or mean value. The objective of a good air distribution
system is to provide, within the occupied space, the proper combination of
temperature, air motion, and relative humidity to keep the occupants comfortable.
The heat balance equation for the human body is the equation of the rate
of heat production to the rate of heat loss. The human body continuously
generates heat. Therefore, the body must dissipate heat to stay within the
comfort range. The total metabolic energy produced within the body is the
metabolic energy required for the persons activity plus that required for shivering. Some of the bodys energy production may be expanded as external
work done by the muscles. The remaining difference is either stored (causing the body temperature to rise) or dissipated to the environment through
the skin surface and respiratory tract. This heat dissipation from the body
occurs by several modes of heat exchange: sensible heat ow from the skin
and during respiration, latent heat ow from the evaporation of sweat and
61
moisture diffused through the skin, and latent heat ow due to evaporation
of moisture during respiration. Sensible heat ow from the skin is a mixture
of conduction, convection, and radiation for a clothed person.
The existing technique for evaluation of the global thermal comfort of the
body is based on an analysis of the heat balance for the human body:
S = M W C R Esk Cres Eres K
(1)
Where:
S
M
W
C
R
Esk
Cres
Eres
K
Heat storage in the body, W/m2 ,

Metabolic heat production, W/m2
External work, W/m2
Heat loss by convection, W/m2
Heat loss by radiation, W/m2
Evaporative heat loss from skin, W/m2
Convective heat loss from respiration, W/m2
Evaporation heat loss from respiration, W/m2
Heat loss by conduction, W/m2
It is assumed that the body is in a state of thermal equilibrium (comfort) so

that no change in heat storage occurs, i.e., thermal storage is 0. The factors
inuencing this heat balance are activity level, thermal resistance of clothing,
vapor permeability of clothing, air temperature, air speed, and water vapor
partial pressure. The activity level for passengers could be in the range of
40 W/m2 , sleeping, to 60 W/m2 , seated.
The heat transfer from the skin surface to the surrounding air is treated
in two sections: from the skin through the clothing through conduction heat
loss, and from the clothing to the environment through a combination of
convective and radiation heat losses. The convective heat loss can be expressed in terms of a heat transfer coefcient and the difference between the
mean temperature of the outer surface of the clothed body and the indoor
air temperature. The transport of heat through clothing strongly depends
on the thermal resistance of clothing ensembles. The thermal resistance of
normal indoor clothing is in the range 0.51.0 clo (1 clo is 0.155 m2 K/W).
Typical values of thermal resistance of clothing ensembles, Iclo , are given in
Table 3 [19].
In case of aircraft passengers, a clothing value equivalent, in clo, of the seat
can be deduced and the total clo value of the passenger can be corrected by
adding the clo of the seat. The correction value is proportional to the amount
of chair surface area in contact with the body [2, 20].
For environmental conditions similar to commercial and residential buildings ASHRAE prepared a chart to predict the thermal comfort of an occupant,
see Fig. 1. The ASHRAE Standard 55-92 species environmental conditions
that are acceptable to 80% or more of the occupants [19]. It is mainly applicable to sedentary occupants with normal winter or summer clothing
62
Table 3 Thermal resistance of clothing ensemble

Ensembles description
Icl (clo)
Walking shorts, short-sleeve shirt

Trousers, short-sleeve shirt
Trousers, long-sleeve shirt
Same as above plus suit jacket
Trousers, long-sleeve shirt, long-sweater, t-shirt
Sweat pants, sweat shirt
Knee-length skirt, short-sleeve shirt, panty hose, sandals
Knee-length skirt, long-sleeve shirt, half slip, panty hose, long-sleeve sweater
Same as above, replace sweater with suit jacket
Ankle-length skirt, long-sleeve shirt, suit jacket, panty hose
0.36
0.57
0.61
0.69
1.01
0.74
0.54
1.1
1.04
1.1
Clothing insulation value maybe expressed in clo units. For clarity, the symbol I is used
with the clo unit instead of the symbol R. The relationship between the two is R = 0.155I,
or 1.0 clo is equivalent to 0.155 m2 K/W.
ensembles (0.81.2 clo winter clothing or 0.60.8 clo summer clothing). The
acceptable ranges of operative temperature and humidity for winter and summer seasons are dened by the shaded areas in the psychrometric chart.
Fig. 1 Acceptable ranges of operative temperature and humidity (1992 ASHRAE

standard55. American Society of Heating, Refrigerating and Air-Conditioning Engineers www.ashrae.org)
63
ASHRAE sets its temperature limits (coupled with relative humidity) to

form a comfort zone, as follows:
Winter:
To
Summer:
To
t0 = 2023.5 C
t0 = 20.524.5 C
t0 = 22.526 C
t0 = 23.527 C
at 18 C wet bulb
at 2 C wet bulb
at 20 C wet bulb
at 2 C wet bulb
These limits are acceptable assuming the occupants are clothed in typical
summer and winter clothing, and with light, mainly sedentary activity.
Haghighat et al. measured the thermal comfort aboard 43 ights on commercial airlines [2]. The measurements were performed continuously during
the whole ight (from the departure gate to the arrival gate), and thermal
comfort parameters monitored were temperature, relative humidity, and carbon dioxide. Figures 25 show the temperature, relative humidity and carbon
dioxide concentration for four ights as a function of time. Under summer
conditions (a cooling season) ASHRAE Standard 55-92 [19] recommends that
the mean ambient temperature should be in the range 2326 C and the minimum level of RH 30%. As indicated in these gures, the air temperature was
often below the recommended range, and the RH was always too low.
4.1
Thermal Comfort Requirements for Elderly Persons
Most thermal comfort studies have dealt with a homogeneous population
with similar physical characteristics (neither ill nor old). Age, race, habits,
health, and origin are some parameters which must be taken into account
when the thermal comfort of the human body is considered [21].
The American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) Standard 55-1992 states in reference to Table 3 (ASHRAE
55-1992, p. 8), For infants, certain elderly people, and individuals who are
physically disabled, the lower limits of Table 3 should be avoided. The statement wishes to accommodate, in a general fashion, the ill or physically handicapped. The manner in which it is treated needs to be addressed, considering
increasing concern over thermal comfort for persons with physical disabilities, which have been brought to light by many health care professionals.
The thermal sensations of persons who suffer from various disabilities,
diseases, genetic disorders, as well as those taking medications, vary from
one population to another. Physically disabled, elderly, and ill persons do
not portray the same impairments and thus thermal sensations also follow
a distinct behavior. Epidemiological studies have shown that the elderly are
particularly susceptible to extremes in temperature [22, 23]. It is generally
accepted that, since the elderly are less active than the young, the elderly
would require more external heat to maintain neutral conditions. Temperature control mechanisms have been shown to deteriorate with age [24, 25].
64
Fig. 2 CO2 concentration, RH and temperature for the Airbus 320 (Flight 6/27)
Fig. 3 CO2 concentration, RH and temperature for the DC9 (Flight 6/30B)
However, since the elderly have a lower level of metabolic heat production,
and a reduced evaporative loss of heat, the two phenomena cancel each other,
implying that the elderly do not prefer higher temperatures than the young. It
has been shown, however, that the elderly may not be able to sense a decrease
in ambient temperature as well as the young since the peripheral temperature perception becomes impaired. Furthermore, it has been shown that the
elderly may have an impaired autonomic function, a lower sudomotor response, an altered sensitivity of the cutaneous vasculature to thermal stress,
and more variable body core temperature control [2530].
65
Fig. 4 CO2 concentration, RH and temperature for the Boeing 767 (Flight 7/10)
Fig. 5 CO2 concentration, RH and temperature for the Airbus 340 (Flight 7/23)
Past research work presents little evidence to suggest that healthy elderly
people preferred a different ambient temperature than that of the young. This
was shown in experimental studies where subjects were exposed to several
thermal environments [3133]. Fanger found that the neutral temperature
was equivalent at 25.6 C for both young and old [34]. McIntyre also found
no differences in preferred temperature after high and low temperature exposures [35]. Collins et al. found that even though elderly people preferred
the same mean comfort temperature as the young, they did so less precisely
than the young when given control over their environment [24]. Yoshida et al.,
66
through detailed measurements, showed that the disabled and the aged need
higher temperatures than their caregivers [36].
Taylor et al. studied age-related differences between young and old males
with respect to their inability to regulate room temperature [37]. A comparison between the measured skin temperature show that it varies between
28.3 and 29 C during the cold changes and between 31.4 and 32.3 C for hot
changes, for elderly and young persons respectively.
5
Health Considerations Relative to Environmental Parameters
Flight attendants and passengers have complained about air quality in airplanes. However, a systematic collection of data related to airplane environmental exposures is not yet available and the effects of the environmental
conditions on occupants health and comfort are yet to be fully exploited.
Complaints reported may be related to the ight environment (e.g., dryness),
chemical or biological contaminants, psychological and physiological stressors, and exacerbation of pre-existing medical conditions. Crew members
and passengers often indicate fatigue, dizziness, headaches, sinus and ear
problems, dry eyes, and sore throats during and after travel. These symptoms, common to many conditions, are often mistaken for infections [38].
Other characteristics, such as stresses unrelated to their time on aircraft, can
make people more susceptible to infections. Cohen et al. (1991) indicate that
psychological stress can be associated with increased susceptibility to the
common cold [39]. Table 4 provides a summary of the key potential pollutants
by class of concern (high, medium, and low) their health effects, frequency of
exposure, and concentrations limits by contaminants as dened by relevant
standards.
According to the 1986 report of the National Academy of Sciences, contaminants and environmental characteristics (e.g., relative humidity and air
pressure) identied in surveys of airline cabin air suggest a diverse set of adverse health effects that could arise from exposure to cabin environment
from acute effects to long term effects [12]. Exposure sources may be categorized as exposures related to incidents such as CO, deicing uids, vapors from
leaks of engine oils; and exposure related to normal operations such as ozone,
CO2 , relative humidity, cabin pressure, bioefuents, and pesticides.
Densely conned occupants and pressurized spaces generically characterize airline cabins and similar enclosed spaces. Operational services in
buildings as well as in airplanes strive for energy efciency and the provision
of healthy and comfortable environments. From 1978 to 1999 the number of
passengers increased from 383 million to 1.462 billion. This change is characterized by older passengers; passengers with health conditions that could
be exacerbated during ights (e.g., cardiovascular diseases); more densely
Medium concern
Airborne
allergens
Ozone (O3 )e
High concern
Cabin pressure
(pressure
altitude
equivalent)
Environmental
factor
Inhalation can result in irritated

eyes and nose, sinusitis, acute
exacerbation of asthma,
or anaphylaxis
Because of decreased oxygen pressure

serious health effects may occur, in
infants and people with cardiovascular
diseases. Temporary pain or discomfort
due to gas expansion (e.g., middle
ear or sinus) may occur
Airway irritation and reduced lung
function may occur at 0.1 ppm
with increasing severity at higher
concentrations, exposure durations
and respiratory rates
Potential
health impact
Frequency and intensity of exposure

sufcient to cause sensitization
or symptoms is not known
Aircraft without O3
converters that y at high altitudes
might have higher concentrations
The frequency and the concentrations
on these ights need to be
studied further.
Nearly all ights
Frequency
of exposure
Few exposure data are available

Only self reported information is
available on hypersensitivity
responses
FAA: 0.25 ppm, sea level equivalent,

any time above 9800 m
0.1 ppm, sea level equivalent,
any time above 9800 m
TWA, during any 3-h interval
ASHRAEa : 0.05 ppm
OSHA PELb : 0.1 ppm
EPA NAAQSc : 0.12 ppm (1 h)
0.08 ppm (8 h)
ACGHIH TWAd : 0.05, 0.08, 0.1 ppm
(heavy, moderate and light work)
Reliable measurements are available

FAA: 75 kPa at 2440 m
With a range from 101 kPa on
the ground at sea level to a minimum of
75 kPa in ight regardless of altitude
Concentrations
limits
Table 4 Key potential pollutants by class of concern, their health effects; frequency of exposure and concentrations limits by contaminants as
dened by relevant standards. (Source: adapted from National Academy of Sciences, 2002 [13])

67
Exposure may have no effect or

cause an infection with or without
symptoms
Health effects, such as skin rashes,
can result from dermal or inhalation
exposure
Infectious
agents
Pesticides
Elevated concentrations are

associated with increased perceptions
of poor air quality. Used as an
indicator of ventilation adequacy
Headaches and lightheadedness occur

at low concentrations. More severe
health effects result from higher
and longer exposures
CO
Low concern
Carbon dioxide
(CO2 )
Potential
health impact
Environmental
factor
Table 4 (continued)
Concentrations are normally below

FAA limits
Occurrence is possible on selected

aircraft used in international
ights
Presence of infectious agents is

likely, but the frequency of
exposure is not known
High concentrations could occur

during air-quality incidents.
Frequency of incidents is highly
uncertain, but believed to be low
Frequency
of exposure
FAA: 5000 ppm

(sea level equivalent)
ASHRAEa : 700 ppm
above ambient airf
OSHA PELb : 5000 ppm
EPA NAAQSc : na
ACGHIH TWAd : 5000 ppm,
30 000 ppm
(Short-term exposure level)
Reliable measurements are available.
FAA: 50 ppm (sea level equivalent)

ASHRAEa :9 ppm (8 h), 35 ppm (1 h)
OSHA PELb : 50 ppm
EPA NAAQSc : 35 ppm
(1 h)-9 ppm (8 h)
ACGHIH TWAd : 25 ppm
No quantitative guidelines.
Little information is available
on the transmission of infectious
agents on airplanes.
N/A Only self reported
information is available on
health effects
Concentrations
limits
68
Inhalation of high concentrations

can cause adverse health effects
Mucosal irritation and annoyance

may occur
Temporary drying of skin, eyes,
and mucous membrane can occur at
low RH (1020%)
Deicing uids
Nuisance odors
Relative
humidity (RH)
Low RH occurs in most ights
Potentially on all ights
Very low
Frequency
of exposure
N/A
No information is available
on incidence of uids entering
airplanes
N/A
Data are available from
survey of cabin occupants
Reliable measurements in airplanes
are available. FAA minimal
design rate: 18%
Concentrations
limits
ASHRAE 62-89 [15]

EPA NAAQS, 40 CFR 50 [18]
c OSHA PEL: permissible exposure limit
d ACGHIH TWA: total weighted average concentration in a normal 8-h workday and 40-h workweek
e National Institute of Occupational Safety and Health (NIOSH) recommend exposure limit not be exceeded at any time for O is 0.10 ppm [40];
3
California Air resources Board ambient air-quality standard for O3 is 0.09 ppm for 1-h exposure; World Health Organization guideline for O3 is
0.06 ppm for 8-h exposure [41]
f Applies to use of carbon dioxide as a proxy for odors from bioefuents
Potential
health impact
Environmental
factor
Table 4 (continued)

69
70
packed economy class; and higher load factor (i.e., ratio of occupied seats to
total number of seats available on ights).
According to the 2002 report of the National Academy of Sciences [13]
a few conclusions can be drawn on the health of passengers and crew members:
Lowered barometric pressure in cabin environments may result in passenger and crew discomfort and health effects in susceptible people. Infants
may also be at greater risk for hypoxia under conditions of reduced oxygen
partial pressure (PO2 ).
The potential for adverse health effects associated with exposure to low
relative humidity needs to be studied further. However, low relative humidity can result in the discomfort of dry mucus membranes. Eye, nose
and respiratory tract irritation symptoms may continue even after the exposure is discontinued.
Ozone levels are greater at higher altitudes. Respiratory problems such as
coughing, wheezing, and asthma have been associated with elevated concentrations of ozone.
High concentrations of aldehydes, such as formaldehyde and phosphate
esters, and CO may cause neurological effects. Associations between presence, concentration, and health effects in passengers and crew members
need to be studied further.
Disinfection of some aircraft is normally done with pyrethroid pesticides,
which have low toxicity for humans; however they are neurotoxins and can
cause adverse health effects.
Stress from ying can exacerbate symptoms in passengers and crew who
have pre-existing illness such anemia, asthma, emphysema, and coronary
arterial diseases.
Infectious diseases can be transmitted from person to person in aircraft
on the ground and in ight. It is also known that people and crew members might travel while infectious with common respiratory diseases. There
are persistent concerns about the transmission of infectious agents, such
as inuenza, tuberculosis, and measles viruses during ights. The perception of this risk has been fueled by a few incidents of transmission of these
agents during ights. Evidence indicates that the concentrations of microorganisms found in aircraft cabins are not higher than in comparable public
spaces. There is little evidence that aircraft cabins enhance transmission of
measles virus [13]. Although it is reasonable to assume that infectious agents
are transmitted during ights, it is not possible to determine conclusively
whether transmission is related to personal contact or environmental conditions specic to passenger cabins [13].
71
6
Indoor Environment Parameters
In terms of the number of people using the airplane as a means of transportation, 600 million passengers in 1997 in the USA was predicted to grow
to a billion by the early 21st century [42]. With this growth, concerns about
indoor air quality have also grown. A number of similarities can be drawn
between the indoor environmental qualities of buildings and airplanes. The
strongest is that both are enclosed spaces occupied by people. In both environments people are exposed to a mixture of outside and recirculated air.
However, unlike buildings, the operation of airplanes is characterized by
operation in extreme temperatures, and varied ambient air quality and pressure. At a cruise altitude of 11 000 m the outside air temperature is usually
about 55 C but can range from about 70 to 30 C). Further, at this altitude the atmospheric pressure is about one fth that at sea level [43] (Table 5).
The practical percentage of oxygen (O2 ) is about the same as at sea level and
the partial pressure of oxygen (PPO2 ) is only about 4.7 kPa compared with
21 kPa at sea level, far below what is necessary for people.
6.1
Pressure (and Altitude Equivalent)
The ve main contributing gases of the atmosphere are nitrogen, oxygen, carbon dioxide, and water vapor, which account for 99.99% of the volume below
90 km altitude. Observations conrm that the atmospheric gases are mixed in
constant proportions up to 50 km [44, 45].
The atmospheric temperature varies with altitude in a manner which suggests that there are atmospheric layers each having particular characteristics.
The lowest layer, the troposphere, is the zone where weather phenomena and
atmospheric turbulence take place. Throughout this layer the temperature decreases linearly from the surface upwards fairly uniformly over the earth at
a mean rate of about 6.5 C/km. This decrease continues regularly until it
ceases at a height of several kilometers, which depends mainly on the latitude.
Table 5 Temperature prole of the US Standard Atmosphere
Property
Symbol
SI
English
Temperature
Pressure
Density
Specic weight
Viscosity
T
P
15 C
101.3 kPa (abs)
1.225 kg/m3
1.789 105 kg/(m s)
59 F
14.696 Psia
0.002377 slug/ft3
0.07651 lbf/ft3
3.737 107 lbf s/ft2
72
For a linear temperature variation with altitude given by T = T0 mZ, the

pressure variation with elevation can be approximated by:
P = P0 (1 m Z/T0 )g/mR = P0 (T/T0 )g/mR
(2)
Where:
T0 and P0
R
m
Z
g
Absolute temperature and pressure at sea level. Sea level conditions of the US Standard Atmosphere are summarized in Table 5
Ideal gas constant, (for air Ra = 287.1 J/kg K)
Mean rate of temperature variation with elevation, C/m
Altitude, m
Gravitational force per unit mass, 9.81 m/s2
The height of the upper boundary of the troposphere, the tropopause, varies
with latitude, season, and weather situation. In general, it has an elevation
of about 810 km in arctic regions in winter and is highest at 1618 km in
tropical and equatorial regions.
The second major atmospheric layer is the stratosphere, which extends
upward from the tropopause to about 50 km above sea level. This layer is
characterized by a temperature that steadily increases with height. The maximum temperature associated with the absorption of the suns ultraviolet
radiation by ozone occurs near the top of the layer where the average temperature is only a little below 0 C. The boundary where the temperature is at
a maximum is called the stratopause.
Commercial ights travel at altitudes of 10 00015 000 m where the air is
almost dry and the temperature is around minus 60 C, and the barometric
pressure is as low as 26 kPa. The air pressure is so low that a person would
become confused and lethargic in less than a minute. FAA stipulates that
cabin pressure should not be lower than 75 kPa, the air pressure that would be
experienced at an altitude of 2440 m (8000 ft) [16]. The purpose of the pressurization is to maintain the PO2 within acceptable levels. The PO2 values at
sea level and at pressure altitude of 2440 m (8000 ft) are 21 kPa and 16 kPa,
respectively. Thus the minimal PO2 allowed in the aircraft cabin at the maximal allowed cabin pressure altitude of 2440 m (8000 ft) is 74% of the sea level
value.
Therefore, the air has to be compressed to the proper pressure, which
heats it, and then cooled to the required cabin temperature, and then conditioned in an Environmental Control Unit (ECU) before it is introduced into
the cabin.
6.2
Ventilation Requirements
During ight the airplane cabin is totally dependent on the air provided
by the installed environmental control system. As described earlier, aircraft
73
environmental control systems must operate under extreme environmental

conditions, unlike those for buildings. Compressed air used for cabin pressurization is also used for ventilation, though it must be emphasized that pressurization and ventilation serve two distinct purposes. The primary purpose
of pressurization is to maintain the PO2 at acceptable levels while ventilation is a way of eliminating air pollutants from aircraft [51]. Environmental
systems must also prevent rapid changes in cabin pressure, which can cause
sudden changes in the volume occupied by gases in the body cavities and
result in discomfort.
Flow rate has almost no impact on PO2 ; the occupants consume only
a small portion of the oxygen in the air and, according to the NAS report,
ventilation rates well below those normally present in aircraft would not
seriously affect PO2 [13]. FAR 25 minimal designs outside air ow rate of
0.25 kg/min (0.55 lb/min) per cabin occupant implies that oxygen is supplied into the cabin at 0.058 kg/min (0.127 lb/min) per person. Assuming
a typical sedentary adult consumes oxygen at about 0.44 g/min [46]; occupants oxygen consumption reduces the PO2 levels by about 0.8% compared with a PO2 reduction of up to 25% to the reduced cabin pressure.
Thus, these conditions provide an adequate oxygen concentration within
the cabin even at lower ow rates, as long as the cabin is adequately
pressurized [51].
Four factors are considered in determining the ow rate requirements:
(a) outside air necessary to remove contaminants; (b) conditioned air to remove heat from the cabin; (c) total ow rate to provide adequate ventilation;
(d) outside air required to provide oxygen and to pressurize the aircraft. In
older aircraft only outside air was supplied to the cabins. Under these conditions the outdoor air ow necessary to comply with all four requirements
called for larger air ow rates than specied in FAR 25. For example, the ow
rate specied to maintain temperature control could be more than twice the
FAR specied for outside air for contaminant control.
This fact led to the practice of air recirculation to achieve higher rates
of total airow independent of outdoor air ow rates. This way the cabin
can benet from good circulation and conditioned air without causing cold
drafts. The temperature of the air to be provided can be determined using the
steady state heat balance as follows:
Tc = Ts + (Q/ms )(1/cp )
(3)
Where
Tc
Ts
Q
Temperature in the cabin, C

Temperature of the air supplied to the cabin, C
Amount of heat that must be added or removed from the cabin,
W(Btu/s)
74
ms Flow rate of conditioned air supplied to the cabin, kg/s

cp Specic heat of the air, 1000 J/kg C
Naturally, this implies that ltration is necessary. In new aircraft this is conducted using high efciency particulate lters (HEPA), with minimal efciency of 99.7% for 0.3 m particles. These lters, while effective in removing
particles from the recirculated air, including bacteria and viruses, need to
be used in combination with activated charcoal or similar lters if chemical absorption of gaseous contaminants is to be accomplished. While it is
understood that recirculation is not a substitute for outside air, outdoor air
ow rates in older aircraft without recirculation range from 5.9 to 9.6 L/s
(12.420.4 cfm) while on aircraft with recirculation range from 3.6 to 7.4 L/s
(7.615.6 cfm). These values reect the responses of Airbus and Boeing to
NAS in 2001 [13].
An outside air change rate of 1015 air changes per hour and a total air
change rate of 2030 air changes per hour are normally adequate to maintain cabin pressure and a comfortable thermal environment at cruise altitude.
Mass ow is the unit used in the FAR to specify the ventilation requirements.
In fact the mass ow rate is more important that the volume ow rate. Liters
per second (L/s) or cubic feet per minute (cfm) are common units to express
ow rates, but this practice can lead to ambiguity if pressure and temperature
are not stated at the same time.
For example, 1 m3 (35 ft3 ) of air at sea level will expand to 3 m3 (106 ft3 )
at cruise altitude pressure and temperature. Considering the FAR 25 minimum ventilation requirement of 0.25 kg/min (0.55 lb/min), at sea level and
atmospheric temperature of 15 C, this corresponds to a volumetric ow
rate of 3.4 L/s, which, at the maximal allowed cabin pressure altitude of
2440 m (8000 ft), and a typical cabin temperature of 22 C (72 F), is 4.7 L/s
(9.9 cfm). At the ambient outside atmospheric pressure at an altitude of
12 000 m (39 300 ft) and an atmospheric temperature of 63 C ( 81 F) the
volumetric ow rate would be 13.0 L/s (27.6 cfm).
Contaminants may be generated inside the cabin or outside the cabin or
may form from chemical reactions occurring inside the cabin. Source emission uctuations and ventilation rates inuence the concentration rates of
contaminants within the cabin. Generation of contaminants within the cabin
can be assessed using the basic steady state ventilation equations. For a particular contaminant i maybe expressed as [47]:
Dc,i = Do,i + Si /Vo
Where:
Dc,i
Do,i
Contaminant density in cabin air, kg/m3

Density of contaminant in outside air used for ventilation, kg/m3
(4)
Si
Vo
75
Strength of contaminant source, kg/s

Ventilation rate of outside air, m3 /s
Concentrations rather than densities are easier to use for gaseous contaminants. Therefore, the above equation can be expressed as follows:
Cc,i = Co,i + (Si MWa )/(mo MW)
(5)
Where:
Cc,i
Co,i
Si
mo
MWa
MW
Volume fraction of contaminant in cabin air

Volume fraction of contaminant in outside air used for ventilation
Strength of contaminant source, kg/s
Ventilation rate of outside air, kg/s
Molecular weight of air (28.96 g/mol)
Molecular weight of contaminant gas (g/mol)
The values of Eq. 5 can be demonstrated using CO2 as an example. CO2

in cabin air is largely anthropogenic, and the amount produced depends
on the level of human activity. From the analysis of past monitoring studies it has been concluded that CO2 concentrations often exceed 1000 ppm,
a level commonly used to judge acceptability in buildings, but are similar to
those measured in other transportation systems such as trains, buses, and
subways [50, 51]. None of the studies reported values above 5000 ppm. FAA
recently proposed that the allowable CO2 concentration in aircraft cabins
should be lowered to 5000 ppm from 30 000 ppm [16].
CO2 concentration is sometimes used to assess the adequacy of ventilation, and sometimes as an indicator of other contaminants [48]. FAR Section 25.831 states: Ventilation: 2. CO2 concentration during ight must be
shown not to exceed 0.5% by volume (sea level equivalent) in compartments
normally occupied by passengers or crew members. In aircraft, CO2 is primarily a bioefuent produced by normal metabolism of the occupants. Under
this condition, it is rather unlikely that CO2 concentrations reach the point
of causing adverse health effects [13]. Other bioefuents generated in proportion to CO2 may be responsible for complaints about odors, stale air,
or stufness. Current studies indicate that CO2 levels measured in aircrafts
ranged from 7001800 ppm [49].
A sedentary person will generate CO2 at about 7.7 106 kg/s (ASHRAE
621999). The concentration of CO2 in clean outdoor air is about 0.037%. The
molecular weight of CO2 is 44.01 /mol. If the occupants are the only source of
CO2 in the cabin, Eq. 5 becomes
Cc,CO2 = 0.00037 + N(7.7 106 )(0.658/mo )
(6)
N is the number of occupants and 0.658 is the ratio of the molecular weights
of air and CO2 . Considering that FAR 25 minimal design ventilation rate is
76
0.0042 kg/s per person (0.25 kg/min) can be estimated as

Cc,CO2 = 0.00037 + N(7.7 106 )(0.658/0.0042)
= 0.00158 = 1580 ppm
(7)
It must be underlined that contaminant concentrations in the cabin are subject to factors such as time to build up to steady state conditions after the
introduction of the source and time to decline after the source is removed.
Also contaminants may react in the cabin air, or on surfaces in the cabin. It
is also specic for airplanes having rapid changes of the cabin air (typically
around 15 min).
6.3
Relative Humidity
As in buildings, humidity is controlled both for human comfort and for integrity of the physical components. High humidity levels (e.g., greater than
70%) in combination with high temperature lead to occupants discomfort,
and potentially create condensation problems on the inside of the aircraft
shell, which might lead to corrosion or biological growth. Control systems are
then designed and operated to remove moisture from the outside air before it
is supplied to the cabins.
However, at cruise altitude relative humidity (RH) tends to be low, typically in the 1020% range. Nagda et al. indicate an average RH of 14.718.5%
depending on the aircraft assessed in each of the nine different studies considered [50]. There is no air humidication system in the aircraft, and humidity is only generated by water vapor from the breath and perspiration of
passengers. Therefore, the level of humidity would be expected to decrease
as the number of passengers decreases or when the ratio of outside air to
re-circulated air increases. Food preparation in the galleys could increase humidity, but air from galleys usually is not exhausted from the aircraft. For
safety, to reduce icing and corrosion, the mechanical systems are designed
to remove as much of the water from the cabin air as possible. The primary
cause of these low levels is the frequent exchange of cabin air with outside air.
A modern airplane is operated to assure 1015 outside air changes per hour
and 2030 total air changes per hour. Since the outdoor temperature at operating altitudes is low, 43 to 65 C, the moisture content is very low (i.e.,
< 1% RH at 23 C). This indicates that RH tends to be below the comfort levels
specied in ASHRAE standard 55-92 [19].
It is possible to design and operate for higher levels of RH to above 25%
and this has been experimentally tested [52]. However, normally, this is not
done. Limiting factors include the extra load to be carried, the potential for
microbial growth associated with humidiers, condensation, and fatigue of
the airplane structure. Minimizing the air exchanges at cruise altitude could
77
also increase humidity, but this practice has an inherent conict with the need
for control of pollutant exposure. By reducing the outdoor airow, RH will
increase at the same time as contaminant concentrations.
Using the basic steady-state ventilation equation for a particular contaminant (i.e., Eq. 5), even if water vapor is not considered a contaminant in this
context, moisture generation by occupants is often estimated using Fangers
relationship: a sedentary person, not sweating, in a low humidity environment, produces 0.013 g/s of moisture [53]. With this amount of moisture
generation, and the FAA minimal design ow rate of outside air of 0.042 kg/s,
the water vapor concentration in the cabin air will be 0.0050, or 0.5%, which
corresponds to a relative humidity of 18% at typical cabin air temperature.
This assessment might be slightly underestimated because of the differences
among individuals level of activities.
Table 6 shows the range of temperature and carbon dioxide and the lowest relative humidity measured for a number of ights, and Figs. 15 show the
variation of relative humidity as a function of time [21].
6.4
Ozone
Ozone (O3 ) is the primary ambient air pollutant of concern at cruise altitudes (900012 000 m). Altitude, latitude, and season are factors inuencing
O3 concentrations. O3 uctuates over a short period of time because of meteorological processes that cause air exchange between the lower stratosphere
and the upper troposphere [54]. Bischof [55] indicates that the highest O3
concentrations in the cabin are experienced during high altitude, long distance ight at high latitudes in the spring. His study on 14 ights over polar
areas reported concentrations greater than 0.1 ppm for 75% of the ight time,
with maximal concentrations of 0.4 ppm averaged over 4 h and 0.6 ppm over
1 h.
At cruise altitudes O3 is naturally present. The suns ultraviolet radiation
disassociates oxygen molecules (O2 ), and the oxygen atoms then tend to associate with other O2 to produce O3 . O3 itself is reactive and decomposes in
the stratosphere either by photodissociation or by catalytic destruction (e.g.,
reaction with nitrogen oxides or chlorine oxides). These phenomena of production and destruction, given the persistence of O3 concentrations, are fairly
well balanced. For example, over North America, O3 concentrations tend to
be higher at higher latitudes (greater than 60 N) than at low latitudes (circa
30 N).
Exposure to low concentrations of O3 has been associated with adverse
health effects. According to Weschler and Shields [56], O3 can also react with
chemicals present in the cabin and produce irritating contaminants such as
reactive radicals, quasi-stable compounds, and stable aldehyde, ketones, and
organic acids. These new substances can be more irritating than their pre-
78
Table 6 Temperature, relative humidity and CO2 for a number of ights and airlines
Airbus
320
DC9
Boeing
767
Airbus
3540
Flight no Altitude
m
Passenger load
per capacity
Temp
C
Lowest RH
%
CO2 level
ppm
6/27
6/28
6/28A
7/1
7/2
7/3
7/3A
7/5
7/5A
7/6
7/25
7/25A
8/2
8/2A
8/2B
8/2C
8/4
8/4A
8/5
8/8
8/9
8/10
6/28
6/29
6/30
6/30A
6/30B
7/6
7/6A
7/30
7/30A
7/30B
8/1
8/1A
8/1B
8/1C
8/9
7/10
7/18
7/22
7/22A
7/23
7/23
32/137
86/137
90/137
65/137
62/137
137/137
49/137
50/137
60/137
4/137
130/137
128/137
57/137
137/137
103/137
105/137
101/137
98/137
63/137
84/92
26/92
60/92
92/92
75/92
60/92
52/92
60/92
49/92
65/92
75/92
66/92
44/92
65/92
187/203
185/203
85/195
70/195
35/195
177/284
2324
2123
2123
2124
2122
2122
2023
2022
1923
2023
2022
1922
2224
2024
2125
2224
2024
2026
2123
2123
2124
2024
2024
2022
2024
2124
2124
2026
2223
2225
2122
2021
2121
2123
2324
2327
2223
2224
2224
2225
2227
1924
5.4
3.3
3.7
1.8
4.9
6.2
5.2
13.1
2.6
2.7
5.8
4.4
18.5
18.2
15.3
7.6
2.5
2.4
2.3
4.3
2.2
11.1
23.0
10.9
8.6
6.9
7.1
7.3
17.0
10.8
9.8
19.4
12.5
12.0
15.5
12.3
2.3
4.5
7.6
2.3
1.8
3.3
7421368
293664
4491016
390938
351997
469781
449840
5661172
430723
390958
6061114
312625
7811446
7811231
6251271
6841622
5081329
5082013
371957
5471075
7811290
4881035
6051211
309703
43011 407
5471250
5671446
567996
3711172
6251055
5081113
430645
5671290
7231309
6251387
6451368
488782
6841348
430820
430977
469801
4691114
11 900
11 300
11 600
11 900
11 300
11 900
11 300
10 700
11 900
11 300
8500
8500
8200
11 300
10 700
11 300
10 700
9450
11 300
10 700
7900
9500
9750
10 050
10 700
10 050
9450
10 050
8500
9450
9350
10 050
8200
10 700
12 200
11 900
10 700
11 900
10 700
11 900
79
cursors [57], which is another reason to limit O3 concentration in airplane

cabins.
The reactions with ozone might occur in the gas phase or on surfaces. To
be signicant, compounds known to react in the gas phase with O3 must react
at a rate as great as the air exchange rate normally used in airplanes [58].
Potential sources of compounds known to react with O3 include solvents,
cleaning uids, and synthetic rubber materials. These compounds include
d-limonene, -pinene, and isoprene.
Although the higher air exchange rates in airplanes, compared to the exchange rates used in ofce or residential buildings, limit the time available
for gas-phase (homogenous) chemical reaction to occur, evaluating the importance of surface (heterogeneous) O3 is less straightforward. In fact these
last reactions tend to occur at the high ventilation rates. In this situation the
reaction with unsaturated carbon-carbon bonds is favored and products with
a range of volatilities are generated. Volatile products desorb from the surface and enter the gas phase, in which they are diluted with ventilation air.
However, the greater production rate is offset by a greater dilution rate.
Rates of volatilization and vapor pressure inuence the deposition of
semivolatile products of heterogeneous O3 chemistry (e.g., formaldehyde).
This implies that volatilization can occur for extended periods after the initial production, even when O3 concentrations are close to zero. Also, surface
concentrations might become large enough for the surface emission rates to
exceed odor thresholds for some compounds [13].
Nazaroff et al. [59] and Weschler [60] have conducted extensive studies to
understand O3 decomposition in buildings. They have identied three important points in this context: (a) rate of reaction depends on humidity, with
higher rates of reaction occurring at higher humidity levels; (b) an aging effect where the rate of reaction slows after a period of constant exposure; (c)
decomposition on surfaces can produce secondary reactions such as production of aldehydes.
The FAA airworthiness standards for O3 are 0.25 ppm at any time above
9800 m and 0.1 ppm time-weighted average during any 3-h interval [16]. Because of the uncertainties associated with O3 concentration measurements,
and little condence that the FAA standard on O3 is regularly met, the NAS
recommends establishing a monitoring program for O3 concentrations in
airplanes [13]. The program should target both concentration trends and effectiveness of control measures. Instrumentation used should be reliable and
accurate and capable of making real time O3 measurements.
80
7
Conclusions
Concentrations in ground level air determine the exposure to various pollutants (e.g., O3 , CO, particulate matter) under normal operating conditions. At
cruise altitude the pollutant of primary concern is O3 . The National Academy
2002 Committee did not nd data that provided condence that O3 standards are regularly met. Data on CO2 , CO, particulate matter, volatile organic
compounds (VOC), and semi volatile organic compounds (SVOC) are sparse,
although CO2 concentrations appear to be below the FAA regulations. CO
and particulate matter concentrations appear to be lower than health-based
standards for ambient air, while VOC and SVOC appear to be in similar concentrations as in other transportation vehicles.
The thermal comfort studies are simply a comparison of the physical
measurements carried out in a limited number of ights with the recommendations in the two ASHRAE standards. However, these are based on health
and comfort considerations for an indoor environment at an atmospheric
pressure of around one atmosphere. As an example, ASHRAE Standard 55-92
on thermal environment conditions for human occupancy is based almost
entirely on data from climatic chamber studies at normal atmospheric pressure. It also derives largely from work with healthy and young subjects. This
suggests there may be some limitations to the use of the ASHRAE standards
in aircraft cabins at altitude. First, not all passengers are healthy and/or
young. The elderly and disabled persons are groups of special concern. During the ight, the cabin pressure is allowed to decrease to around 0.8 bar or
lower. There is no data in the literature to prove the validity of these standards
to groups of passengers who may be neither young nor healthy, and who are
at sub-atmospheric pressure, for both thermal comfort and health. It is, however, well established that the effect of various indoor air contaminants on the
human body may be intensied under sub-atmospheric conditions.
References
1.
2.
3.
4.
Hocking MB (2000) Chemosphere 41:603

Haghighat F, Allard F, Megri A, Blondeau P (1999) Indoor Built Environ 8(1):58
Haghighat F, Donnini G (1998) Building Environ 34:479
Woods JE, Boschi N (1995) Trends and perspectives in healthy buildings in research
and industry. Proceedings 13th international CIB world building congress, Amsterdam (The Netherlands), Publication 186, p 43
5. Vitruvius (First century BC) Ten books of architecture. Morgan MH (Translation)
Dover, NY, 1961
6. WHO (1948) Constitution of the World Health Organization. In: Basic Documents.
World Health Organization, Geneva, Switzerland
81
7. UN (1948) Resolution 217A-III, Universal Declaration of Human Rights, Article 25.

United Nations, NY
8. Murray CJL, Lopez AD (eds) (1996) The global burden of disease: vol 1. World Health
Organization, Harvard School of Public Health, and The World Bank, Geneva
9. ASHRAE (2001) Ventilation for acceptable indoor air quality; standard 622001.
American Society of Heating, Refrigerating and Air-Conditioning Engineers, Atlanta,
GA
10. Hunt ED, Reid DH, Space DR, Tilton FE (1995) Commercial airliner environmental
control system-engineering aspects of cabin air quality. Proceedings of the annual
meeting of the Aerospace Medical Association, 1995, Anaheim, CA, pp 18
11. Space DR, Johnson RA, Rankin WL, Nagda NL (2000) The airline cabin environment: past present, and future research. In: Nagda NL (ed) Air quality and comfort
in airliner cabins: ASTM STP 1393. ASTM, West Conshohocken, PA, 189
12. National Academy of Sciences National Research Council (1986) The airliner cabin
environment: air quality and safety. National Academy Press, Washington, DC
13. National Academy of Sciences National Research Council (2002) The airliner cabin
environment and the health of passengers and crew. National Academy Press, Washington, DC
14. ACGIH (1998) Threshold limit values for chemical substances and physical agents and
biological exposure. American Conference of Governmental Industrial Hygienists,
Cincinnati, OH
15. ASHRAE (1989) Ventilation for acceptable indoor air quality, ANSI/ASHRAE 62
1989. American Society of Heating, Refrigerating and Air-Conditioning Engineers,
Atlanta, GA
16. 14 CFR (1998) Code of federal regulations, airworthiness standards, Title 14 25.
Washington, DC
17. 29 CFR 1910 (1998) Code of federal regulations, occupational safety and health standards, Title 29, 1910. Washington, DC
18. 40 CFR 50 (1998) Code of federal regulations, national primary and secondary ambient air quality standards, Title 40, 50. Washington, DC
19. ASHRAE (1992) Thermal environmental conditions for human occupancy, ANSI/
ASHRAE 55-1992. American Society of Heating, Refrigerating and Air-Conditioning
Engineers, Atlanta, GA
20. Donnini G, Molina J, Martello C, Lai DHC, Lai HK, Chang CY, Laamme M, Nguyen
VH, Haghighat F (1997) Field study of occupant comfort and ofce thermal environments in a cold climate. ASHRAE Trans 103(2):205
21. Haghighat F, Megri AC, Donnini G, Giorgi G (2000) Responses of disabled, temporarily ill, and elderly persons to thermal environments. ASHRAE Trans 106(2):329
22. Keatinge WR (1986) Seasonal mortality among elderly people with unrestricted home
heating. Br Med J 293:732
23. Taylor NAS, Grifths RF, Cotter JD (1994) Aust NZ J Med 24:705
24. Collins KJ, Exton-Smith AN, James MH, Oliver DJ (1994) Age Aging 9:17
25. Collins, KJ, Dore C, Exton-Smith AN, Fox RH, MacDonald IC, Woodward PM (1977)
Accidental hypothermia and impaired temperature homeostasis in the elderly. Br Med
J 1:353356
26. Foster KG, Ellis FP, Dore C, Exton-Smith AN, Weiner JS (1976) Age Aging 5:91
27. Kenney WL, Fowler SR (1988) Appl Physiol 65:1082
28. Collins KJ, Easton JC, Beleld-Smith H, Exton-Smith AN, Pluck RA (1985) Clin
Sci 69:465
29. Wagner JA, Horvath SM (1985) Appl Physiol 58:187
82
30. Marion GS, McGann KP, Camp DL (1989) Gerontology 37:225

31. Rohles FH, Johnson MA (1972) Thermal comfort in the elderly. ASHRAE Trans
78(1):131
32. Grifths LD, McIntyre DA (1973) Environ Res 6:382
33. Fanger PO (1970) Thermal comfort analysis and applications in environmental engineering. McGraw-Hill, New York
34. Fanger PO (1973) Thermal comfort. McGraw-Hill, New York
35. McIntyre DA (1975) Determination of individual preferred temperatures. ASHRAE
Trans 81:131
36. Yoshida JA, Banhidi L, Polinsky et al (1993) J Thermal Biol 18(5/6):363
37. Taylor NAS, Allsopp NK, Parkes DG (1995) Preferred room temperature of young vs
aged males. Gerontology: Med Sci 50A(4):216
38. IEH (2001) Consultation on the possible effects on health, comfort and safety on aircraft cabin environments, IEH Web report W5. Institute for Environment and Health,
Leicester, UK http://www.le.ac.uk/ieh
39. Cohen S, Tyrrell DA, Smith AP (1991) Psychological stress and susceptibility to the
common cold. N Eng J Med 325(9):606
40. NIOSH (1997) Ozone. In: Pocket guide to chemical hazards, DHHS (NIOSH) 97-140.
National Institute of Occupational Safety and Health, Cincinnati, OH, pp 238239
41. WHO (2000) Guidelines for air quality. Cluster of sustainable development and
healthy environment, Dept. of Protection of the Human Environment, Occupational
and Environmental Health Program, World Health Organization, Geneva
42. NTSB (1998) We are all safer: NTSB-inspired improvements in transportation safety.
National Transportation Safety Board, Washington, DC
43. ASHRAE (1999) Aircraft. In: ASHRAE handbook fundamentals, chap 9. American
Society of Heating, Refrigerating and Air-Conditioning Engineers, Atlanta, GA
44. Wallace JM, Hobbs P (1977) Atmospheric science: an introductory survey. Academic,
New York
45. HMSO (1971) Handbook of aviation meteorology. Her Majestys Stationary Ofce,
London, UK
46. Nishi Y (1981) Measurement of thermal balance of man. In: Cena K, Clark JA (eds)
Bioengineering thermal physiology, and comfort. Elsevier, New York, pp 2940
47. ASHRAE (1997) Ventilation and inltration. In: ASHRAE handbook fundamentals,
chap 25. American Society of Heating, Refrigerating and Air-Conditioning Engineers,
Atlanta, GA
48. ASHRAE (1999) Ventilation for acceptable indoor air quality, ANSI/ASHRAE 62
1999. American Society of Heating, Refrigerating and Air-Conditioning Engineers,
Atlanta, GA
49. Nagda NL, Rector HE (2003) Indoor Air 13(3):292
50. Nagda NL, Rector HE, Li Z, Space DR (2000) Aircraft cabin air quality: a critical
review of past monitoring studies. In: Nagda NL (ed) Air quality and comfort in
airlines cabins: ASTM STP 1393. American Society for Testing and Materials, West
Conshohocken, PA, pp 215239
51. Hocking MB (2002) Trends in cabin air quality of commercial aircraft: industry and
passenger perspectives. Rev Environ Health 17:49
52. Hocking MB, Foster HD (2004) Environ Health Res 3(1):7
53. Fanger PO (1982) Thermal comfort: analysis and applications in environmental engineering. Krieger, Malbar, FL
54. Seinfeld JH, Pandis SN (1998) The atmosphere. In: Atmospheric chemistry and
physics: from air pollution to climate change. Wiley, New York, pp 148

55.
56.
57.
58.
59.
83
Bishof W (1973) O3 measurements in jet airliner cabin air. Water Air Soil Poll 2(1):3
Weschler CJ, Shields HC (1997) Atms Environ 31(21):3487
Wolkoff PT, Clausen PA, Wilkins CK, Nielsen GD (2000) Indoor Air 10(2):82
Weschler CJ, Shields HC (2000) Indoor Air 10(2):92
Nazaroff WW, Gadgil AJ, Weshler CJ (1993) Critique of the use of deposition velocity in modeling indoor air quality. In: Nagda NL(ed) Modeling of indoor air quality
and exposure, ASTM STP 1205. American Society for Testing and Materials, West
Conshohocken, PA, pp 81104
60. Weschler CJ (2000) Indoor Air 10(2):269

DOI 10.1007/b107238
Human Physical Stresses

at Normal and Abnormal Cabin Pressures
Neal W. Pollock
Center for Hyperbaric Medicine and Environmental Physiology, Department of
Anesthesiology, Duke University Medical Center, Durham, NC 27710, USA
neal.pollock@duke.edu
1
1.1
1.2
1.3
1.4
Physics of Altitude .
Atmospheric Pressure
Temperature . . . . .
Humidity . . . . . . .
Oxygen Content . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
88
88
90
91
91
2
2.1
2.2
2.3
Human Physiology and Oxygen

Oxygen Transport . . . . . . . .
Hypoxia . . . . . . . . . . . . .
Time of Useful Consciousness .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
91
92
93
94
3
3.1
3.2
3.2.1
3.2.2
3.2.3
3.3
Aircraft Capabilities . . . . . . . .
Flight Altitudes . . . . . . . . . . .
Cabin Pressurization . . . . . . . .
Evolution . . . . . . . . . . . . . . .
Engineering . . . . . . . . . . . . .
The Future of Cabin Pressurization
Normal Cabin Altitude . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
95
95
96
96
97
97
97
4
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
Health at Normal Cabin Pressure . . . .

Hypoxia . . . . . . . . . . . . . . . . . .
Passive Smoking . . . . . . . . . . . . . .
Low Humidity . . . . . . . . . . . . . . .
Transmission of Infectious Diseases . . .
Ergonomic Restrictions . . . . . . . . .
Chemical Contaminants . . . . . . . . . .
Tolerance with Underlying Health Issues
Inight Medical Emergencies . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
98
98
99
99
100
100
100
101
101
5
5.1
5.2
5.2.1
5.2.2
5.2.3
5.2.4
Loss of Cabin Pressure . . . . . . . . . . . .

History of Cabin Depressurization . . . . . .
Health Effects of Cabin Decompression . . .
Barotrauma . . . . . . . . . . . . . . . . . .
Hypoxia . . . . . . . . . . . . . . . . . . . .
Decompression Sickness (DCS) . . . . . . .
The Future of Decompression Sickness Risk
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
101
102
103
104
104
104
106
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
107
88
N.W. Pollock
Abstract Atmospheric pressure is reduced as a function of altitude, thus making hypoxia,

the condition of oxygen deciency, a concern for aviation. The effects of low grade hypoxia are often subtle and may be missed by both ight crews and passengers. The most
severe effects are widely appreciated when high prole incidents occur. The international
collective was stirred after October 25, 1999 when Payne Stewart, a professional golfer,
and ve other persons were killed when a Learjet ew on unmonitored autopilot for
approximately four hours before running out of fuel and crashing in South Dakota. Radio contact was lost after the aircrew acknowledged clearance to an altitude of 11 900 m
(39 000 ft). The accident investigation determined that the crew was incapacitated when
inadequate supplemental oxygen delivery followed a loss of cabin pressurization. This
chapter will discuss the physics and physiology of hypoxia, describe cabin pressurization and discuss the health effects at normal cabin pressure and following unplanned
depressurization.
Keywords Hypoxia Pressurization Supplemental oxygen Altitude decompression
sickness
Abbreviations
ATA atmospheres absolute
DCS decompression sickness
EPT effective performance time
FAD ying after diving
FAR federal aviation regulations (U.S.)
FI O2 fraction of inspired oxygen
HbO2 hemoglobin saturation
ICAO international civil aviation organization
ISA
international standard atmosphere
Pa O2 partial pressure of arterial oxygen
PA O2 partial pressure of alveolar oxygen
PI O2 partial pressure of inspired oxygen
Sp O2 arterial oxygen saturation estimated with pulse oximetry
TUC time of useful consciousness
1
Physics of Altitude
Environmental conditions play an important role in physiological function.
Normal atmospheric conditions vary as a function of altitude.
1.1
Atmospheric Pressure
Air pressure is exerted by the weight of air molecules held in the atmosphere
by gravity. Unlike water, which has tightly packed molecules, air has a large
amount of space between molecules and is highly compressible. The standard reference for atmospheric pressure is mean pressure at sea level. Sea
Human Physical Stresses at Normal and Abnormal Cabin Pressures
89
level pressure can be expressed in a variety of units: 101.3 kilopascals (kPa),

1.0 atmosphere absolute (ATA), 14.7 pounds per square inch (psi), 1013 millibars (mb), 760 millimeters of mercury (mm Hg), or 29.92 inches of mercury
(in Hg). Atmospheric pressure decreases in a curvilinear manner as altitude
increases.
The U.S. standard atmosphere [1] employs the most widely-used formula
for estimating pressure at any given altitude:
U.S. Standard Atm (mm Hg)

5.25588
288.15/(288.15 (6.5 Altitude in km))
760
=
(1)
The U.S. standard atmosphere is identical to the International Civil Aviation

Organization (ICAO) standard for altitudes below 32 km [2].
The U.S. standard atmosphere assumes a uniform atmosphere around the
globe. In actuality, atmospheric density varies as a function of latitude, which
is expressed as a thinner atmospheric layer in polar regions and a thicker atmospheric layer in equatorial regions. A less well-known formula has been
developed to estimate pressure for equatorial altitudes [3]:
Equatorial Region Predictor (mm Hg)

= exp 6.63268 0.1112 Alt in km 0.00149 (Alt in km)
(2)
A depiction of the two computations appears in Fig. 1. The magnitude of

the difference in barometric pressure computed by the two formulae increases with altitude. The equatorial region predictor estimates a less than
Fig. 1 Estimation of atmospheric pressure at different altitudes
90
N.W. Pollock
Table 1 Comparison of barometric pressure computed for different altitudes with standard atmosphere and equatorial region predictor formulae
Altitude
(m)
(ft)
Barometric
Pressure
computed by
standard
atmosphere (SA)
equation
(kPa)
(mm Hg)
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10 000
11 000
12 000
13 000
14 000
15 000
16 000
17 000
18 000
19 000
20 000
21 000
22 000
0
3300
6550
9850
13 100
16 400
19 700
22 950
26 250
29 550
32 800
36 100
39 350
42 650
45 950
49 200
52 500
55 750
59 050
62 350
65 600
68 900
72 200
101.3
89.9
79.5
70.1
61.6
54.0
47.2
41.1
35.6
30.7
26.4
22.6
19.3
16.3
13.8
11.6
9.6
8.0
6.6
5.3
4.3
3.5
2.8
760
674
596
526
462
405
354
308
267
231
198
170
145
123
103
87
72
60
49
40
32
26
21
Barometric
Pressure
computed by
equatorial region
predictor (ERP)
(kPa)
(mm Hg)
101.3
90.5
80.6
71.6
63.4
55.9
49.2
43.2
37.8
33.0
28.7
24.9
21.5
18.5
15.9
13.7
11.7
9.9
8.4
7.1
6.0
5.1
4.3
760
679
604
537
475
420
369
324
284
247
215
187
161
139
120
102
88
75
63
54
45
38
32
Percent
difference
(ERP/SA)
0.1
+ 0.7
+ 1.4
+ 2.1
+ 2.8
+ 3.6
+ 4.4
+ 5.3
+ 6.2
+ 7.3
+ 8.5
+ 9.9
+ 11.6
+ 13.4
+ 15.6
+ 18.2
+ 21.2
+ 24.7
+ 28.8
+ 33.7
+ 39.5
+ 46.4
+ 54.6
one percent greater atmospheric pressure at 1000 m, a 10% greater pressure at

11 000 m, and an almost 40% greater pressure at 20 000 m. Additional values
appear in Table 1.
Unless otherwise stated, the U.S. standard atmosphere will be used to compute the values derived in this chapter.
1.2
Temperature
The ICAO, a United Nations agency, denes the sea level international standard atmosphere (ISA) as 101.325 kPa pressure, 15 C (288 K) temperature
and zero percent humidity [2]. The ISA has a lapse rate of 6.5 C per km
91
through the troposphere (topping out around ve miles above the surface in
polar regions and as high as 10 miles above the surface in equatorial regions).
The troposphere is dened by a pattern of linear decrease in temperature
with increasing altitude. The tropopause separates the troposphere from the
stratosphere. The temperature of the lower stratosphere is stable at 217 K
( 56 C). The temperature increases in a linear manner through the majority
of the stratosphere due to ozone reactions. High-energy ultraviolet radiation
breaks the bonds of normal oxygen (O2 ), producing free oxygen (O) that can
combine with normal oxygen to form ozone (O3 ). Ozone absorbs more ultraviolet energy, releasing heat energy and splitting into one molecule of normal
oxygen and one free oxygen atom. The free oxygen can form more ozone or
normal oxygen depending on the element it reacts with rst. The greatest
amount of ultraviolet energy is absorbed in the highest reaches of the stratosphere where the ultraviolet radiation is most abundant. The greatest concentration of ozone is found at the approximate midpoint of the stratosphere.
1.3
Humidity
The troposphere contains virtually all of the water vapor in the atmosphere.
The atmosphere above this zone is almost completely dry.
1.4
Oxygen Content
Air is comprised of approximately 0.78 nitrogen, 0.21 oxygen and 0.003 carbon
dioxide by volume plus other trace gases. While the percentage of individual
gases in a mixture is independent of altitude, the absolute content of each, and
thus of the mixture, decreases as a function of altitude. The partial pressure
of ambient oxygen (ambient PO2 ) can be computed for any altitude with the
following equation:
Ambient PO2 = Barometric Pressure 0.2093
(3)
where 0.2093 is the more accurate volume fraction of oxygen.

The ambient PO2 at sea level is approximately 21.2 kPa (159 mm Hg).
2
Human Physiology and Oxygen
Oxygen is essential to our existence. Without oxygen to serve as the nal electron acceptor in the respiratory chain and then combine with hydrogen to
form water, the process of energy metabolism cannot be continued.
92
N.W. Pollock
2.1
Oxygen Transport
The partial pressure of inspired oxygen (PI O2 ) is lower than ambient PO2
due to the presence of water vapor as air is warmed and saturated during
inhalation. Water vapor pressure is 6.3 kPa (47 mm Hg) at body temperature
regardless of barometric pressure. Since the addition of water vapor does not
affect the proportion of gases in air the computation is straightforward:
PI O2 in kPa = (Barometric Pressure [kPa] 6.3) 0.2093
(4)
PI O2 in mm Hg = (Barometric Pressure [mm Hg] 47) 0.2093
(5)
The PI O2 at sea level is approximately 19.9 kPa (149 mm Hg). The amount of
oxygen reaching the alveolar level of the lung (PA O2 ) where gas exchange occurs with the blood is reduced to a mean of 13.9 kPa (104 mm Hg) due to
the mixing of fresh gas with that remaining in the airways at the end of expiration. The arterial PO2 (Pa O2 ) is further reduced to a mean of 13.3 kPa
(100 mm Hg) due to a small ventilation-perfusion mismatch as some blood
in the alveolar capillaries passes through poorly ventilated alveoli and the
addition of relatively deoxygenated blood from the bronchial and cardiac circulations (venous admixture).
Molecular oxygen is carried in the bloodstream in two ways. The vast majority (> 97%) is bound to the hemoglobin found in huge concentrations in
red blood cells. The unbound portion of oxygen is dissolved in plasma. The
binding of oxygen to hemoglobin is both rapid and reversible. The oxygen
tension (partial pressure) in the surrounding tissue drives the uptake and
release of oxygen by hemoglobin. Association occurs when and where the
oxygen tension is high and dissociation occurs when and where the oxygen
tension is relatively low.
A depiction of the oxyhemoglobin (Hb02 ) dissociation curve appears in
Fig. 2.
Efcient gas exchange allows hemoglobin in the pulmonary capillary beds
to be almost fully saturated with oxygen at a PaO2 of 13.3 kPa (100 mm Hg).
Increasing the oxygen tension has no effect on the carrying capacity of fullysaturated hemoglobin. The only increase in oxygen transport under these
conditions results from an elevation of the dissolved oxygen, a relatively inconsequential effect outside of the hyperbaric environment.
A reduction in tissue oxygen tension has a powerful and immediate effect on HbO2 . Oxygen is quickly released from hemoglobin when the oxygen tension is low. Such is the case in the peripheral capillary beds where
working tissues have extracted oxygen. The mean oxygen tissue tension of
5.3 kPa (40 mm Hg) is met with an immediate dissociation of oxygen from the
hemoglobin, replenishing tissue levels. The steep part of the oxyhemoglobin
93
Fig. 2 Oxyhemoglobin dissociation curve
dissociation curves indicates the region where a small change in the tissue
tension of oxygen will have a dramatic effect on the amount of dissociation.
The oxyhemoglobin association-dissociation process is even more dynamic than may be evident from the above discussion. In reality, the balance
is constantly shifting in response to a host of factors. A relatively rightward
shift (favoring dissociation) results from an increase in the tissue tension of
carbon dioxide, an increase in tissue temperature, an increase in the concentration of 2,3-diphosphoglycerate (a byproduct of anaerobic metabolism),
and/or a decrease in pH. Such changes, for example, would develop in the
capillary beds of exercising muscle. Less extreme metabolic conditions in the
pulmonary capillary bed would leave it in a more leftward zone, thus favoring
association (uptake) of oxygen. Recovery from exercise in the working muscle
tissue would result in a relatively leftward shift as metabolic factors return to
resting values. Voluntary hyperventilation, as another example, would drive
down the tissue tension of carbon dioxide and effect a strong leftward shift as
long as the hyperventilation was maintained.
The dynamic nature of oxyhemoglobin association-dissociation is an important foundation of our physiological function. It ensures oxygen uptake
when the supply is plentiful and promotes delivery to the tissues in proportion to the need.
2.2
Hypoxia
The preceding discussion demonstrates that hypoxia, the condition of less
than normal oxygen content, is managed at the tissue level as part of a normal, continuous cycle. A critical element of successful management is an ade-
94
N.W. Pollock
quate supply of oxygen. The aviation environment affects the ambient supply
and can, under the right combination of circumstances, produce a signicant
compromise. Aviation-related hypoxia has long been categorized to reect
different mechanisms [4]. Hypoxic hypoxia, or altitude hypoxia, results from
the reduction in oxygen tension associated with increasing altitude, and is
the principal focus of this discussion. Hypoxia begins when PA O2 falls below 13.3 kPa (100 mm Hg), potentially at an altitude as low as 450 m (1500 ft).
While only the performance of elite athletes participating in high-intensity
exercise has been shown to be affected at that low an altitude, it does establish an effect threshold. Subtle compromise under resting conditions may be
measured at altitudes as low as 1200 m (4000 ft). Night vision is measurably
impaired at that altitude [5]. Reaction time to novel complex tasks has been
shown to be impaired at an altitude of 1500 m (5000 ft) while performance on
previously learned tasks was not impaired at 2450 m [6]. Similar decrements
were reported by other investigators [7]. Hypoxic compromise is more likely
to occur in the presence of potentiating factors such as certain drugs and/or
medications, poor health or ambient contaminants such as carbon monoxide.
The presentation of hypoxia can be categorized into signs (evident to an
observer), symptoms (felt by the individual) and combined effects. Signs of
mild to moderate hypoxia include: increase in rate and depth of respiration,
delayed reaction time, impaired judgment, impaired muscle coordination,
and cyanosis. Symptoms include: paresthesias, lightheadedness, dizziness,
fatigue, air hunger (sense of not enough breathable air), feelings of apprehension or anxiety, visual impairment (affecting color/night vision, blurring, or
tunnel vision), mental confusion, nausea, headache, hot/cold ashes, and tingling. Combined effects may appear as a marked behavior change, possibly
in the form of an inappropriate euphoria or belligerence [810]. The impact
of hypoxic compromise is clearly important if experienced by individuals in
control of aircraft.
2.3
Time of Useful Consciousness
Signs and symptoms may have little time to become manifest if the degree
of hypoxic stress is extreme, particularly if the rate of onset is rapid. Time
of useful consciousness (TUC), also known as effective performance time
(EPT), provides an estimate of how long an individual is expected to be able
to function effectively during acute exposure to reduced ambient pressure
without supplemental oxygen and/or recent acclimatization (protective adaptation made possible by repeated prior exposures). Such a situation could
arise during depressurization of a pressurized cabin or rapid ascent in an unpressurized cockpit. Estimates of time of useful consciousness for different
altitudes used in North American ight training are found in Table 2 [11].
95
Table 2 Estimated time of useful consciousness breathing ambient air at different exposure altitudes [11]
Exposure altitude
(m)
(feet)
Time of useful consciousness/

Effective performance time
(min)
6100
6700
7600
8550
9150
10 650
12 200
13 700
20 000
22 000
25 000
28 000
30 000
35 000
40 000
45 000
30 +
510
35
2.53
12
0.51
0.250.33
0.150.25
Actual TUC is inversely related to rate of ascent and physical activity [12]
and inuenced by day-to-day factors and inter-individual differences. Factors
that may reduce TUC include: fatigue, increased age [13], dehydration, hypoglycemia, alcohol intake [14], medications and recent smoking. TUC will
increase with slower depressurization rates and recent acclimatization [15] to
hypoxia.
3
Aircraft Capabilities
3.1
Flight Altitudes
Early aircraft and the majority of current general aviation (i.e., noncommercial) aircraft employ normally aspirated engines. The performance
of these engines is strongly inuenced by altitude, practically limiting acceptable aircraft performance to altitudes of less than 3650 m (12 000 ft). The
evolution of aviation engine systems, however, from turbocharging small
engines to producing powerful jet engines, has allowed aircraft to perform effectively at much higher altitudes. Commercial aircraft now regularly cruise
at altitudes in excess of 13 700 m (45 000 ft) and military aircraft in excess
of 24 400 m (80 000 ft) [16]. A new step in the evolution of ight was taken
on June 21, 2004, when SpaceShipOne (Scaled Composites, Mojave, CA) became the rst privately-developed aircraft to reach space when it climbed to
an altitude of 100 124 m (328 491 ft) [17]. Richard Bransons Virgin Galactic
announced September 27, 2004 at the Royal Aeronautical Society in Lon-
96
N.W. Pollock
don that it will offer commercial space ights by 2007 with the technology
used in SpaceShipOne. The plan is to provide two to three hour suborbital
ights, with 90 min spent at approximately (328 100 ft) 100 000 m altitude.
The group expects to accommodate 3000 passengers in the rst ve years of
operation [18].
The ability of aircraft to attain high altitude requires additional engineering solutions to protect those on board.
3.2
Cabin Pressurization
Mechanical systems are used to maintain pressure within the passenger and
crew compartments (cabin altitude) at pre-selected levels. Cabin pressure is
generally not maintained at sea level pressure because of the weight required
to provide structural integrity sufcient to support such a large pressure differential at high altitudes. Increasing the weight of the aircraft would decrease
the payload.
3.2.1
Evolution
Attempts to develop pressurized aircraft in the U.S. commenced in 1920. After
several efforts and signicant failures, a successful model (Lockheed XC-35)
was delivered in 1937 [19]. The ability for sustained ight at high altitudes
was realized with the development of the jet engine. The rst ight of a turbojet powered aircraft (German) took place on August 27, 1939 (the Heinkel
He-178; engine designer Hans von Ohain). The rst ight of a British turbojet aircraft took place on May 15, 1941 (the Gloster E28/39; engine designer
Frank Whittle). The rst production jet aircraft was the Messerschmitt Me
262 ghter (rst ight July 18, 1942; engine designer Anselm Franz). While
concerns over hypoxia at high altitude were well established at this point,
concerns over sudden depressurization related to wartime operations slowed
acceptance of the pressurized cabin [19].
The advantages of cruise altitudes at and above 6100 m (20 000 ft) for
weather and obstacle avoidance became evident as long-duration ights became more common. Improvements in comfort and safety when cabin pressurization replaced oxygen masks made the widespread shift to pressurized
cabins inevitable [19, 20].
Boeing produced a B307 Stratoliner in 1940 that was able to maintain
a pressure differential of 18.1 kPa (2.62 psi) at 4550 m (15 000 ft) thereby providing a cabin pressure of 2450 m (8000 ft). The differential decreased to
17.2 kPa (2.5 psi) at a maximum cruise altitude of 6100 m (20 000 ft), providing a cabin pressure of 3800 m (12 400 ft). While successful, the vertical range
was limited. Boeing then developed the B-29 with the ability to maintain
97
a pressure differential of 45.2 kPa (6.55 psi), able to provide a cabin altitude
of 2450 m at an actual altitude of 9150 m (30 000 ft) [19].
Modern pressurized aircraft generally maintain a pressure differential between 27.669.0 kPa (4.010.0 psi). A 55.2 kPa (8.0 psi) differential, for example, would provide a cabin altitude of 3050 m (10 000 ft) at an actual altitude of 13 700 m (45 000 ft), or a 2450 m (8000 ft) cabin altitude at an actual
altitude of 12 900 m (42 250 ft).
3.2.2
Engineering
The engineering specics used to maintain cabin pressure vary with aircraft
type. Fundamentally, ambient air is collected, compressed, conditioned and
distributed through the aircraft cabin to maintain pressurization. The amount
of recyling of existing cabin air is one of the variables. Older commercial aircraft models frequently provide 100% fresh outside air to the cabin space.
These include the B-27, DC-9, and some DC-10s. Newer jets combine fresh
outside air with up to 50% re-circulated air to conserve fuel. These include the
A-300, B-737, B-747, B-757, some DC-10s, and the MD-80.
Cabin pressure is automatically regulated by controlling instrumentation
that is set to maintain a pressure differential between ambient and cabin
pressure and to vary the rate of pressure change during ascent and descent.
Automatic outow valves are employed to ensure that cabin overpressurization does not occur. Should an unexpected leak in the fuselage develop, such
as might be the case from the partial loss of a window seal, the outow valves
would close to compensate for the additional loss of pressure through the new
opening.
3.2.3
The Future of Cabin Pressurization
Advances in materials and systems engineering make it likely that the costs
associated with maintaining lower cabin altitudes will be reduced in the future. One example can be seen in SpaceShipOne. This craft was designed
to reach suborbital space at vacuum pressure. The cabin altitude was to be
maintained at 1800 m (5900 ft) throughout high-altitude ight [21]. This represents a maximum pressure differential approaching 76 kPa (11 psi). It is
likely that a similar evolution will continue to increase the pressure differential that can be maintained in regular commercial aircraft.
98
N.W. Pollock
3.3
Normal Cabin Altitude
The U.S. Federal Aviation Regulations (FAR) state that the maximum operating altitude in commercial, pressurized aircraft be limited to 2450 m (8000 ft)
(FAR 25.841). This limit was accepted by the U.S. Civil Aeronautics Board in
1957. This cabin altitude limit now serves as the international standard for air
carriers. The selection of the limit was somewhat arbitrary. One practical consideration was that there are major destination airports nearing this altitude.
Early studies also found little evidence of signicant compromise in healthy
subjects.
While commercial aircraft are required to follow aviation regulations and
limit cabin altitude to 2450 m, cabin pressure proles do not have to be
recorded. A small number of studies of cabin altitude have been published.
One measured cabin altitudes during 22 scheduled, commercial ights of
75180 min duration in either 707 or DC-9 aircraft. Peak cabin altitudes ranging between 1900 m (6050 ft) and 2600 m (8450 ft) with an unpredictable
variability in the pattern of pressure differential [22]. A subsequent study
sampled a wider variety of aircraft 16 types operated by 28 different airlines
completing 204 scheduled, commercial ights. The median cabin altitude was
1900 m (6200 ft), ranging from sea level to 2700 m (8900 ft). Approximately
71% of these ights were in B-727, DC-9 and B-737 aircraft. Higher mean (
standard deviation [SD]) altitudes were observed with newer aircraft, 2135
418 m (7004 1373 ft) for 14 B-767 and A-310 ights. Flight durations were
not reported [23]. More recent work evaluated cabin altitude in 21 ights
with mean ight duration of 239 38 (SD) min (mean standard deviation).
While ve different types of aircraft were included, 62% of the ights were in
the B737-300. The mean ( SD) peak cabin altitude was 2320 213 m (7610
700 ft) [24].
4
Health at Normal Cabin Pressure
Normal cabin altitude is well tolerated by healthy individuals. However, the
expanding reliance on air transportation has increased the age and health
spectra of air travelers and, in some cases, ight crews.
4.1
Hypoxia
A recent observational study of commercial airline pilots during 21 ights
with mean cabin altitude of 2320 213 m noted that over 50% of the subjects had minimum arterial saturation measures (Sp O2 estimated by pulse
99
oximetry) of less than 90% [24]. It is likely that the extreme low values were
transient. The graphic presentation provided for a single individual indicated
very brief excursions below 90% and an average Sp O2 of approximately 94%.
No adverse effects were reported.
A very recent study has developed a model which predicts an inverse relationship between age and arterial oxygen saturation in healthy individuals at
2450 m [25]. An age effect was not found in the previous study discussed [24].
Sensitive tests of performance have indicated that postural control is
slightly impaired at 2450 m (8000 ft) [26]. While there are implications for
compromise of spatial orientation, there is no evidence of compromised functional performance. The degree of hypoxia found at 2450 m is mild enough
that it will likely not be noticed by a healthy individual maintaining a close
to resting physical activity. This is not to say that these are benign exposures, however. Certainly, early recognition of hypoxia signs and symptoms
is important to avoid more serious escalation.
While symptoms of hypoxia are not expected to be severe in a healthy
population at normal cabin altitude, a number of complaints are frequently
reported during ight exposure. These include: increased susceptibility to
colds, irritation of eyes, skin and throat, intestinal complaints, nausea, lower
back pain, headache, dizziness, faintness and fatigue [27, 28]. Most of these
symptoms can be attributed to a few key factors.
4.2
Passive Smoking
The greatest insult to health and comfort during commercial aircraft travel is
likely the effect of secondhand smoke [29]. The mean respirable particle level
is reduced by approximately 94 percent during non-smoking ights [3032].
The increasingly common ban on smoking represents a signicant advance
in air cabin health. Complaints of eye and throat irritation, dizziness, nausea,
headache and faintness are commonly associated with exposure to secondhand smoke.
4.3
Low Humidity
A relative humidity of 3545% is normally considered comfortable. The extremely low water content of high altitude air results in a negligible relative
humidity when this gas is warmed (but not humidied) to normal aircraft
cabin temperature. The humidity that is found in the cabin primarily arises
from human respiration and food preparation. Levels will vary dramatically
as a function of the number of persons onboard, galley activity and the cabin
air exchange rate [33]. One study (DC-8 and DC-10 aircraft) reported a mean
relative humidity of 12% (range 816%) [34]. Another (Boeing 767-300) re-
100
N.W. Pollock
ported a mean relative humidity of 4.9% (range 0.930%) [32]. Low relative
humidity is associated with increased complaints of eye, skin and throat irritation. These effects are most noticeable with ight duration in excess of
three to four hours [35]. It has been recommended that relative humidity be
maintained as close to 30% as possible [36], but there is resistance to this
on the part of the carriers concerned with increased corrosion of the aircraft
structure.
4.4
Transmission of Infectious Diseases
The symptoms associated with prolonged breathing of dry air may mimic
infection but may also increase susceptibility to infection if mucociliary clearance is impaired. Investigators have shown that mucociliary clearance is
reduced when breathing dry air [37, 38]. Another group has reported that mucociliary clearance is reduced at altitudes greater than 5000 m (16 400 ft) [39],
although the results were confounded as the trials were conducted during
a mountain expedition during which both cold and altitude were simultaneously present. The literature concerning one study evaluating the transmission of the common cold in commercial aircraft exemplies the difculties
in this work [4043]. In addition to the debate regarding interpretation of
the ndings, the relatively short duration of the ights may represent a lower
degree of stress than expected with longer ights.
4.5
Ergonomic Restrictions
Cramped space, restricted mobility and seat design contribute to many physical complaints associated with commercial air travel. The risk of deep vein
thrombosis and/or pulmonary embolism is elevated by prolonged immobility [44], regardless of the mode of transport. One recent study demonstrated
a three-fold increase in deep vein thromboses in 964 passengers following
an eight-hour ight in comparison with a non-traveling control group of
1213 [45]. Only individuals with established risk factors for venous thromboses were shown to develop any.
4.6
Chemical Contaminants
While concerns over volatile organic compounds and ozone levels in aircraft
cabins are common, there is very little evidence that problematic levels exist
on a systematic basis [4648]. The concentration of most organic compounds
are lower than those found in typical home and ofce environments [48].
101
Additional studies are required to conrm this to the satisfaction of both passengers and ight crews [49].
4.7
Tolerance with Underlying Health Issues
While healthy individuals have little difculty tolerating standard cabin pressure, those with underlying health issues may be more susceptible. Common
respiratory conditions that increase risk include: chronic obstructive pulmonary disease, asthma, cystic brosis, coronary artery disease, pulmonary
tuberculosis, and a pre-existing requirement for oxygen or ventilator support [50]. Common cardiac conditions that increase risk include: new onset
or unstable angina, poorly controlled congestive heart failure, frequent/high
grade ventricular ectopy and severe or poorly-controlled hypertension [51].
Some infants may also be at elevated risk [52]. While the preight screening
of potentially at risk individuals is promoted by some investigators [53], the
efcacy of the testing can be questionable [54], and the effort would certainly
be far more expensive than providing an ample supply of available oxygen for
inight management of cases [55, 56].
4.8
Inflight Medical Emergencies
Medical emergencies during air travel are uncommon, occurring at rates of
1 per 14 000 to 40 000 passengers [57]. Syncope, cardiac and gastrointestinal
problems are the most frequent events [58, 59].
Incapacitation of aircraft crew is fortunately a rare event. A study of Air
France pilots and ight engineers ying between 1968 and 1988 identied
ten cases of sudden in-ight incapacitation [60]. Two of the ten were cardiac problems, a common problem identied by other investigators [61]. It
is possible that the reduced oxygen content at normal cabin pressure may
contribute to the stress precipitating these events.
5
Loss of Cabin Pressure
While pressurization systems can compensate for small leaks, more serious
breaches, such as the loss of a window, door seal, or other structural failure, will produce rapid depressurization of the cabin. Mechanical failures can
also compromise the ability of the pressurization system to regulate pressure,
particularly in older aircraft. Finally, crew action can lead to cabin depressurization, for example, to clear the cabin of smoke in the case of re.
102
N.W. Pollock
5.1
History of Cabin Depressurization
The failure of pressurization systems is generally a manageable problem if
adequate supplemental oxygen systems are available and functioning properly. Rare high prole cases, however, demonstrate the outcome of multiple
failures in a critical chain of events.
A recent incident occurred on October 25, 1999 when Payne Stewart, an
Australian professional golfer, and ve other persons were killed when a Learjet, Model 35 ew on unmonitored autopilot for approximately four hours
before running out of fuel and crashing in South Dakota, United States. Radio contact was lost after the aircrew acknowledged clearance to an altitude of
11 900 m (39 000 ft) [62]. The accident investigation determined that the crew
was incapacitated when inadequate supplemental oxygen delivery followed
a loss of cabin pressurization. The details of this incident are very similar to
another high prole case that occurred 20 years earlier in September, 1979.
Radio contact was lost with a Beechcraft Super King Air 200 en route to Exeter, UK, after the pilot acknowledged clearance to carry out an emergency
descent from 9450 m (31 000 ft) [9, 63]. The aircraft ew on unmonitored autopilot for approximately 5.5 hours before crashing in France. The accident
investigation determined that the commander had a practice of depressurizing the cabin prior to commencing emergency descent drills. The oxygen
masks were apparently donned after depressurization, but hypoxic incapacitation quickly resulted because the masks were not connected to the oxygen
supply.
Fortunately, cabin decompression incidents involving fatal outcomes are
relatively rare. A picture of the typical patterns and outcomes is evident from
military aviation records.
Canadian Forces ejection seat aircraft experienced 47 cases of serious
loss in cabin pressurization from 1962 through 1982. These occurred at
a variety of altitudes ranging from 457216 459 m (15 00054 000 ft) [64].
Twenty-nine of the incidents involved explosive depressurization, ve involved intermediate-rate depressurizations, and 13 involved slow depressurization. Compromise of the canopy (loss of seal, failure of latching system,
structural failure and inadvertent jettison) was the most common problem.
Three cases of hypoxia developed, the symptoms recognized by each of the
three pilots involved. No deaths or permanent injuries resulted.
Canadian Forces transport aircraft experienced 47 cases of serious loss in
cabin pressurization from 1963 through 1984. These occurred at altitudes
ranging from 210011 300 m (700037 000 ft) [65]. Two of the incidents involved explosive depressurization, ve involved intermediate-rate depressurizations, and 40 involved slow depressurization. Approximately 70% of these
cases resulted from mechanical failure. Failures of the door or ramp seals
accounted for another 21% of the cases. Two denite cases of hypoxia were
103
reported. Improperly maintained or functioning equipment and inadequate

passenger brieng were identied as contributing factors. No deaths or serious injuries resulted.
U.S. Navy aircraft experienced 205 reported cases of depressurization from
1969 through 1990. These occurred at altitudes ranging from 305012 200 m
(10 00040 000 ft) [66]. Of these, 90% (184 cases) were considered accidental;
73% caused by mechanical failure and 27% by structural damage. Ten percent
(29 cases) were crew-initiated, in most (95%) to clear smoke and fumes from
cockpit. Hypoxia was reported in approximately 20% of all the cases: 23 cases
of mild hypoxia, 18 with loss of consciousness and four fatal cases.
The U.S. Air Force received reports of 298 aircrewmen experiencing symptomatic hypoxia during ight operations between 1970-1980 [8]. Incidents
were reported across a wide range of altitudes, from 3050 to > 12 200 m
(10 000 to > 40 000 ft). Approximately 28% of the cases were attributed to operator error. More than 50% of the events occurred in unpressurized trainer
aircraft. No deaths or serious injuries resulted. A summary of the reported
causes appears in Table 3. Similar patterns are reported by others [9].
The Australian Defence Force had 27 reports of hypoxia led between 1990
and 2001 [10]. The majority of cases occurred at altitudes of 30505800 m
(10 00019 000 ft). A large proportion (63%) were attributed to failure within
the oxygen circuit, the mask or regulator, or a mask leak. There were no cases
of accidental decompression reported. Symptoms were recognized by the affected individuals in 76% of the cases; by another individual in another 10%
of the cases. No deaths or serious injuries resulted.
Table 3 Reported causes of hypoxia in 298 reported cases [8]

Reported cause
Percent of cases
unknown
cabin/cockpit decompression
regulator malfunction
oxygen hose disconnected at quick connect junction
poor mask t
regulator in OFF position
hole in oxygen hose
mask removal in ight
sticking inhalation/exhalation valve
oxygen supply depleted
leak due to mask/microphone improperly housed
missing seal, leaking
disconnection of line within system
33
19
16
9
4
3
3
3
2
2
2
2
1
104
N.W. Pollock
5.2
Health Effects of Cabin Decompression
5.2.1
Barotrauma
Barotrauma is a common but typically not serious problem associated with
depressurization. This is primarily due to the relatively narrow range of atmospheric pressure exposure. The absolute pressure differential across the
entire atmospheric column is equal to the difference between sea level and
10 m (33 ft) of sea water. Only very rapid depressurizations would be expected
to cause signicant problems for individuals familiar with equalizing techniques. A somewhat greater risk of discomfort would be expected for lay
persons.
Only one of the 47 cases of depressurization in Canadian Forces Air Transport aircraft included descriptions of barotrauma and this was limited to
difculty equalizing middle ear pressure, with no reported effect on ight
management [65].
5.2.2
Hypoxia
Hypoxia is the single greatest hazard associated with cabin decompression.
The high cruise altitudes of modern aircraft demand a quick response time
to activate oxygen breathing systems for the ight crew and passengers and
initiate emergency descent procedures should depressurization occur. The
recognition of the high risk associated with hypoxic compromise have resulted in good programs to train pilots to recognize symptoms of hypoxia
early and to document cases with the intent of avoiding future incidents.
Careful investigation and open reporting facilitate rapid corrections in
equipment or procedure, a critical strength in world aviation.
5.2.3
Decompression Sickness (DCS)
The risk of DCS is relatively low for most ying. The threshold altitude associated with a signicant risk of developing DCS may be between 4900 and
6400 m (16 00021 000 ft) [6769]. While DCS has been reported at 3350 m
(11 000 ft), this was with direct ascent from near sea level and physical exercise, both events somewhat atypical for a normal ight exposure [70]. While
the onset time for symptoms is inversely correlated with exposure altitude,
the rate of decompression does not appear to affect the risk up to rates
of 24 400 m min1 (80 000 ft min1 ) [71]. The ambient pressure at 18 000 ft is
approximately 0.5 atm abs. Most unpressurized aircraft y at altitudes less
105
than 3650 m (12 000 ft or approximately 0.65 atm abs). The U.S. FAR allow the
cabin pressure of pressurized aircraft to fall as low as a 2450 m (8000 ft or
approximately 0.76 atm abs) equivalent.
DCS is generally classied by presentation as Type I (pain only, typically
joint pain) and Type II (neurological involvement). Type I symptoms may
produce acute discomfort, but typically resolve with recompression. Type II
symptoms can be much more disruptive of physical performance and consciousness and may resolve less readily.
DCS was reported in two of the 47 cases of serious cabin depressurization
in Canadian Forces ejection seat aircraft when the canopy seal was compromised [64]. Both cases presented as Type I symptoms (joint pain) which
developed at cabin pressures of 7600 m (25 000 ft) and 9150 m (30 000 ft).
Flight control was not compromised in either case. No DCS was reported in
the 47 cases of depressurization in Canadian Forces transport aircraft [65].
Eleven crewmen reported Type I (pain-only) DCS in the 205 depressurizations recorded in U.S. Navy aircraft [66]. Most cases involved mild to
moderate joint pain and none were incapacitating.
A greater risk of DCS is associated with ying after diving exposure due
to the excess inert gas dissolved in the tissues during diving. The U.S. FAR,
stopping short of regulation, issues guidelines for the minimum surface interval between diving and ying: (1) a 12 hour surface interval after nodecompression dives (maximum 2450 m [8000 ft] cabin altitude); (2) a 24
hour surface interval after decompression dives (maximum 2450 m cabin altitude); and (3) a 24 hour surface interval after any dive if cabin altitude
exceeds 2450 m [72].
Recommendations for restrictions on ying after diving have been cautiously advanced since experimental evidence is limited. The rst North
American ying after diving workshop was held in 1989, sponsored by the
Undersea and Hyperbaric Medical Society and the U.S. National Oceanic and
Atmospheric Administration. Consensus recommendations following this
workshop were: (1) wait for 12 hours before ying after up to two hours of
no-stop diving within the previous 48 hours; (2) wait for 24 hours before ying after multi-day, unlimited no-stop diving; (3) wait for 2448 hours before
ying after dives that required decompression stops; and (4) do not y with
DCS symptoms unless necessary to obtain hyperbaric treatment [73].
A second North American ying after diving workshop was held in 2002,
sponsored by the Divers Alert Network and the Undersea and Hyperbaric
Medical Society. The consensus guidelines released after this meeting reected the increasing complexity of diving practices. The guidelines were
restricted to dives during which air was breathed and when post-dive ight
was conducted at cabin altitudes between 6002450 m (20008000 ft). The
recommendations were: (1) for a single no-decompression dive, a minimum
preight surface interval of 12 hours; (2) for multiple dives per day or multiple days of diving, a minimum preight surface interval of 18 hours; and
106
N.W. Pollock
(3) for dives requiring decompression stops, there is little evidence on which
to base a recommendation and a preight surface interval substantially longer
than 18 hours appears prudent [74].
The U.S. Special Operations Command requires a minimum 24 hour preight surface interval after diving and at least 30 min of oxygen breathing for
altitude exposures of 7600 m (25 000 ft) or higher. The rule was established as
a conservative practice in the absence of direct research data. A recent study
to evaluate the surface interval rule found no difference in DCS risk with preight surface intervals between 12 and 24 hours following a 60 min dive at
18.3 m (60 ft). It must be noted, however, that the dry, resting simulated dive
may not produce an equivalent risk to a typical dive [75]).
The U.S. National Aeronautics and Space Administration have produced
ying after diving tables that allow different minimal preight surface intervals depending on whether air or oxygen is breathed post-dive [76]. The scope
of the current NASA ying after diving table is limited. It was designed solely
to address the demands of in-water austronaut training activity, accommodating exposure to a maximum physiologic pressure (ambient water pressure
plus suit pressure) equivalent to a depth of 15 m (50 ft).
5.2.4
The Future of Decompression Sickness Risk
The risk of DCS has the potential to become a more important factor in commercial aviation in the future. Current practice in the case of commercial
aircraft depressurization is to provide emergency oxygen to all passengers
and crew by mask and to reduce altitude as fast as practicable to 3050 m
(10 000 ft) or lower with the current limitations on available oxygen supply.
This produces signicant operational challenges for airline operators. Flying at lower altitudes demands a much greater fuel consumption than ight
at high altitudes. Rerouting aircraft to alternate airports plays havoc with
scheduling and passenger satisfaction. Improvements in the technology of
onboard oxygen-generating systems (OBOGS) developed in the 1970s [77]
promises to resolve the problems with limited oxygen supply. This development would make it possible to consider sustained oxygen provision to
passengers and crew and the possibility of continued ight at a more efcient
higher altitude that could minimize schedule disruption.
The possibility exists that fairly high altitudes will be considered for very
long unpressurized ights. This combination could make the risk of DCS
a much more important factor than it currently is in aviation.
107
References
1. U.S. Standard Atmosphere, 1976 (1976) United States Committee on Extension to the
Standard Atmosphere. National Oceanic and Atmospheric Administration, Washington, DC (NOAA-S/T 76-15672): Supt. of Docs., US Gov Print Ofce (Stock No.
003-017-00323-0)
2. International Civil Aviation Organization (1993) Manual of the ICAO Standard Atmosphere extended to 80 kilometers [262,500 ft] Doc 7488. International Civil
Aviation Organization, Quebec, Canada
3. West JB (1996) J Appl Physiol 81:1850
4. Barcroft J (1914) The Respiratory Function of the Blood. University Press, Cambridge, England
5. McFarland RA, Evans JN (1939) Am J Physiol 127:37
6. Denison DM, Ledwith F, Poulton EC (1966) Aerosp Med 37:1010
7. Ernsting J (1978) Aviat Space Environ Med 49:495
8. Rayman RR, McNaughton GB (1983) Aviat Space Environ Med 54:357
9. Ernsting J (1984) Aviat Space Environ Med 55:407
10. Cable GG (2003) Aviat Space Environ Med 74:169
11. Willits P (ed) (1998) Private Pilot Manual. Jeppesen Sanderson, Englewood, CO
12. Busby DE, Higgins EA, Funkhouser GE (1976) Aviat Space Environ Med 47:117
13. Yoneda I, Tomoda M, Tokumaru O, Sato T, Watanabe Y (2000) Aviat Space Environ
Med 71:72
14. Nettles JL, Olson RN (1965) JAMA 194:1193
15. Sen Gupta J, Mathew L, Gopinath PM (1979) Aviat Space Environ Med 50:714
16. Pilmanis AA (2003) Lancet 362:16
17. Scaled Composites Tier One Private Manned Space Program Website, June 21,
2004 Press Release. http://www.scaled.com/projects/tierone/062104-2.htm. Cited 29
September 2004
18. Virgin Galactic Program Website, September 27, 2004 Press Release. http://www.
virgingalactic.com/news.html. Cited 29 September 2004
19. McFarland RA (1971) Aerosp Med 42:1303
20. Tomlinson DW (1941) J Aviat Med 12:136
21. Encyclopedia Astronautica: SpaceShipOne. Webpage updated September 29, 2004.
http://www.astronautix.com/craft/spaipone.htm. Cited 29 September 2004
22. Aldrete JA, Aldrete LE (1983) South Med J 76:12
23. Cottrell JJ (1988) Chest 93:81
24. Cottrell JJ, Leboviz BL, Fennell RG, Kohn GM (1995) Aviat Space Environ Med 66:126
25. Muhm JM (2004) Aviat Space Environ Med 75:905
26. Nordahl SHG, Aasen T, Owe JO, Molvaer OI (1998) Aviat Space Environ Med 69:590
27. Lyznicki JM, Williams MA, Deitchman SD, Howe JP III, Council on Scientic Affairs,
American Medical Association (2000) Aviat Space Environ Med 71:832
28. Rayman RB (2002) Aviat Space Environ Med 73:211
29. Mattson ME, Boyd G, Byar D, Brown C, Callahan JF, Corle D, Cullen JW, Greenblatt
J, Haley NJ, Hammond K, Lewtas J, Reeves W (1989) JAMA 261:867
30. Lee SC, Lam S, Luk F (2000) Proc Healthy Build 1:471
31. Lindgren T, Norback D (2002) Indoor Air 12:263
32. Lindgren T, Norback D, Andersson K, Dammstrom BG (2000) Aviat Space Environ
Med 71:774
33. Hocking MB (1998) Am Indust Hyg Assoc J 59:446
34. Vieillefond H, Fourn P, Aufret R (1977) Aviat Space Environ Med 48:503
108
N.W. Pollock
35. Nagda NL, Hodgson M (2001) Indoor Air 11:200

36. Air Transport Medicine Committee, Aerospace Medical Association (1997) Aviat
Space Environ Med 68:943
37. Salah B, Dinh Xuan AT, Fouilladieu JL, Lockhart A, Regnard J (1988) Eur Resp J
1:852
38. Daviskas E, Anderson SD, Gonda I, Chan HK, Cook P, Fulton R (1995) Eur Resp J
8:742
39. Barry P, Mason N, OCallaghan C (1997) Eur Resp J 10:35
40. Zitter JN, Mazonson PD, Miller DP, Hulley SB, Balmes JR (2002) JAMA 288:483
41. Hocking M, Foster HD (2002) JAMA 288:2972
42. Zitter JN (2002) JAMA 288:2972
43. Hocking MB, Foster HD (2004) J Environ Health Res 3:7
44. Symington IS, Stack BHR (1977) Br J Dis Chest 71:138
45. Schwarz T, Siegert G, Oettler W, Halbritter K, Beyer J, Frommhold R, Gehrisch S,
Lenz F, Kuhlisch E, Schroeder HE, Schellong SM (2003) Arch Intern Med 163:2759
46. Hetrick SM, Gould WD, Christensen DE (2000) Aviat Space Environ Med 71:408
47. Nagda NL, Koontz MD (2003) Aviat Space Environ Med 74:101
48. Nagda NL, Rector HE (2003) Indoor Air 13:292
49. Rayman RB (2002) Aviat Space Environ Med 73:211
50. British Thoracic Society Standards of Care Committee (2002) Thorax 57:289
51. Alexander JK (1995) Cardiol Clin 13:271
52. Parkins KJ, Poets CF, OBrien LM, Stebbens VA, Southall DP (1998) Brit Med J
316:887
53. Robson AG, Hartung TK, Innes JA (2000) Eur Respir J 16:214
54. Christensen CC, Ryg M, Refvem OK, Skjonsberg OH (2000) Eur Respir J 15:635
55. Lyznicki JM, Williams MA, Deitchman SD, Howe JP III, Council on Scientic Affairs,
American Medical Association (2000) Aviat Space Environ Med 71:827
56. Naeije R (2000) Eur Respir J 16:197
57. Mortazavi A, Eisenberg MJ, Langleben D, Ernst P, Schiff RL (2003) Aviat Space
Environ Med 74:922
58. Donaldson E, Pearn J (1996) Aust N Z J Surg 66:431
59. Gendreau MA, DeJohn C (2002) N Engl J Med 346:1067
60. Martin-Saint-Laurent A, Lavernhe J, Casano G, Simkoff A (1990) Aviat Space Environ Med 61:256
61. Froom P, Benbassat J, Gross M, Ribak J, Lewis BS (1988) Aviat Space Environ Med
59:278
62. Newman DG (2000) Flight Safety Australia Mar-Apr:42
63. Underwood Ground KE (1982) Aviat Space Environ Med 53:24
64. Brooks CJ (1984) Aviat Space Environ Med 55:1154
65. Brooks CJ (1987) Aviat Space Environ Med 58:268
66. Bason R, Yacavone DW (1992) Aviat Space Environ Med 63:341
67. Houston CS (1947) J Aviat Med 18:165
68. Davis JC, Tager R, Polkovitz HP, Workman RD (1971) Aerosp Med 42:85
69. Haske TL, Pilmanis AA (2002) Aviat Space Environ Med 73:1059
70. Kumar KV, Waligora JM, Calkins DS (1990) Aviat Space Environ Med 61:685
71. Pilmanis AA, Webb JT, Kannan N, Balldin UI (2003) Aviat Space Environ Med
74:1052
72. Federal Aviation Regulations. Website access: http://www.airweb.faa.gov/Regulatory_
and_Guidance_Library/rgFAR.nsf/MainFrame?OpenFrameSet
109
73. Shefeld P (ed) (1989) Flying after diving. Vol. 77(FLYDIV) 12/1/89. Undersea and
Hyperbaric Medical Society, Bethesda, MD
74. Shefeld P, Vann RD (ed) (2004) Flying After Diving Workshop. Divers Alert Network, Durham, NC, p 8
75. Pollock NW, Natoli MJ, Gerth WA, Thalmann ED, Vann RD (2003) Aviat Space Environ Med 74:1163
76. Pollock, NW, Fitzpatrick DT (2004) In: Shefeld P, Vann RD (ed) DAN Flying After
Diving Workshop Proceedings. Divers Alert Network, Durham, NC, p 59
77. Manatt SA (1981) Onboard oxygen generation systems. Aviat Space Environ Med
52:645

DOI 10.1007/b107239
Air Travel and Thrombosis

Paul L. F. Giangrande
Oxford Haemophilia Centre and Thrombosis Unit,
Churchill Hospital, Oxford OX3 7LJ, UK
paul.giangrande@ndm.ox.ac.uk
1
Deep Vein Thrombosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
112
Air Travel and Thrombosis: Is There a Link? . . . . . . . . . . . . . . . . .
112
How Common Is It? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
113
What Causes It? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
114
Prevention of Venous Thromboembolism Associated with Flight . . . . .
117
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
119
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
120
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
120
Abstract A deep vein thrombosis (DVT) is essentially the formation of a clot in the veins
of the leg. This causes obstruction to the normal ow of blood in the limb which can result in pain and swelling of the leg. Occasionally, a fragment of the clot can break off and
pass in the blood stream to the heart or major blood vessels leading into the lungs from
the heart. This phenomenon is known as pulmonary embolism and has been estimated
to occur in approximately 1% of cases of deep vein thrombosis. The long-term consequences of venous thromboembolism are not insignicant and include risk of recurrence
and post-phlebitic syndrome. It is now generally accepted that there is an association between any form of long-distance travel and venous thromboembolism and therefore the
alternative term of travellers thrombosis has been suggested as an alternative to the
term economy class syndrome. Thromboembolism is rarely observed after ights of less
than 5 h duration and, typically, the ights are of 12 h or more. Stasis in the venous circulation of the lower limbs is undoubtedly the major factor in promoting the development
of venous thromboembolism associated with travel. Some individuals may be particularly
predisposed to develop venous thrombosis because of congenital (inherited) deciencies
of natural anticoagulants, such as antithrombin, protein C or protein S. However, routine
screening of passengers for these abnormalities is not justied or cost effective but may
be of value in selected cases. It has also recently been suggested that exposure to mild
hypobaric hypoxia in pressurized aircraft may also result in activation of the coagulation
cascade but the data are conicting. The risk of venous thromboembolism is largely conned to those with recognized additional risk factors for venous thromboembolism. Leg
exercises whilst seated help to reduce the risk of DVT. There is also clear evidence from
prospective and randomized clinical trials to support the use of compression hosiery as
a preventative measure. By contrast, there is no rm evidence to support the indiscriminate use of aspirin as a routine prophylactic measure. Airlines have recently taken positive
112
P.L.F. Giangrande
steps to address the issue of air travel and thrombosis. At the same time, the travelling
public needs to be more aware of the issues and assume some responsibility for ensuring
tness to y and the choice (and therefore cost) of their seats.
Keywords Thrombosis Thromboembolism Thrombophilia Air travel
1
Deep Vein Thrombosis
A deep vein thrombosis (DVT) is essentially the formation of a clot in the
veins of the leg. This causes obstruction to the normal ow of blood in the
limb which can result in pain and swelling of the leg. Occasionally, a fragment
of the clot can break off and pass in the blood stream to the heart or major blood vessels leading into the lungs from the heart. This phenomenon is
known as pulmonary embolism and has been estimated to occur in approximately 1% of cases of deep vein thrombosis. The consequences of pulmonary
embolism depend on the size of the clot fragment. If relatively small, the fragment can pass to the periphery of the lung where it can cause chest pain and
haemoptysis (coughing up blood). If the embolus is large, it can completely
obstruct the ow of blood through the lungs with fatal consequences. The
mortality associated with pulmonary embolism rises with increasing age, but
is in the range 215% of cases.
The long-term consequences of venous thromboembolism are not insignicant. Quite apart from the pain and discomfort associated with an
acute thrombotic episode, the inconvenience and side-effects of anticoagulant
treatment with warfarin should also not be overlooked. Even with appropriate treatment, there is often some permanent damage to the veins in
the leg and this can lead to permanent circulatory problems with persistent
swelling of the limb or even chronic ulceration (post-phlebitic syndrome).
Approximately 60% of patients will develop post-phlebitic syndrome within
2 years despite appropriate anticoagulant therapy. A previous medical history
of thrombosis will also preclude future prescription of hormone replacement
therapy (HRT) or oestrogen-containing oral contraceptives for women, and
make it difcult to secure travel insurance in future because of the high risk
of recurrence.
2
Air Travel and Thrombosis: Is There a Link?
The subject of air travel and thrombosis has been the subject of much debate in both the lay and medical press in recent years. Although a possible
link has been recognized for many years the very rst report concerned that
113
of a physician who travelled from Boston to Venezuela in 1946 [1]. The consensus at a meeting of experts convened in March 2001 by the World Health
Organization to review the evidence was that there probably is a link, although the incidence is low and mainly involves passengers with additional
risk factors for venous thromboembolism [2]. However, venous thromboembolism is not exclusively associated with air travel, and it has also been
documented following long car, bus or even train journeys. A case control
study of 160 consecutive patients with DVT showed that 39/160 (24.5%) had
recently completed a journey by car, train or plane of longer than 4 h (of
which nine involved air travel) [3]. When the patients with DVT were compared with the control group, a history of recent travel was reported four
times more frequently in the subjects with venous thromboembolism (odds
ratio = 4). Thrombosis associated with ight is also by no means restricted
to those in the relatively conned conditions of economy class, and thus the
alternative term of travellers thrombosis has been suggested.
It is possible to derive some general conclusions from published cases of
venous thromboembolism associated with travel [4]. Thromboembolism is
rarely observed after ights of less than 5 h duration and, typically, the ights
are of 12 h or more. The risk rises with age: subjects over the age of 50 are
more at risk whilst those under the age of 40 are less vulnerable. Symptoms
of thromboembolism do not usually develop during or immediately after the
ight, but tend to appear within 3 days of arrival, when the patient may
present far away from the airport and thus the causal link may not be immediately apparent. Symptoms of thrombosis or pulmonary embolism have been
reported up to 2 weeks after a long ight. Pulmonary embolism may also be
the rst manifestation, without any symptoms in the lower limbs. Whilst most
case reports and studies involve deep vein thrombosis (DVT) in the lower
limbs, there are also reports of cerebral venous thrombosis [5] and arterial
thrombosis [68] associated with long ights. There is even a case report of
DVT in a patient with haemophilia who presented direct from an airport after
a ight of around 20 h [9].
3
How Common Is It?
The precise incidence of thromboembolism in relation to air travel is uncertain, though it has been estimated that at least 5% of all cases of deep
vein thrombosis (DVT) may be linked to air travel. A study based on 56 conrmed cases of pulmonary embolism amongst 135.3 million passengers passing through one airport in the period 19932000, clearly demonstrated an
association between duration of travel and risk of pulmonary embolism [10].
The incidence of pulmonary embolism was signicantly higher (1.5 cases per
million) for passengers travelling more than 5000 km when compared with
114
P.L.F. Giangrande
a risk of only 0.01 cases per million amongst passengers travelling less than
5000 km. Cases of pulmonary embolism clearly only represent the tip of the
iceberg of cases of DVT. A recent observational analysis from New Zealand,
based on the review of 878 passengers who travelled extensively (at least 10 h
within a 6 week period: mean 39 h) reported an incidence of venous thromboembolism of 1%, including four cases of pulmonary embolism and ve of
DVT [11]. However, the incidence of latent, asymptomatic thrombosis is likely
to be even higher. A prospective study of long-haul air passengers over the
age of 50 reported that 12/116 passengers (10%) were found by ultrasound
scanning to have asymptomatic DVT conned to the calf [12].
4
What Causes It?
The aetiology of venous thrombosis is usually multifactorial, with a combination of both constitutional and environmental factors responsible for causing
a thrombosis in an individual at a given time [13]. In this section, the possible
contributory causes will be considered and particular consideration given to
the environment of the typical jet airliner.
The three underlying causes of thrombosis are classically dened as Virchows triad: stasis, hypercoagulability of the blood, and vessel wall disease.
The risk factors for venous thrombosis are quite different from those for arterial thrombosis, and vessel wall disease is a major predisposing cause of
arterial thrombosis. Vessel wall disease is the most important factor in the development of arterial thrombosis and such thrombi are typically pale and rich
in platelets on histological examination. By contrast, thrombi in the venous
circulation consist primarily of red cells enmeshed in brin strands.
Stasis in the venous circulation of the lower limbs is undoubtedly the major
factor in promoting the development of venous thromboembolism associated with travel. The potential danger of connement in cramped conditions
has been recognized for some years. An increase in the incidence of fatal
pulmonary embolism was reported during the Blitz in London during the
Second World War [14]. Simpson recognized that the primary cause was mechanical impairment of venous circulation due to squatting for a prolonged
period in air raid shelters, and he recommended that bunks should be installed. The term economy class syndrome has been coined to describe
the phenomenon and this also emphasizes the role of impairment of venous
circulation due to prolonged immobility in a cramped position [15, 16]. Ingestion of alcohol will also encourage immobility during a ight and the use
of strong sedative medication may also be associated with an increased risk
of venous thrombosis [1719].
115
Table 1 Risk factors for venous thromboembolism

Age greater than 40 years (but especially > 65)
Previous thrombotic episode (especially pulmonary embolism)
Documented thrombophilic abnormality (e.g. antithrombin deciency)
Other haematological disorders (polycythaemia & thrombocythaemia)
Pregnancy and puerperium
Malignancy
Congestive heart failure or recent myocardial infarction
Recent surgery (especially lower limb)
Chronic venous insufciency
Oestrogen therapy (e.g. oral contraceptive pill, hormone replacement therapy)
Obesity
Prolonged recent immobility (e.g. after recent stroke)
Dehydration (diarrhoea)
A number of other risk factors that predispose to venous thromboembolism are now also recognized, primarily through clinical experience in the
setting of surgery [20]. These are listed in Table 1.
The effect of age was highlighted in a recent study from Australia, which
concluded that the annual risk of venous thromboembolism is increased by
12% if one long-haul ight is undertaken annually [21]. Although the incidence of thromboembolism was less than 1/100 000 arriving passengers
under the age of 40 it rose steadily to exceed 14/100 000 in those aged 75
or over. A haematological abnormality may exist in an individual which predisposes to the development of venous thromboembolism. Such disorders
include the relatively rare congenital (inherited) deciencies of natural anticoagulants, such as antithrombin, protein C or protein S. Routine screening
of passengers for these abnormalities is not justied or cost effective but
may be of value in selected individuals who have had an episode of venous
thromboembolism, or where there is a strong family history. A recent study
demonstrated that an inherited thrombophilic defect or use of an oral contraceptive pill increased the risk of thrombosis associated with air travel 16or 14-fold, respectively [22].
The value of screening passengers for thrombophilic defects prior to longhaul ights has been raised. Such screening is not, of course, routinely offered
in other circumstances associated with an increased risk of thrombosis (e.g.
prior to starting on an oral contraceptive pill, pregnancy, prior to orthopaedic
surgery) and no case has yet been established for air travel to be treated differently from current practice for thrombophilia screening in other elds.
In one small uncontrolled, retrospective study of patients with ight-related
DVT, 6/20 (30%) subjects had a thrombophilic defect (factor V Leiden in
ve) [23]. Four subjects had a history of a previous episode of thrombosis, and other potential risk factors were identied in ten subjects (including
116
P.L.F. Giangrande
malignancy, leg in plaster, use of pill or HRT). Five of the 20 patients had
a negative thrombophilia screen and no other identiable risk factor.
Aircraft typically y at altitudes of between 35 00040 000 feet in order
to avoid turbulence and reduce drag, thus beneting fuel consumption. The
cabin air is derived from the outside atmospheric air, which is drawn in and
compressed. The maximum pressure in the cabin at cruising altitude is inuenced by the allowable differential pressure across the wall of the cabin. This
varies with aircraft design but the lowest pressure permitted by the regulatory
authorities for civil aircraft is equivalent to an altitude of 8000 feet. Although
the percentage of oxygen in the cabin remains unchanged at around 21%, the
partial pressure of oxygen is reduced to around 74% of the sea level value.
The very cold air at this altitude (typically around minus 50 C) contains only
negligible water vapour and the humidity in the cabin is thus typically very
low.
It has also recently been suggested that exposure to mild hypobaric hypoxia in pressurized aircraft may also result in activation of the coagulation
cascade, but the data are conicting. Markers of activation of coagulation
were transiently elevated in an uncontrolled study of 20 healthy male volunteers who were exposed to a hypobaric environment designed to simulate
the conditions of an aeroplane cabin [24]. The plasma levels of prothrombin
fragments 1 and 2, thrombinantithrombin complex and activated coagulation factor VII increased signicantly although the D-dimer level remained
unchanged. Treatment with heparin was able to inhibit the development of
this apparent activation of the coagulation cascade [25]. Another study of
eight subjects who ascended rapidly to altitudes of 3940 and 5060 m (with
an estimated ambient air pressure of 51 KPa) by helicopter in Nepal documented increases in the levels of prothrombin fragments 1 and 2 and PAI-1
(plasminogen activator inhibitor-1, a key inhibitor of brinolysis) [26]. There
was no observed change in the plasma levels of thrombinantithrombin complex, D-dimers, factor VII and von Willebrand factor antigen. The authors of
this study postulated that the short-term increase in the level of PAI-1, reversed upon return to a lower altitude, was due to the release of the molecule
from vascular endothelial cells due to hypoxia [27]. By contrast, another small
study involving exposure of eight volunteers to 8 h of isocapnic hypoxia in
a purpose-built chamber showed no evidence of activation of the coagulation
cascade, with no signicant changes in the plasma levels of activated coagulation factor VII, prothrombin fragments 1, thrombinantithrombin complexes
or D-dimer levels [28]. This well-designed study included a control arm in
which volunteers were also exposed to air. Furthermore, end-tidal PO2 was
held at 50 Torr (as opposed to an estimated 60 Torr in the Bendz et al. [24]
study) during the hypoxic exposure and the end-tidal PCO2 was tightly regulated to permit examination of the effects of pure hypoxia without any
concomitant respiratory alkalosis. An important conclusion of this study was
that whilst no signicant changes were observed in this small study, the re-
117
sults do not exclude the possibility that activation of the coagulation cascade
might occur in susceptible individuals (e.g. those with thrombophilia). However, blood was sampled only at the end of the 8-h hypoxic period, and the
possibility that transient hypoxia occurs after only 12 h was not excluded
by this study [29]. Further work is clearly needed in this area in order to
determine whether the mildly hypoxic environment of the cabin of civilian
airliners inuences activation of the coagulation cascade.
Contrary to the widespread belief that passengers on long-haul ights may
develop dehydration through increased insensible loss of water across the
skin and mucous surfaces, it has been calculated that the maximum possible
increase in insensible loss of water over an 8-h period in such conditions is
only around 100 mL [30]. Although systemic dehydration is not a signicant
factor in healthy individual, the low humidity can certainly lead to dryness of
the mucous membranes and a sensation of thirst. An enquiry into various aspects of air travel and health in the UK concluded that: We are satised that
the low cabin humidity is not intrinsically harmful. Any uncomfortable dryness of the skin, mouth, nose and throat can be alleviated simply by a sip of
water or other local application of moisture and is not a threat to health. On
a long ight, assuming normal uid intake, one glass of water can more than
offset any additional loss due to cabin humidity. The common advice to drink
a little more water than usual is thus sound [31]. However, excessive consumption of alcohol or gastrointestinal infections associated with vomiting
and diarrhoea may also contribute to the development of signicant dehydration.
5
Prevention of Venous Thromboembolism Associated with Flight
A number of general measures may be taken to minimize the risk of thrombosis associated with long ights [32]. Perhaps the most important step is to
consider at the outset whether the patient is actually t to y in the rst place.
For example, it is probably wise to defer long-haul travel after recent major
orthopaedic surgery. Passengers should be encouraged to carry out leg exercises from time to time whilst seated (e.g. exion, extension and rotation
of the ankles will help to promote circulation in the lower limbs). However,
many airlines discourage unnecessary walking about in the cabin as there is
always the possibility of encountering unexpected clear-air turbulence. Hand
luggage stowed under seats will also restrict movement. Luggage should be
stowed in racks above the seats and not below the seat where it could limit
movement even more. Passengers should take advantage of refueling stops on
long-haul ights to get off the plane and walk around for a while. Adequate
hydration should be ensured during the ight. It is not necessary to abstain
from alcohol, but excessive consumption should be avoided as this will both
118
P.L.F. Giangrande
promote diuresis and discourage mobility. Similarly, the use of sedatives is

best avoided. Although oestrogen-containing oral contraceptive pills as well
as HRT in older women are recognized risk factors for venous thrombosis,
I would not advocate interrupting such hormonal medication for the period
of travel.
A number of prospective studies have shown a clear benet from the use
of compression hosiery (ight socks). In the rst relevant study, 231 passengers were recruited prior to long-haul ights and randomized into two
groups. 12/116 (10%) who did not wear compression hosiery were diagnosed
after the ight as having asymptomatic calf DVT with duplex ultrasonography, but none of the 115 who wore compression hosiery were affected [12].
In the LONFLIT-4 study of 372 passengers considered to be at medium to
high risk of thromboembolism, 0/179 subjects wearing compression hosiery
developed DVT, but 6/179 (3.35%) controls developed asymptomatic DVT
(four DVT, two supercial) (p < 0.002) [33]. In the subsequent LONFLIT-5
study of 224 high-risk passengers who went on an even longer ight, DVT
was observed in 6/102 (5.8%) control subject and only 1/103 (0.97%) subjects
wearing compression hosiery (p < 0.0025) [34]. Quite apart from reducing the
risk of thrombosis, compression hosiery help to prevent oedema of the legs
and feet which can itself cause discomfort after a long ight. Flight socks
have the advantage of being readily available without prescription and are
washable and thus reusable. They apply graduated pressure to the leg which
is maximal at the ankle, thus encouraging venous return. It is important to
note that the usual full-length stockings used in hospital for prophylaxis of
thromboembolism in patients undergoing surgery are not suitable for use
in ight as they provide a lower pressure at the ankle (UK Class I standard: 1417 mm Hg) as they are designed for recumbent patients. It is also
important that the patient is provided with the correct type and size of compression stocking: unfortunately, there is no internationally agreed standard
with regard to the degree of compression. The stockings also need to be worn
correctly taking care to ensure that there is no constriction in the popliteal
area (behind the knee). Stockings are contraindicated in cases of peripheral
vascular disease as the additional compression could provoke ischaemia. The
wearing of stockings may precipitate supercial thrombophlebitis in subjects
with varicose veins [12].
Aspirin has been advocated by some in the general prophylaxis of thrombosis associated with travel. The benecial effect is weak in absolute terms
and it has been estimated that if the rate of travel-related DVT is 20 per
100 000 travellers, then 17 000 people would need to be treated with aspirin
in order to prevent just one episode of DVT [35]. Furthermore, there is a potential for side effects such as allergic reaction or gastrointestinal bleeding:
13% of subjects taking aspirin in a study to evaluate its potential in preventing
venous thrombosis associated with air travel reported mild gastrointestinal
symptoms [36]. The use of heparin may be considered in the relatively few
119
passengers considered to be at particularly high risk of thrombosis (e.g. history of more than one thrombotic episode and an identied thrombophilic
abnormality), although many such subjects are already likely to be on longterm oral anticoagulation anyway.
The Select Committee on Science and Technology of the House of Lords in
the UK has conducted an enquiry into air travel and health [31]. With regard
to thrombosis, the Committee recognized that the airline industry is a highly
competitive market and governments are therefore understandably reluctant
to impose new regulations on a national carrier. Suggestions proposed by
the committee included the development of standardized denitions for seat
sizes, with the option of pre-booking specic seats. It must be acknowledged
that airlines have recently taken positive steps to address the issue of air travel
and thrombosis. For example, attention is now being drawn to the problem
through the use of leaets and articles in in-ight magazines or videos which
promote leg exercises whilst seated. Some airlines are now providing more
leg-room in economy class. As the Committee points out in its report, it is
quite possible that any such positive changes introduced by individual airlines will actually prove to be a marketing advantage. At the same time, the
travelling public needs to be more aware of the issues and assume some responsibility for ensuring tness to y and the choice (and therefore cost) of
their seats. As the Committee pointed out: We were struck by the fact that
some of the more sensational journalism about DVT and seating appeared
in the travel section of newspapers alongside pages of advertisements for the
cheapest ights in what must be, almost by denition, minimum standard
seating [31].
6
Conclusions
In summary, it is now generally accepted that there is an association between long-distance air travel (as well as other forms of long-distance travel)
and venous thromboembolism. The risk is largely conned to those with recognized additional risk factors for venous thromboembolism. Leg exercises
whilst seated help to reduce the risk of DVT. There is also clear evidence from
prospective and randomized clinical trials to support the use of compression
hosiery as a preventative measure. By contrast, there is no rm evidence to
support the indiscriminate use of aspirin as a routine prophylactic measure.
120
P.L.F. Giangrande
Glossary
D-Dimer: A polypeptide derived from the degradation of brin in a blood clot (thrombus). The detection of signicant quantities in the blood implies the presence of
a thrombus somewhere in the blood vessels.
Embolism: Obstruction of a blood vessel, usually in the lung (pulmonary embolism),
due to detachment of a fragment of thrombus.
Haemophilia: Congenital bleeding disorder due to deciency of coagulation factor VIII.
Hypoxia: Low oxygen tension (content) in the blood.
Ischaemia: Lack of blood ow in a limb or organ, due to partial or total obstruction.
Orthopaedic surgery: Surgery of the bones and joints, e.g. total knee or hip replacement.
Puerperium: The rst month after the birth of a child.
Prophylaxis (adj. prophylactic): Measure taken to prevent a condition developing, e.g.
the use of compression hosiery or use of an anticoagulant to prevent thrombosis.
Thrombophilia: Genetic predisposition towards thrombosis, usually due to a hereditary
deciency of natural circulating anticoagulants in the blood, such as antithrombin.
Thrombophlebitis: Inammation of the veins, usually the ones on the surface of the leg,
resulting in pain and redness of the limb.
Thrombosis: Coagulation within a blood vessel to form a solid clot, impeding blood
ow.
Thrombus: Solid clot within the blood vessel.
Warfarin: Drug used as an anticoagulant to treat established cases of thrombosis.
References
1. Homans J (1954) Thrombosis of the deep leg veins due to prolonged sitting. New Engl
J Med 250:148149
2. World Health Organisation (WHO) consultation on air travel and thromboembolism:
Geneva 1213th March, 2001: http://www.who.int/ncd/cvd/dvt.htm
3. Ferrari E, Chevallier T, Chapelier A, Baudouy M (1999) Travel as a risk factor for venous thromboembolic disease: a case-control study. Chest 115:440444
4. Giangrande PLF (2002) Air travel and thrombosis. Br J Haematol 117:509512
5. Pfausler B, Vollert H, Bosch S, Schmutzhard E (1996) Cerebral venous thrombosis
a new diagnosis in travel medicine. J Travel Med 3:165167
6. Collins REC, Field S, Castleden WM (1979) Thrombosis of leg arteries after prolonged
travel. BMJ iv:147148
7. Teenen RP, MacKay AJ (1992) Peripheral arterial thrombosis related to commercial
airline ights: another manifestation of the economy class syndrome. Br J Clin Pract
46:165166
8. Ashkan K, Nassim A, Dennis MJS, Sayers RD (1998) Acute arterial thrombosis after
a long-haul ight. J R Soc Med 91:324
9. Stewart AJ, Manson LM, Dennis R, Allan PL, Ludlam CA (2000) Thrombosis in a duplicated supercial femoral vein in a patient with haemophilia A. Haemophilia
6:4749
10. Lapostolle F, Surget V, Borron SW, Desmaizieres M, Sordelet D, Lapandry C, Cupa M,
Adnet F (2001) Severe pulmonary embolism associated with air travel. New Eng J Med
345:779783
121
11. Hughes RJ, Hopkins J, Hill S, Weatherall M, Van de Water N, Nowitz M, Milne D,
Ayling J, Wilsher M, Beasley R (2003) Frequency of venous thromboembolism in low
to moderate risk long distance air travellers: the New Zealand air travellers thrombosis (NZATT) study. Lancet 362:20392044
12. Scurr JH, Machin SJ, Bailey-King S, Mackie IJ, McDonald S, Smith PD (2001) Frequency and prevention of symptomless deep-vein thrombosis in long-haul ights:
a randomized trial. Lancet 357:14851489
13. Rosendaal FR (1999) Venous thrombosis: a multicausal disease. Lancet 353:11671173
14. Simpson K (1940) Shelter deaths from pulmonary embolism. Lancet ii:744
15. Cruickshank JM, Gorlin R, Jennett B (1988) Air travel and thrombotic episodes: the
economy class syndrome. Lancet ii:497498
16. Sahiar F, Mohler SR (1994) Economy class syndrome. Aviat Space Environ Med
65:957960
17. Zornberg GL, Jick H (2000) Antipsychotic drug use and risk of rst-time idiopathic
venous thromboembolism: a case-control study. Lancet 356:12191223
18. Thomassen R, Vandenbroucke JP, Rosendaal FR (2001) Antipsychotic drugs and venous thrombosis. Br J Psychiatry 179:6366
19. Hagg S, Spigset O (2002) Antipsychotic-induced venous thromboembolism: a review
of the evidence. CNS Drugs 16:765776
20. Arfvidsson B, Eklof B, Kistner RL, Masuda EM, Sato DT (2000) Risk factors for venous
thromboembolism following prolonged air travel. Hematol Oncol Clin N Am 14:391
400
21. Kelman CW, Kortt MA, Becker NG, Li Z, Mathews JD, Guest CS, Holman CDJ (2003)
Deep vein thrombosis and air travel: record linkage study. BMJ 327:10721075
22. Martinelli I, Taioli E, Battaglioli T, Podda GM, Passamonti SM, Pedotti P, Mannucci PM (2003) Risk of venous thromboembolism after air travel: interaction with
thrombophilia and oral contraceptives. Arch Intern Med 163:26742676
23. Rege KP, Bevan DH, Chitolie A, Shannon MS (1999) Risk factors and thrombosis after
airline ight. Thromb Haemost 81:995996
24. Bendz B, Rostrup M, Sevre K, Andersen TO, Sandset PM (2000) Association between
hypobaric hypoxia and activation of coagulation in human beings. Lancet 356:1657
1658
25. Bendz B, Sevre K, Andersen TO, Sandset M (2001) Low molecular weight heparin
prevents activation of coagulation in a hypobaric environment. Blood Coagul Fibrinolysis 12:371374
26. Mannucci PM, Gringeri A, Peyvandi F, Di Paolantonio T, Mariani G (2002) Short-term
exposure to high altitude causes coagulation activation and inhibits brinolysis.
Thromb Haemost 87:342343
27. Gertler JP, Perry L, LItalien G, Chung-Welch N, Cambria RP, Orkin R, Abbott WM
(1993) Ambient oxygen tension modulates endothelial brinolysis. J Vasc Surg
18:939945
28. Crosby A, Talbot NP, Harrison P, Keeling D, Robbins PA (2003). Relation between
acute hypoxia and activation of coagulation in human beings. Lancet 361:22072208
29. Bendz B, Sandset PM (2003) Acute hypoxia and activation of coagulation [letter].
Lancet 362:997998
30. Nicholson AN (1998) Dehydration and long haul ights. Travel Med Intl 16:177181
31. Select Committee on Science and Technology (House of Lords, UK): Air travel and
health. HMSO, London (2000), section on humidity p 14, comment on passenger
choice of seats p 52
32. UK Department of Health (2001): www.doh.gov.uk/dvt/
122
P.L.F. Giangrande
33. Belcaro G, Cesarone MR, Shah SS, Nicolaides AN, Geroulakos G, Ippolito E, Winford M, Lennox A, Pellegrini L, Brandolini R, Myers KA, Simeone E, Bavera P, Dugall M,
Di Renzo A, Moia M (2002) Prevention of edema, ight microangiopathy and venous
thrombosis in long ights with elastic stockings. A randomized trial: The LONFLIT-4
Concorde Edema-SSL Study. Angiology 53:63545
34. Belcaro G, Cesarone MR, Nicolaides AN, Ricci A, Geroulakos G, Shah SS, Ippolito E,
Myers KA, Bavera P, Dugall M, Moia M, Di Renzo A, Errichi BM, Brandolini R,
Dugall M, Grifn M, Rufni I, Ricci A, Acerbi G (2003) Prevention of venous thrombosis with elastic stockings during long-haul ights: the LONFLIT-5 JAP study. Clin
Appl Thromb Hemost 9:197201
35. Loke YK, Derry S (2002) Air travel and venous thrombosis: how much help might
aspirin be? Medscape Gen Med 4:4
36. Cesarone MR, Belcaro G, Nicoliades AN, Incandela L De S, Geroulakos G, Lennox A,
Myers KA, Ippolito E, Winford M (2002) Venous thrombosis from airtravel: the
LONFLIT-3 study-prevention with aspirin vs. low molecular weight heparin in high
risk subjects: a randomised trial. Angiology 53:16

DOI 10.1007/b107240
Disruptive Behaviour (Air Rage): Possible Aetiology

J. T. B. Moyle H. Muir (u)
Human Factors Group, School of Engineering, Craneld University, Craneld UK
h.muir@craneld.ac.uk
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
123
Causes of Aggressive Behaviour . . . . . . . . . . . . . . . . . . . . . . . .
124
3
3.1
3.2
3.3
3.4
3.5
Environmental Triggers of Aggressive Behaviour

Barometric Pressure . . . . . . . . . . . . . . . . .
Carbon Dioxide . . . . . . . . . . . . . . . . . . .
Relative Humidity . . . . . . . . . . . . . . . . . .
Temperature . . . . . . . . . . . . . . . . . . . . .
Noise . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
127
127
133
134
135
135
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
135
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
135
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Abstract The incidence of disruptive behaviour by passengers in civil aircraft is unknown

due to under-reporting. The theoretical pathophysiology is discussed. Underlying aggressive personality and life stress predispose to disruptive behaviour. The use and misuse
of certain drugs including alcohol further sensitise susceptible individuals who are then
triggered by conditions induced by the aircraft cabin environment, especially mild hypoxia and nicotine withdrawal.
Keywords Disruptive Aggression Altitude Aircraft Stress Hypoxia Nicotine
Abbreviations
ARDS Adult respiratory distress syndrome
CCF Congestive cardiac failure
COPD Chronic obstructive pulmonary disease
JAR
Joint Aviation Regulations
ODC Oxygen dissociation curve
SaO2 Arterial blood sample
SpO2 Arterial oxyhaemoglobin saturation
1
Introduction
Many articles in the popular press and even a complete book [1] have
been published about disruptive passenger behaviour in aircraft. However the
J. T. B. Moyle H. Muir
124
vast majority of this has been about behavioural and legal aspects and about
the management and punishment of offenders. There has been little research
about possible aetiology from a scientic basis. It is quite possible that environmental factors may provide the basic cause or may trigger such behaviour.
2
Causes of Aggressive Behaviour
Aggressive behaviour usually stems from the synergism of a combination of
factors, namely a prior disposition caused by an aggressive personality plus
added stress and then a triggering event or events. This combination may be
amplied by the addition of certain drugs or chemicals or, conversely, even as
a result of their sudden withdrawal (Table 1) [2].
The underlying personality ranges from a very calm, stable, passive, altruistic type to the other extreme of being highly assertive and aggressive.
Examples would range from uncomplaining long-term care-givers to the hireand-re manager who has little regard for the individual and has a short
fuse. The type of personality more likely to become a disruptive passenger
is of the aggressive/assertive type.
Table 1 Contributors to Aggressive Behaviour
Table 2 Contributors to Stress

Before journey
Life stresses
Reason for ight
Business
Pleasure
Competition
Sickness
Bereavement
Fear of ying
Safety concerns
Journey to Airport
At Airport
During ight
Parting/Farewells
Driving stresses
Lack of sleep
Trafc delays
Difculty parking
Fear of ying
Queuing
Crowding
Noise
Delays
Poor information
Fear of ying
Fear of terrorism
Crowding
Lack of knowledge
Lack of information
Noise
Enforced company
Fear of ying
Fear of terrorism
Physiological
stresses
125
The next component to be considered is stress. This may be categorised in

reference to air travel e.g., as shown in Table 2.
There have been many attempts to quantify life stress [3]. Holmes and
Rahe developed a scale quoted in Trauers book on stress, which is shown in
Table 3 Life Events Stress Table
Death of spouse
Divorce
Marital separation
Jail term
Death of close family member
Personal injury/illness
Marriage
Fired from work
Marital reconciliation
Illness family member
Pregnancy
Sex difculties
New family member
Business readjustment
Change in nancial state
Death of a close friend
Change to different line of work
Change in number of arguments with spouse
Taking on a large mortgage
Foreclosure on mortgage/loan
Change in responsibilities at work
Son/daughter leaving home
Trouble with in-laws
Outstanding personal achievement
Spouse begins/stops work
Children begin/end school
Change in living conditions
Revision of personal habits
Trouble with boss
Change in work hours/conditions
Change in residence
Change in schools
Change in recreation patterns
Change in social activities
Taking-on small mortage/loan
Change in sleeping habits
Change in number of family get-togethers
Change in eating habits
Holiday
Christmas
Minor violations of the law
100
73
65
63
63
53
50
47
45
44
40
39
39
39
38
37
36
35
31
30
29
29
29
28
26
26
25
24
23
20
20
20
19
18
17
16
15
15
13
12
11
126
Table 3 [4]. They based their table upon the abilities of various life events to
effect change in our lives. A score is assigned to a number of common lifechange events; the higher the score over a given period of time, the higher the
stress level is likely to be in an individual.
Increases in the score from life stresses lead to many physiological and
psychological symptoms and signs (Table 4).
Of relevance to so-called Air Rage are anxiety attacks and anger which may
turn to violence. As this anger may be unusual to a particular individual, it
may then be exacerbated by guilt [5].
A major component of the risk of inducing aggressive behaviour is the use,
either legitimately or otherwise, of drugs and chemicals. Tables 5 and 6 list
common drugs and chemicals which could make the likelihood of aggressive
behaviour more likely or less likely. The reason for noting those which may
reduce aggression is that sudden withdrawal of one of these agents (such as
nicotine) may also increase the risk of sudden aggressive behaviour.
Table 4 Symptoms and Signs of Stress

Breathlessness; palpitations; nausea and vomiting; dizziness; need for alcohol;
excessive smoking; anorexia or craving for food; insomnia; nightmares; lethargy;
indigestion; headaches; anxiety attacks; nail-biting; chronic indigestion;
nger/foot tapping; neck ache; back ache; ulcers; becoming accident-prone;
exacerbation of asthma, eczema or psoriasis; addiction to medication or drugs;
impotence; high blood pressure; anger/violence; phobias; heart disease;
strokes; increased risk of malignant disease
Table 5 Agents That May Induce Aggressive Behaviour
Alcohol
Amphetamines
Anabolic steroids
Aspartame
Caffeine
Cocaine
Marijuana
Phencyclidine
Solvents (including gasoline fumes)
Substances that raise norepinephrine (noradrenaline) in the
Central Nervous System (CNS)
e.g., many antidepressants
Substances that reduce serotonin (5-hydroxytryptamine) in the CNS
e.g., amphetamines
Any substance that raises acetylcholine concentration in the CNS
e.g., anticholinesterase insecticides
(Benzodiazepines and Fluoxetine may induce or reduce aggressive behaviour)
127
Table 6 Drugs That May Reduce Aggression

Major tranquillisers
Nicotine
Anticonvulsants
Marijuana
Benzodiazepines
Anabolic steroids
Lithium
Medroxyprogesterone
Although normally sedating, these drugs may paradoxically induce an excited state in
34% [6].
The so-called major tranquillizers normally used to control schizophrenia and psychosis are also known to be anti-psychotics or neuroleptics.
They fall into three main groups: phenothiazines (including chlorpromazine,
prochlorperazinene), butyrophenones (such as haloperidol, droperidol) or
thioxenthines. They all have a quieting, though not necessarily sedating,
effect on the brain.
Of the agents listed, the most common problems which are actually induced by commercial aviation are the administration of alcohol [7] and caffeine and the withdrawal of nicotine.
3
Environmental Triggers of Aggressive Behaviour
Apart from the psychological effects of enforced reduction in personal space
and possibly having to sit in close proximity to some undesirable character,
there are certain environmental changes which are not obvious to passengers and crew but that are known to lead to aggression. Some of the more
important of these are:
Barometric pressure
Carbon dioxide
Humidity
Temperature
Noise
3.1
Barometric Pressure
Despite all modern commercial airliners being pressurised, it is not commonly known that they are pressurised to a cabin altitude of around 8000 feet
128
above sea level. Although 8000 feet is a relatively low elevation, it is certainly
signicant to persons with respiratory disease, for reasons that will become
apparent later. Also, the rate at which change in cabin altitude occurs makes
an enormous difference to the physiological effect. Although mankind may
live in a healthy state at 20 000 feet with no problems if acclimatisation has
occurred that is if altitude is attained slowly, allowing the physiology to
change, especially an increase in the oxygen carrying capacity (in the form of
increased haemoglobin) of the blood and an increase in cardiac output this
is a completely different state of affairs from the rapid ascent in a few minutes
that commonly occurs in aircraft.
Most of our understanding of human physiology in commercial aircraft is
based upon research carried out in the early 1940s on t young air force pilots. Little work has been done since on civilian physiology and certainly none
on the effects of cabin altitude on unhealthy individuals.
The principles behind barometric pressure effects on the human body are
based upon Daltons Law of Partial Pressure and the Oxyhaemoglobin Dissociation Curve.
Daltons Law of Partial Pressure states that, in a mixture of gases, each gas
exerts the pressure that it would exert if it were in that volume alone. This
pressure is known as the partial pressure and the sum of the partial pressures
equals the total pressure of the mixture. Thus at sea level the partial pressure
of oxygen in dry air is 21.2 kPa (159 mm Hg). As altitude increases barometric
pressure falls and so the partial pressure of oxygen decreases too (Table 7).
The Oxyhaemoglobin Dissociation Curve, ODC, is the relationship between
the partial pressure of oxygen in contact with haemoglobin in the erythrocytes (red blood cells) and the percentage saturation of the erythrocytes with
oxygen. The ODC has a sigmoidal shape as shown in Fig. 1.
Table 7 Dependence of Barometric Pressure and Partial Pressure of Oxygen on Altitude
Altitude
Barometric Pressure
Partial Pressure of Oxygen
30 000 ft
20 000
18 000
16 000
14 000
12 000
10 000
8000
6000
4000
2000
Sea level
33 kPa
47
50
55
60
64
70
75
81
88
94
101
5.6 kPa
8.4
9.2
10
11
12
13
14
16
17
18
21
247 mm Hg
349
379
412
446
483
523
564
609
659
707
760
42 mm Hg
63
69
76
83
91
100
108
118
127
138
159
129
Fig. 1 Oxygen Dissociation Curve
It must always be remembered that the partial pressure of the inhaled oxygen will be greater than the oxygen actually in contact with the haemoglobin
in the capillaries in the lung, as oxygen transport is entirely by partial pres-
Table 8 Conditions Which May Cause a Shift of the ODC to the Right, Thus Worsening
Hypoxia
Increase in:
2,3-diphospoglycerate in red blood cells
arterial pH
body temperature
increase in arterial PCO2
Even minor degrees of congestive cardiac failure (CCF)
Chronic obstructive pulmonary disease (COPD)
Pregnancy
Cirrhosis of the liver
-blockers (used in the treatment of high blood pressure and CCF)
Sickle cell disease
Adult respiratory distress syndrome (ARDS)
130
sure gradient from the air, through the respiratory tract, the alveolar gas, the
haemoglobin in the lung capillaries, the arterial systemic blood to the tissues, the cells and eventually to the mitochondria in every cell in the body,
Fig. 2 [8].
Thus one cannot simplistically relate altitude to the oxyhaemoglobin saturation of arterial blood, especially as extremes of age and deteriorating health
lead to a greater difference between the partial pressure oxygen in inhaled air
and that which is in contact with the haemoglobin in the blood in the lung
Fig. 2 Oxygen Cascade from Atmospheric to Cytoplasmic PO2 . Original gure available
from Nunns Applied Respiratory Physiology [8]
131
capillaries. The maximum cabin altitude for commercial aircraft was chosen
as 8000 feet altitude equivalent, which corresponds theoretically to the highest altitude before there is a sudden decrease in saturation as the shape of the
ODC changes (Fig. 3).
There may also be a shift in the ODC to the right due to physiological
changes or pathology (see Table 8).
Either a decrease in the partial pressure of oxygen presented to the blood
in the lungs or a shift in the ODC to the right will induce a type of hypoxia or
a decrease in the level of oxygen carried by the haemoglobin. Hypoxia may be
classied in four different ways, any of which may occur in humans ying in
aircraft, especially in unt passengers (Table 9).
Fig. 3 Erroneous Oxyhaemoglobin Dissociation Curve

Table 9 Classication of Types of Hypoxia
Type
Possible Causes
Hypoxic
Reduction of partial pressure of oxygen in arterial blood,

for example, due to: low PiO2 , hypoventilation,
impairment of gas exchange in the lungs, ventilation/blood
perfusion mismatches in the lungs.
Anaemic
Reduction in oxygen carrying capacity in the blood.

Anaemia of any cause
Any cause of reduction in blood ow through the lungs
AND to the tissues
Inability of the cells to make use of the oxygen when delivered
to them. Classically caused by cyanide poisoning but also by
vitamin B1 deciency.
Ischaemic
Histotoxic
132
Fig. 4 Theoretical SpO2 during Flight in Commercial Aircraft
It must of course be remembered that the air in the cabin is not fresh, as
approximately 50% is re-circulated; this minimises the fuel needed to compress external air. Therefore, the partial pressure of cabin air oxygen is lower
than one would expect for the cabin barometric pressure at a particular cabin
altitude.
Prior to the 1980s, the only method of measuring the oxyhaemoglobin saturation of arterial blood available was to actually take an arterial blood sample (SaO2 ), an unpleasant procedure not without risk, and inject it into a cooximeter. In the early 1980s the non-invasive technique of pulse-oximetry
was developed [9]. Pulse oximetry allows continuous monitoring of arterial
oxyhaemoglobin saturation (SpO2 ), even whilst carrying out other tasks. Fig-
Fig. 5 Actual SpO2 Recording from Pilot of a Commercial Flight. Original gure available
from Cottrell et al. [10]
133
ure 4 shows a theoretical plot of SpO2 from take-off to landing. Note that the
SpO2 starts at the normal level of 98%, decreases to 93% at a cabin altitude
of 8000 feet and returns to 98% again when the aircraft has landed, assuming that the initial and destination airports are at sea level. Figure 5 shows the
actual recorded SpO2 level from the pilot of a routine commercial ight [10].
Cottrell et al. [10] also monitored the in-ight SpO2 of 38 commercial air-crew
who were reasonably t, in ve different aircraft types. The enormous variation in minimum SpO2 during ight is shown in Fig. 6. It can be seen that
there is poor correlation between theoretical and actual minimum SpO2 .
The effect of even mild hypoxia is dependent upon its intensity, the rate of
change and the duration, age and intercurrent illness, tness and whether the
subject abuses their body with tobacco or alcohol.
Initial signs and symptoms of hypoxia include a slowing of cognition, difculty with new tasks, confusion, and aggression, very similar to intoxication
with alcohol. In fact synergism with two agents producing similar effects often produces unexpectedly magnied symptoms.
The permitted rate of change in cabin altitude by regulation is 500
feet/minute in ascent and 300 feet/minute in descent. However, this is not
based upon respiratory physiology but on ear discomfort.
Gupta et al. [11] in 1997 showed that if SpO2 in healthy adult volunteers
was decreased over the course of an hour there was a sudden change in cerebral physiology when the SpO2 decreased to 90%. This is depicted in Fig. 7 as
a sudden increase in cerebral blood ow. It is as though the brain is suddenly
stressed and is demanding more oxygen. This may declare itself as a sudden
change in cognitive function and may well be the most important trigger of
aggressive behaviour in susceptible subjects. The Human Factors Group at
Craneld University has made a number of attempts to gain funding in order
to advance research into this area but to no avail.
Fig. 6 SpO2 in 38 Commercial Air Crew. Original gure available from Cottrell et al. [10]
134
Fig. 7 Sudden Change in Cerebral Blood Flow when SpO2 Reduced Below 90%. Original
gure available from Gupta et al. [11]
3.2
Carbon Dioxide
The permitted level of carbon dioxide has been reduced from 3% to < 0.5%
in the new JARs (Joint Aviation Regulations). Again the partial pressure of
cabin carbon dioxide is higher than normal fresh air because of recirculation
of ventilation air. Mildly raised carbon dioxide levels make passengers feel
uncomfortable and can induce aggression.
3.3
Relative Humidity
The air at increasing altitude contains less and less water vapour due to the
drop in temperature. When an aircraft is at high cruising altitude the air
compressed and pumped into the cabin is virtually dry. This means that the
humidity of the cabin atmosphere falls from 5070% at sea level to around
10% when cruise altitude is attained. A humidity of 5070% is healthy and
comfortable, whereas 10% is uncomfortable [12]. It causes drying of the respiratory tract and dehydration of the skin and general discomfort, and therefore
is likely to contribute to a tendency towards aggression. The general dehydration of the body is made worse by alcohol and caffeine-containing beverages,
as these have a diuretic effect, making the body lose more water than it gains
from the beverage itself.
135
3.4
Temperature
The comfortable temperature for most people in normal clothing is 20 C.
High ambient temperature has been shown to increase aggressive tendencies,
especially in those brought up in a temperate climate. Baron and Ransberger
found that the incidence of rioting increased with ambient temperature [13].
3.5
Noise
High noise levels are known to induce stress and aggression [14]. The mechanism for this effect is that psychological arousal is increased, which may
induce aggression, especially if the level of underlying stress is higher than
normal. However, those who had been informed of this effect and about the
noise before it occurred were found to exhibit less aggression [15].
4
Conclusions
Theoretical causes for disruptive behaviour have been discussed. Regrettably,
the regulatory bodies in the aviation industry have not seen t to fund the
research necessary to further investigate our theoretical ndings. Further research requires the design and construction of an environmental chamber
with different specications from those already available for high altitude research. Existing chambers have small area/volume ratios and are thus able to
be evacuated to simulate very high altitudes. The chamber required for this
research only needs to be evacuated to the equivalent of 10 000 feet but does
need to be of sufcient size to allow simulation of various seating congurations and to be equipped to exactly simulate modern aircraft cabins.
References
1. Dahlberg A (2001) Air Rage: the underestimated safety risk. Ashgate, Aldershot, UK
2. Renfrew JW (1997) Aggression and its causes: A biopsychosocial approach. Oxford
University Press, New York
3. Davies G (1988) Stress The challenge to Christian caring. Kingsway, Eastbourne, UK
4. Trauer T (1986) Coping with stress. Salamander, London
5. Markham U (2003) Managing stress - The stress survival guide for today. Chrysalis
Books, London
6. Salzman C, Kochansky GE, Shader RI, Porrino LJ, Hormatz JS, Sweet CPJ (1974)
Chlordiazepoxide-induced hostility in a small group setting. Arch Gen Psychiatry
31:401405
136
7. Badaway AA-B (1986) Alcohol as a psychopharmacological agent. In: Brain PF (ed)

Alcohol and aggression. Croom Helm, Dover, NH, 55:83
8. Lumb AB (2000) Nunns respiratory physiology, 5th edn. Butterworth-Heinemann,
Edinburgh
9. Moyle JTB (2002) Pulse oximetry. 2nd edn. BMJ, London
10. Cottrell JJ, Lebovitz BL, Fennell RG, Kohn GM (1995) Inight arterial saturation. Continuous monitoring by pulse oximetry. Aviat Space Environ Med 66(2):12630
11. Gupta AK, Menon DK, Czosnyka M, Smielewski P, Jones JG (1997) Thresholds for hypoxic cerebral vasodilatation in volunteers. Anesth Analg 85:817820
12. Hawkins FH (1987) Human factors in ight. Ashgate, Aldershot, UK
13. Baron RA, Ransberger VM (1978) Ambient temperature and the occurrence of collective violence: The long hot summer revisited. J Pers Soc Psychol 36:351360
14. Hutchinson RR, Pierce GE, Emley GS, Proni TJ, Sauer RA (1977) The laboratory measurement of human anger. Behav Sci 1:241259
15. Harris M, Huang L (1974) Aggression and the attribution process. J Soc Psychol
92:209216

DOI 10.1007/b107241
Airplanes and Infectious Disease

Harriet A. Burge
Environmental Microbiology Laboratory, 1150 Bayhill Drive, Suite 100,
San Bruno, CA 94066, USA
hburge@emlab.com
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
138
2
2.1
2.2
2.3
2.3.1
2.3.2
2.3.3
2.3.4
2.4
2.5
Contagious Disease . . . . . . .
Presence of Infected Individuals
Stage of the Disease . . . . . . .
Pathways for Agent Transfer . .
Aerosols . . . . . . . . . . . . .
Droplets and Fomites . . . . . .
Direct Contact . . . . . . . . . .
Role of Ventilation . . . . . . . .
Role of Host Sensitivity . . . . .
Time Spent in the Environment
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
139
140
140
140
141
141
141
142
142
142
3
3.1
3.2
3.3
3.4
3.5
3.6
3.7
Reports of Outbreaks . . . .
Upper Respiratory Infections
Inuenza . . . . . . . . . . .
Tuberculosis . . . . . . . . .
SARS . . . . . . . . . . . . .
Measles, Chicken Pox . . . .
Bacterial Meningitis . . . . .
Hemorrhagic Fevers . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
143
143
144
144
144
145
145
145
4
4.1
4.2
4.2.1
4.2.2
Research Designed to Address Concerns

Microbial Investigations . . . . . . . . . .
Models and Risks . . . . . . . . . . . . .
General Infectious Disease Models . . . .
Models Applied to Aircraft . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
145
146
146
146
146
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
146
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
147
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Abstract Air travel is associated with crowded conditions that can facilitate the transmission of airborne infectious diseases. The risk of contracting such diseases depends on the
presence of an infected person who is shedding infectious particles and sufcient exposure of a sensitive person to achieve an adequate dose to cause disease. Proximity to the
infectious person and the length of time spent near the person are the most important
risks for contracting a disease. Ventilation patterns play a lesser role in disease transmission. Well-documented outbreaks of inuenza, severe acute respiratory syndrome (SARS),
and tuberculosis have occurred. Other common respiratory illnesses have probably also
138
H.A. Burge
been spread via aircraft, but outbreaks remain unrecognized. Research on the spread of
infectious disease in aircraft has focused on sampling for microorganisms in air (which
has little relevance), and on the development of models to predict the risks for specic
diseases.
Keywords Aircraft Infection Contagious disease Airborne Tuberculosis Inuenza
SARS
Abbreviations
TB
Tuberculosis
SARS Severe acute respiratory syndrome
1
Introduction
Commercial air transport has provided access to the world for those able to
undertake such travel. In fact, 1.7 million passengers travel each year, representing 600 passenger miles of travel [1]. There are risks associated with all
forms of travel and many of them are similar across transport types [2]. Thus,
stress, accidents, and exposure to disease agents are associated with all forms
of transportation. However, because of the distances that can be traversed
in a relatively short time, the crowded conditions on most commercial aircraft, and the inability to escape, concern regarding the risk of contracting
infectious disease during air travel has become signicant [3, 4].
It is important to remember that there are other infection-associated
concerns associated with air travel. Historically, transport of disease vectors has been of particular concern. Many studies could be cited regarding
transport of mosquitoes that subsequently led to outbreaks of malaria [5].
Control of these problems has focused on disinsection of aircraft during
ight [6].
Food-borne outbreaks of disease have also occurred on aircraft [79].
These are uncommon, and control depends on appropriate food handling
rather than any change in the aircraft environment. Aircraft sewage also contains human pathogens that could be transported over long distances, but the
risk of transmission from this source is probably low [10].
The movement of infectious agents from one part of the world to another via aircraft is another important concern [11]. People traveling in
an infectious state may contribute to the spread of epidemics. This phenomenon is of concern for the spread of weaponized organisms that cause
contagious disease. On the other hand, modeling studies have indicated that
halting air travel in order to interrupt the transmission pathway even for
very serious diseases such as smallpox is likely to cause greater disruption than the disease itself [12]. The possibility that long-distance ights
139
could temporarily damage the immune system has also been raised [13].
Such damage could lead to recurrent infections acquired on the ground after
a ight. Further research will be needed to document the validity of these
concerns.
The primary concern today appears to be person-to-person spread of contagious disease in the aircraft cabin, and that will be the focus of this chapter.
2
Contagious Disease
Contagious diseases are infections that are spread from an infected person to an uninfected sensitive person. Contagious diseases can be spread
by aerosols, by large droplets, by direct contact with materials or surfaces
that have been contaminated by an infectious person (fomites) or by direct person-to-person contact [14]. All of these modes of transmission could
occur in a crowded aircraft cabin.
Table 1 Examples of common contagious diseases and their modes of transmission
Disease
Agent
Transmission
References
Inuenza
Tuberculosis
Chicken pox
Measles
SARS
Common colds
Virus
Bacterium
Virus
Virus
Virus
Viruses
Aerosols
Aerosols
Aerosols
Aerosols
Droplets, aerosols
Droplets, aerosols,
fomites
[15, 16]
[17]
[18]
[19, 20]
[21, 22]
[23, 24]
Aerosol-transmitted diseases are generally caused by agents that are resistant to environmental stresses, and diseases that lead to environmental
release of agent-containing secretions. Thus, respiratory illnesses are most
commonly airborne, although many other diseases could be spread articially
or accidentally via the airborne route. Examples in this latter category include
hepatitis, some gastrointestinal viruses and smallpox.
For a contagious disease to be transmitted from one person to another
there must be an infectious person in the environment (i.e., a source for the
agent). There must be an exposure pathway so that the agent can travel from
the infectious person to another person, and there must be a person who is
sensitive (i.e., not resistant) to the agent. Thus, the risk of infectious disease
transmission is related to the probability that:
1. There is a person infected with the agent in the environment
2. The infected person is actively shedding the disease agent
140
H.A. Burge
3. There is an uninterrupted pathway from the infected person to other people

4. There are one or more sensitive persons in the environment
5. These sensitive people are in the aerosol long enough for sufcient exposure to occur.
2.1
Presence of Infected Individuals
In order for contagious disease transmission to occur, one or more infected
individuals must be present, or have been present recently in the case of
fomite transmission (fomites are inanimate objects that act as reservoirs for
disease agents). The risk of disease transmission is related to the probability that one or more individuals infected with the agent will be present in the
environment. The exception to this is, of course, bioterrorism, for which an
inanimate reservoir may have been prepared.
The probability that an infected person will be in the environment is related to the incidence of the disease in the population likely to be in the
environment. Thus, agents of the common cold are probably in most aircraft.
During the inuenza season, there is a reasonably high probability that one
or more persons harboring the inuenza virus will be on board. On the other
hand, there is only a small probability that a person with tuberculosis will
be on board any individual aircraft. In the USA, tuberculosis is not uncommon, but is present primarily in populations that do not routinely y (e.g.,
prison and homeless populations) [25, 26]. In other parts of the world, TB is
very common, but, again, most of those with the disease do not y. For diseases that are rare in the general population (e.g., SARS, hemorrhagic fevers,
plague), it is extremely unlikely that an infectious individual will be aboard
any individual aircraft.
2.2
Stage of the Disease
Not only must the infected individual be present, but the disease must be in
a stage where agents are being shed into the environment in sufcient quantity that transmission could occur. For some diseases, this stage occurs before
symptoms appear. For most, however, active disease must be present resulting in symptoms that lead to agent release. In the case of TB, for example, the
majority of people with the disease do not shed large quantities of organisms,
further reducing the probability of disease transmission.
141
2.3
Pathways for Agent Transfer
2.3.1
Aerosols
Bacterial and viral aerosol particles are essentially droplet nuclei. The organisms are shed from the infected host in wet droplets containing one or more
agents embedded in wet mucous secretions. Once in the air, the droplets dry
rapidly, shrinking to the size of the contained agents surrounded by dried
secretions. These dried secretions tend to protect the organisms from environmental damage.
Bacteria and viruses in aerosols act as small particles, settling and/or traveling on air currents and in response to electrical charges in the same way as
other small particles [27]. True aerosol-transmitted disease agents can remain
airborne, travel relatively long distances and remain infective [28, 29].
Distribution of the aerosols depends on air movement patterns within the
space. Although theoretically air movement in aircraft is in a circular motion
within rows, actually there is some general transfer of air backward within the
aircraft (see Sect. 2.1). Aerosols generally decay (become less concentrated)
logarithmically with distance both physically and biologically. Needless to
say, the closer one is to the source, the greater the risk of sufcient exposure
to cause illness. For diseases such as measles, where inhalation and deposition of very few virions can cause illness, the risk of contracting the disease
decreases less with distance than, for example, for anthrax, where a signicant number of organisms must reach an appropriate site before illness
is likely.
2.3.2
Droplets and Fomites
Droplet-borne diseases are thought to spread up to one meter from the
source patient, and hospital infection control practices reect this belief. The
droplets may impact directly in the respiratory tract, or on the hands or other
skin surfaces, or even on nearby inanimate surfaces. Touching contaminated
surfaces could transfer organisms to the hosts hands, and infection could
occur with hand transfer to the respiratory tract. A study of rhinovirus colds
transfer by this method, however, indicated that secretions would have to
be transferred while still damp, which in the aircraft environment would be
a very short time [30].
142
H.A. Burge
2.3.3
Direct Contact
Direct contact means just what it says: touching an infected person in some
way. Some diseases can be transmitted with minimal contact (many common
colds). Others require intimate contact of a sort unlikely to occur accidentally.
2.3.4
Role of Ventilation
Ventilation can only interrupt airborne infectious diseases, and then only for
individuals relatively distant from the source. Increasing clean air ventilation
rates has been shown to reduce the incidence of upper respiratory infections
in large ofce buildings [31, 32]. High ventilation rates combined with good
ltration will reduce exposure over time. In aircraft where recirculation is
used, the ltration systems are probably adequate to remove the vast majority
of droplet nuclei. Even with good ltration and very high air exchange rates,
however, ventilation cannot signicantly reduce the risk of exposure for those
very close to the source (e.g., sitting in the same row in an aircraft).
2.4
Role of Host Sensitivity
In order to develop an infectious disease, the host must be sensitive [33]. Infections can only occur in people who do not have either natural or specic
acquired immunity to the agent. There is a range of natural immunity in the
population, with some people being highly resistant while others are highly
sensitive, and it is not a given that any individual without specic immunity
to a disease agent will become infected. The risk of illness following exposure
along this immunity distribution ranges from near zero for the most resistant to very high for the most sensitive. In addition to this natural sensitivity
range, acquired immunodeciency is becoming more and more common in
the population, and some of these individuals are exquisitely sensitive even to
agents of relatively low virulence.
For many of the common contagious childhood diseases, a large majority
of the US population has naturally or articially acquired immunity. For tuberculosis, many countries other than the US immunize for this very common
disease. TB is sufciently uncommon in the US that immunization is relatively
rare.
143
2.5
Time Spent in the Environment
The time spent in an aerosol is another critical factor in calculating risk of
a disease-causing exposure. The longer one is in the presence of an infectious
aerosol, the higher the risk that a potentially infectious dose will be inhaled.
The lower the aerosol concentration (or the number of agents on a surface
or in each respiratory droplet) the longer one must remain in contact with
the aerosol. Clearly, time spent in contact with TB patients is critical in determining the risk of contracting the disease [34]. Although less well studied for
other diseases, this is probably a universal phenomenon.
3
Reports of Outbreaks
Theoretically, any contagious disease that is airborne could cause an outbreak
on an aircraft, and those that are droplet borne could spread to those in adjacent seats. Obviously, diseases that t these categories and are common world
wide are the most likely candidates for such spread. Thus, common colds, inuenza, tuberculosis, and measles have been spread on aircraft, and many
cases have probably occurred that have gone unreported because of follow
up failures. The rare or newly emergent diseases are much less likely to be
present on aircraft. On the other hand, follow up is likely to be intensive,
so that these are over-reported in relation to the more common illnesses.
In addition, there is the problem of separating infections acquired in ight
from those acquired at other points in travel [3]. Thus it is impossible to say
whether or not a specic case of the common cold was contracted while staying away from home, on the airplane going to or from home, or, in fact, at
home before travel began. Given these cautions, the following is a brief discussion of some of the outbreaks that have been documented on commercial
aircraft.
3.1
Upper Respiratory Infections
Zitter et al. evaluated the development of upper respiratory symptoms in passengers on jets with 50% recirculation versus 100% fresh air. There was no
difference between the two populations. It should be noted, however, that
the trips were relatively short (San Francisco to Denver) [35]. Given that
some very common upper respiratory infections are airborne, and others
produce copious droplets that could easily infect adjacent passengers, it is
likely that such disease transmission does occur routinely on aircraft as it
does in any other crowded environment. Given the frequency of these dis-
144
H.A. Burge
eases, it would be nearly impossible to trace an outbreak to the aircraft

environment.
3.2
Influenza
Moser et al. reported an outbreak of inuenza on board a commercial aircraft grounded for three hours with no ventilation. The index case remained
aboard, and the other passengers who remained on board were the most
likely to develop the illness [36]. This is a rare case where the passengers
all disembarked in a relatively small city and many were seen by the same
physician. If the plane had landed in New York or Washington (for example)
the passengers would have dispersed, and the outbreak would have gone unrecognized. Given the infectiousness of the inuenza virus and its airborne
transmission ability, the incidence of outbreaks of this disease linked to commercial airliners has probably been underestimated.
3.3
Tuberculosis
Outbreaks of tuberculosis (TB) have occurred following transport of the organisms through ventilation systems in settings other than aircraft [37]. One
of the best documented cases of TB transmission aboard a commercial airliner involved a crew member with active TB. Time spent with this crew
member was the principal risk factor, and at least two other crew members
acquired positive skin tests. Frequent ying passengers may also have been
affected [38].
Another well-documented case involved a multiply drug-resistant strain of
Mycobacterium tuberculosis, and indicated that the presence of a highly infectious person, a long ight, and close proximity to the infectious person
are the primary risk factors for transmission [39]. A study using retrospective TB testing following transport of an infected patient indicated (but did
not prove) transmission to several other passengers [40]. On the other hand,
a pilot with active TB failed to infect any other of the pilots with whom he
ew [41]. Although this may conrm the low risk for transmission, it may also
represent a case with little shedding of infectious bacteria, or the very high
ventilation rate in the cockpit compared to the passenger cabin.
3.4
SARS
Although generally considered droplet-borne, airborne precautions are also
recommended in the clinical setting, and one building-related outbreak was
determined to be airborne [21, 22]. In fact, it seems clear that airborne trans-
145
mission can occur. The rst 10 patients in the Taiwan SARS epidemic were
closely associated with an infected person on an airplane [42]. Likewise, SARS
may have been introduced to France by a patient who contracted the disease
in Hanoi and infected others on the HanoiParis ight [43]. In an interview
survey, Olsen et al. [44]. documented transmission of SARS to airline passengers seated within the three rows in front of the index case. Other studies have
estimated a low risk of SARS transmission on aircraft, possibly related to the
stage of the illness in the SARS patient, or the number of agent units being
released [45].
3.5
Measles, Chicken Pox
Measles is another disease for which transmission through a ventilation system has been documented [46]. On the other hand, one study revealed no
new measles cases following a seven-hour ight with an index patient [47].
Although measles is a common disease, most people have some level of immunity, and children traveling with active infectious measles are probably
rare. Chicken pox transmission has not been reported related to airliner cabins, probably for the same reason that measles outbreaks are rare in this
environment. For shingles, also caused by the chicken pox virus, the long delay between infection and symptoms makes tracing exposure sources nearly
impossible.
3.6
Bacterial Meningitis
The Centers for Disease Control have reported one case of bacterial meningitis acquired during travel on a commercial airliner [48]. While this disease
agent, Neisseria meningitidis, is the most common cause of bacterial meningitis in the USA, it, nevertheless is a relatively rare disease, and is likely to
present only a small risk for air travelers.
3.7
Hemorrhagic Fevers
The loviruses Ebola and Marburg are able to transmit disease via the airborne route in the laboratory, but to date have not been documented to do
so in the natural environment. Thus, the risks are probably low for the aircraft environment. However, the possibility of their use as biological weapons
remains of potential concern.
146
H.A. Burge
4
Research Designed to Address Concerns
4.1
Microbial Investigations
Several studies have evaluated bacterial levels in commercial airliners during ight. Bacterial levels have either been lower than ground-based interiors
or higher [49, 50]. It is important to remember, however, that the organisms
found in airliner cabin air are rarely (if ever) those likely to cause disease.
The chance of actually collecting a disease agent on any individual ight
is the product of the probabilities of an infectious person being on board,
this person releasing infectious organisms, the samples being collected in
close proximity to this person, and the organisms being identiable using
the sampling and analytical method chosen. These factors apply to other environments as well, making air sampling not the most logical approach for
documenting (or monitoring) airborne infectious disease.
4.2
Models and Risks
4.2.1
General Infectious Disease Models
Rudnick et al. [51] developed equations that could be used to model infection risk on aircraft. Their models depend on knowing the air supply rate and
having it more or less constant, both of which are achievable in aircraft. They
use carbon dioxide measurements, assuming that the CO2 is exhaled from
the occupants. Beggs et al. [52] evaluated several different models and determined that Gammiatoni and Nuccis [53] generalized formulation is most
appropriate for ventilated spaces.
4.2.2
Models Applied to Aircraft
Ko et al. used single and sequential box models to estimate the risk of TB
transmission. These authors concluded that the risk is small except for those
in close contact with the infective patient [34]. Rydock et al. used tracer gas
measurements and came to the same conclusion. This study also documented
the minimal effect of ventilation rate on the potential for disease transmission [54].
147
5
Conclusions
While contagious disease could be acquired during travel on commercial airliners, the risk of contracting anything more serious than the common cold is
low. Excluding persons from aircraft who are obviously suffering upper and
lower respiratory symptoms (sneezing, coughing continuously) would help to
prevent transmission of some diseases. However, the risks of this approach
would have to be balanced with the actual risks of transmission and of the
diseases outcomes. Increasing quality of ltration and/or ventilation rates
is unlikely to signicantly lower the risk of disease transmission in the aircraft environment where each passenger spends such a relatively short time.
The risk for crew members would have to be calculated, but probably remains
quite low.
Further research that applies some of the good infectious disease transmission models to the aircraft environment would elucidate actual risks for
specic diseases [34]. The use of molecular epidemiological techniques in
tracking outbreaks would also contribute [55]. Until further studies can document a signicant risk for infectious disease transmission in the aircraft
environment, cost-increasing measures to reduce already minimal risks seem
unwarranted.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
DeHart RL (2003) Annu Rev Public Health 24:133

Lamar JE 2nd, Malakooti MA (2003) Mil Med 168:523
Al-Jahdali H, Memish ZA, Menzies D (2003) Int J Antimicrob Agents 21:125
Brown TP, Shuker LK, Rushton L, Warren F, Stevens J (2001) J R Soc Health 121:177
Guillet P, Germain MC, Giacomini T, Chandre F, Akogbeto M, Faye O, Kone A,
Manga L, Mouchet J (1998) Trop Med Int Health 3:700
Russell RC, Paton R (1989) Bull World Health Organ 67:543
Hedberg CW, Levine WC, White KE, Carlson RH, Winsor DK, Cameron DN, MacDonald KL, Osterholm MT (1992) Jama 268:3208
Back E, Romanus V, Sjoberg L, Svenungsson B, Bottiger M, Kallings LO (1977) Scand
J Infect Dis 9:175
Eisenberg MS, Gaarslev K, Brown W, Horwitz M, Hill D (1975) Lancet 2:595
Shieh YS, Baric RS, Sobsey MD (1997) Appl Environ Microbiol 63:4401
Grais RF, Ellis JH, Kress A, Glass GE (2004) Health Care Manag Sci 7:127
Grais RF, Ellis JH, Glass GE (2003) Epidemiol Infect 131:849
Rose DM, Jung D, Parera D, Konietzko J (1999) Z Arztl Fortbild Qualitatssich 93:481
Rheinbaben F, Schunemann S, Gross T, Wolff MH (2000) J Hosp Infect 46:61
Bridges CB, Kuehnert MJ, Hall CB (2003) Clin Infect Dis 37:1094
Regan SF, Fowler C (2002) J Gerontol Nurs 28:30
Fennelly KP, Martyny JW, Fulton KE, Orme IM, Cave DM, Heifets LB (2004) Am J
Respir Crit Care Med 169:604
148
H.A. Burge
18. (2004) MMWR Morb Mortal Wkly Rep 53:389

19. Paunio M, Peltola H, Valle M, Davidkin I, Virtanen M, Heinonen OP (1998) Am J Epidemiol 148:1103
20. Ehresmann KR, Hedberg CW, Grimm MB, Norton CA, MacDonald KL, Osterholm MT
(1995) J Infect Dis 171:679
21. Yu IT, Li Y, Wong TW, Tam W, Chan AT, Lee JH, Leung DY, Ho T (2004) N Engl J Med
350:1731
22. Keeler N, Lingappa J (2004) Curr Opin Pediatr 16:61
23. Brundage JF, Scott RM, Lednar WM, Smith DW, Miller RN (1988) Jama 259:2108
24. Goldmann DA (2000) Pediatr Infect Dis J 19:97
25. Baillargeon J, Black SA, Leach CT, Jenson H, Pulvino J, Bradshaw P, Murray O (2004)
Prev Med 38:607
26. White MC, Tulsky JP, Portillo CJ, Menendez E, Cruz E, Goldenson J (2001) Int J Tuberc
Lung Dis 5:400
27. Utrup LJ, Frey AH (2004) Exp Biol Med (Maywood) 229:345
28. Ko G, First MW, Burge HA (2000) Tuber Lung Dis 80:217
29. Ko G, First MW, Burge HA (2002) Environ Health Perspect 110:95
30. Reed SE (1975) J Hyg (Lond) 75:249
31. Menzies D, Adhikari N, Arietta M, Loo V (2003) Infect Control Hosp Epidemiol 24:483
32. Alani A, Barton IE, Seymour MJ, Wrobel LC (2001) Int J Environ Health Res 11:219
33. Mileno MD, Bia FJ (1998) Infect Dis Clin North Am 12:369
34. Ko G, Thompson KM, Nardell EA (2004) Risk Anal 24:379
35. Zitter JN, Mazonson PD, Miller DP, Hulley SB, Balmes JR (2002) Jama 288:483
36. Moser MR, Bender TR, Margolis HS, Noble GR, Kendal AP, Ritter DG (1979) Am J
Epidemiol 110:1
37. Houk VN (1980) Ann N Y Acad Sci 353:10
38. Driver CR, Valway SE, Morgan WM, Onorato IM, Castro KG (1994) Jama 272:1031
39. Kenyon TA, Valway SE, Ihle WW, Onorato IM, Castro KG (1996) N Engl J Med 334:933
40. Wang PD (2000) Am J Infect Control 28:233
41. Parmet AJ (1999) Aviat Space Environ Med 70:817
42. Hsueh PR, Chen PJ, Hsiao CH, Yeh SH, Cheng WC, Wang JL, Chiang BL, Chang SC,
Chang FY, Wong WW, Kao CL, Yang PC (2004) Emerg Infect Dis 10:489
43. Desenclos JC, van der Werf S, Bonmarin I, Levy-Bruhl D, Yazdanpanah Y, Hoen B,
Emmanuelli J, Lesens O, Dupon M, Natali F, Michelet C, Reynes J, Guery B, Larsen C,
Semaille C, Mouton D, Andre M, Escriou N, Burguiere A, Manuguerra JC, Coignard B,
Lepoutre A, Meffre C, Bitar D, Decludt B, Capek I, Antona D, Che D, Herida M, Infuso A, Sauri C, Brucker G, Hubert B, LeGoff D, Scheidegger S (2004) Emerg Infect Dis
10:195
44. Olsen SJ, Chang HL, Cheung TY, Tang AF, Fisk TL, Ooi SP, Kuo HW, Jiang DD,
Chen KT, Lando J, Hsu KH, Chen TJ, Dowell SF (2003) N Engl J Med 349:2416
45. Wilder-Smith A, Paton NI, Goh KT (2003) Trop Med Int Health 8:1035
46. Bloch AB, Orenstein WA, Ewing WM, Spain WH, Mallison GF, Herrmann KL, Hinman AR (1985) Pediatrics 75:676
47. Amornkul, PN, Takahashi, H, Bogard AK, Nakata M, Harpaz R, Efer PV (2004) J Infect Dis 189 Suppl 1:81
48. (2001) MMWR Morb Mortal Wkly Rep 50:485
49. Wick RL Jr, Irvine LA (1995) Aviat Space Environ Med 66:220
50. Dechow M, Sohn H, Steinhanses J (1997) Chemosphere 35:21
51. Rudnick SN, Milton DK (2003) Indoor Air 13:237
149
52. Beggs CB, Noakes CJ, Sleigh PA, Fletcher LA, Siddiqi K (2003) Int J Tuberc Lung Dis
7:1015
53. Gammaitoni L, Nucci MC (1997) Emerg Infect Dis 3:335
54. Rydock JP (2004) Aviat Space Environ Med 75:168
55. Daley CL, Kawamura LM (2003) Int J Tuberc Lung Dis 7:458

DOI 10.1007/b107242
Microbial Contamination in Airplane Cabins:

Health Effects and Remediation
Dick Menzies
Montreal Chest Institute, Montreal, Canada
dick.menzies@mcgill.ca
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
152
Environmental Sources within the Indoor Environment . . . . . . . . . .
153
Mechanisms of Health Effects of Microbial Contamination

in the Indoor Environment . . . . . . . . . . . . . . . . . . . . . . . . . . .
153
Specic Health Effects of Microbial Contamination

of the Indoor Environment . . . . . . . . . . . . . . . . . . . . . . . . . . .
155
Role of Ventilation in Microbial Effects on Health . . . . . . . . . . . . . .
156
Control of Microbial Contamination . . . . . . . . . . . . . . . . . . . . .
157
Microbes within Aircraft Sources and Health Effects . . . . . . . . . . .
160
Microbial Control in Aircraft . . . . . . . . . . . . . . . . . . . . . . . . .
161
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
162
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
163
Abstract Microorganisms that affect human health are found in all indoor environments,
including cabins of commercial aircraft. Those that arise from human sources can be
transmitted by direct contact, droplets, or the airborne route. Infections from human
sources include Inuenza, Rhinovirus, SARS and tuberculosis. Transmission by the airborne route can be reduced by sterilizing the air with ultraviolet germicidal irradiation,
or by diluting the contaminated air with outdoor air through ventilation. Microbes arising
from environmental sources include bacteria, fungi and other organisms such as protozoa. These usually have very simple requirements for growth water and a simple substrate such as dust. They cause health effects through direct infection rarely (one example
is Legionnella), but more commonly cause immune reactions resulting in hypersensitivity or allergy mediated diseases. Environmental sources of microbial contamination are
best prevented, but can be remediated through cleaning, germicidal chemicals, or ultraviolet germicidal irradiation. Airborne microbial substances including toxins, antigens
and viable organisms can be removed by outdoor air ventilation or ltration. In aircraft
cabins transmission of pathogens from human sources is difcult to control, but airborne transmission can be reduced through increased outdoor air ventilation or ltration.
Environmental microbial contamination can, and does occur in aircraft cabins. These microbial sources are best prevented but, if detected, can be removed through cleaning or
152
D. Menzies
disinfection. Ultraviolet germicidal irradiation is an under-utilized technology that may

be useful for sterilizing air as well as potential environmental sources.
Abbreviations
TB
Tuberculosis
UVGI Ultraviolet germicidal irradiation
ACH Air changes per hour
HVAC Heating ventilation and air conditioning
SARS Severe acute respiratory syndrome
HEPA High efciency particulate air
1
Introduction
As air travel increases globally health risks of air travel become of greater
concern. This is because of the direct impact on the health and well-being of
passengers (and crew), the potential economic consequences (including legal
compensation for damages), and the broader public health impact. The recent
SARS outbreak demonstrated how rapidly an epidemic can spread globally
though modern air travel. This review will examine the types of microbes that
are found in this environment, their sources, health effects, and remediation
or prevention.
Microbes that affect human health within the indoor environment originate from two sources humans and the environment. Microbes from human
sources typically cause disease through infectious mechanisms. They include
common viruses such as rhinovirus, inuenza viruses and measles. Human to
human transmission of viruses within the indoor environment has been well
documented [13]. SARS (severe acute respiratory syndrome) is a new Coronavirus that quickly achieved global recognition as it appears to be highly
transmissible within the indoor environment [4, 5], and causes severe manifestations with a high case fatality rate [68]. Transmission of certain forms
of pneumonia such as that caused by the Adenovirus [9] and Streptococcus
Pneumoniae [10] has been demonstrated within indoor environments. Tuberculosis, a pathogen which can remain viable airborne for more than 24
hours within the indoor environment, has been transmitted in a wide range
of indoor environments [11, 12], including aircraft cabins [1315].
Microbes from environmental sources must be capable of growth within
the environment. This growth results in the generation of aerosols of microorganisms, or of microbial particles or toxins. These are then inhaled by
humans. Such micro-organisms include many species of bacteria, fungi and
other micro-organisms such as protozoans [16]. Most of these are not considered highly pathogenic, as most do not cause direct infections in humans.
Rather they cause health effects through allergic or immune mechanisms.
Microbial Contamination in Airplane Cabins: Health Effects and Remediation
153
2
Environmental Sources within the Indoor Environment
Most environmental fungi and bacteria require only water, and a simple substrate for growth. Substrates can be dust, furnishings, or building materials.
Many organisms grow well in the absence of light (in fact natural sunlight
will often kill them). Given these simple requirements for growth, it should be
no surprise that environmental micro-organisms are ubiquitous in the indoor
environment. Abundant growth, leading to high concentrations, can occur on
any surface with sufcient water. This includes building materials that have
been damaged by ooding, ground water or spillage [17], anywhere that condensation of water occurs, such as air conditioning systems or where there
is standing water, such as water cooling towers, and humidication systems.
Microbial contamination has been identied within all parts of modern ventilation systems, including lters [18, 19], air conditioning chillers [2022],
drip pans [20], humidiers [23], and ducts [24, 25]. In addition, mold and
bacterial contamination is common in areas of food preparation or consumption.
Microbial growth on surfaces does not directly affect human health, except rarely if they are ingested or contact skin directly. In almost all instances
health effects result from inhalation of airborne microbial organisms, antigens, or toxins. This means the microbial substances must be aerosolized
in one of several ways. The most obvious is disturbance by human activity, including normal work activities, and cleaning. The latter can cause
bursts of very high concentrations of microbes, creating a particular hazard for those present [26, 27]. In the heating ventilation and air conditioning
(HVAC) systems the moving currents of air can act to aerosolize, and then
efciently disperse the microbes or their products throughout the occupied
spaces. Once aerosolized the microbes, and their antigens or toxins, will remain airborne for some time and therefore can be inhaled by humans in that
environment.
3
Mechanisms of Health Effects of Microbial Contamination
in the Indoor Environment
Micro-organisms cause health effects in humans through three general mechanisms: direct infections, toxins, and immune mediated reactions [28]. Direct
infectious complications are those that result from true infection of the human host by a microbial pathogen. Acquisition of infection is almost always
by inhalation, although the infectious particles can impact on the upper or
lower airways, with different resultant clinical manifestations. For example
rhinoviruses that cause the common cold and inuenza viruses, preferen-
154
D. Menzies
tially impact in the upper airways perhaps because they are bound to small
airborne dust particles. By contrast, airborne TB bacteria are in droplet nuclei of 15 microns size. Particles of this size will reach the alveolar level of
the lung and therefore cause a pneumonia-like condition. Other examples of
potentially airborne microbes that cause direct infectious health effects include measles and SARS. Some organisms can cause infectious diseases, or
immune mediated syndromes. For example Legionella pneumophila can result in pneumonia, termed Legionnaires disease, [29, 30] but can also be
a systemic illness that is immune mediated, termed Pontiac Fever [29, 31].
It is unknown if these manifestations reect differences in the infecting dose,
or the host response.
Micro-organisms produce a substantial array of toxins, but to date their
health effects are poorly understood. These include endotoxin produced
by certain bacteria, and mycotoxins produced by certain fungi. Experimental exposure to high concentrations of endotoxin has produced fever, difculty breathing, and short-lived changes in lung function. [3234]. In crosssectional population based studies, exposure to relatively low levels of endotoxin has been linked to non-specic building related symptoms, and sickness
absence [3537]. However, in these cross-sectional studies, the ndings could
also have been due to substantial confounding, since no relationship between
endotoxin levels and health effects was found in several other studies. The
health effects of mycotoxins, produced by certain fungi, such as Stachybotrys
Atra, are more controversial [38]. Mycotoxin exposure has been linked to
health effects in uncontrolled case reports and case series [39]. In one casecontrol study this toxin was linked to pulmonary haemorrhage in infants [40].
However in a subsequent publication, this studys methods were heavily criticized, ndings reversed in re-analysis, and conclusions withdrawn by the
same agency that conducted the original case-control study [41]. In summary,
although many microbes produce toxins, and intuitively toxins can not be
good, there is inconclusive evidence they actually cause health effects in the
concentrations found in the indoor environment.
Microbial health effects can also be mediated by immune reactions,
through allergic or cell mediated mechanisms [42]. Allergic manifestations
are mediated by mast cells that release histamine upon exposure to microbial
products such as proteins, or components of the cell wall. Symptoms occur
soon after exposure, and range from itchy watery eyes, with nasal stufness,
congestion and discharge typical of allergic rhinitis to chest tightness, wheezing and difculty breathing typical of asthma. Persons with a personal or
family history of allergy or atopic illnesses, are more likely to manifest allergic
responses to airborne microbial contaminants.
The other major mechanism of immune reaction is a cell mediated response to inhaled allergens. This response, mediated by lymphocytes, often manifests only hours after exposure, making the diagnosis less obvious
than with immediate allergic responses. Manifestations include pneumoni-
155
tis, which resembles pneumonia, causing dyspnea, chest tightness, hypoxia in

severe cases, and systemic symptoms of fever, chills and sweats. This hypersensitivity pneumonitis has also been described with several other occupational exposures, such as farmers lung [43, 44]. Milder reactions involving
systemic symptoms only, have been termed Pontiac fever, or humidier
fever. These terms reect the geographic locale or source in early descriptions of this problem, rather than the pathogenetic mechanism. Therefore,
this problem shall be referred to as hypersensitivity systemic syndrome in
this chapter. Interestingly non-smokers are at particularly high risk for this
type of immune response [22, 43, 44].
4
Specific Health Effects of Microbial Contamination
of the Indoor Environment
Outbreak investigations have been the most frequent method to identify
the health effects of microbial contamination of the indoor environment.
Typical outbreaks involve a large number of affected individuals who have
similar clinical manifestations [21, 4547]. Subsequent investigation revealed
substantial microbial contamination by one, or multiple, microbes. In all outbreaks, improvement occurred when the source was eliminated. Interestingly,
in a few outbreaks that were carefully investigated, exposed persons exhibited
a range of clinical manifestations ranging from severe disease requiring treatment, to mild symptoms [45, 46]. Those with milder manifestations might
normally have been overlooked or considered to have symptoms unrelated
to the indoor environment. The microbial cause for these mildly affected
persons was established only because others within the same indoor environment were more seriously affected. The variability of clinical manifestations
in these outbreaks is unexplained, but most likely reects variation in individual susceptibility due to age, gender, co-morbid illnesses, cigarette smoking,
or other factors [48]. This variability of response has been repeatedly demonstrated, including under carefully controlled conditions [4952].
Outbreaks have served to demonstrate the pathogenic role of certain
organisms, and the potential importance of certain indoor environment
sources. Examples include contamination of water cooling towers by Legionella pneumophila causing pneumonia [30, 53] or a hypersensitivity
systemic syndrome [31, 53], fungal contamination of air conditioning systems
or water damaged building materials leading to hypersensitivity pneumonitis
and asthma [23, 45, 46], and contamination by multiple organisms of stagnant
water in humidiers [21] causing hypersensitivity pneumonitis or systemic
syndrome.
The role of microbial contamination of the indoor environment and health
effects at a population level is undened. This is because in non-outbreak
156
D. Menzies
situations microbial levels are lower [5457], and the link between these
low microbial levels and health effects is much harder to establish. At lower
concentrations of organisms, only the most susceptible of the exposed individuals will be affected, and their clinical manifestations may be mild and
non-specic [48, 54]. As well, lower levels of microbial contamination are
more difcult to measure accurately because of the variability of concentrations related to time of day, human activity, ambient temperature and
humidity [58, 59]. When microbial concentrations are very high, these variations are relatively unimportant, but become progressively more important at
lower concentrations. There is some evidence of the population impact of microbial exposures in the home environment from surveys of the health effects
of bacteria and fungi in homes or residences [6067]. Although not directly
applicable, this is the most important source of information on the health
effects of microbes. Even here the most consistent relationships with health
effects have been with markers of bacterial and fungal growth, such as visible mold, or damp damage, and not with actual measured airborne microbial
concentrations [61, 63, 66, 6870].
In one study in the non-residential environment, low concentrations
of Alternaria in the lters of some HVAC systems of large ofce buildings were linked to respiratory symptoms and positive allergy skin tests
to Alternaria [71]. However this relationship was detected only in 2% of
the total study population, and only by means of a complex series of
investigations.
5
Role of Ventilation in Microbial Effects on Health
The term ventilation of the indoor environment generally refers to mechanical exchange of indoor air with outdoor air. The primary objective of this
exchange is to remove indoor air pollutants. The mechanical systems can also
heat or cool, humidify or de-humidify, and lter the air being delivered to
the indoor space. As such these mechanical systems can act to disseminate
airborne microbial contaminants, but can also reduce the concentration of
microbes in indoor air, through dilution.
A number of population-based studies have demonstrated a clear link between ventilation levels and microbial transmission. These include Inuenza
and rhinovirus outbreaks [1, 3], excess occurrence of adeno-viral pneumonia among military recruits housed in mechanically ventilated barracks [9]
compared to naturally ventilated barracks and excess occurrence of Strep
pneumonia in prison inmates where there was greater crowding and less
ventilation [10]. A recent study documented that within ofce buildings,
airborne rhinovirus concentrations were higher when building outdoor air
supply was lower [72].
157
The best evidence for a link between ventilation levels and microbial transmission has come from studies of tuberculosis. In one study, an ofce worker
with contagious TB infected many of her co-workers, some of whom had no
direct contact but worked in ofces ventilated with the same re-circulated
air as the affected worker. This implied that the ventilation system, by recirculating germ-laden air, acted to disseminate the infection throughout the
building. Transmission was mathematically related to ventilation level in this
study [73]. A second study detected signicantly higher transmission of TB
infection to hospital workers on clinical units with lower levels of ventilation [74].
Given the recent intense interest in SARS, it is important to mention that
the mode of transmission of SARS is somewhat unclear. There is evidence
that SARS is transmitted by means of droplets, or direct contact from person to person [4, 5], but given that it is a respiratory pathogen, there remains
a possibility that airborne transmission also occurs. Therefore, at this time it
is prudent to consider that SARS may be transmitted by the airborne route,
in which case increasing levels of ventilation could help reduce the risk of
transmission.
6
Control of Microbial Contamination
There are two general approaches to microbial control in any indoor environment. The most effective long term solution is elimination of all locations
of microbial growth (source control) [75]. The alternative, if elimination of
all possible sources is impossible, is to eliminate airborne microbes through
sterilization, ltration, or dilution.
Source control, the preferred approach, can be further sub-divided into
prevention or remediation [76]. Prevention implies preventing the conditions
that favour microbial growth. Given the dependence of microbes on water,
the most successful and practical approach is to prevent water accumulation, condensation, or inltration, as well as any subsequent water damage.
This means installation of dehumidication systems where humidity levels
are high such as environments at or below ground level (obviously not a problem on aircraft!). Prevention of water inltration means water proong the
building shell, particularly at or below ground level [17]. Prevention of condensation with air-conditioning systems is not possible. Instead the objective
is rapid removal of all condensate, because if water accumulates it will quickly
become contaminated. Humidication systems should be designed without
a reservoir of standing water. The best systems use steam humidication,
rather than ultrasonic or other forms of nebulization of cool water [77].
Prevention also includes careful selection of equipment, furnishings, and
building materials that will not act as media for microbial growth. In the
158
D. Menzies
indoor environment, carpets are an important and common source of microbial contamination [78], particularly where food or drinks are prepared
or consumed. Having no carpets, or selecting low-pile carpets is an important preventive measure. Other furnishings and equipment should be selected
with similar criteria they should have smooth surfaces, reducing accumulation of microbial substrate (dirt), reducing surface area for microbial growth,
and facilitating cleaning.
Remediation is the term applied to elimination of microbial sources after
contamination has occurred. The most effective method is to completely
remove contaminated sources such as damp damaged carpets, furnishings,
draperies, insulation or other building materials. Cleaning can be effective,
but is often a less permanent solution. In some cases, such as cleaning ventilation ducts, or changing lters, these actions can result in important release
of microbial products, resulting in very high, albeit transient microbial exposures [26, 27]. Therefore these activities should be performed when the
occupants are not present, and signicant care must be taken to prevent an
occupational hazard to those performing these tasks [27]. Microbial reduction can also be accomplished with germicidal chemicals, but it is important
to recognize that these chemicals may themselves be associated with health
effects. Therefore chemical cleaning must be done when the occupants are absent, and appropriate precautions must be taken to ensure the workers doing
the cleaning are not exposed. Furthermore, sufcient time must have elapsed
before the occupants so return, that all traces of the chemicals have dissipated.
Reduction of airborne concentration of microbes can be achieved by direct
sterilization of air through use of natural sunlight or ultraviolet germicidal irradiation (UVGI). Because the sterilizing ultraviolet rays of natural sunlight
are largely eliminated by glass, natural sunlight is not a practical option for
air sterilization within most indoor environments. UVGI has been used to
sterilize air and thereby prevent airborne transmission of certain diseases,
most notably measles transmission within schools and tuberculosis transmission within hospitals and other health care facilities [79]. UVGI is also used
for sterilizing air in meat packing plants, pharmaceutical manufacturing, and
operating rooms [80].
The efcacy of UVGI in sterilizing air is therefore unquestioned, but its
application is limited by certain potential hazards. These include eye irritation, and a theoretical risk of skin cancer. Therefore UVGI cannot be used
in occupied spaces, or if used, only to irradiate the upper air of the room
with the xtures constructed and mounted to prevent direct irradiation of
the human occupants [79]. However, in other areas, such as within the HVAC
system, UVGI is potentially highly useful within the ducts to sterilize the
air, or to irradiate the air-conditioning systems to eliminate condensaterelated microbial contamination [81]. In one recent study UVGI irradiation
of air-conditioning systems in 3 large ofce buildings resulted in a signi-
159
cant reduction of allergic, and systemic syndrome-type symptoms in 771

ofce workers [82]. The most signicant improvement was seen in the most
susceptible i.e. workers who had a history of atopy or allergy, as well as
non-smokers. In summary, it would appear that UVGI is an under-utilized
technology that may be a useful addition to the microbial control armamentarium.
A major limitation of airborne sterilization is that the killed airborne microbes, although no longer able to cause infections, may still be immunogenic
and result in allergic or hypersensitivity manifestations. Filters can be effective to eliminate viable and non-viable microbes, as well as microbial particles.
However, the efciency of the lter is an important determinant of the benecial effects. Although lters found in most HVAC systems will trap microbes
that are adherent to airborne dust particles, they will not trap particles of
15 microns size [83]. These are termed respirable particles, as they can reach
the alveoli when inhaled. Particles of this size may contain microbial antigens,
such as proteins or cell wall parts, or viable bacteria, such as tuberculosis.
These pathogenetically important particles will only be trapped by more efcient HEPA lters (high efciency particulate air lters). HEPA lters will
trap more than 99% of particles of 0.3 micron size or larger [83]. However
these lters are much more costly, and because of their greater resistance they
require greater fan strength and energy to operate, further increasing their
cost, and limiting their applicability. An additional limitation of lters is that
they must be changed frequently, as they become contaminated. Several reports have documented that poorly maintained lters can become a source
of airborne microbial contamination [19, 84]. At the time of changing the lters, there can be a substantial airborne burst of microbes released which
can pose signicant hazards to those in the occupied space [85, 86]. Therefore it is recommended that lters are changed when the indoor environment
is not occupied and the occupational hazard for those changing the lters is
prevented by suitable protection [85, 86].
Dilution of indoor airborne pollutants of all types including viable microbial organisms, toxins, or antigens through exchange with outdoor air
(i.e. ventilation) is a commonly used solution. At very low indoor air exchange rates, very high concentrations of indoor contaminants can be seen.
As ventilation is increased, these concentrations will decline rapidly at rst,
but the benets become progressively less, with progressively higher air exchange rates, as demonstrated in Fig. 1. Above six air exchanges per hour,
further increases in ventilation will have little additional benet. The disadvantages of increased ventilation include the increased fan strength required
to achieve the ventilation, and the substantial energy costs to heat or cool as
well as to humidify or dehumidify the outdoor air brought in [71]. It is because of these energy costs that outdoor air exchange rates have been reduced
over the past 20 years in modern buildings and aircraft.
160
D. Menzies
Fig. 1 Effects on pollutant concentration at the end of 1 hour of exchange with unpolluted
outdoor air at varying rates of exchange
Microbial transmission from human to human that is mediated by droplets

or direct contact will not be affected by environmental conditions. Increasing
levels of outdoor air supply will have little benecial effect. However organisms that are transmitted from human to human by the airborne route, such
as tuberculosis, measles, or inuenza, will be affected by changes in ventilation [1, 3, 9, 10, 73]. Increasing the outdoor air supply will increase the
removal of the airborne microbes, by replacing contaminated air with clean
outdoor air. This will result in reduced probability of inhalation by humans.
However, as with all other indoor pollutants, dilution of infectious microbes
by outdoor air will be most effective if pre-existing ventilation levels are low,
but will have diminishing returns above a certain level.
7
Microbes within Aircraft Sources and Health Effects
Although it seems highly likely that transmission of microbes from passenger
to passenger through droplets or direct contact does occur during commercial airight, there is little documentation of this phenomenon [87]. There is
some evidence of airborne transmission of microbes within aircraft, mostly
from investigations of potential transmission of tuberculosis [88]. Tuberculosis has received more attention than other pathogens for several reasons. The
disease is serious. Tuberculosis bacteria can survive in airborne droplet nuclei
161
for more than 24 hours potentially important during prolonged airights.

The condition is rare in many populations nowadays, making transmission
events easier to detect than for common viruses such as the rhinovirus. Nine
investigations of potential TB transmission aboard commercial aircraft have
been published [1315, 89]. In only two of the nine was evidence of transmission found [13, 15]. One was related to a ight attendant who was infectious
for a prolonged period of time; despite this, the risk to passengers was very
low, and far less than the risk for household contacts of the same contagious cases [13]. These investigations provide useful information because this
transmission of TB can be considered a proxy of airborne transmission of
many other pathogens whose transmission is much more difcult to measure [88].
Other health effects of microbial exposures such as allergic or hypersensitivity disorders have not been investigated at all among passengers on
commercial air ights. These problems could occur in this environment. For
example, given the regular consumption of food and drink, microbial contamination of carpets, and furnishings seems plausible. As well, microbial
contamination of aircraft ventilation system lters has been documented [84].
However, because human subjects are exposed infrequently, and for relatively
brief periods, the health effects resulting from these exposures would be very
difcult to detect, without a highly systematic investigation directed to detect
them.
8
Microbial Control in Aircraft
Prevention of human to human transmission mediated by droplets or contact can only be prevented by controlling the source in this case the
humans themselves. This means screening the passengers as well as the
crew. During the height of the SARS epidemic passengers were screened for
fever and cough, although often only after arrival in certain countries. To
prevent SARS transmission (and also prevent transmission of many other
respiratory pathogens) one would have to screen all passengers and crew
for fever and cough. Embarkation would be denied to anyone failing the
screening. Apart from the logistic difculties of screening all passengers
rapidly prior to embarkation, this would also have substantial cost implications because of delayed travel. Other approaches to limiting transmission of potentially contagious respiratory disease would be to require wearing of masks by such passengers. However, wearing masks may be seen as
branding the passengers, creating stigma and substantial anxiety for the affected passenger and all around them. This would also raise the possibility
of later legal repercussions. Nonetheless, in an epidemic of a serious airborne transmitted illness such as SARS, these measures may be justied to
162
D. Menzies
protect other passengers, and to diminish the international spread of disease.

Prevention of immune mediated health effects is best achieved by source
control. This means elimination of potential microbial sources such as contaminated or water damaged carpets or upholstery and scrupulous attention
to maintenance with frequent changes of lters in the ventilation systems. An
additional option is to install UVGI to irradiate the lters to prevent their
becoming a source of microbial contamination.
Using ltration to eliminate microbes has many limitations, as noted
above. These include failure to trap respirable pathogenic or immunogenic
particles from microbes with standard lters, and becoming sources themselves. High efciency HEPA lters may be impractical because of the increased ventilation fan strength and energy requirements to overcome the
lter resistance.
Outdoor air ventilation can be increased to dilute and remove airborne
contaminants. A limitation of this approach in commercial aircraft is that the
outdoor air is at low pressure, and is very dry and cold. Therefore this air
must be compressed and heated to aircraft cabin pressure, which incurs substantial energy costs. These energy costs have driven aircraft manufacturers
and operators to minimize outdoor air supplied to aircraft cabins over the
last two decades. However certain minimum standards must be maintained
in order to minimize the risk to passengers. It has been suggested that the
low relative humidity contributes to a higher incidence of upper respiratory
tract infections in the week following air travel [90], which could be prevented
through modest humidication.
Another untested option would be install ultraviolet germicidal irradiation within the aircraft ventilation systems. This UVGI could irradiate, and
effectively sterilize recirculated air, reducing the potential for airborne transmission of certain pathogens. This could reduce the need for outdoor air
exchange rate, thereby conserving energy. However this technology would
have no effect on airborne microbial antigens and their associated health effects, nor on transmission mediated through droplets or direct contact. This
application of UVGI appears theoretically sound, but has not been evaluated.
9
Summary
In summary, exposures to many different microbes can occur within aircraft
cabins. Most of these are from human sources, and can result in transmission of infectious diseases, particularly viral illnesses such as common colds,
measles, inuenza or even SARS. These are spread by direct contact or by
droplets, so environmental control measures are limited. For these situations
source control is the only effective method, but this implies screening passen-
163
gers prior to embarkation, with denial to ill passengers on grounds of possible

contagiousness. This would only be justied in an outbreak situation of a serious pathogen such as SARS. For pathogens, such as tuberculosis, that are
transmissible by the airborne route, the risk can be reduced by greater ventilation, meaning greater outdoor air exchange rate.
The other microbial contaminants that are commonly seen in homes and
non-industrial environments and result in immune mediated disorders such
as asthma, rhinitis, or hypersensitivity syndromes are less likely within this
environment. However, contamination can and does occur. Much of this
should be preventable by careful selection of appropriate furnishings and
materials, as well as thorough and regular cleaning and maintenance. Adequate ventilation can play some role in reducing exposure to any indoor air
pollutants, including airborne microbes or their antigens and products. The
potential role of ultraviolet germicidal irradiation requires further evaluation,
although for certain purposes this technology should be safe and effective.
Acknowledgements Dr. Menzies receives a research salary award from the Fond de
Recherche en Sant du Qubec.
References
1. Warshauer DW, Dick EC, Madel AD, Flynn TC, Jerde RS (1989) Rhinovirus Infections
in an isolated Antarctic Station: Transmission of the virus and susceptibility of the
population. Am J Epidemiol 129:319
2. Jaakkola JJK, Heinonen OP (1995) Share Ofce Space and the Risk of the Common
Cold. Eur J Epidemiol 11:213
3. Drinka PJ, Krause P, Schilling M, Miller BA, Shult P, Gravenstein S (1996) Clinical Investigation: Reporting of an Outbreak: Nursing Home Architecture and Inuenza-A
Attack Rates. Journal of American Geriatrics Society 44
4. Varia M, Wilson S, Sarwal S, McGeer A, Gournis E, Galanis E, Henry B (2003) Investigation of a nosocomial outbreak of severe acute respiratory syndrome (SARS) in
Toronto, Canada. CMAJ 169:285
5. Chan-Yeung M, Yu WC (2003) Outbreak of severe acute respiratory syndrome in Hong
Kong Special Administrative Region: case report. BMJ 326:850
6. Choi KW, Chau TN, Tsang O, Tso E, Chiu MC, Tong WL, Lee PO, Ng TK, Ng WF,
Lee KC, Lam W, Yu WC, Lai JY, Lai ST (2003) Outcomes and prognostic factors in
267 patients with severe acute respiratory syndrome in Hong Kong. Ann.Intern Med
139:715
7. Gomersall CD, Joynt GM, Lam P, Li T, Yap F, Lam D, Buckley TA, Sung JJ, Hui DS, Antonio GE, Ahuja AT, Leung P (2004) Short-term outcome of critically ill patients with
severe acute respiratory syndrome. Intensive Care Med 30:381
8. Avendano M, Derkach P, Swan S (2003) Clinical course and management of SARS in
health care workers in Toronto: a case series. CMAJ 168:1649
9. Brundage JF, Scott RMcN, Lednar WM, Smith DW, Miller RN (1988) Buildingassociated risk of febrile acute respiratory diseases in army trainees. JAMA 259:2108
164
D. Menzies
10. Hodge CW, Reichler MR, Dominguez EA, Bremer JC, Mastro TD, Hendricks KA,
Musher DM, Elliott JA, Facklam RR, Breiman RF (1994) An epidemic of pneumococcal disease in overcrowded, inadequately ventilated jail. N Engl J Med 331:643
11. Raffalli J, Sepkowitz KA, Armstrong D (1996) Community-based outbreaks of tuberculosis. Arch Intern Med 156:1053
12. Lincoln EM (1965) Epidemics of tuberculosis. Adv Tuberc Res 14:157
13. Driver CR, Valway SE, Morgan WM, Onorato IM, Castro KG (1994) Transmission of
Mycobacterium tuberculosis Associated With Air Travel. JAMA 272:1031
14. Miller MA, Valway S, Onorato IM (1996) Tuberculosis risk after exposure on airplanes. Tuberc Lung Dis 77:414
15. Kenyon TA, Valway SE, Ihle WW, Onorato IM, Castro KG (1996) Transmission of
Multidrug-Resistant Mycobacterium Tuberculosis During a Long Airplane Flight. The
New England Journal of Medicine 334:933
16. Finnegan MJ, Pickering CAC, Davies PS, Austwick PK, Warthurst DC (1987) Amoebae
and humidier fever. Clinical Allergy 17:235
17. Hoffman RE, Wood RC, Kreiss K (1993) Building-related asthma in Denver ofce
workers. Am J Public Health 83:89
18. Elixmann JH, Schata M, Jorde W. Fungi in lters of air-conditioning systems cause the
building related illness. Walkinshaw D. 1, 193-195. (1990) Toronto, Canadian Mortgage & Housing Corporation. Proceedings of the Fifth International Conference on
Indoor Climate and Air Quality
19. Elixman J (1996) Investigation of alleric potential induced by fungi on air lters of
HVAC systems. The Seventh International Conference on Indoor Climate and Air
Quality 125
20. Hugenholtz P, Fuerst JA (1992) Heterotrophic Bacteria in an Air-Handling System.
Applied and Environmental Microbiology 58:3914
21. Banaszak EF, Thiede WH, Fink JN (1970) Hypersensitivity pneumonitis due to contamination of an air conditioner. N Engl J Med 283:271
22. Baur X, Richter G, Pethran A, Czuppon AB, Schwaiblmair M (1992) Increased Prevalence of IgG -Induced Sensitization and Hypersensitivity Pneumonitis (Humidier
Lung) in Nonsmokers exposed to Aerosols of a Contaminated Air Conditioner. Respiration 59:211
23. Burge PS, Finnegan MJ, Horseld N, Emery D, Austwick P, Davies PS, Pickering CAC
(1985) Occupational asthma in a factory with a contaminated humidier. Thorax
40:248
24. Pasanen P, Pasanen AL, Jantunen M (1993) Water Condensation promotes Fungal
Growth in Ventilation Ducts. Indoor Air 3:106
25. Smoragiewicz W, Cossette B, Boutard A, Krzystyniak K (1993) Trichothecene mycotoxins in the dust of ventilation systems in ofce buildings. Int Arch Occup Environ
Health 65:113-117
26. Anderson K, Morris G, Kennedy H, Croall J, Michie J, Richardson MD, Gibson B
(1996) Aspergellosis in immunocompromised paediatric patients: associations with
building hygiene, design, and indoor air. Thorax 51:256
27. Rautiala S, Reponen T, Nevalainen A, Husman T, Kalliokoski P (1998) Control of Exposure to Airborne Viable Microorganisms During Remediation of Moldy Buildings:
Repot of Three Case Studies. American Industrial Hygene Association Journal 59:455
28. Husman T (1996) Health Effects of Indoor-Air Microorganisms. Scand J Work Environ Health 22:5
29. Dennis PJ, Taylor JA, Fitzgerald RB, Bartlett CLR, Barnow GI (1982) Legionnella pneumophilia in water plumbing systems. Lancet 1:949
165
30. Fraser DW, Tsai TR, Orenstein W, Parkin WE, Beecham HJ, Sharrar RG et al. (1977)
Legionnaires disease: description of an epidemic of pneumonia. N Engl J Med
297:1189
31. Glick TH, Gregg MB, Berman B, Mallison G, Rhodes WW, Kassanoff I (1978) An Epidemic of Unknown Etiology in a Health Department: I. Clinical and Epidemiologic
Aspects. Am J Epidemiol 107:149
32. Rylander R, Bake B, Fischer JJ, Helander IM (1989) Pulmonary Function and Symptoms after Inhalation of Endotoxin. Am Rev Respir Dis 140:981986
33. Fogelmark B, Goto H, Yuasa K, Marchat B, Rylander R (1992) Acute pulmonary toxicity of inhaled Beta-1, 3-Glucan and endotoxin. Agents Actions 35:50
34. Smid T, Heederik D, Houba R, Quanjer PH (1994) Dust- and endotoxin-related acute
lung function changes and work-related symptoms in workers in the animal feed
industry. American Journal of Industrial Medicine 877888
35. Teeuw KB, Vandenbroucke-Grauls CMJE, Verhoef J (1994) Airborne Gram-negative
Bacteria and Endotoxin in Sick Building Syndrome. Arch Intern Med 154:2339
36. Milton DK, Wypij D, Kriebel D, Walters MD, Hammond SK, Evans JS (1996) Endotoxin
exposure-response in a gerglass manufacturing facility. Am J Ind Med 29:3
37. Milton DK, Amsel J, Reed CE, Enright PL, Brown LR, Aughenbaugh GL, Morey PR
(1995) Cross-sectional follow-up of a u-like respiratory illness among berglass
manufacturing employees: endotoxin exposure associated with two distinct sequelae.
Am J Ind Med 28:469
38. Page E, Trout D (1998) Mycotoxins and Building-Related Illness. JOEM 40:761
39. Jarvis BB, Zhou Y (1996) Toxigenic molds in water damaged buildings. Journal of
Natural Products 59:553
40. Centers for Disease Control (CDC) (1994) Acute Pulmonary hemorrhage/hemosiderosis
among infants Cleveland. Morbidity and Mortality Weekly Report 43:881
41. Center for Disease Control (1997) Update: pulmonary hemorrhage/hemosiderosis
among infants-Cleveland, Ohio 19931996. 43:881883
42. Salvaggio J (1981) Mould induced asthma. J Allergy Clin Immun 68:327
43. Maimberg P, Rask-Anderson A, Palmgren U, Hoglund S, Kolmodin-Hedman B, Stalenheim G (1985) Exposure to microorganisms, febrile and airway-obstructive symptoms, immune status and lung function of Swedish farmers. Scand J Work Environ
Health 11:287
44. Malmberg P, Rask-Andersen A, Rosenhall L (1993) Exposure to microorganisms associated with allergic alveolitis and febrile reactions to mold dust in farmers. Chest
103:1202
45. Woodard ED, Friedlander B, Lesher RJ, Font WF, Kinsey R, Hearne FT (1988) Outbreak of hypersensitivity pneumonitis on an industrial setting. JAMA 259:1965
46. Bernstein RS, Sorenson WG, Garabrant D, Reaux C, Treitman RD (1983) Exposures to
respirable, airborne Penicillium from a contaminanted ventilation system: clinical,
environmental and epidemiologic aspects. Am Ind Hyg Assoc J 44:161
47. Arnow PM, Fink JN, Schuelter DP et al. (1978) Early detection of hypersensitivity
pneumonotis in ofce workers. Am J Med 64:236
48. Menzies D, Bourbeau J (1997) Building-related illnesses. New Engl J Med 337:1524
49. Doeland HJ, Nauta JJP, Van Zandbergen JB, Van Der Eerden HAM, Van Dieman NGJ,
Bertelsmann FW, Heimans JJ (1989) The Relationship of Cold and Warmth Cutaneous
Sensation to Age and Gender. Muscle and Nerve 12:712
50. Grivel F, Candas V (1991) Ambient temperatures preferred by young European males
and females at rest. Ergonomics 34:365
166
D. Menzies
51. Kjaergaard S, Molhave L, Pedersen OF (1989) Human exposures to indoor air pollutants: n-decane. Environ Int 15:473
52. Kjaergaard S, Molhave L, Pedersen OF (1991) Human reactions to a mixture of indoor
air volatile organic compounds. Atmos Environ 25A:1417
53. Apte MG, Fisk WJ, Daisey JM (2000) Association between indoor CO2 Concentrations
and Sick Building Syndrome Symptoms in U.S. Ofce Buildings: An Analysis of the
1994-1996 BASE Study Data. Indoor Air 10:246
54. Harrison J, Pickering CA, Faragher EB, Austwick PK, Little SA, Lawton L (1992) An investigation of the relationship between microbial and particulate indoor air pollution
and the sick building syndrome. Respiratory Medicine 86:225
55. Skov P, Valbjorn O, Pedersen BV, Danish Indoor Climate Study Group (1990) Inuence
of indoor climate on the sick building syndrome in an ofce building. Scand J Work
Environ Health 16:363
56. Nelson NA, Kaufman JD, Burt J, Karr C (1995) Health symptoms and the work environment in four nonproblem United States oce buildings. Scand J Work Environ
Health 21:51
57. Menzies R, Tamblyn RM, Nunes F, Hanley J, Tamblyn RT (1996) Exposure to Varying Levels of Contaminants and Symptoms Among Workers in Two Ofce Buildings.
American Journal of Public Health 86:1629
58. American Conference of Governmental Industrial Hygienists (1989) Guidelines for
the assessment of bioaerosols in the indoor environment Cincinnati, Ohio
59. Buttner MP, Stetzenbach LD (1993) Monitoring airborne fungal spores in an experimental indoor environment to evaluate sampling methods and the effects of human
activity on air sampling. Applied and Environmental Microbiology 59:219
60. Roby RR, Sneller MR (1979) Incidence of fungal spores at the homes of allergic
patients in an agricultural community, II. Correlations of skin tests with mold frequency. Ann Aller 43:286
61. Yang CY, Tien YC, Hsieh HJ, Kao WY, Lin MC (1998) Indoor Environmental Risk Factors and Childhood Asthma: A Case-Control Study in a Subtropical Area. Pediatric
Pulmonology 26:120
62. Strachan DP, Flannigan B, McCabe EM, McGarry F (1990) Quantication of airborne
moulds in the homes of children with and without wheeze. Thorax 45:383
63. Spengler J, Neas L, Nakai S, Dockery D, Speizer F, Ware J, Raizenne M (1994) Respiratory Symptoms and Housing Characteristics. Indoor Air 4:72
64. Verhoeff AP, Van Wijnen JH, Van Reenen-Hoekstra ES, Samson RA, Van Strien RT,
Brunekreef B (1994) Fungal Propagules in House Dust 2. Relation with Residential
Characteristics and Respiratory Symptoms. Allergy 49:540
65. Holst PE, Coleman ED, Sheridan JE, ODonnell TV, Sutthoff PT (1986) Asthma and
fungi in the home. NZ Med J 96:718
66. Brunekreef B, Dockery DW, Speizer FE, Ware JH, Spengler JD, Ferris BG (1989) Home
dampness and respiratory morbidity in children. Am Rev Resp Dis 140:1363
67. Dales RE, Burnett R, Zwanenburg H (1991) Adverse health effects among adults exposed to home dampness and molds. Am Rev Resp Dis 143:505
68. Andrae S, Axelson O, Bjorksten B, Frediksson M, Kjellman N (1988) Symptoms of
bronchial hypersensitivity and asthma in relation to environmental factors. Arch Dis
Child 63:473
69. Infante-Rivard C (1993) Childhood asthma and indoor environmental risk factors.
Am J Epidemiol 137:834
70. Dales RE, Zwanenburg H, Burnett R, Franklin CA (1991) Respiratory health effects of
home dampness and molds among Canadian children. Am J Epidemiol 134:196
167
71. Menzies D, Comtois P, Pasztor J, Nunes F, Hanley JA (1998) Aeroallergens and workrelated respiratory symptoms among ofce workers. J Allergy Clin Immunol 101:38
72. Myatt TA, Johnston SL, Zuo Z, Wand M, Kabadze T, Rudnick S, Milton DK (2004)
Detection of airborne rhinovirus and its relation to outdoor air supply in ofce environments. Am J Resp Crit Care Medicine 169:1187
73. Nardell EA, Keegan J, Cheney SA, Etkind SC (1991) Airborne infection: Theoretical
limits of protection achievable by building ventilation. Am Rev Resp Dis 144:302
74. Menzies RI, Fanning A, Yuan L, FitzGerald JM (2000) Hospital ventilation and risk of
tuberculous infection in Canadian Health Care Workers. Ann Intern Med 133:779
75. Pope AM, Patterson R, Burge H (1993) Agents, Sources, Source Controls, and Disease.
Washington DC, National Academy Press. Indoor Allergies: Assessing and Controlling
Adverse Health Effects. pp 86130
76. Morey P (1996) Mold Growth in buildings: removal and prevention. IAQ 2:27
77. Solomon WR (1974) Fungus aerosols arising from cold-mist vaporizers. J. Allergy
Clinical Immunology 54:222
78. Gravesen S, Larsen L, Gyntelberg F, Skov P (1986) Demonstration of microorganisms
and dust in schools and ofces. Allergy 41:520
79. Nardell EA (1995) Interrupting transmission from patients with unsuspected tuberculosis: A unique role for upper-room ultraviolet air disinfection. Am J Infect Control
23:156
80. Berg M, Bergman BR, Hoborn J (1989) Shortwave ultraviolet radiation in operating
rooms. J Bone Joint Surg 71:483
81. Menzies D, Pasztor J, Rand T, Bourbeau J (1999) Germicidal ultravioletirradiation in
air conditioning systems: effect on ofce worker health and wellbeing: a pilot study.
Occup Environ Med 56:397
82. Menzies D, Popa J, Hanley JA, Rand T, Milton DK (2003) Effect of ultraviolet germicidal lights installed in ofce ventilation systems on workers health and wellbeing:
double-blind multiple crossover trial. Lancet 362:1785
83. Nicas M (1995) Respiratory protection and the risk of mycobacterium tuberculosis
infection. Am J Ind Med 27:317
84. Moorehead J (2000) Contaminated Cabins Can Spread Disease. Air Safety Week: Air
Safety & Security Trends, Policy and Regulation 14(11). Phillips Business Information,
Inc.
85. Centers for Disease Control and Prevention (1994) Recommendations and Reports.
Guidelines for preventing the transmission of mycobacterium tuberculosis in healthcare facilities. MMWR 43:1
86. National Institute for Occupational Safety and Health. NIOSH recommended guidelines for personal respiratory protection of workers in health-care facilities potentially
exposed to tuberculosis. 155. (1992) Atlanta, Georgia, US Department of Health and
Human Services
87. Wenzel RP (2003) Airline Travel and Infection. N Engl J Med 334:981
88. Anonymous. Guidelines enable health authorities to assess risk of tuberculosis transmission aboard aircraft. Cabin Crew Safety 33 [4-5]. (1998) Flight Safety Foundation
89. Whitlock G, Calder L, Perry H (2001) A case of infectious tuberculosis on two longhaul aircraft ights: contact investigation. N Z.Med J 114:353
90. Hocking MB, Foster HD (2004) Common cold transmission in commercial aircraft:
Industry and passenger implications. Journal of Environmental Health Research 3:7

DOI 10.1007/b107243
Insecticide Use in Occupied Areas of Aircraft

Judith Murawski
Association of Flight Attendants-CWA, 501 3rd St NW, 2nd Flr., Washington, DC 20001,
USA
jmurawski@afanet.org
1
1.1
1.2
1.3
1.4
Insecticide Application to Control Domestic Insects . .

For Routine Control of Cockroaches and Other Insects
In Response to Reported Insect Sightings . . . . . . . .
For Seasonal Control of Particular Domestic Insects . .
In the Cargo Hold . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
170
170
171
171
171
2
2.1
2.2
2.3
2.4
Insecticide Application to Comply with Foreign Quarantine Regulations

Background and Justication for Spraying . . . . . . . . . . . . . . . . .
Contents of Sprays and Application Methods . . . . . . . . . . . . . . . .
Standards Intended to Limit Aircraft Occupants Exposure to Insecticides
Potential For Exposure to Insecticides on Aircraft . . . . . . . . . . . . .
.
.
.
.
.
172
172
174
175
177
3
3.1
3.2
3.3
3.4
3.4.1
3.4.2
3.4.3
Health Impact of Insecticide Application on Aircraft . . . . . .

Reports of Adverse Health Effects Filed by Aircraft Occupants .
Physiological Factors That Increase Individual Susceptibility . .
Health Impact of Exposure to Mixtures of Particular Insecticides
Toxicological Data . . . . . . . . . . . . . . . . . . . . . . . . . .
Pyrethroids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Permethrin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Phenothrin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
179
179
180
181
183
184
185
185
4
4.1
4.2
Potential Alternative Control Methods . . . . . . . . . . . . . . . . . . . .

Mechanical Disinsection Method to Control the Spread of Domestic Insects
Proposed Mechanical Disinsection Methods
to Satisfy Foreign Quarantine Regulations . . . . . . . . . . . . . . . . . .
185
185
186
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
188
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Abstract Insecticides are applied in the aircraft cabin for four key reasons: (1) to comply
with foreign quarantine regulations applicable to certain international ights; (2) to control insects in the aircraft galleys where food and food waste are stored; (3) to respond to
insect sightings reported by passengers or crew; and (4) to combat seasonal insect populations. Insecticide application related to foreign quarantine regulations has generated
the most controversy and concern for crew and passenger health. Forty seven countries
require that the cabin and cockpit of commercial aircraft are sprayed with insecticides,
either prior to or upon arrival, to protect against importing insects that may be on board
and may carry disease or damage the environment. Spraying practices vary widely between countries and airlines. Although the World Health Organization describes these
practices as safe if carried out with the recommended precautions, little or no attention
is paid to exposure control practices. Government agencies, labor unions, airlines, and
170
J. Murawski
environmental groups have received reports of ill health from passengers and crew, with
symptoms that range from rash to anaphylaxis. The current focus is on developing mechanical methods of disinsection that will satisfy countries quarantine concerns without
compromising the health of aircraft occupants.
Keywords Chemical disinsection DDT Mechanical disinsection Permethrin
Phenothrin
Abbreviations
DDT dichlorodiphenyltrichloroethane;
DEET diethyl m-toluamide;
DOT US Department of Transportation;
EPA US Environmental Protection Agency;
ICAO International Civil Aviation Organization;
IHR International Health Regulations;
USDA United States Department of Agriculture;
WHO World Health Organization
1
Insecticide Application to Control Domestic Insects
1.1
For Routine Control of Cockroaches and Other Insects
There is little available information on the insecticide products and application methods employed by airlines for routine domestic insect control. In the
US, the Environmental Protection Agency (EPA) does not approve any insecticides that are labeled for spray application in the cabin or cockpit, whether
occupied or unoccupied [1, 2]. However, there is evidence that products not
registered for aircraft application are applied on aircraft anyway. Maintenance
workers at one major US airline are instructed to remove galley supplies and
spray a product that contains resmethrin into cracks and crevices to drive
any insects into the open [3]. They must then activate a bug bomb that contains permethrin and piperonyl butoxide on the galley oor, and leave the
aircraft sealed and unoccupied for two hours, if possible [3]. Any dead insects are collected, and exposed surfaces in the galleys are washed with soap
and water before the galley supplies are replaced. Information is not available
on insecticides and application methods at other airlines, although it is unlikely that these are isolated practices. Airlines are not precluded from using
gel baits (e.g., abamectin, hydramethylon) and cockroach traps on aircraft because the approved uses are vague due to the perceived negligible exposure
risk. Several major airlines have reported the use of these products, particularly in the aircraft galleys.
171
1.2
In Response to Reported Insect Sightings
Airline industry ofcials have acknowledged the practice of nonroutine
spraying in response to insect sightings in the cabin. There are no data available on the frequency and nature of spraying; presumably, these will vary
considerably with season and destination. In response to insect sightings,
maintenance workers at one major airline are instructed to fog the cabin with
a product that contains 3% pyrethrum and 6% piperonyl butoxide, starting at
the aft end and moving forwards [4]. The lavatories, galleys, and coat closet
are sprayed, but not the cockpit. The cabin is left unoccupied for at least two
hours if possible, after which maintenance workers collect any dead insects,
and replace blankets, pillows, and galley supplies. Information is not available on insecticides and application methods at other airlines, although it is
unlikely that these are isolated practices.
1.3
For Seasonal Control of Particular Domestic Insects
Historically, passengers and crew on commercial aircraft operating on particular domestic ights in the US have had the potential for exposure to
a mixture of DDT (dichlorodiphenyltrichloroethane) and Sevin (1-napthyl
N-methylcarbamate). In 1963, the US Department of Agriculture (USDA) instituted requisite spraying of passenger and cargo compartments to control
the spread of ying Japanese beetles [5]. The policy applied to aircraft that
ew to the state of California and other agriculture-dependent states, departing from areas of the US with documented Japanese beetle populations
during the summer months. This policy was enforced beyond the 1972 US ban
of DDT until the early to mid-1980s. It was stopped in response to court action [5] and then replaced by a mechanical method of insect control (see also
Sect. 4.1).
1.4
In the Cargo Hold
There is limited available information on the specic insecticide products applied in the cargo hold. Baggage handlers and airline maintenance workers
may be exposed during the course of their workday. Although there is some
air exchange between the cargo hold and the cabin and cockpit due to air
pressure changes during ight, insecticide exposure in the cabin and cockpit as a result of contaminated cargo-hold air is expected to be minimal (see
also Table 1). However, on some regional aircraft, the cargo hold can be on the
same level and separated only by a curtain.
172
J. Murawski
2
Insecticide Application to Comply with Foreign Quarantine Regulations
2.1
Background and Justification for Spraying
Countries with established spraying requirements believe that the health of
their plants, animals, or people are at risk from insects that may be imported
via the passenger cabin and cockpit of commercial aircraft. Currently, 47
countries issue disinsection requirements, either for all or selected incoming ights [6, 7]. Cases of tropical disease in nonendemic (i.e., nonnative)
countries can be especially serious, because local doctors do not expect to
encounter such cases so diagnosis and treatment is slowed [8]. However, exposing crew and passengers to potentially high levels of insecticides in an
enclosed space has been shown to carry health risks of its own.
The practice of treating aircraft with insecticides to kill any hitchhiking insects that might carry tropical disease or damage plants aircraft
disinsection is not new. For example, Australia published its Quarantine
Act of 1908 with incoming ships in mind, and then later, incoming aircraft.
Similarly, in the 1930s, India ruled that incoming aircraft must be sprayed
with insecticides to prevent the importation of mosquito vectors of yellow
fever [8].
Industrialized countries are not typically considered to be sources of tropical disease vectors, but cases of tropical disease in these nonendemic countries have been described. For example, there were 89 reports of conrmed or
probable airport malaria in 12 nonendemic countries over a 30-year period
ending in August 1999 [8]. These people had recently traveled by air between
nonendemic countries but perhaps landed in an endemic country en route, or
lived near an airport, suggesting that aircraft were somehow to blame. There
are a variety of potential sources for the tropical diseases that have infected
travelers and nontravelers alike (Table 1).
Aircraft-related reports of invasive alien species describe insects in packing material, passenger baggage and other cargo, the cargo hold, the aircraft
cabin, and imported animals [9]. These imported insects would not necessarily have survived in the host country. Countries that have described
problems with invasive alien species (whether plants, animals, or viruses)
all report that they have detailed border entry controls and/or management
programs including: brochures handed out to passengers (both in-ight and
on disembarkation); websites; media advertisements; airport notices; quarantine declaration on arrival cards; information workshops and awareness
campaigns in schools and villages; detector dogs; disinsection and disinfection of aircraft; passenger, cargo, and baggage searches; shoe disinfectant
mats; light traps at airports; and post ofce mail inspection [9].
173
Table 1 Sources of invasive alien species and the impact of aircraft disinsection on risk
management
Source of invasive alien species
Impact of aircraft disinsection
The passenger cabin and cockpit of

commercial aircraft are potential sources
of insects that can infect aircraft occupants.
In theory, such insects can also escape
the aircraft and infect people or damage
plant life in neighboring communities,
either directly or via successful mating
with local insects. Insects can enter
the cabin and cockpit at the departure
point, or during a short stopover
in an endemic country in transit
between two nonendemic countries
(runway malaria) [8].
People that live near an international
airport or border may get infected
by local mosquitoes that were infected
by biting local travelers, migrant
workers, or immigrants whose blood
was infected elsewhere.
The cargo hold is another source of
insects that may infect people at the
airport or in neighboring communities,
or damage local plant life. For example,
a mosquito that can carry dengue fever,
yellow fever, and other diseases was found
in cargo containers of lucky bamboo that
arrived by ship in California in early 2000.
The bamboo was stored in a few inches
of standing water that provided a perfect
breeding ground for the mosquitoes [11].
Spraying insecticides in the cabin and

cockpit may reduce the risk of this route
of disease transmission, but will not
eliminate it. Other options are to
implement programs for insect trapping
and sterile insect release. Aerial
spraying near major international airports
is not an uncommon practice.
These types of imported cases of tropical

disease are increasing [10], but
spraying insecticides in the cabin and
cockpit will not reduce this risk.
Spraying insecticides in the cabin

and cockpit will not reduce the risk of
importing insects in the cargo hold.
However, it is not uncommon or
unreasonable to inspect and spray the
unoccupied cargo holds, and restrict the
import of products that are associated
with particular pests.
The necessity and appropriateness of aircraft disinsection measures have

been sources of debate. Proponents say that the practice is necessary to prevent the introduction of tropical disease and insects that could damage crops
or compromise the host habitat [12]. The World Health Organization (WHO)
asserts that the spraying procedures are safe and necessary, if carried out
with the proper precautions [13]. One publication reports that spraying during three test ights achieved 100% mortality among mosquitoes and ies
kept in cages throughout the aircraft cabin, even when the air conditioning
system was operating [14].
174
J. Murawski
Opponents of aircraft disinsection say that it is inappropriate to spray

without requirements to ensure passenger and crew exposure to the insecticides is minimized, that passengers should be informed prior to ticket
purchase, and that requisite visitor inoculations and integrated pest management programs at airports may be more effective than spraying insecticides
in the occupied areas of the aircraft.
A scientic report commissioned by a government with disinsection requirements on all incoming ights commented that disinsection is an excellent example of quarantine being applied in response to a general biological
principle, with little or no attempt to assess the risk [15]. Another report
recommended that the practice of aircraft disinsection be discontinued, noting that many international travelers regard disinsection as an unacceptable
imposition, adding that, aircraft disinsection will not address the disease
burden caused by large numbers of people carrying organisms of malaria
and a range of viruses that arrive every month [16].
Complicating matters further, the justication for spraying is often not
documented, and most countries enforce their spraying requirements for all
arriving aircraft, without concern for risk related to season or point of departure. Complexity for airline operators and quarantine personnel has been
cited as one reason for blanket spraying requirements [17]. The fact that risks
can change over time with the migration of species is presumably another
consideration. In the process, however, thousands of crewmembers and passengers may be exposed to insecticides each day with little or no justication.
The emphasis now is on developing mechanical methods of aircraft disinsection that will satisfy countries concerns without compromising the health
of passengers and crew [18, 19] (see Sect. 4.2).
2.2
Contents of Sprays and Application Methods
The vast majority of countries with disinsection requirements specify that the
aircraft must be treated with a solution that contains a pyrethroid insecticide
(typically permethrin), either in the occupied or soon-to-be-occupied aircraft
cabin and cockpit [7].
In 1985, the WHO formally recommended that aircraft disinsection products be free from human toxicity with insecticidal action not less than
a product that contains 3% DDT, 1.6% pyrethrum, 7.5% xylene, 2.9% petroleum distillate, and 85% CFCs [20]. To this end, the WHO classied permethrin and phenothrin as effective and safe. A recent survey conducted by the
International Civil Aviation Organization (ICAO) indicated that, with few exceptions, these are the insecticides of choice [7]. An analysis of the sprays
applied on a major airline indicated that the sprays also contain solvents such
as methylene chloride and benzene-based chemicals [21].
175
The primary insecticide application method is referred to as in-ight, topof-descent, or upon arrival. Some foreign quarantine authorities recommend
that a cabin crewmember rst announce the procedure, typically minutes before they must walk through the cabin spraying over passengers heads. For
example, the announcement at one airline informs passengers that the cabin
is about to be treated using a nontoxic spray recommended for this purpose by the World Health Organization. This airline used to require its cabin
crew to inform passengers that the spray contains no harmful chemicals, but
[they] may prefer to cover food and beverages [22].
After the announcement, members of the cabin crew are required to slowly
walk down the aisles, releasing the spray from a pressurized can over the
passengers heads. If cans are not available on-board, then agriculture agents
spray the occupied cabin upon arrival while the aircraft doors are kept
closed. Alternatively, cabin crewmembers are instructed to spray the occupied cabin prior to departure after the doors have been closed (blocks away
application).
Countries that require disinsection endorse some variation of in-ight application, but some will accept an alternative residual treatment. The residual
application method involves insecticide applicators that board the aircraft
and spray every surface the passenger seats, jumpseats, carpets, galley
surfaces, overhead bins, cockpit, crew rest bunks, and lavatories with an
insecticide and solvent concentrate that has been diluted with water in large
drums. The process takes place shortly before crew and passengers board,
without their knowledge. Once an aircraft has been residually treated, quarantine ofcials will allow it to land without additional insecticide application
for the next 56 days [12].
Although the ICAO survey published in 2001 suggests that pyrethroids are
now almost exclusively the insecticides of choice, there is evidence that DDT
was applied on aircraft to comply with foreign quarantine regulations from
the late 1950s until the late 1970s [23, 24], and perhaps longer particularly in
developing countries given the effective endorsement of on-aircraft application of DDT by the WHO in 1985 [20].
2.3
Standards Intended to Limit Aircraft Occupants Exposure to Insecticides
There are no domestic regulations to limit exposure to insecticides on aircraft. In the US, for example, the EPA does not approve any insecticides
specically for application on aircraft, having expressed doubt that the benets exceed the risks [1, 2]. However, insecticide products registered by
another countrys environmental agency are applied outside of EPA jurisdiction, on US-registered aircraft carrying US-based crew and passengers, to
comply with foreign quarantine regulations. These insecticide-treated aircraft
are also routed on domestic ights. The US Federal Aviation Administration
176
J. Murawski
assumed authority of occupational safety and health standards for crewmembers [25] but has not published any exposure control standards for aircraft
disinsection. This regulatory quagmire is not unique to the US; worldwide,
insecticides are applied on aircraft outside of the jurisdiction of the relevant
environmental or agricultural authority, and no government agency takes
direct responsibility for the health of crew or passengers in ight. The US
Department of Transportation (DOT) did propose a rule that would have
required passenger notication prior to ticket purchase, but it met strong
resistance from the travel industry and was never published [26].
ICAO, sister agency to the WHO under the United Nations, publishes minimum standards for aviation safety, health, and security. Key disinsection
standards are listed in Table 2 [27]. ICAO standards are internationally recognized, but they are not accompanied by guidance material, they tend to be
general in nature, and there is no enforcement mechanism.
The WHO also publishes nonenforceable standards for aircraft disinsection. The WHO International Health Regulations (IHR) require that disinsecting, deratting, and other sanitary operations shall be carried out so as
not to cause undue discomfort to any person, or injury to his health [28].
A WHO committee recently recommended that the IHR denition of aircraft
disinsection be enlarged to include procedures that control (not just kill) insect vectors of human diseases. This is a signicant shift, as it would allow
approved means of mechanical disinfection [72]. The IHR also dictate when
Table 2 Key international disinsection standards published by ICAO under Annex 9
2.22Contracting States shall limit any routine requirement for the disinsecting of
aircraft cabins and ight decks with an aerosol while passengers and crews are on
board to same-aircraft operations originating in, or operating via, territories that
they consider to pose a threat to their public health, agriculture, or environment.
2.23Contracting States that require disinsection of aircraft shall periodically review
their requirements and modify them as appropriate, in light of all available evidence
relating to the transmission of insects to their respective territories via aircraft.
2.24When disinsection is required, a Contracting State shall authorize and accept only
those methods and insecticides, whether chemical or nonchemical, which are
recommended by the World Health Organization. Note: This provision does not
preclude the trial and testing of other methods for ultimate approval by the
World Health Organization.
2.25Contracting States shall ensure that their procedures for disinsecting or any other
remedial measure are not injurious to the health of passengers and crew and cause the
minimum of discomfort to them.
2.26Contracting States shall, upon request, provide appropriate information, in plain
language, for air crew and passengers, explaining the pertinent national regulations,
the reasons for the requirement, and the safety of properly performed aircraft
disinsection.
177
disinsection is necessary, such as when an aircraft or ship is departing a country where mosquito-borne disease is endemic, and is bound for a nonendemic
country [29]. However, countries can and do opt out of this provision, without
penalty [30].
2.4
Potential For Exposure to Insecticides on Aircraft
The majority of the reports submitted by passengers, crewmembers, and
their physicians describe symptoms (1) during and after in-ight spraying, or
(2) during and after one of the two ight legs that follow residual treatment of
the cabin and cockpit. Many of the complaints cite damp surfaces and a distinct odor of insecticides in the crew rest compartments [31, 32]. The many
possible routes of entry into the body lungs, skin, and stomach suggest
that exposure to the insecticides and solvents can be considerable, especially
on long-haul ights.
The potential for airborne exposure to permethrin extends beyond the
time immediately following application because carpeting and upholstered
furniture can absorb and later release permethrin back into the air. A study of
permethrin-contaminated carpet found that carpet removal had a signicant
impact on health improvement relative to nonremoval, even when the carpet
had not been treated for 10 years [33]. The researchers concluded that indoor contamination of permethrin is highly persistent and may be the cause
of adverse health effects.
Dermal exposure, either during in-ight spraying or after residual spraying if surfaces are not dry when passengers and crewmembers board the
aircraft, is another concern. The absorbed dose of both crewmembers and
passengers (including an assessment of the impact of frequent, routine, and
prolonged exposures) has not been addressed, and may account for some of
the major differences in the expectation and experience of safe versus toxic.
A 2-year investigation into doctors reports of insecticide-related illness
among crewmembers that work in a residually sprayed cabin concluded that
post-disinsection aircraft ventilation procedures and administrative measures did not effectively limit exposure, and that current assumptions about
the human health impacts of residual disinsection underestimate the risks
of this procedure [34]. A mathematical model estimated that 45 minutes
after the residual application ended, when crewmembers were expected to
board, the airborne concentration of permethrin in the cabin would approach
6 mg/m3 if there were no mechanical ventilation, which was not an unusual
practice at the airline in question [34].
An exposure assessment during in-ight spraying was commissioned
to assist the German Federal Institute for Risk Assessment in developing
a method of disinsection that is effective, but properly protects passengers
and crew (Table 3) [35]. Between 107 and 204 g of spray containing 1.25%
178
J. Murawski
Table 3 Exposure data for in-ight spraying with 107204 grams of a standard reference
aerosol containing 0.31% pyrethrin and 2.6% piperonyl butoxide
Type of sampling
Agent
Concentration
Airborne
Airborne
Aircraft surfaces
Aircraft surfaces
Inhalation
Inhalation
Pyrethrin
Piperonyl butoxide
Pyrethrin
Piperonyl butoxide
Pyrethrin
Piperonyl butoxide
Median 21 g/m3 (380 g/m3 )

285 g/m3 (54581 g/m3 )
38.5 ng/cm2
22 ng/cm2
20 g
116 g
pyrethrum extract (insecticide), of which 0.31% was pyrethrin and 2.6%

piperonyl butoxide (synergist), were released into the cabin of an A310 aircraft in six separate trials, each time with the air conditioning system operating. The exposure monitoring revealed that the upper and lower arms of
the person applying the spray were most strongly contaminated and that
the dermal exposure of the passengers was about half that of the crewmember assigned to spray the cabin, being most pronounced on their heads and
thighs. The in-service insecticide exposures may be underestimated depending on the destination; Australia and New Zealand, for example, require that
an A310 cabin be treated with a smaller quantity (100 g) of a more concentrated aerosol (2% phenothrin), and that for pre-embarkation or on-arrival
spraying, the air conditioning system be turned off, both during the spraying
and for 5 to 15 min afterwards [12].
The WHOs ofcial position is that, if carried out with the recommended
precautions, aircraft disinsection methods should not carry any relevant
health risk [13]. Notwithstanding the range of individual sensitivity to insecticide products, there is evidence that application methods and exposure
control practices can vary dramatically between airlines and countries. The
precautions recommended by the WHO and ICAO are not always met.
For residual treatment, the WHO recommends that the cabin and cargo
compartment of a B747 aircraft be treated with 25 L of a 2% emulsion of
permethrin [36]. The Australian quarantine authority recommends 33 L of
a 2% solution for the same aircraft [12]. A recent investigation into evidence of insecticide-related ill health among cabin crew at one major US
airline describes how the airline applies 35.4 L of a 2.2% permethrin solution to each 747 aircraft [34], depositing 56% and 18% more permethrin
than recommended by the WHO or Australian quarantine authority, respectively. Reports from cabin crewmembers at different airlines indicate
that the ventilation period between the end of the spraying and the time
that crewmembers are required to board the aircraft can vary between
179
45 min and 24 h. ICAO Standard 2.25 is intended to limit insecticide exposure on aircraft, but it does not dictate specics, such as how long an
aircraft must be ventilated following residual treatment prior to reentry.
For in-ight spraying, both the WHO [36] and the Australian quarantine authority [11] recommend a maximum 400 g of aerosolized spray
that contains 2% of d-phenothrin on a 747-400 aircraft; the Australian
authority explicitly states that this amount must not be exceeded. However, one major US airline requires its cabin crewmembers to empty two
341-g cans of a 2% d-phenothrin spray over the passengers heads, delivering 59% more d-phenothrin than necessary or recommended [37]. In
contrast, cabin crewmembers at some airlines describe their practice of
deliberately emptying the cans of spray down the lavatory, for example, to
avoid having to spray the passengers and walk through a plume of insecticide mist.
Overexposure may carry serious implications, not only for occupant
health, but for aviation safety: one commercial pilot reported such a severe
allergic reaction during a freshly sprayed ight that he was having difculty
concentrating and was making mistakes upon landing such that the safety
of ight was compromised [31].
Another exposure risk factor, particularly for crewmembers, is the impact
of being exposed repeatedly because this can provoke a magnied physiological response [38]. Crewmembers bid for their trips by seniority, and tend to
keep the same schedule when possible. Long-haul international ights are the
most popular because a crewmember spends less time away from home in
a given month and makes more money.
Exposure control measures are available. These include: (1) mechanical
means of disinsection; (2) not spraying the cabin and cockpit when occupied;
(3) purchasing an extra set of crew bunk mattresses and treating them off the
aircraft in advance so that they can dry properly before being used; (4) ensuring that the cabin and cockpit are truly dry and odor free prior to boarding
crewmembers and passengers; and (5) notifying passengers of any spraying
requirements prior to ticket purchase.
3
Health Impact of Insecticide Application on Aircraft
3.1
Reports of Adverse Health Effects Filed by Aircraft Occupants
There are few epidemiological studies into either the acute or chronic effects
of exposure to insecticides applied on aircraft specically. Unfortunately, although pyrethroid metabolites can be measured in urine one to three days
180
J. Murawski
following exposure, detection only conrms exposure, not health effects.

Also, the levels of metabolites do not correlate with the severity of health effects [73]. Similarly, blood biomarkers may be measured, although the number of facilities that conduct this testing is limited and the levels are unlikely
to correlate predictably with health effects [73, 74].
The most common acute symptoms recently reported by crewmembers
exposed to pyrethroids were cardiovascular, dermatological, gastrointestinal, neurological, ocular, and respiratory [34]. In addition to the potential
for acute illness, the potential for cumulative or chronic health problems
has been acknowledged [34]. In administrative legal decisions, judges have
awarded total and permanent disability status to crewmembers exposed to
high levels of insecticides in the aircraft cabin [32]. Judges have ruled that
crewmembers sustain[ed] injury, internal and neurological, as a result of exposure to toxic substances arising out of and occurring in the course of their
employment. The only explicit reference in the literature to a chronic health
effect and aircraft disinsection specically is a moderate association between
self-reported application of DDT on aircraft and an elevated risk of breast
cancer among female ight attendants [39].
Government agencies, labor unions, airlines, and environmental groups
have received reports of symptoms that passengers and crewmembers, or
their physicians, have attributed to insecticide exposure for many years.
When the US Centers for Disease Control amended its aircraft disinsection
rule to discontinue routine spraying on ights arriving in Hawaii, it referred
to reports of anaphylaxis and other severe allergic reactions from people
exposed to the insecticides applied on aircraft [40]. The EPA has received reports of symptoms ranging from nausea to neurological decits, noting that
it is impossible to prevent dermal or inhalation exposures during an inight cabin iniction [41]. The Northwest Coalition for Alternatives to Pesticides [41] and the National Pesticide Telecommunications Network describe
similar reports. In 1995, when the DOT proposed a regulation to ensure that
passengers be informed of in-ight insecticide spraying requirements prior to
paying for their tickets, it referred to numerous letters with complaints, and
highlighted the discrepancy between the intended use of in-ight spray and
the health hazard warnings on the label [26].
The Association of Flight Attendants labor union received reports of
insecticide-related ill health from crewmembers and passengers on more
than 200 ights in one year (20002001) at a single airline [31]. The Airline Pilots Association labor union has also received reports of ill health and
formally acknowledged the detrimental health effects that have been documented by their members [42].
181
3.2
Physiological Factors That Increase Individual Susceptibility
People with preexisting immune system disease, as well as infants and children, may be more sensitive to the permethrin, which is the active ingredient
in the residual sprays [17]. There is also evidence that certain pyrethroids
may affect early neurologic and reproductive development [43]. A signicant
association between exposure to indoor insecticides in utero and childhood
cancer has been described [44]. Pyrethroid-induced dysfunction of the permeability of the blood-brain barrier in rats in utero may raise concerns for
pregnant iers who are not informed of insecticide spraying requirements
prior to a given ight [75]. This is a concern because babies and toddlers are
known for their tendencies to play on the oor and insert objects in their
mouths, including upholstery.
Butylcholinesterase (also called pseudocholinesterase or serum cholinesterase) is an important enzyme in the metabolism of pyrethroid insecticides. Reduced levels of this enzyme have been observed in people that
have various congenital deciencies, acute infection, heart disease, or use
oral contraceptives [45, 46], putting them at increased risk for the adverse effects associated with exposure to pyrethroids. Similarly, menstruation [47],
pregnancy, age, obesity, some drug therapy, and liver disease [48] have been
associated with a reduction in the circulating levels of butylcholinesterase.
Inter-individual variation in levels of other relevant enzymes have been identied, differences that may have profound effects on susceptibility to toxic
effects [76].
Passengers and crewmembers are not informed of insecticide application
in advance, so are left to their own devices to protect themselves during
a ight. Passengers are typically not even permitted to leave the aircraft prior
to insecticide application in the occupied cabin when the aircraft is still at
the gate. Even if people are aware that particular countries enforce insecticide spraying rules, residually treated aircraft are own on both domestic and
international routes where spraying is not required.
3.3
Health Impact of Exposure to Mixtures of Particular Insecticides
Coexposure to compounds that inhibit the enzyme carboxyesterase (e.g., tricresylphosphates) can signicantly increase the toxicity of pyrethroids [21,
38, 49]. Tricresylphosphates are ingredients in aircraft engine oils and most
hydraulic uids, and can contaminate the aircraft air supply systems (see also
Chap. 1.3 and Chap. 7). Coexposure to compounds that inhibit the mixedfunction oxidase systems (e.g., piperonyl butoxide) can also signicantly increase the toxicity of pyrethroids [21, 38]. Piperonyl butoxide is a popular
182
J. Murawski
Table 4 Basic chemical information and chemical structures

Chemical name and information
Carbaryl (Sevin)
carbamate insecticide C12 H11 NO2
1-naphthalenol methylcarbamate
CAS No. 63-25-2
DDT
organochlorine insecticide C14 H9 Cl5
dichlorodiphenyltrichloroethane
CAS No. 50-29-3
Diethyl m-toluamide (DEET)

insect repellent C12 H17 NO
N, N-diethyl-3-methylbenzamide
CAS No. 134-62-3
Malathion
organophosphate insecticide C10 H19 O6 PS2
1,2-di(ethoxycarbonyl)ethyl O, O-dimethylphosphorodithioate
CAS No. 121-75-5
Permethrin
pyrethroid insecticide C21 H20 Cl2 O3
3-(2,2-dichloroethenyl)-2,2dimethylcyclopropanecarboxylic acid
(3-phenoxyphenyl)methyl ester
CAS No. 52645-53-1
Chemical structure
183
Table 4 (continued)
Chemical name and information
Chemical structure
Phenothrin
pyrethroid insecticide C23 H26 O3
2,2-dimethyl-3-(2-methyl-1propenyl)cyclopropanecarboxylic acid
(3-phenoxyphenyl)methyl ester
CAS No. 26002-80-2
Piperonyl butoxide
unclassied, synergist C19 H30 O5
5-[2-(2-butoxyethoxy)ethoxymethyl]-6
-propyl-1,3-benzodioxole
CAS No. 51-03-6
additive in aircraft insecticide formulations because it is inexpensive and intensies the insecticidal activity of pyrethroids.
Laboratory animals exposed to a combination of permethrin, malathion,
and the insect repellent DEET (diethyl m-toluamide) exhibited greater
impairments in neurobehavioral tests than when exposed to permethrin
alone [50], suggesting the possibility for a magnied physiological response
when exposed to permethrin and other insecticides or repellents that may be
applied on either domestic or international ights. Stress may exacerbate the
effects of coexposure to low doses of permethrin and other chemicals [51].
3.4
Toxicological Data
Basic chemical information for the insecticides and insect repellent that are
referenced in this section is presented in Table 4 [52]. Overexposure to solvents and propellants has received very little attention and, alone or in combination, may explain some of the symptoms described.
Information on the relative toxicities of the insecticides and insect repellent that are referenced in this chapter is presented elsewhere [53].
Pyrethroids are generally characterized as having high insecticide toxicity
and low mammalian toxicity, and are therefore considered preferable to other
broad-spectrum insecticides. Despite this, a number of acute and chronic
184
J. Murawski
health effects attributed to exposure to pyrethroids generally, or permethrin

or phenothrin specically, have been identied.
3.4.1
Pyrethroids
A WHO report on the safety of pyrethroid use across industries summarizes
reports of burning skin, fatigue, itching, dizziness, headache, and nausea
following accidental overexposure in the workplace [13]. The last three symptoms are attributed to organic solvents. The WHO summary of animal studies
describes carcinogenic, endocrine, immunotoxic, neurobehavioral, neurotoxic, and reproductive effects, usually at doses described as above those
recommended by the WHO.
A series of case reports that describe chronic effects of acute pyrethroid intoxication reported that the most frequent symptoms to remain 2 years after
exposure were autoimmune disease, immune system deciencies, and neurological symptoms (e.g., reduced intellectual performance and endurance
during mental work, personality disorder, visual disturbances, and neuropathy) [54]. Regular exposure to insecticides has been associated with an increased risk of mild cognitive dysfunction [55], although the specic insecticide products were not dened. Severe neurologic effects such as seizures
and loss of consciousness can follow heavy exposures to pyrethroids [56]. Two
cases of hemoptysis following exposure to pyrethroids have been described in
the literature; one patient reported severe anemia, the other, impaired respiratory function [77, 78].
Two-thirds of surveyed workers assigned to package pyrethroids in
a manufacturing plant reported abnormal facial sensations, described as
feelings of burning, tingling, itching, tightness, or numbness [57]. These
symptoms typically developed within 30 min of exposure and did not last
longer than 24 h. One third of the workers reported upper respiratory complaints, and on examination, 14% had a skin rash, mainly on the face and
chest. Physical examinations and laboratory tests did not identify other abnormalities.
Sixty-four cases of chronic pyrethroid intoxication were reported to the
German Federal Health Ofce in one year, six of which presented symptoms
classied as multiple chemical sensitivity syndrome with which a causal
link with pyrethroid exposure was either established or could not be ruled
out [58, 59]. Chronic illness as a result of low-level exposure to pyrethroids
has been a source of ongoing debate in Germany [79].
Five of nine workers that entered their ofce building two days after a solution of 0.25% pyrethroid had been injected into the ventilation ducts experienced dizziness, headaches, nausea, vertigo, and shortness of breath [60].
In vitro tests have indicated the ability of pyrethroids to be endocrine disruptors, relating to estrogen, progesterone [61, 62], and androgen action [63].
185
Estrogens inuence reproductive development, aging, and carcinogenesis.

Androgens inuence the development of male sex characteristics.
3.4.2
Permethrin
Permethrin has been recognized as an irritant to the eyes, skin, and respiratory tract [64].
The results of a recent animal study of dermal exposure to permethrin
suggest that, while exposure to real-life doses may produce no overt signs of
neurotoxicity, signicant neurobehavioral decits and neuronal degeneration
are induced in the brain [50]. Animal studies have implicated permethrin in
damage to the parts of the brain that are selectively affected in Parkinsons
disease [65]. A 59-year-old man who sprayed two cans of ant-killer aerosol
containing a total of 9 g chlordane and 454 mg permethrin in an unventilated
room lost consciousness for half an hour and then, two weeks later, developed
motor neuron disease [66]. The specic cause of his disease was not identied, but the authors recommended further investigation.
3.4.3
Phenothrin
Phenothrin is a suspected kidney toxicant and a suspected neurotoxicant [67].
Antiandrogenic effects of phenothrin have been described, and exposure to
phenothrin-containing delousing spray used on clothes and bedding has been
suggested as the cause of an epidemic of gynecomastia (breast development)
among Haitian refugees in a US detention center [68].
4
Potential Alternative Control Methods
Mechanical disinsection methods are dened as those that utilize physical
means to keep aircraft free of insects of concern. Mechanical means must
address concerns over importing insects that spread disease or damage the
environment, without exposing crewmembers and passengers to insecticides,
solvents, and propellants.
Even though the WHO endorses pyrethroid insecticides for aircraft application, it has actively investigated insecticide alternatives off aircraft due
to insecticide resistance, a decreased acceptance of spraying, environmental
concerns, and the rising costs of insecticides [69].
186
J. Murawski
4.1
Mechanical Disinsection Method to Control the Spread of Domestic Insects
Since the mid-1980s in the US, mechanical disinsection has been successfully
applied on cargo aircraft that operate during the summer months between
certain states to control the spread of Japanese beetles, replacing the application of a mixture of DDT and Sevin (see also Sect. 1.3). Specically, during
the beetle season, the USDA requires the airlines to hang heavy curtains made
of overlapping strips of clear plastic at the top of equipment used to load
cargo and board pilots on cargo aircraft located in states that host Japanese
beetle populations, and intended to y to western states that are agriculture
dependent and Japanese beetle free [70]. These plastic curtains (excluders)
have proven highly successful at keeping these ying insects out of the aircraft. The USDA reserves the right to spray with pyrethroid insecticides if
Japanese beetles are spotted on board. However, ofcials report that sightings
are extremely rare.
Alternative materials may be better suited to keeping other pests, such as
mosquitoes, off the aircraft. Curtains made of overlapping strips of chemically treated mosquito netting may be the most effective option at aircraft
service doors. Air blowers stationed in the passengers boarding bridge and
blowing into the aircraft may be the most suitable option to prevent any ying insects from leaving the aircraft trough the passenger boarding door (see
Sect. 4.2).
4.2
Proposed Mechanical Disinsection Methods
to Satisfy Foreign Quarantine Regulations
Air blowers intended to prevent ying insects from entering or leaving
aircraft have recently been tested under laboratory conditions at a USDA
laboratory [18]. Preliminary results indicate a success rate of 99% at excluding a combination of three species of mosquitoes and the common
housey [71].
The results of the USDA testing suggest that such air blowers may be
used at the aircraft end of the passenger boarding bridge to blow air into
the aircraft, thereby preventing any ying insects from deplaning upon arrival in a country with disinsection rules. The forced air must be provided
at a suitable angle and velocity to repel any mosquitoes (or other insects
of concern) that may y towards the opening. A means to ensure that the
airblowers do not pressurize the aircraft must also be implemented. Keeping the service door open, but covered with a simple net screen that can
be stowed during ight and quickly mounted upon arrival, is one viable
option.
187
Similar blower equipment could be installed in passenger boarding bridges

prior to departing for a country with aircraft disinsection rules to remove any
errant insects that may be hiding on the clothing or skin of passengers or
crew.
A suitable mechanical disinsection technology, whether air blowers or
treated net curtains will also need to be tested at aircraft service doors because all doors must be protected. Necessary structural modications to
loading equipment (e.g., passenger boarding bridges, staircases, scissor lifts,
and conveyors for loading supplies) will have to be made in advance.
Mechanical means of aircraft disinsection may be considered as part of an
integrated pest management program. Complementary pest control measures include import restrictions, prompt removal and retention of food
waste, and insect surveillance at airport facilities [17]. In some cases, requisite
vaccinations for visitors may also be appropriate.
Aside from the obvious health benets to passengers and crew, mechanical
means of disinsection boast many benets compared to chemical spraying.
For example, after initial development and equipment cost, air curtains are
inexpensive to maintain, and it would no longer be necessary to purchase
chemical spray products. Airlines need not budget expensive ground time for
ventilating the aircraft, which is necessary after residual spraying, nor must
they enforce in-ight or pre-ight spraying that is so unpopular with passengers and crew. Finally, mechanical means are tourism friendly and would
not generate the complaints from crewmembers and passengers, the lostwork time, the workers compensation costs, and the lawsuits, all generated
by chemical spraying.
For countries to adopt these methods, it will be necessary for their specialists to test equipment and materials suitable for the particular insects of
concern, and a means to certify aircraft as protected prior to arrival. To this
end, ICAO intends to coordinate with the countries in question and the WHO
to evaluate and document assessments of the efcacy, practicability, and cost
effectiveness of alternatives to the currently endorsed chemical methods of
aircraft disinsection, and where appropriate, to record the approval of such
alternatives [18]. One option is for a country to accept incoming aircraft protected with mechanical means of disinsection but, similar to the US policy
for ights arriving in Hawaii [40], reserve the right to spray on individual
high-risk ights when deemed necessary.
Acknowledgements The author thanks crewmembers for their invaluable insights into
their working environment, especially G. Lerno, D. Brown-Dodson, C. Sawyer, D. Casey,
G. McCann, J. Popko, P. Johnson, and K. Bray. As well, thanks to Dr. Martin Hocking at
the University of Victoria, Canada, for his expert editorial work.
188
J. Murawski
References
1. US Environmental Protection Agency (14 May 1996) Pesticide reregistration notice
96-3. Ofce of Pesticide Programs, Washington, DC
2. US Environmental Protection Agency (19 Apr 2004) Personal communication with
Mr. M. Johnson, Chief, Insecticide Branch, Ofce of Pesticide Programs Registration
Division, Arlington, VA
3. US Airways (2004) Maintenance card no. J012-32-2 (galley)
4. US Airways (2004) Maintenance card no. J012-32-1 (cabin)
5. Aviation Consumer Action Project et al. vs. United States Department of Agriculture
(10 Nov 1977) Complaint in civil court, civil action no. 77-1941, US District Court for
the District of Columbia
6. International Civil Aviation Organization (2001) FALP/3-WP/12
7. International Civil Aviation Organization (2001) FALP/3-IP/1
8. Gratz NG, Steffen R, Cocksedge W (2000) Bulletin of the WHO 78:995
9. International Civil Aviation Organization (2004) FAL/12-WP/29
10. Schlagenhauf P, Steffen R, Loutan L (2003) J Travel Med 10:106
11. Kramer V (2001) Presentation: Mosquitoes and public health in California. Vector
Borne Disease Section, California Department of Health Services, Oakland, CA
12. Australian Quarantine & Inspection Service and New Zealand Ministry for Agriculture and Forestry Quarantine (18 Mar 2004) Schedule of disinsection procedures
13. World Health Organization (2003) Safety of pyrethroids for public health use an
update (draft report). WHO, Geneva
14. Russell RC, Paton R (1987) Bulletin of the WHO 67:543
15. Australian Department of Primary Industries and Energy (1988) Lindsay Review,
Canberra, Australia
16. Nairn ME, Allen PG, Inglis AR et al. (1996) Australian quarantine a shared responsibility. Department of Primary Industries and Energy, Canberra. Australian
Quarantine Review Secretariat, Commonwealth of Australia
17. Naumann ID, McLachlan K (1999) Aircraft disinsection a report commissioned
by the Australian Quarantine & Inspection Service. Plant Protection Branch, National Ofce of Animal and Plant Health, Department of Agriculture, Fisheries, and
Forestry, Australia
18. International Civil Aviation Organization (2004) FAL/12-WP/117
19. International Civil Aviation Organization (2004) FAL/12-WP61
20. World Health Organization (15 Feb 1985) Weekly Epidemiological Record 60:4552
21. van Netten C (2002) Sci Total Environ 293:257
22. United Airlines (2002, 1994) Flight attendant announcement books. Chicago, IL
23. Wartenberg D, Stapleton CP (Aug 1997) Abstract presented at the 9th Annual Conference of the International Society of Environmental Epidemiology, Taipei, Taiwan
24. Hinman AR (1994) Statement before the Subcommittee on Aviation, Committee on
Public Works and Transportation, US House of Representatives, 103rd Congress, 2nd
Session. In: Airliner Cabin Air Quality No. 103-61. US Government Printing Ofce,
Washington, DC
25. US Federal Register (1975) 40:29114
26. US Federal Register (18 Jan 1995) 60 FR 35963598
27. International Civil Aviation Organization (2002) Annex 9 to the convention on international civil aviation
28. World Health Organization (1969) International health regulations, part IV, chapter 1,
article 25 (1)(a)
189
29. World Health Organization (1969) International health regulations, articles 67 and 83
30. World Health Organization (1969) International health regulations, annex 2
31. Murawski J (2002) Paper presented at the 19th Annual SCSI International Aircraft
Cabin Safety Symposium, Burbank, CA
32. International Transport Workers Federation Civil Aviation Section (2004) Aviation
Safety & Health Database International, London
33. Prhl A, Boge KP, Alsen-Hinrichs C (1997) Environ Health Perspect 105:844
34. California Department of Health Services Occupational Health Branch (2003) Investigative report: occupational illness among ight attendants due to aircraft disinsection, Oakland, CA
35. Berger-Preiss E, Koch W, Behnke W et al. (2004) Int J Hyg Environ Health 207:419
430
36. World Health Organization (1995) Report on the informal consultation on aircraft
disinsection. WHO/PCS/95.51 International Program On Chemical Safety
37. Association of Flight Attendants, AFL-CIO (15 Dec 2000) Personal correspondence
with United Airlines contact person for aircraft disinsection, San Francisco, CA
38. US National Research Council Committee on Air Quality in Passenger Cabins of Commercial Aircraft (2002) The airliner cabin environment and the health of passengers
and crew. National Academy Press, Washington, DC
39. Wartenberg D, Stapleton CP (1998) Br Med J 316:1902
40. US Federal Register (12 Oct 1979) 44 FR 5891158912
41. US Congress (1994) Hearing before the Subcommittee on Aviation Committee on
Public Works and Transportation, US House of Representatives, 103rd Congress, 2nd
Session. In: Airliner cabin air quality no. 103-61. US Government Printing Ofce,
Washington, DC
42. ALPA International 88th Regular Executive Board Meeting (Oct 2001) Airline
Pilots Association, Delegate Committee Recommendation: Disinsection of aircraft by
foreign contracting states. Delegate Committee #4, Agenda Item #5
43. Landrigan PJ, Claudio L, Markowitz SB et al. (1999) Environ Health Perspect 107:431
44. Ma X, Bufer PA, Gunier RB et al. (2002) Environ Health Perspect 110:955
45. US National Library of Medicine and National Institutes of Health (2003) Medline
plus: serum cholinesterase. Bethesda, MD
46. Pasquariello CA, Schwartz RE (1993) Can J Anaesth 40:529
47. Davis ME, Yu EA, Fugo NW (1948) J Clin Epidemiol 666
48. Genc S, Gurdol F, Guvenc S, Kargi Y (1997) Eur J Clin Chem Clin Biochem 35:239
49. Ray DE, Forshaw PJ (2000) J Toxicol Clin Toxicol 38:95
50. Abdel-Rahman A, Dechkovskaia AM, Goldstein LB et al. (2004) J Toxicol Environ
Health A 67:331
51. Abdel-Rahman A, Abou-Donia S, El-Masry E et al. (2004) J Toxicol Environ Health A
67:163
52. Cambridge Soft Corporation (2004) Chemnder.com. Cambridge, MA
53. Patty FA, Clayton GD, Clayton FE, Battigelli MC et al. (2001) Pattys industrial hygiene
and toxicology, 3rd edn. Wiley Interscience, New York
54. Muhler-Mohnssen H (1999) Toxicol Lett 107:161
55. Bosma H, van Boxtel MPJ, Ponds RWHM et al. (2000) Lancet 356:912
56. He F, Wang S, Liu L et al. (1989) Arch Toxicol 63:54
57. He F, Sun J, Han K et al. (1988) Br J Ind Med 45:548
58. Altenkirch H (2000) Neurotoxicology 21:589
59. Altenkirch H, Hopmann D, Brockmeier B et al. (1996) Neurotoxicology 17:645
60. Lessenger JE (1992) J Toxicol Environ Health 35:261
190
J. Murawski
61.
62.
63.
64.
Go V, Garey J, Wolff MS et al. (1999) Environ Health Perspect 107:173

Garey J, Wolff MS (1998) Biochem Biophys Res Commun 251:855
Eil C, Nisula BC (1990) J Steroid Biochem 35:409
World Health Organization International Program on Chemical Safety/International
Labor Organization (2001) International chemical safety card 0312
Bloomquist JR, Barlow RL, Gillette JS et al. (2002) Neurotoxicology 23:537
Pall HS, Williams AC, Waring R et al. (1987) Lancet 9:685
Registry of Toxic Effects of Chemical Substances (1997) US National Institute for
Occupational Safety and Health, Cincinnati, OH
Brody SA, Loriaux DL (2003) Endocr Pract 9:370
World Health Organization (7 Nov 1986) Weekly Epidemiological Record 61:345
7 United States Code of Federal Regulations (1 Nov 1996) 301.48-4
US Department of Agriculture (9 Mar 2004) Air curtain technology for aircraft disinsection, presentation at Miami International Airport. Mosquito and Fly Research
Unit, Center for Medical, Agricultural, and Veterinary Entomology, Gainesville, FL
World Health Organization (2005) Third report of Committee A, 58th Assembly.
WHO, Geneva
Personal communication with Dr. R. Das, Public Health Medical Ofcer, Occupational
Health Branch (2005) California Department of Health Sevices, USA
Leng, Kuhn, Idel (1997) Sci Total Environ 12:173181
Sinha C, Agrawal AK, Islam F et al. (2004) Int J Dev Neurosci 22:3137
Mutch E, Blain PG, Williams FM (1992) Hum Exp Toxicol 11:109116
Kayser K, Plodziszewska M, Waitr E et al. (1998) Respirations 65:2148
Zaleska M, Zych J, Oniszh K (2001) Pneumonol Alergol Pol 69:20610
Kolaczinski JH, Curtis CF (2004) Food Chem Toxicol 42:697706
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.

DOI 10.1007/b107244
Aircraft Air Quality Incidents, Symptoms, Exposures

and Possible Solutions
Chris van Netten
Department of Health Care and Epidemiology, Division of Public, Occupational, and
Environmental Health, Faculty of Medicine, University of British Columbia, 5804
Fairview Avenue, Vancouver, V6T 1Z3, Canada
chris.van.netten@ubc.ca
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
194
Frequency of Air Quality Incidents in the Airline Industry . . . . . . . . .
196
Problems Associated with Obtaining Objective and Comparable Data . . .
196
Symptoms Associated with Air Quality Incidents . . . . . . . . . . . . . .
198
5
5.1
5.2
Source of Air in Aircraft . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Ram Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Bleed Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
198
199
200
6
6.1
6.2
6.3
Oil and Hydraulic Fluid Contaminants

Jet Engine Lubricating Oils . . . . . . .
Hydraulic Fluids . . . . . . . . . . . . .
Pyrolysis Products . . . . . . . . . . . .
.
.
.
.
200
200
200
201
7
7.1
7.2
Available Data and Required Data . . . . . . . . . . . . . . . . . . . . . . .

Problems in Capturing Rare Air Quality Events in Aircraft . . . . . . . . .
Benets to the Industry of Monitoring Changes in Air Quality . . . . . . .
202
202
203
8
8.1
8.2
8.3
8.4
Available Avenues of Obtaining Air Quality Measurements

Aircraft Filter Analysis . . . . . . . . . . . . . . . . . . . .
Coalescer Bag Analysis . . . . . . . . . . . . . . . . . . . .
Monitoring Aircraft Air Quality During Flight . . . . . . .
Indicators of Air Quality During Incidents . . . . . . . . .
.
.
.
.
.
204
204
204
204
205
9
9.1
Development of a New Air Monitor . . . . . . . . . . . . . . . . . . . . . .

Benets to the Industry from Monitoring for CO . . . . . . . . . . . . . . .
205
206
10
Standards of Exposure . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
206
11
Identication of Potential Synergistic Agents . . . . . . . . . . . . . . . .
208
12
Exposure Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . .
208
13
Alternate Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
209
14
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
209
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
209
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
in Aircraft
. . . . . . .
. . . . . . .
. . . . . . .
. . . . . . .
.
.
.
.
.
.
.
.
.
194
C. van Netten
Abstract Although air quality incidents in aircraft occur at low frequencies, ranging from
1 per 10 000 ights to 3.8 per 1000 ights depending on aircraft type, these are not rare
events considering there are close to 30 000 ights per day in the USA alone. An analysis
of the reports by pilots and ight attendants indicates that the majority of reported symptoms fall into the category of central nervous system impairment, followed by problems
with the respiratory system. In addition, the majority of mechanical problems that were
identied as the cause of these incidents were associated with oil contamination of the air
compressor stages of the engine and the auxiliary power unit (APU). In addition, in some
aircraft types, hydraulic uid contamination of the APU air intake was also frequently reported. Analysis of jet engine lubrication oils and hydraulic uids indicates these agents
can be a source of carbon monoxide and tricresyl phosphates. Exposure to either of these
agents has been linked to central nervous system impairment. Identication of contaminants released into the air during such incidents is virtually non-existent as it would
require a large number of air quality monitors to be placed on aircraft in order to capture
these rare events. As a solution to this problem a small inexpensive air sampler has been
developed that is self-contained and can be activated by anyone. This sampler also has
a direct-reading CO monitor that can be used to provide an objective criterion for triggering the air sampler during an event. The exposed sampler can then be forwarded to the
laboratory for analysis of oil contaminants using gas chromatographymass spectrometry
(GC-MS). In this fashion a data base can be accumulated that provides an objective measure of exposures during these incidents and whether these exposures can be linked to the
symptoms that have been reported by ight crew personnel. A GC-MS analysis has the
additional benet of identifying potential synergistic agents, such as the pesticides used
to disinsect aircraft.
Keywords Aircraft Air quality Tricresyl phosphates Carbon monoxide
Air sampling Hydraulic uids Jet engine oils Neurological symptoms
Synergistic effects
Abbreviations
ACGIH American Conference of Governmental Industrial Hygienists
APU
Auxiliary Power Unit
ATC
Air trafc controller
CAS no. Chemical Abstracts identication number
COHb Carboxyhaemoglobin
NAS
National Academy of Sciences
ppm
Parts per million on a volume/volume basis
STEL
Short-term exposure level
TCP
Tricresyl phosphate
TWA
Time-weighted average
1
Introduction
In any mode of transportation, events occur that inuence the air quality to
which passengers and crew members are exposed. The aircraft industry is no
Aircraft Air Quality Incidents, Symptoms, Exposures and Possible Solutions
195
exception to this [1]. In this industry these abnormal events can similarly be
traced to external sources, such as the aircraft taking in exhaust air from another aircraft on the tarmac, or from internal sources such as the malfunction
of components in the air supply and handling system. On occasion, exposures to air contaminants from different sources can combine, resulting in
unexpected synergistic effects. When this occurs exposed individuals could
experience severe symptoms at exposure levels considered quite acceptable
for each of the individual agents alone. Exposure levels to individual agents
are well established in the occupational setting: exposure levels for mixtures,
however, are rarely addressed [2] or are non-existent.
This chapter will limit itself to air quality incidents that are a result of some
malfunction within the aircraft air handling system. In addition, some reference will be made to potential synergists that one might also encounter in the
aircraft environment.
Poisoned pilots almost crashed. It started with an insidious feeling of discomfort. The cabin attendants felt strange, experiencing an incredible pressure
in their heads and bodies. One person described the feeling as like doing
a moonwalk. Another person detected a barely perceptible odour.
On the next ight, the discomfort returned and was now also experienced by
the two pilots. During the third ight, the crew realized that there was something unusual in the air inside the aeroplane. And when the cabin manager
went into the cockpit prior to landing, she discovered that both the pilots were
wearing their oxygen masks. The captain felt so bad that he had handed over
the controls for his rst ofcer to land the plane.
We broke out the oxygen masks. From the onset of the feeling of sickness,
I very rapidly became worse and worse, feeling dizzy and groggy despite the
oxygen. After about two minutes I slowly began to recover. As the rst ofcer
was feeling much better, he took over the controls.
So writes Captain Niels Gomer in his report regarding these incidents [3].
These incidents took place on one BAe146-200 aircraft operated by Braathens Malm Aviation on November 12, 1999, on three ights between
Bromma and Sturup. They were designated by the Swedish Board of Accident Investigation, the airline company, and the aeroplane manufacturer as
extremely unusual and serious.
On November 7, 2000, the ight crew of a B757 ying from London
Heathrow to Copenhagen noticed an oily metallic smell in the cabin. On the
approach back to Heathrow, the air trafc controller (ATC) became concerned
that the pilots were not responding to his communication as the aircraft had
not reduced its speed in order to prepare for landing. Finally at 3.5 nautical
miles from the aireld the pilots responded to the question from the ATC if
everything was all right. The aircraft landed safely. It was discovered later that
there was an oil leak in the auxiliary power unit (APU), allowing engine oil to
be heated and released into the ventilation system of the aircraft [4]. The con-
196
C. van Netten
clusion was that this was a serious incident, and the ight crew were partially
incapacitated.
Santa Barbara News-Press, Saturday, March 13, 2004. A Bombardier CRJ
200 had to make an emergency landing at the Santa Barbara Airport on
a ight to Las Vegas after the cabin lled with smoke after take off. Thirtyeight passengers had to be evacuated. No serious health problems with passengers and crew. Cause of the problem, a malfunctioning APU [5].
2
Frequency of Air Quality Incidents in the Airline Industry
The previously described air quality incidents are not isolated cases but occur
at a regular frequency in the aircraft industry. A recent study of three aircraft companies, based on air quality incident reports submitted by crew
members to the airline company, identied the frequency of these air quality
incidents [6].
3
Problems Associated with Obtaining Objective and Comparable Data
The frequency of incidents can vary considerably between aircraft operators
as this is highly dependent on the type of the aircraft and its maintenance.
As an example, the aircraft company operating the BAe-146 aircraft, referred
to in Table 1, has since made great improvements in maintenance procedures
preventing many of these incidents from occurring.
There also appears to be a degree of reluctance of the ight crew members to ll out and submit an air quality incident report. This reluctance can
be traced to a number of factors. The most important one is a lack of objective environmental input. Since most aircraft obtain their cabin air from the
main engines as well as their APU, there is a temporary bad smell when the
engines are started. This smell disappears within about 1 min and is related
to the pooling of small quantities of oil in the engine when it is not operating. The ight crew members are somewhat used to this smell and do not
report it, as it is a common occurrence. Sometimes this smell persists and, at
the extreme, a blue haze or smoke appears in the cabin. Since there is a wide
spectrum of severity of the exposure, one may well ask at what point does
one decide that this is not a normal event but an incident. There is no problem deciding whether to submit a complaint on either side of this exposure
spectrum. The large number of events that fall in between these exposures are
a problem since the ight crew has no guidance or objective measurement to
assess the severity of the event. Often the triggering event in these instances is
197
Table 1 Air quality incident frequencies, based on ight incident reports submitted to
three North American air carriers by ight crew members (C. van Netten, R.H.S Brands,
S. Hoption-Cann, V. Lentino, 2001,unpublished report to the US National Academy of
Sciences) [7]
Aircraft type
Average number of incidents per aircraft

Per year
Per 1000 cycles
BAe-146
MD-80
A-320
B-747
DC-10
B-767
B-737
6.4
1.01
1.67
0.34
0.38
0.21
0.07
3.88
1.02
1.29
1.25
1.04
0.63
0.09
cycle is dened as a take-off, ight, and landing sequence.

on incomplete data and estimates
Based
directly related to a health issue. These can range from an ill feeling to nausea,
dizziness, to incapacitation.
Another reason why the reporting system is inconsistent between aircraft
companies and between crew members is an intimidation factor. Compared
with ight attendants, pilots are generally less likely to report an incident
when it produces mild symptoms. As in any population of individuals, one
can observe differences in susceptibility to a particular exposure and pilots
are no exception. Unlike ight attendants, however, pilots are totally dependent on their medical certicate to allow them to follow their chosen career.
Since, during an exposure event, no objective feedback is provided to the pilots, they are totally dependent on making a subjective assessment. Under
these circumstances a pilot does not want to complain about a health issue
based on an exposure when his colleague sitting next to him does not experience it because it would target his health status and his medical certicate.
Consequently the companies receive more air quality incident reports from
the ight attendants, who are usually well protected from intimidation by
means of their union afliation.
In order to obtain reliable information one has to remove the burden of
this subjective assessment of the air quality within the aircraft from the pilots.
This should be replaced with sensitive environmental monitoring that provides an objective evaluation of the air quality. In addition, strict guidelines
should be provided by an external agency as to when to report these incidents. Only with these criteria in place would pilots feel free to provide the
feedback that is required to create an objective data base.
Another important link in obtaining an objective assessment of these air
quality incidents is to be able to correlate them to the mechanical records
198
C. van Netten
of the aircraft in question. This would provide information, for instance, of

whether a reported exposure event coincides with leaking oil seals in the engine. Mechanical records are most difcult to obtain and are generally not
voluntarily parted with by the aircraft company [8].
4
Symptoms Associated with Air Quality Incidents
Within the aircraft not all areas are exposed to the same quality of air. The
cockpit, for instance, gets 100% fresh air compared with the 60/40% recirculated air that is present in the cabin (US NAS). Since the source of this
fresh air is the engines, any contamination event would be rst felt by the
pilots in the cockpit. This has serious consequences, i.e. they are exposed
sooner and to a higher degree, hence the observed problems in the Malm
and Birmingham incidents when the ight attendants discovered that the pilots were on oxygen in the cockpit.
Although pilots are highly vulnerable to air quality incidents, they have reported few symptoms, likely owing to the problems identied earlier. Most
of the symptoms that are experienced by ight crew members have been reported by the ight attendants. These results are summarized in Table 2.
In addition to the acute effects described previously, there are also longterm chronic effects. These health problems are even more difcult to trace to
an exposure event, or events, as these chronic effects are often the combined
result of a number of exposures that have been present at low concentrations
over many years. By the time a pilot experiences these often poorly reversible
symptoms, it is too late, he/she fails the medical and a career as a pilot is
gone. In addition, since no data were collected during these air quality incidents, it is very difcult for that pilot to substantiate any claim that his/her
poor health was related to occupational exposures. As indicated in Table 2
there seems to be a high incidence of neurological problems, some of which
have been classied as Parkinsons disease like.
5
Source of Air in Aircraft
In order to obtain an understanding of the nature and extent of these possible
exposures, and their associated symptoms, it is important to investigate the
source of the air that ventilates the cabin in aircraft.
199
Table 2 Frequency and type of symptoms reported by ight crew members. Of company
X, (MD-80 aircraft) and company Y (mixed eet of aircraft) (C. van Netten, R.H.S Brands,
S. Hoption-Cann, V. Lentino, 2001,unpublished report to the US National Academy of
Sciences)
Number of
incidents
Company
Any symptom
Eyes, ears, nose,
throat
Eye
Nose
Throat
Ear
Central nervous
system
Intoxication
Neuropsychiatric
Other
Respiratory system
Cardiovascular
system
Gastrointestinal
system
Skin
Other
Percentage of incidents Percentage of all

with symptoms
incidents
X
467
Y
244
X
78.1
Y
81.6
93
62
19
45
6
76
26
31
22
19
19.9
13.3
4.1
9.6
1.3
31.1
10.7
12.7
9.0
7.8
15.6
10.4
3.2
7.5
1
25.4
8.7
10.4
7.4
6.4
428
419
19
113
94
192
188
6
23
83
91.6
89.7
4.1
24.2
20.1
78.7
77.0
2.5
9.4
34.0
71.6
70.1
3.2
18.9
15.7
64.2
62.9
2.0
7.7
27.8
19
4.1
2.9
3.2
2.3
117
47
43
62
10
22
37.9
10.1
9.2
25.4
4.1
9.0
29.6
7.9
7.2
20.7
3.3
7.4
Not all air quality incidents that have been reported were associated with symptoms
and one individual could have multiple symptoms.
5.1
Ram Air
The earlier Douglas and Boeing aircraft used outside air that was introduced
into the cabin by means of an air-scoop which was located on the outside of
the fuselage allowing air to force itself into the aircraft during ight, hence
the name ram air [9]. Although the source of this air is clean and generally
only vulnerable to contamination from external sources, it was discontinued
on later models in favour of a bleed air source for economic reasons.
200
C. van Netten
5.2
Bleed Air
Bleed air comes from the jet engine. Since the jet engine operates on the basis
of compressing outside air to a high degree prior to entering the combustion chambers it was decided that some of this highly compressed air could
be extracted, i.e. bled off and used to ventilate the cabin. This bleed air is
not only vulnerable to potential contaminants present in the outside air, but
is also vulnerable to potential contaminating events in the compressor stages
of the engine. The temporary smell of oil when the engines are started, as
described earlier, is a direct consequence of this.
6
Oil and Hydraulic Fluid Contaminants
6.1
Jet Engine Lubricating Oils
When an oil seal in the compressor stage of a jet engine is not sealing
properly, jet engine lubricating oil enters the airstream, is aerosolized, compressed, and heated to a high degree before it enters airpack units (environmental control systems) and enters the cabin. Temperatures in excess of
450 C have been reported [1012] at pressures as high as 175 psi [7].
6.2
Hydraulic Fluids
MD-80 aircraft often experience one additional source of exposure owing to
the location of the air intake of the APU. This air intake is located immediately
behind a small hole in the aircraft fuselage that allows any hydraulic uid
and/or oil that has accumulated in the bilge of the aircraft to be dumped overboard. In the MD-80 these agents are sucked into the air intake of the APU
when it is operating and supplies air to the cabin during certain phases of the
ight where maximum engine power is required, such as during take-off.
In order to nd a connection between air quality incidents and symptoms
experienced by passengers and crew, one needs to know the individual constituents of these oils and hydraulic uids.
Table 3 summarizes the constituents in a number of oils and uids that are
reported in the material safety data sheets supplied by the manufacturer.
201
Table 3 Composition of some jet engine lubricating oils and hydraulic uids as reported
in their material safety data sheets
Type of oil/uid
Engine oils.
Mobil jet oil 254
Mobil jet oil II
Mobil jet oil 291

Hydraulic uids.
Skydrol 500B-4
Skydrol LD-4
HyJet IV
HyJet IV-A+
HyJet XL
Reported composition
CAS no.
Tricresyl phosphate (15%)

Tricresyl phosphate (15%)
1-Naphthalamine,
N-phenyl (1%) (Mobil)
No reportable ingredients
(Mobil Australia Ltd.)
1330-78-5
1330-78-5
90-30-2
Tributyl phosphate
Dibutyl phenyl phosphate
Butyl diphenyl phosphate
(Monsanto Company St Louis)
Tributyl phosphate
Dibutyl phenyl phosphate
Butyl diphenyl phosphate
2,6-ditert-butyl-p-cresol
(Monsanto Company St. Louis)
Epoxy modied alkyl esters
Tributyl phosphate (7080%)
(Chevron)
Trialkylphenyl phosphate (12%)
Cyclic aliphatic epoxide (2%)
Additives (7%)
(Chevron)
126-73-8
2528-36-1
2752-95-6
126-73-8
2528-36-1
2752-95-6
128-37-0
Not provided
126-73-8
126-73-8
126-73-8
68937-41-7
3388-03-2
CAS no. Chemical Abstracts unique identication number for the actual compound
6.3
Pyrolysis Products
When these oils and uids were exposed to simulated temperature conditions present in the aircraft it was reported that, among other compounds,
carbon monoxide (CO) was released into the atmosphere, indicating that pyrolysis of some of the constituents had taken place. Engine lubricating oils
generated more CO than hydraulic uids under the same temperature conditions [1012].
In addition, it was observed that the tricresyl phosphates (TCPs) and other
oil constituents could be captured from the air at 25 C. It appears that these
low volatility compounds condense out of the air but remain airborne as an
202
C. van Netten
aerosol or are associated with particulate matter, making these compounds

accessible to the respiratory route of exposure.
These ndings have identied potential exposures to agents which represent two main categories, i.e. those resulting in acute effects such as incapacitation from exposure to CO, and those resulting in long-term chronic effects,
such as delayed neurological problems from exposure to TCP and its isomers
as well as CO.
7
Available Data and Required Data
At this point in time one has a data base of symptoms that are experienced by
passengers and crew members. There also is an extensive data base in the scientic literature on the effects associated with exposures to CO and TCP. The
symptoms reported by aircraft crew members appear very consistent with the
known symptoms of CO and TCP. At this time the missing link is a data base
of exposure measurements in aircraft during these air quality incidents. Such
information is crucial in connecting the observed symptoms experienced by
ight crew members to those that have been reported for these agents in the
scientic literature.
7.1
Problems in Capturing Rare Air Quality Events in Aircraft
One of the main reasons why these exposure measurements are not available
at present is due to the sporadic nature of these incidents and a lack of appreciation as to what to measure. Although CO was identied as an agent
of interest to be monitored and recommended by the US NAS committee on
Airline Cabin Environment and Health of Passengers and Crew [7], little has
been done to date and CO monitoring in aircraft is virtually absent.
Another reason why exposure data are not available is the reluctance of the
industry to do the monitoring in their aircraft. The data that are provided by
the industry generally relate to non-incident ights on aircraft that are wellserviced and maintained.
The argument is that, given the sporadic nature of these incidents, it would
be too cumbersome and expensive to place a set of instruments on each aircraft and wait for an incident to occur. Although this is a legitimate argument
based on the very elaborate instrumentation that was used to monitor air
quality in non-incident ights, which usually tries to measure very low concentrations, this should not be used as an excuse for not trying to capture
these incidents.
This argument is therefore not acceptable when efforts are made to capture incidents which are associated with exposures well above normal making
203
them much easier to capture with relatively inexpensive instrumentation. The

only additional problem is to ensure that such instrumentation is available at
the time when an incident is taking place.
At present data-logging CO monitors are available which could be provided by the airline company to a designated crew member to be turned on
during an event. Although this was one of the NAS recommendations, at this
point in time the industry has been rather reluctant to provide aircraft or
a crew member with such an instrument. As a consequence crew members
have resorted to carrying their own instruments to be able to obtain objective
data during an incident.
The data obtained from one pilot using such a data-logging CO monitor
(Drger) over a 2-month period when a crew member smelt fumes and/or felt
unwell in BAe 146 aircraft that he was ying, identied consistent peak levels
of CO ranging from 15 to 21 ppm. These values compare poorly with the CO
levels that were monitored on similar non-incident ights which consistently
read 0 ppm. These ndings also clearly indicate that air virtually devoid of CO
is possible with proper maintenance. The fact that this pilot was not willing
to be identied is a sad reection of the intimidation factor that prevents pilots from coming forward and providing the industry with an early-warning
system before an incident becomes an accident.
Although CO monitoring appears to be a good indicator of the aircraft
air handling system based on the performance of the APU and engine, it
does not address the long-term chronic symptoms associated with possible
organophosphate exposure such as TCP from engine oils and hydraulic uids.
7.2
Benefits to the Industry of Monitoring Changes in Air Quality
Routine monitoring for CO on aircraft, besides identifying a health hazard,
has an additional benet. A change in CO levels within the aircraft, although
in the acceptable range that is consistent with, for instance, the activation of
the APU or the engines, provides an early-warning system of a possible seal
failure. This was observed in a BAe-146 which showed elevated levels of CO
in the 9 ppm range 1 week before a seal failure in the APU resulted in an air
quality incident [13].
204
C. van Netten
8
Available Avenues of Obtaining Air Quality Measurements in Aircraft
8.1
Aircraft Filter Analysis
Currently, one way of obtaining a rough qualitative measure of the constituents that might be present in the cabin air is to obtain air lters that have
been used within the aircraft. A small number of these have been analysed
(personal observation). A set of lters from a Boeing 737 galley and lavatory,
for instance, did not show any evidence of TCP and its isomers by the analytical procedure used, but did show the presence of many other constituents,
such as caffeine, in both locations. In addition, the lavatory lter also showed
the presence of cocaine and amphetamines. A modication of the analysis applied to another, similar lter from a lavatory did clearly indicate the presence
of TCP isomers, indicating potential respiratory exposure. The data provided
by these analyses indicate integrated exposure to these agents since the lters
were installed, but does not provide the data that are needed to describe the
acute events that result in symptoms.
8.2
Coalescer Bag Analysis
Another source of information regarding the quality of the air entering the
cabin is to analyse the coalescer bags. These bags are woven cloth socks
located in the air supply system prior to the air entering the cabin. Their function is to extract excess water from the bleed air supply. Analysis of extracts
from an MD-80 coalescer bag showed the presence of TCP and its isomers.
This indicates the release of these oil constituents into the ventilation air but
does not necessarily show exposure, as the argument could be made that the
coalescer bag actually lters out these contaminants before the air enters the
cabin.
Although these analyses indicate an interesting capability of these lters
and coalascer bags to reect the past history of the aircraft and potential
exposure, they do not tell when these events occurred and for how long.
8.3
Monitoring Aircraft Air Quality During Flight
During aircraft air quality incidents that are related to APU and engine problems, many gases and agents appear to be present in the smoke and/or
fumes that have been observed by ight crew members [14]. Existing protocols for monitoring gases released during these events require the use of
205
electronic sensors specic to each gas of interest. These direct-reading, datalogging, instruments provide a clear description of what is present in the
air, at what concentration, and for how long. Existing protocols for monitoring constituents in aerosols and/or particulate matter involve an air ltering
arrangement using lter cassettes and an appropriate pumping device to provide an airow in the range 12 l/min through the lter [15]. This lter can
then be sent to an appropriate analytical laboratory for analysis.
Data-logging gas-sensing instruments can be quite costly and in general
are currently too expensive to be deployed on a large scale as is required to
capture sporadic events. Similarly, currently available cassette lter systems
are also prohibitively expensive and, in addition, usually require the presence
of an industrial hygienist.
8.4
Indicators of Air Quality During Incidents
If one could reduce the number of agents monitored for to only a few that
are representative of the incident, then costs could be reduced, allowing larger
numbers to be available.
It appears from the information available to date that the most likely candidates that could serve as reliable indicators of air quality during these
sporadic events are CO and oil components, such as TCPs. Exposure to CO
represents acute toxic effects, whereas exposure to TCPs appears to be an indicator for chronic effects. Nevertheless, monitoring for these two agents still
needs elaborate and expensive equipment, making it too unrealistic for the
large-scale use that is required to capture these sporadic events.
9
Development of a New Air Monitor
In order to address this issue, a new type of lter-based air sampler has been
developed that is small, i.e. a plastic cylinder 7-cm tall and 5 cm in diameter,
is self-contained, self-sealing and, above all, inexpensive, allowing many to be
placed in the occupational setting [16]. In order to activate this sampler all
one has to do is twist the cap 45 , which activates the centrifugal pump allowing air to ow through the lter at a rate of 2 l/min for a 20-min period of
time. After exposure the cap is rotated in the opposite direction, which turns
off the pump and seals the lter from further exposure. The whole unit is
forwarded to the laboratory for analysis.
This sampler has been used on several ights between Canada and Australia in order to capture the nature and the extent of pesticides used for
disinsection purposes.
206
C. van Netten
9.1
Benefits to the Industry from Monitoring for CO
There are two reasons why we need CO monitoring. First and foremost are
safety reasons, especially when the aircraft has been outtted with an activated carbon lter, as any tell-tale odour, visible smoke, or aerosol would
initially likely be adsorbed, preventing early detection. The second reason is
to monitor any deterioration in bleed air quality, specically how it is dependent on the efciency of the oil seals to prevent engine oil from entering the
ventilation system. CO spikes when the APU is turned on during ight, only
to disappear when this equipment is turned off, indicating possible oil seal
deterioration.
Although both reasons are somewhat different they can be addressed effectively with a common approach.
Since in many aircraft the pilots get more outside air, their location in the
aircraft makes them more vulnerable to any malfunction in the air supply system. For this reason alone the pilots should have access to an ambient CO
level at all times. In addition, the pilots know when certain types of equipment are turned on or off, such as an APU, and consequently can make the
correlation with changing CO levels. Pilots also have access to information
that allows them to discriminate between a possible internal source and an
external source of CO, i.e. they would know whether they are in the exhaust
stream of another aircraft.
Most aircraft have at least two air supply systems and different sections of
the aircraft receive air from different engines. The cockpit in the BAe 146, for
instance, is provided with air from jet engines 1 and 2, whereas the cabin gets
air from engines 3 and 4. In this instance a CO monitor in the cockpit would
not be representative of the cabin air quality. It would therefore be prudent
to monitor for CO in each section of the aircraft that is serviced by different
engines. The readout of these monitors should be in the cockpit in order to
provide feedback to the pilots and other ight crew members regarding the
status of these vulnerable components of the aircraft.
10
Standards of Exposure
Alarm levels could be set at the current airworthiness standard set by the US
Federal Aviation Administration of 50 ppm. The problem is that the period of
time is not specied, making this difcult to interpret when spikes of varying duration occur, i.e. does one become concerned when 50 ppm of CO is
measured for 1 s or when this level is present for 15 min? In other words this
standard is neither enforceable nor practical.
207
The American Conference of Governmental Industrial Hygienists (ACGIH)

has a time-weighted average (TWA) (8 h) of 25 ppm [2]. A short-term exposure level (STEL) would be useful in order to suggest a spike duration;
however, a STEL is not available from this or other regulatory agencies.
Another approach in identifying what an appropriate sampling time
should be, in order to set a time limit for spikes, has been put forward by
Roach and others [17, 18]. This approach is realistic and defendable based on
the biological half-life of the agent of interest, which is a measurable entity.
With respect to CO, the half-life of carboxyhaemoglobin (COHb) is not clearly
dened as it depends on conditions such as altitude (i.e. partial pressure of
oxygen) and the concentration of CO2 . The former agent will decrease the
half-life to 80 min, whereas the latter will compound the problem by increasing acidosis [19]. The half-life for COHb has been estimated to be around
2 h [17] to 4 h [19]. The suggested sampling duration is one tenth of this
period, which amounts to 1224 min. Roach [17] suggests 10 min. If a spike
lasts for this period of time and exceeds an acceptable standard, an alarm
signal could be communicated to the ight crew.
The ACGIH TWA (8 h) of 25 ppm [2] standard could be argued with, as it
relates to the occupational setting and to healthy workers (595th percentile)
and not to the general public. Several groups in the population have been
identied as being at risk from exposures to low levels of CO (i.e. even at outdoor levels as low as 3 ppm) and include those individuals with heart disease,
chronic respiratory disease or that are pregnant [20]. The US Environmental
Protection Agency has an 8-h limit of 9 ppm [7], which would be more appropriate to protect the general public. A 10-min spike at this concentration
could be used as a triggering event and indicates some inherent problem after
outside sources have been eliminated. This level is also quite reasonable, since
properly maintained aircraft are capable of having CO levels close to 0.0 ppm.
Currently there are CO monitors available that are small, data-logging,
cannot be switched off, and will last for 6 months. The continuous monitoring
function of these CO detectors can be used to provide an objective measure of air quality at any time. When, for instance, a 10-min spike of 9 ppm is
encountered, this can be an objective triggering event and can be used to activate the air sampler described before. Both devices, in fact, can be combined
into one functional air monitoring unit with CO measuring acute events and
effects and the lter providing exposure measurements that are likely associated with more chronic effects.
208
C. van Netten
11
Identification of Potential Synergistic Agents
The use of indicators to monitor the extent and character of rare air quality
incidents can be very useful as it reduces the complexity of tracking the event.
At the same time there is always a tendency by those using these indicators to
forget that they are only a surrogate measure of a complex event and that they
provide an oversimplication of the event itself. The presence of other agents
that are able to alter the toxicity of a particular exposure could be ignored.
In this respect a comprehensive chemical analysis of the exposed lters from
the monitors that were activated during an event will provide a data base of
other agents that might be present in the cabin air. Some of these agents might
have a synergistic toxic effect with other agents present. These bad combinations of exposures might explain why certain individuals show symptoms at
reported levels of exposure well below the limits for each of the individual
agents. The use of insecticides in aircraft is a typical example. Permethrins
are required by certain countries for the disinsection of aircraft and can be
found in most of the larger aircraft capable of international ight. On the basis of the scientic literature, such synergistic effects were postulated between
organophosphates and permethrins [7, 21]. Experience of the Gulf War syndrome and experiments performed by Abou-Donia [22, 23] have conrmed
this.
An additional synergistic effect between CO and permethrins has been
postulated [21] and needs to be evaluated in the aircraft environment as there
appears to be a potential for these agents to act synergistically, resulting in
a loss of nighttime vision.
12
Exposure Data Acquisition
In order to obtain a data base that accurately reects the nature and extent
of air quality incidents, one needs to have many monitors out with ight
crews in order to be able to capture an event when it occurs. As an example
there were 27 501 domestic airline ights per day in October 2003, in the
USA [24]. If one were to use a conservative frequency of one incident per 2000
ights, then there are approximately 13 incidents per day. If 2000 monitors
were available to the industry one should be able to capture on average at least
one event per day. At this rate, a reliable exposure data base would become
available to the public, ight crews, and the industry within a very reasonable
time span, allowing one to identify the extent of, and the connection between,
the symptoms experienced by ight crews and the contaminants in the air
supply of aircraft.
209
13
Alternate Solutions
It should be emphasized that most of the air quality incidents are directly
linked to the poor quality of the bleed air that is supplied to the cabin either
from the APU or the engines [7]. These problems could easily be eliminated
if ram air is used along with compressors and a heat-exchange system. Although this approach was terminated, for economic reasons, in favour of
bleed air, it is interesting to note that the new Boeing 7E7 (currently called the
787), also for economic reasons, has designed an air supply system that is not
based on bleed air [25].
14
Conclusion
A majority of aircraft air quality incidents can be traced to contamination
of the ventilation system from jet engine oils and/or hydraulic uids. The
symptoms reported by ight crew members often identify the central nervous system as being affected, followed by the respiratory system. Exposure to
agents that could explain these symptoms and which are likely to be present
during an incident have been identied as CO and TCPs. Exposure measurements of these agents during these incidents has not been done as it requires
many units of expensive equipment to be present on aircraft in order to capture these sporadic events. A new air sampler has been developed that has the
ability to address this issue and which could quantify exposures during these
incidents and the role these oil and uid components might play in explaining the symptoms experienced by ight crew members. An alternate solution
to the health problems that have been associated with bleed air ventilation
systems in aircraft is to use another source of air within the cabin.
References
1. Hocking MB (2002) Rev Environ Health 17(1):149
2. TLVs and BEIs (2004) Threshold limit values of chemical substances and physical
agents and biological exposure indices. American Conference of Governmental Industrial Hygienists, Cincinnati, OH
3. Gromer N (1999) The Bromma and Malm incident. Braathens Malm Aviation
4. UK mandatory occurrence reporting system (2004) Safety Investigation Group occurrence number 200008363
5. Medina H (2004) It was the scariest experience in my life. Santa Barbara News-Press
6. van Netten C, Brands RHS, Hoption-Cann S, Lentino V (2001) Descriptive epidemiology of air quality incidents experienced in aircraft from three airline companies.
Report to the National Academy of Sciences (unpublished)
210
C. van Netten
7. National Research Council (2002) The airliner cabin environment and the health of
passengers and crew. National Academy Press, Washington, DC
8. Alaska Airlines arbitration hearings (2002) Association of ight attendants vs Alaska
Airlines, Seattle
9. Lorengo D, Porter A (1989) Federal Aviation Administration, US Department of
Transportation
10. van Netten C, Leung V (2000) J Appl Occup Environ Hyg 15(3):277283
11. van Netten C, Leung V (2001) Arch Environ Health 56(2)
12. van Netten C (2000) In: Nagda N (ed) Air quality and comfort in airliner cabins. STP
1393AST, West Conchohocken, PA, pp 6175
13. van Netten C (2000) Flight attendants personal stealth monitoring observation
14. van Netten C, Hilliard NB (1998) J Appl Occup Environ Hyg 13(10):733739
15. Laboratory analytical methods (1989) Workers Compensation Board of British
Columbia. Richmond, BC, Canada
16. van Netten C (2001) Personal and environmental air sampling apparatus. US Patent
application no 20040045376, March 11, 2004
17. Roach SA (1966) AIHA J 112
18. Salzman BE (1970) J Air Pollut Assoc 20:10:660665AIHA
19. Amdur MO, Doull J, Klaassen CD (1991) Casarett and Doulls the basic science of
poison, 4th edn. Pergamon Press, New York, p 268
20. Rom WN (1992) Environmental and occupational medicine, 2nd edn. Little Brown
and Company, Boston, p 1229
21. van Netten C (2002) Analysis and implications of aircraft disinsectants. Sci Total Environ 239(13):257262
22. Abou-Donia MB, Wilmarth KR, Jensen KF (1996) J Toxicol Environ Health 48:3556
23. Abou-Donia MB (2003) Arch Environ Health 58(8):484497
24. US Air Trafc Control (2003) Federal Aviation Administration, Washington, DC, October 2003
25. Anonymous (2004) Aircraft technology engineering and maintenance. Flug Rev 8:30

DOI 10.1007/b107245
Published online:
Aircraft Air Quality Malfunction Incidents:

Causation, Regulatory, Reporting and Rates
C. Winder (u) S. Michaelis
School of Safety Science, The University of New South Wales, NSW 2052 Sydney,
Australia
c.winder@unsw.edu.au, s.michaelis@student.unsw.edu.au
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2
2.1
2.2
2.3
2.4
2.5
Engine Lubricating Oil and Hydraulic System Malfunctions

Sources of the Problem . . . . . . . . . . . . . . . . . . . . .
Regulatory Requirements . . . . . . . . . . . . . . . . . . . .
Reporting Requirements . . . . . . . . . . . . . . . . . . . .
Evidence of Reporting of Defects . . . . . . . . . . . . . . . .
Under-Reporting . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
214
214
216
216
217
222
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
223
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
225
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
212
Abstract The issue of aircraft air contamination due to oils and hydraulic uids leaking
into the aircraft air supply is a known problem in the aviation industry. There are a range
of regulations that are in place to ensure all cases of fume contamination are reported and
therefore investigated. However, there is strong evidence that the reporting system to regulatory agencies is not working and, consequently, under-reporting occurs and the fume
events taking place are considerably higher than the aviation industry admits. There are
a variety of reasons for this including commercial pressures, fatalism about long-standing
and apparently insurmountable engineering problems, operational procedures that focus
on keeping aircraft ying and a culture to minimise health and safety risks. These have
signicant health and safety implications for crew and passengers.
Keywords Aircraft air contamination Fumes Reporting FAR/JAR 25.831 ASR MOR
Abbreviations
APU
BAe
British Aerospace
BALPA British Airline Pilots Association
CAA
Civil Aviation Authority of the United Kingdom
CASA Civil Aviation Safety Authority of Australia
FAR
JAR
MOR Mandatory Occurrence Reporting
212
S. Michaelis C. Winder
1
Introduction
There is no question that air safety is an important issue. Unlike many other
types of transportation accidents, the loss of a passenger airplane in ight
is a catastrophe. There are a range of factors that can lead to airplane accidents, including problems of language [1], problems of communication [2],
problems with technology [3], and problems with attitudes to safety [4, 5].
One major aircraft manufacturer recently stated that its denition of aircraft
safety was based upon the aircraft not having had a fatality due to a technical
problem [22].
As commercial imperatives drive any business, the need to keep aircraft
ying is critical and malfunctions in aircraft could be seen as a threat to
business activities. A system needs to be developed that identies such malfunctions, assesses their signicance and efciently resolves the problems
they create. One such problem, discussed below, deals with malfunctions in
the engine or hydraulic system that lead to ight deck and passenger cabin air
quality problems.
Studies indicate [7, 8] that it is common that all modes of transport have
ventilation rates less than current ASHRAE 62 guidelines for commercial
buildings [9]. For example, a Canadian study of one aircraft type and airline
found that 25 of 33 commercial ights did not satisfy the ASHRAE air ventilation criteria of 15 ft3 per occupant and that 18 of 33 ights had less than
10 ft3 per occupant [10]. This nding, of itself, does not imply poor air quality. However, it suggests that initiatives to reduce air quality should be resisted
and indicates that opportunities to improve air quality should be encouraged.
The cabin of an airplane is a specialised working environment and should
be considered as such. Recommendations for pressurisation of airplane cabins (to an equivalent of 8000 ft) were established in the 1960s using healthy
male volunteers. This is sufcient to lower the partial pressure of oxygen
(from 159 mm Hg at sea level to 118 mm Hg), that is, a level that may have an
impact on physiological function (itself dependent on blood O2 saturation)
of some individuals and an impact on physiological function is more likely
where individuals are undertaking effort.
The oils and hydraulics used in airplane engines are toxic, and specic ingredients of oils are irritating, sensitising and neurotoxic [11]. When oil or
hydraulic uids leak, they can contaminate the air supplied from the engines
or APU and may be in the form of unchanged oil/uid, degraded oil/uid
from long use in the engine, combusted oil/uid or pyrolised oil/uid. A leak
may be in the form of gases, vapours, mists and particulate matter. If leak incidents occur and the oil/uid is ingested into the air being used for the cabin
(bleed air) and passed to the ight deck and passenger cabin, exposed staff
and passengers may be exposed to contaminants that can affect their health
and safety and they do not have access to appropriate information that can
Aircraft Air Quality Malfunction Incidents
213
advise them as to hazard, risk or control of exposure. Where leak incidents

are known to be mixed forms of contaminants, an additional component of
toxicity exists whereby irritant or toxic vapours or gases may be adsorbed
onto the surface of mists or particulates. Under such circumstances, the dose
response characteristics of the gas or vapour may be altered. Therefore, the
use of risk acceptability criteria for chemical exposures, such as exposure
standards or threshold limit values (TLVs), to conclude that exposures are acceptable is inapplicable in certain situations in the aviation industry [11, 12].
Such standards should not be applied at altitude, or in other situations where
the possibility of escape to fresh air is lacking. Acceptability criteria for chemical exposures at altitude must consider the interaction of reduced oxygen,
skin exposure to mists, and interactions with other contaminant exposures.
Contaminants may be well below current recommended safety standards,
yet generate complaints due to the synergistic effect. Some standards are outdated having not incorporated more recent medical and scientic evidence.
Additionally, extenuating factors onboard aircraft including humidity and
cabin pressure have not been studied to the extent that new standards can be
proposed incorporating these factors or interactions between them [13, 14].
There is currently no agreement amongst aviation toxicologists on
whether the threshold limit values (TLVs) or NASA spacecraft maximum
allowable concentrations for airborne contaminants (SMACS) are the most
appropriate toxicological standard [15]. Symptoms of immediate nature and
reported by exposed staff in single or small numbers of repetitive incidents
are consistent with the development of irritation and discomfort. Symptoms
of a short term nature (that is, continuing symptoms for up to 6 months) reported by some exposed staff following escalating numbers of leak incidents
are consistent with the development of initially temporary but eventually
irreversible health problems in a number of body systems. Additionally,
symptoms of a long term nature (that is, sustained symptoms for at least
6 months) reported by some exposed staff following an intense or signicant number of low level leak incidents are consistent with the development
of an irreversible discrete occupational health condition, termed aerotoxic
syndrome [14, 16]. Where contamination of air in ight deck and passenger
cabin occurs that is sufcient to cause symptoms of discomfort, fatigue, irritation or toxicity, this contravenes air quality provisions of the Federal Aviation
Regulations (FAR), most notably FAR 25.831a/b [17].
This is a signicant aviation safety matter to pilots, cabin crew and passengers where leak incidents affect the ability of pilots to y planes safely or the
ability of cabin crew to perform their duties as expected in either normal or
in emergency conditions. Also, this is a signicant health and safety matter to
airline staff and passengers where leak incidents affect their health.
Information provided by oil manufacturers to airplane manufacturers understates the toxicity of their oil products [11]. This has been accepted uncritically by aircraft manufacturers and airline operators and is used by them
214
in a manner that misleadingly understates risk. Additionally, all studies that

have been carried out to measure atmospheric contamination in airplanes
from leak events are sufciently awed on methodological inadequacies as to
render their conclusions invalid [11].
Evidence is available that suggests that there are a substantial number of
leak incidents on airplanes, especially on certain models of aircraft. Many
of these leaks go unreported to aircraft operators. Of those leak incidents
that are reported to aircraft operators, many are not reported to regulatory authorities and of those leak incidents that are reported to regulatory authorities, not all are added to relevant databases. Additionally, only
a very small number of leak incidents are investigated fully and are available for review; however, most of these investigations have been found to be
inadequate.
2
Engine Lubricating Oil and Hydraulic System Malfunctions
2.1
Sources of the Problem
The aviation industry itself acknowledges that air quality exposure events
are primarily due to oil leaking into the air supply. For example, company
memos, industry and government submissions to previous Government Inquiries, and other documentation indicate:
Society of Automotive Engineers (SAE) aerospace information report [18]
Engine compressor bearings upstream of the bleed ports are the most
likely sources of lube oil entry in the engine air system and thence into the
bleed system contaminating the cabin/cockpit air conditioning systems.
Mobil Oil (manufacturer of Mobil Jet Oil II) [19]
If cabin air becomes contaminated with any lubricant and/or its decomposition products, in sufcient quantities, some degree of discomfort due
to eye, nose and throat irritation could be experienced. Problems like
these can be generally traced to improper design, improper maintenance
or malfunctioning of the aircraft.
Allied Signal (manufacturer of airplane auxiliary power units [APU]) [20]
Several BAe 146 aircraft are having reports of objectionable odours described as dirty socks or musty smells. Very little work has been done
in the aviation industry to pinpoint the chemical compounds causing such
odours... the odour appears to be coming from breakdown products of the
oil, either through incomplete combustion on the catalytic converter, or
by chemical or biological reaction occurring in the environmental control
system of the aircraft.
215
British Aerospace (BAe, manufacturer of airplanes)

Every engine leaks oil from its seals and bearings. [22].
The air supply is protected from contamination by seals, which achieve
maximum efciency during steady state operation. However, they may
be less efcient during transients (engine acceleration or deceleration)
or whilst engine is still achieving an optimum operating temperature.
Improvements in seal design continue to increase efciency, and when
available, modications are provided for the engines and APU [73].
Reports of cabin air odours have been received from time to time and
have predominantly been determined to be due to minor systems failures
such as leaks from oil seals on the aircraft engines or APU [22].
BAe Service information leaet 21-45, Troubleshooting - operator experience of oil contamination of the engine/APU bleed air [74]
Ansett Australia (former airline)
The source of the odours has been identied as primarily Mobil Jet Oil II
leaking past oil seals in the engines and or APU unit into the air conditioning system[23].
the short-term symptoms associated with odours that have been reported on the BAe 146 and other types are substantiated. These odours
have been generally linked with inadequate ventilation together with aircraft system defects [24].
Civil Aviation Safety Authority of Australia (CASA)
All engines and APUs leak oil and suffer fumes as a feature of the design
of air conditioning systems using bleed air [25].
Civil Aviation Authority of the United Kingdom (CAA)
Although the exact cause of crew incapacitation is not yet known, the
most probable source is oil leaking from the engines or APU and contaminating the air supply to the cabin and cockpit through the air conditioning
system [26].
Although the immediate investigations were not able to nd a denitive
cause of the symptoms experienced, circumstantial evidence suggested
potential contamination of cabin air by abnormal concentrations of noxious gases or vapours [27].
Evidence from these incidents indicated that contamination of the ventilation systems by engine oil fumes was the most likely cause [28].
There is a paradox in that all parties acknowledge that a problem exists,
but then deny that it is a serious matter. Many deny that it is an air safety issue, rather an occupational health and safety (OHS) general health or comfort
issue [2225, 29].
Dozens of in-cabin leak/smoke events are documented annually, often correlated to aircraft uid leak events. However, leak incidents are much more
frequent, correlated to less obvious aircraft uid leaks and residual contamination that are seen by many as a normal part of ying [11, 30].
216
2.2
Regulatory Requirements
National aviation safety regulations such as the FARs and JARs cover areas
of airplane performance, and include ventilation airworthiness requirements
that require a sufcient amount of uncontaminated air to be supplied so that
the crew can operate without undue discomfort or fatigue and so that the
cabin be free of harmful or hazardous levels of gases or vapours [17].
While the term undue discomfort may be interpreted subjectively, the
presence of contaminants in airplane air sufcient to impair ight crew capability, or the ability of cabin crew to perform their duties effectively as
expected under the legislation, would seem to be an apparent example of
a breach of these regulations.
While the term harmful or hazardous levels of gases or vapours may
also be subject to misinterpretation, especially in the use of measures of risk
acceptability such as exposure standards, at least these offer the potential
to clarify minimum sea level equivalences of what constitutes harmful or
hazardous levels. Lack of or inadequate monitoring cannot imply there are
no harmful or hazardous contaminants present if reports are consistently being made.
The aviation industry refers to ozone, carbon monoxide and carbon dioxide when considering contaminants in terms of the airworthiness requirement, [31] and has until recently ignored all other contaminants.
2.3
Reporting Requirements
There is a spectrum of defects and malfunctions in an airplane engine ranging from the trivial to the serious, to the catastrophic. As trivial malfunctions
can escalate into serious events, it is necessary to ensure that all types of malfunctions are identied, investigated and rectied.
FAR/JAR regulations impose strict guidelines on how aircraft defects are
dened, must be reported, investigated and dealt with. Of necessity, these are
based upon those airworthiness standards taken from the FARs and JARs that
cover aircraft design and operation.
The regulations are clear on maintenance and reporting. For instance
in the UK, the aircraft commander must report all technical defects in the
aircraft technical log [32]. Reportable occurrences are incidents or defects
which, if not corrected, would endanger the aircraft, its occupants or any
other persons and are to be made to the aviation regulator under the Mandatory Occurrence Reporting (MOR) scheme. These must be led by the Captain as an MOR with the CAA within 96 h so as to advise of hazardous
or potentially hazardous incidents and defects [33, 34]. A few examples include re; explosion; smoke or toxic or noxious fumes that resulted in the
217
use of emergency equipment or procedures; incapacitation of any member

of the ight crew or incapacitation of any member of the cabin crew that
rendered them unable to perform essential emergency duties; leakage of hydraulic uids, fuel or oil that resulted in possible hazardous contamination of
the aircraft structure, systems or equipment or risk to occupants.
In Australia, reports required include reports of major defects and defects. A major defect is a defect of such a kind that it may affect the safety
of the aircraft or cause the aircraft to become a danger to person or property [37], or smoke, toxic or noxious fumes inside the aircraft [36]. All
defects must be reported in the aircraft technical log by the pilot by the termination of the ight [37], with a defect being seen as an imperfection that
impairs the structure, composition or function of an object or system [36].
Reports on major defects such as oil contamination must be made and investigated in a variety of ways and reported to CASA within 2 days [37, 38] as
well as the accumulation or circulation of toxic or noxious gases in the crew
compartment or passenger cabin [39]. Air safety reports must be made to
the Australian Transport Safety Bureau (ATSB) within 72 h for any occurrence
that could affect the safety of the operation of the aircraft [40].
2.4
Evidence of Reporting of Defects
The reporting systems documented under the International Civil Aviation Organization (ICAO) protocols and legislated by national aviation safety regulations are established so that information arising from incident events passes
from the aircraft operator to the regulator and manufacturer, such that modications can be made where necessary and so that the information is shared
by all parties. These must be adhered to for the information to be utilised
effectively.
However, there are many different types of mandatory and non-mandatory
report formats available. Some of the mandatory reports include: defect reports in the aircraft technical log, defects and major defects sent to the aviation regulator, and air safety incident reports. Some of the non-mandatory
reports include: airline and crew internal reports/information; reports sent
between the manufacturer, regulator and operator; condential reports to the
regulators or bureau of air safety; union reports; crew surveys; medical/legal
reports; passenger reports and so on.
In fact, for such a heavily regulated industry, there is a surprising lack of
conformity in the ways in which malfunctions and defects can be reported in
the various national systems.
Other possible sources of data that can be used to suggest that incidents
are occurring include manufacturers service bulletins (SB), service information leaets (SIL) and the airworthiness directives (AD) that are issued to deal
with problems identied in the operation of aircraft.
218
Despite the fact that there are over 240 advisory service bulletins, service
information leaets and other manufacturer and operator communications
for two aircraft types relating to the specic issue of oil leaks and fume contamination from 19842003 [41], the CAA and CASA have only issued three
ADs in support of fumes (see paper by Best and Michaelis in this volume).
An AD is issued by a regulator to compel the aircraft operator to comply
with manufacturers service bulletins in the case where a safety threat exists
or could exist. Until recently, oil fumes in Australia were not seen by CASA
as a major defect and were not forwarded to CASA, despite the regulations
necessitating this [42, 43].
Table 1 shows a small fraction of the known incidents, which are based
on reported and accessible information. This information must be looked at
whilst bearing in mind the scale of under-reporting, which is examined later.
The information available is clearly greatly dependent on the source. It can
be seen that there are a substantial number of reports on particular types of
aircraft. Some of the more signicant ones are:
One BAe 146 operator reported oil/fumes every 66 ights in 1992, reducing to every 131 ights in 1999; and 775 mandatory aircraft technical log
reports in two and a half years [23].
The British Airline Pilots Association (BALPA) survey of B757 pilots
showed that 106 pilots reported in excess of 1667 fume events, mostly
thought to be associated with oil contamination of the air supply [44].
FAA service difculty reports search (SDRS) shows 8268 cases of smell,
fume, odour, gas, toxic fume, or toxic gas from 1986 to 2000 [45].
There were 760 reports of contamination at one US airline on the MD80
aircraft from 1989 1998 [45].
For BAe 146, 791 optional odour occurrences were reported [46].
146 BAe aircraft operators made 439 reports from 1985 to 2000, including
212 from one operator over 3 years [47].
Despite even the very limited numbers in Table 1 that are high, particularly
in the case of the Ansett Australia Airlines BAe 146, the aviation industry
regulators report that fumes/oil contamination is a rare event.
Ansett Australia Airlines claimed that fume events are a very very rare
occurrence [23] but at the same time encouraged its crews to report
odour occurrence events (yet this was acknowledged as still widely underreported [46]). The crews who worked on a eet of 1315 aircraft, operating an average number of sectors per day, reported one fume-related event
every 66 ights in 1992, reducing to one every 131 ights in 1999 [23]. The
odour/fume reports were primarily associated with leaking oil [23]. Therefore, this very, very rare occurrence could amount to a fume/oil related
defect report every day or two.
In the UK, the CAA state that smoke, gas or leak incidents occur once every
22 265 ights (128 events from 1989 to 1999) [48, 49] and the CAA say they
BAe 146
19852000
BAE - UK
Aircraft defect
19911999
reports - Australia
BAe 146
20002002
AAIB - UK
775
439
19
47
23
128
85+
104+
1667+
Number
1667+ reports of smoke or fumes mostly thought to be from oil

in air conditioning system
Smoke and or fumes oil/smoke/fumes/de-icing/hydraulic uid
16 reports 19881998
88 reports 1999January 2004
Smoke and or fumes oil/smoke/fumes/de-icing/hydraulic uid
11 reports 19851995
68 reports 19962003
Reports sent via email or airline reports (not on CAA data base)
Airline reports not on CAA data base
Smoke/gas fumes (non-mandatory)
1 event every 22 265 ights
B757 (21), BAe 146 (17)
Smoke/fumes incidents,
B757 10
BAe 146 9+
36 operators reported 227 cases of contaminated air -19852000
1 operator reported 212 cases of tainted cabin air 19961999
Mandatory reports in aircraft technical log. Number of reports
1992 418 reports = 1 every 66 ights
1997 189
1999 (6 months)168 reports = 1 every 131 ights
Comment
Some MOR reports not available for review and others referred to as defects only with no MOR
BAe 146/B757
B757
BAe 146
5 Jet types
19982004
20022004
19891999
Other UK data
Other UK data
CAA - UK
BAe 146
19852003
B 757
B757
Aircraft
UK CAA MOR
1988
Jan 2004
2001
BALPA - UK
UK CAA MOR
Dates
Type of report
and country
Table 1 Rates of aircraft smoke/fume/oil and other uid contamination
[23]
[47]
[50]
[56]
[56]
[48, 49]
[51]
[51]
[44]
Source

219
19922000
19902000
1999
FAA - US
NTSB - US
TSB - Sweden
Jet
transport
BAe 146
1
4360
22
32
8268
760
167
791
Number
All crew members temporarily affected by

probably polluted cabin air
Optional BAe 146 odour occurrence reports

(predominantly Mobil jet oil II leaking into air supply)
Examples of oil seal bearing defects, fumes and crew impairment
Oil/hydraulic fume - smoke or odour incidents
SDRS - Smell, fume, odour, gas, toxic fume, or toxic gas
900 reports at 1 airline (73% on MD80)
Accidents and Incidents Data Systems (AIDS)
23 (14%) air quality events connected to air contaminants
in ventilation system 1 every 3 590 000 departures
60 events of ventilation toxic contaminant events
+ Smoke in cockpit/cabin(19781999)
Fumes generated by engine/APU clearly present
hazard level 02
events with no denitive cause not included
less than 1 in 1,000 events were serious threat to ight safety
or immediate serious physical harm
Smoke/fumes
Comment
Some MOR reports not available for review and others referred to as defects only with no MOR
hazard level 0 consequences with no safety effect fumes/smoke have no effect on crew or passenger beyond noticing them,
hazard level 2 signicant consequences smoke or toxic fumes that cause minor impairment or injuries to crew or passengers.
BAe 146
BAe 146
Various
MD80
Various
19962002
19912002
19862000
19891998
19891999
Various
BAe 146
19912000
Odour occurrence
reports -Australia
CASA - Australia
ATSB - Australia
FAA - US
AFA - US
FAA - US
Aircraft
Dates
Type of report
and country
Table 1 (continued)
[47]
[52]
[77]
[54]
[55]
[45]
[45]
[53]
[46]
Source
220
221
have 189 MOR reports on two aircraft types (162 from 1996 to 2004) [51]. The
UK Air Accidents Investigation Bureau (AAIB) had 19 reports of smoke/fume
incidents from 2000 to 2002 on the BAe 146 and B757 [50].
In the US, the FAA state that there is one air quality incident every
3 590 000 departures (23 related to toxic contamination in ventilation systems) [53] and the FAA AIDS database has 60 cases of ventilation toxic
contaminant events from 1978 to 1999 [53]. However later data [77] reports
one fume event per 10,000 ights or less.
In Australia, CASA states there have been 22 events in 6 years [54] (despite evidence showing defect reports occurring up to every 131 ights on
the BAe 146 eet [23], i.e. almost every day). Fume events are also thought
to be to be 50% greater than reported [46], with others suggesting a 90%
under-reporting rate with fumes seen as a normal part of ight [30]. The Australian Bureau of Air Safety (ATSB) had 32 BAe 146 incident reports of oil or
hydraulic fumes/smoke or odour incidents from 1991 to 2002 [55].
Some data that is known to have been reported fails for various reasons to
actually be present on regulator databases. BALPA has had 47 Boeing 757 reports sent direct from crews via email or submitted to airlines which did not
get entered into the UK CAA database, as well as 22 BAe 146 airline reports
(all from one airline) which are not on the UK CAA database [56].
Another example of how many regulator databases lack accuracy in relation to fume events is that there are 775 mandatory Australian BAe 146
aircraft log reports [23] and 791 optional BAe 146 odour occurrence reports [46] which were mostly reported to Ansett Australia, yet only 32 were
received by the Australian Transport Safety Bureau (ATSB) [55], and a very
small number appear on the Australian CASA database [54].
Use of information from within one source is often inconsistent and can
vary greatly. An example is the UK CAA data which lists 56 fume events from
19962003, 66 cases where crew and passengers suffered symptoms of discomfort, while the MOR database shows 162 reports during this period. This
does not even take into account the incomplete database and under-reporting
factors [51, 56, 57].
The differing databases and lack of real understanding of the scale of the
problem led one BAe 146 operator to state that events were increasing over
a period of time, while the regulator stated that there was a decrease in reports [58, 59].
Other examples of how defects and fumes are reported include:
BAe complaint of difculty report: report 27803.BAe 146 reported by
B Rogers of BAe regarding Dan Air: Can Hateld (British Aerospace) provide a denitive statement on the medical implications of fumes/smells in
the cabin ... Dan Air cabin crew have complained of headaches and nausea
... Here we have a reported case of fumes and nausea and despite a 2 year
wait we still have no statement on health and safety. Can you please hasten
an answer at this point (February 1991) [60].
222
Ansett Australia BAe 146 odour occurrence report: All three ight attendants had tightness in chest, sore throats, headaches, slurred speech from
purser during P/A (May 1995) [46].
UK ASR B757: Toxic fumes in ight deck. Aircraft had two previous
ights with oil fumes in ight deck reported. Suggests air conditioning
ducting needs to be cleaned before further ight. Captain felt giddy and
ill, while First Ofcer, ground staff and cabin crew all reported headaches
and feeling unwell (1998). This ASR was not passed to the UK CAA, despite the MOR box being ticked requiring the report to be forwarded to
UK CAA [56].
UK CAA MOR 200007913 B757: Fumes on ight deck and in cabin. Recurring fault considered to be residual engine oil contamination in the
bleed ducts. Reporter conrms that similar incident had been reported
on previous sector and that the aircraft has a history of oil leaks ... although there were no written reports as such. After take-off thrust was set,
a strong smell likened to burning rotten socks was apparent on ight
deck ... during climb, smell was still evident on ight deck each pilot in turn breathed 100% oxygen because they both felt light-headed ...
on shut down both pilots still felt light-headed and also shaky (October
2000 [51]).
CASA major defect database BAe 146: No 1 engine No 9 bearing seal
leaking. Suspect fumes entering cabin and causing crew problems (August 2001 [54]).
Indeed, the difference between statistics due to under-reporting, varying
data on internal databases, reporting to operators, and ofcial reporting to
regulators allows all parties to use awed data to perpetuate well-entrenched
positions with important health and safety trends ignored.
2.5
Under-Reporting
The Australian Senate inquiry into the BAe 146 cabin air quality recognised
that under-reporting was a major problem [61]. The 2001 BALPA Boeing 757
survey reported 1667 fume/smoke incidents [44], while the UK CAA database
shows only 104 Boeing 757 reports over the same period [51]. These gures
highlight the problem of relying on regulator databases for accuracy concerning the scale of the problem whilst under-reporting continues to occur.
The reasons for under-reporting are complex. There is a long-standing culture existing in some airlines of crews not reporting fumes or reporting leak
incidents verbally [62] and some crews may be discouraged from writing reports in the aircraft log [63]. It must also be remembered that fumes and their
effects are poorly understood by crews and dismissed by many in the aviation
industry as not being an aircraft safety issue but a health problem [22, 23, 25].
223
Crews are advised that inhalation of aircraft oil/uids is not harmful to their
health and that their symptoms are not related to aircraft air [44]. Crews may
be fearful of reporting fumes due to awareness that some crews have been harassed, stood down and or terminated after reporting fumes [58, 64] and that
others have lost their medical licences [65, 66]. Others have continued their
rostered duty after fume events as the effects are poorly understood, or they
have been advised or felt the pressure to continue ying [44, 46, 51, 61, 68].
Others report fear of being branded as troublemakers as they would be reporting fumes too often if all cases of fumes were to be reported as aircraft
defects [44]. Additionally leak incidents that do not effect all crew members
equally are not viewed by some as an aircraft defect [67].
Oil seals are not as efcient in certain stages of ight and therefore
the problem may be seen as being intermittent and part of normal operations [73]. Failure of some airline engineers to rectify leak problems or to
comply with ventilation regulations such as FAR/JAR 25.831 does not encourage crews to report fumes, especially when leak incidents are often reported
to be rectied at company convenience [68], not safety of ight, for information only, no fault found, report further or similar [30, 51, 69].
Leak incidents may occur over numerous sectors and are often ongoing
over days, sometimes months [51, 58] with residual contamination being an
important problem on some aircraft [11, 69], which also fails to generate
reports. Additionally there is an accepted practice in the industry of only reporting non-vital defects at the end of the day or duty.
Engineers may have difculties in tracing and isolating the source, which
may result in the aircraft being returned to service with no fault found and
the leak unresolved [29, 51, 69].
3
Conclusions
It can be seen that there are engine oil and hydraulic uid leaks occurring on
aircraft due to reasons which include the design issue that some engine seals
are not as efcient in transient operations, residual contamination events and
more major contamination events due to part or full system malfunctions.
This, combined with the fact that fume events have been under- recognized
and under-reported and seen as more of a nuisance, raises a number of signicant concerns.
It is clear that these fume events and the medical effects experienced by
crews and passengers, occur a lot more frequently than the industry and regulators are prepared to publicly accept. In some cases the regulator actually
denies that pilots could conceivably fail to report all fume events, yet this is
factually known to occur [44, 56, 62, 70].
224
Even if collated fully, the documentation will not collect together the majority of incidents actually occurring because of the under-reporting problem, but it could at least show important trends. Despite fume events relating
to oil contamination being dismissed by the CAA as being of no risk to
health or safety [71] the lack of accurate data is of concern due to the health
and safety ramications from the medical effects of crew breathing contaminated air. Crew symptoms of feeling unwell or irritation are not seen as a regulator responsibility unless classied as partial impairment or greater such
that the safety of ight and landing is affected [57, 70]. Regulatory agencies
and manufacturers usually claim that the issue is one of OHS importance and
not one of ight safety [25], despite acknowledging that this is outside their
eld of expertise [22, 25]. Conversely, the OHS authorities claim the problem
is not within their responsibility as it is an aviation regulator problem [67].
Airlines, not surprisingly, usually claim it is neither a health nor safety issue.
While fumes have generally been dismissed as a non-event [72], one
manufacturer has acknowledged that fumes were previously seen as a nuisance rather than as a potential threat to ight safety [73, 74]. Aviation safety
notes that use of oxygen is a serious incident [75], but crews generally do
not using oxygen even though advised that it is required when fume events
are suspected [26].
The same source of data may give conicting information [77] and additionally the exclusion of fume events without denitive links to the engine or
APU may reduce the true level of incidents and hence the degree of the problem. The selective interpretation of fume events deemed to be of concern may
also downplay the scale of the problem.
Fume and smoke events that are listed at the time of the incident to have no
reported effects on crew or passengers have led many to downgrade the full
range of signicant health & safety implications of exposure to contaminated
air from leaking oil & hydraulic uids. Events not reported to cause immidiate
serious threat to ight safety are not being given the attention they deserve,
particularly given the known problems of under and incomplete reporting of
many fume events.
The true extent of the problem remains largely unknown. For the full scale
of the problem to be better understood the regulators need to enforce regulations that require leak incidents to be reported, and the reports that are made
need to be forwarded to the regulators as required by the legislation.
In general, the regulations surrounding contaminated air defects on
aircraft are not being followed. While low numbers of major incident
leak reports get reported and investigated, this process is often inadequate [50, 51, 76]. Most others slip between the cracks and a lot of objective
information is deemed anecdotal by the industry. This allows an inaccurate
picture of the real situation to develop, which is then accepted as reality,
adopted as practice and defended with the rigor that only incorrect dogma
can produce.
225
Whilst civil aviation has denied, and continues to deny, the scale and effect
of these issues from both an under-reporting and medical effect perspective
for over 30 years, the military now accepts that the occurrence of smoke
and/or toxic fumes in the aircraft cockpit or cabin is more common than is
generally realised and there is some evidence that continued exposure to
small amounts of certain contaminants may produce chronic, long term, and
irreversible damage to humans [12].
References
1. Faith N (1966) Black box: why air safety is no accident. Boxtree, London
2. Cushing S (1994) Fatal words: communication clashes and aircraft crashes. University
of Chicago Press, Chicago
3. Forman P (1990) Flying into danger: the hidden facts about air safety. Mandarin,
London
4. Schiavo M (1997) Flying blind, ying safe. Avon, New York
5. Andersen N (1997) Broken wings: a ight attendants journey. Avia, Coquitlam,
Canada
6. BAe (2000) British Aerospace submission and evidence by British Aerospace systems,
to the Australian Senate inquiry into air safety (19992000) BAe 146 cabin air quality.
Parliament of Australia, Canberra
7. Spengler J, Burge H, Dumyahn T, Muilenburg M, Foresterm D (1997) Environmental
survey on aircraft and ground-based commercial transportation vehicles. Harvard
School of Public Health, Boston
8. Hocking MB (2002) Trends in cabin air quality of commercial aircraft: industry and
passenger perspectives. Rev Environ Health 17:149
9. ASHRAE (1990) Ventilation for acceptable indoor air quality: ANSI/ASHRAE Standard 62. American Society of Heating, Refrigerating and Air-Conditioning Engineers,
Atlanta
10. ODonnell A, Donnini G, Nguyen V (1991) Air quality, ventilation, temperature and
humidity in aircraft. ASHRAE J 4:4246
11. Winder C, Balouet JC (2002) The toxicology of commercial jet oils. Environ Res
89:146164
12. Singh B (2004) In ight smoke and fumes. Australia Aviation Safety Spotlight
13. Fox R (1997) Air-quality testing aboard Ansett Airlines BAe 146 aircraft, August 1997.
Allied Signal Aerospace Corporation
14. Senate of Australia (2000) Air safety and cabin air quality in the BAe 146 aircraft. Senate Rural and Regional Affairs and Transport References Committee, Final
report. Parliament of Australia, Canberra
15. Fox RB (2000) Air quality and comfort measurement aboard a commuter aircraft and
solutions to improve perceived occupant comfort levels. In: Nagda NL (ed) Air quality and comfort in airliner cabins, ASTM STP 1393. American Society for Testing and
Materials, West Conshohocken, PA
16. Winder C, Fonteyn P, Balouet JC (2003) Aerotoxic syndrome: A descriptive epidemiological survey of aircrew exposed to in-cabin airborne contaminants. J Occup Health
Safety Australia and New Zealand 18:321338
226
17. FAA (1965) Federal aviation regulations, transport category airplanes, airworthiness
standardsventilation section 25.831, February 1965 (amendment 2587 effective
7 May 1996). US Federal Aviation Authority, Washington
18. SAE (1981) Aerospace information report AIR 1539, 30/1/81, environmental control system contamination: sources of vaporous contamination. Society of Automotive
Engineers, Warrendale, PA
19. Mobil Oil Corporation (1983) Mobil Jet Oil II. Environmental Affairs and Toxicology
Department, New York, Correspondence
20. Allied Signal Aerospace (1991) Results of air quality testing for Dan-Air London
21. BAe (2001) BAe 146 manufacturers operations manual: notice to aircrew, operational
notice: No OP 16 and 43 (issue 1). British Aerospace Systems, Hateld
22. BAe (2002) Service information leaet SL 2145: Cabin air quality trouble shooting
advice and relevant modications. British Aerospace Systems, Hateld
23. Ansett Australia (2000) Submission and evidence by Ansett Australia to the Australian Senate inquiry into air safety (19992000) BAe 146 cabin air quality. Parliament of Australia, Canberra
24. Ansett Australia (1998) Consensus statement: external panel of specialists, BAe 146
odour occurrences. Ansett Australia, Brisbane
25. CASA (2000) Submission and evidence of the Civil Aviation Safety Authority of Australia to the Australian Senate inquiry into air safety (19992000) BAe 146 cabin air
quality. Parliament of Australia, Canberra
26. UK CAA (2002) Flight operations department communication (FODCOM) 21/2002:
smoke/fume occurrences. UK Civil Aviation Authority, London
27. UK CAA (2003) Safety initiative: hazardous contamination of ight deck cabin air. UK
Civil Aviation Authority, London
28. UK CAA (2004) Safety regulation group: cabin air quality paper. UK Civil Aviation
Authority, London
29. NJS (2000) Submission and evidence by National Jet Systems, Adelaide to the Australian Senate Inquiry into air safety (19992000) BAe 146 cabin air quality. Parliament of Australia, Canberra
30. AOPIS (2003) Aircraft air contamination: an ongoing health and safety issue, documentary. Aviation organophosphate information site. At: www.aopis.org
31. Best R (2001) Aviation air quality. certication of aircraft in Australia. In: Winder C,
Michaelis S, Weber RO (eds) Proceedings of the aviation air quality symposium, Australian Defence Force Academy, University of New South Wales, Sydney, 7 December
2000, pp 7882
32. JAA (2001) Joint aviation requirements: operations 1.420 occurrence reporting;
amendment 3, December 2001. Joint Aviation Authority
33. UK CAA (2001) CAP 382: Mandatory occurrence reporting scheme. UK Civil Aviation Authority, London. Available at http://www.caa.co.uk/publications/publicationdetails.asp?id=214
34. JAA (2003) Occurrence reporting 21.3, amendment 5, June 2003, AMC 208 and GAI
ACJ 20.X8. Joint Aviation Authority
35. CASA (1988) Australian civil aviation regulation 2. Civil Aviation Safety Authority of
Australia, Canberra
36. CASA (2001) Defect reports, civil aviation advisory publication 511(1). Civil Aviation Safety Authority of Australia, Canberra
37. CASA (1988) Australian Civil Aviation Regulation 248. Civil Aviation Safety Authoritity of Australia, Canberra
227
38. CASA (1990) Australian civil aviation regulation 50, 51, 52A. Civil Aviation Safety
Authority of Australia, Canberra
39. CASA (1998) Australian civil aviation regulations 21.3, 21.3(4)C. Civil Aviation Safety
Authority of Australia, Canberra
40. ATSB (1920) Air navigation act 1920 and editorial: new air safety accident
or incident report. Australian Transport Safety Bureau, Canberra. Available at
http://www.atsb.gov.au/aviation/editorial/forms_article.cfm
41. AOPIS (2004) Industry service bulletin data. Aviation organophosphate information
site at www.aopis.org
42. OBrien K (2002) BAe 146 aircraft: Q398. Hansard, Australian Parliament House, Canberra
43. Vamvakinou M (2004) Air safety cabin air quality: Q3011. Hansard, Australian Parliament House
44. Michaelis S (2003) A survey of health symptoms in BALPA Boeing 757 pilots. J Occup
Health Safety, Australia and New Zealand 19:253261
45. AFA (2003) Aircraft air quality: whats wrong with it and what needs to be done.
Submission by the American Association of Flight Attendants to the Aviation Subcommittee of the Transportation and Infrastructure Committee of the US House of
Representatives, Washington
46. FAAA (2000) Submission and evidence by Ansett Australia by the Flight Attendant
Association of Australia to the Australian Senate inquiry into air safety (19992000)
BAe 146 cabin air quality. Parliament of Australia, Canberra
47. SHK (1999) Report RL 2001:41e: Incident onboard aircraft SE-DRE during ight between Stockholm and Malm, M County, Sweden, on 12 November 1999. Swedish
Statens Haverkommission (Swedish Board of Accident Investigation), Stockholm
48. UK CAA (1999) Unpublished data, CAA library. Civil Aviation Authority, London
49. DETR, DOH (1999) Written evidence of UK Department Environment Transport and
Regions (DETR) and Department of Health (DOH). House of Lords, Select Committee
on Science and Technology report. Air Travel and Health, 19992000
50. DOT AAIB (2004) Aircraft accident report no 1/2004 (EW/C2000/11/4) BAe 146
G-JEAK, 1/2004. UK Department of Transport, Aircraft Air Accidents Investigation
Branch, London. At: http://www.dft.gov.uk/stellent/groups/dft_avsafety/documents/
page/dft_avsafety_029646-01.hcsp#P22_457
51. UK CAA (2004) Mandatory occurrence reporting database. UK Civil Aviation Authority, London
52. NTSB (2004) Aircraft incident data base. US National Transportation Safety Bureau
53. NRC/FAA (2002) The airliner cabin environment and the health of passengers and
crew. Airliner Cabin Environment Response Team (ACERRT), National Research
Council/US Federal Aviation Authority, Washington
54. CASA (2004) Major defects. Civil Aviation Safety Authority of Australia, Canberra.
Available at: http://www.casa.gov.au/avreg/aircraft/sdr/index.htm.
55. ATSB (2000) Submission and evidence by the Australian Safety Transport Bureau to
the Australian Senate inquiry into air safety (19992000) BAe 146 cabin air quality.
56. BALPA (2004) B757 and BAe 146 fume event reports (2004). British Airline Pilots
Association, London
57. James S (2003) Flight deck occurrences from MORs. Safety Regulation Group/
Building Research Establishment Cabin Air Conference, London
58. Pavlinovich N (2003) Witness statement to Australian Industrial Relations Commission. U2002/6475 N. Industrial Relations Commission, Perth
228
59. Knowles S (2003) Aircraft: Q 462. Hansard, Australian Parliament House

60. BAe (1991) Complaint of difculty report 27803. British Aerospace Systems, Hateld
61. Australian Senate (2000) Air safety and cabin air quality in the BAe 146 aircraft.
62. ATSB (2001) Occurrence brief 200002431, BAe 146, VH NJL, fumes in cabin from
oil leak in APU, 30 April 2000. Australian Transport Safety Bureau, Canberra. At:
http://www.atsb.gov.au/aviation/occurs/occurs_detail.cfm?ID=93
63. NJS(1999) Memorandum: FAAA oil fumes report. National Jet Systems, Adelaide
64. Devine R (2000) Captains statement in submission by Australian Federation of Air Pilots, Melbourne, to the Australian Senate inquiry into air safety (19992000) BAe 146
cabin air quality. Parliament of Australia, Canberra
65. Michaelis S, May R (2000) Submission and evidence by Michaelis S, First Ofcer,
May R, First Ofcer to the Australian Senate inquiry into air safety (19992000)
BAe 146 cabin air quality.
66. Janvrin S (2000) Chief Medical Ofcer, Medical Division, Safety Regulation Group, UK
Civil Aviation Authority, London. Correspondence to BALPA pilot member
67. Borger H (2003) High on the ight deck. Occupational Health and Safety Magazine.
CCH, Sydney
68. ATSB (1999) Occurrence brief 199702276, BAe 146, VH NJF, fumes in cabin from
oil leak in APU. Australian Transport Safety Bureau, Canberra, 1999. At: http://
www.atsb.gov.au/aviation/occurs/occurs_detail.cfm?ID=256
69. ATSB (2003) Occurrence brief 200205865, BAe 146, VH NJX, Fumes in cabin from
oil leak in APU, 2 December 2002. Australian Transport Safety Bureau. At: http://
70. Hamilton T (2004) UK CAA. Correspondence to British Airline Pilots Association
Aircraft Environment Task Group
71. Mullin M (2001) Aircraft fumes. House of Commons, Hansard
72. TSBC (2003) SR111 aviation investigation report In-ight re leading to collision with
water Swissair Transport Limited McDonnell Douglas MD-11 HB-IWF Peggys Cove,
Nova Scotia 5 nm SW, 2 September 1998: report no A98H0003. Transportation Safety
Board of Canada, Gatineau, Quebec
73. BAe (2003) Service bulletin SB 21150: Air conditioning - to inspect engine oil seals,
APU and ECS jet pump and air conditioning pack for signs of oil contamination.
British Aerospace Systems, Hateld
74. BAe (2004) Service bulletin SB 21156: Air conditioning - to inspect air conditioning
sound-attenuating ducts for signs of oil contamination. British Aerospace Systems,
Hateld
75. ICAO (2001) Aircraft accident and incident investigation, 9th edn, Annex 13. International Civil Aviation Organisation, Paris
76. AOPIS (2004) Commercial bias and use of misinformation within commercial aviation in relation to cabin air quality. Unpublished report October. At: www.aopis.org
77. FAA (2005) Paper presented by A. Azevedo, federal aviation authority, to the BALPA
2005 contaminated air protection conference. Data review of instances of in-cabin
fumes from propulsian system events

DOI 10.1007/b107246
Published online:
Crew Effects from Toxic Exposures on Aircraft

C. Winder (u) S. Michaelis
School of Safety Science, The University of New South Wales, NSW 2052 Sydney,
Australia
c.winder@unsw.edu.au, s.michaelis@student.unsw.edu.au
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
230
2
2.1
2.2
2.3
Toxic Ingredients of Jet Oils . .

The Substituted Diphenylamine
N-Phenyl-alpha-naphthylamine
Tricresyl Phosphate . . . . . . .
.
.
.
.
231
232
233
234
Effects of Aircraft Oil Leaks on Crew . . . . . . . . . . . . . . . . . . . . .
237
Other Factors of Importance to the Aviation Industry . . . . . . . . . . .
240
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
242
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
245
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Abstract The cabin of an airplane is a specialised working environment and should be

considered as such. The oils and hydraulics used in airplane engines are toxic, and specic ingredients of such materials are irritating, sensitising and neurotoxic. If oil or
hydraulic uids leak out of engines, this contamination may be in the form of unchanged
oil/uid, degraded oil/uid from long use in the engine, combusted oil/uid or pyrolised
oil/uid, in the form of gases, vapours, mists and particulate matter. If leak incidents
occur and the oil/uid is ingested into bleed air and is passed to the ight deck and
passenger cabins of airplanes in ight, aircrew and passengers may be exposed to contaminants that can affect their health and safety. Where contamination of air in the
ight deck and passenger cabin occurs that is sufcient to cause symptoms of discomfort, fatigue, irritation or toxicity, this contravenes the air quality provisions of Federal
Aviation Regulations, most notably FAR 25.831. Symptoms of immediate or short-term
nature and reported by exposed staff in single or few leak incidents are consistent with
the development of irritation and discomfort. Symptoms of a long-term nature (that is,
sustained symptoms for at least six months) reported by some exposed staff following
small to moderate numbers of leak incidents are consistent with the development of an
irreversible discrete occupational health condition, termed aerotoxic syndrome. Features
of this syndrome are that it is associated with air crew exposure at altitude to atmospheric contaminants from engine oil or other aircraft uids, temporarily juxtaposed
by the development of a consistent symptomology including short-term skin, gastrointestinal, respiratory and nervous system effects, and long-term central nervous and
immunological effects.
Keywords Aircraft air contamination Substituted diphenylamine
Phenyl-alpha-naphthylamine Tricresyl phosphate Triorthocresyl phopshate
Organophosphate induced chronic neurotoxicity (OPICN) Aerotoxic syndrome
230
C. Winder S. Michaelis
Abbreviations
CAS
Chemical Abstracts Service
COPIND Chronic organophosphate-induced neuropsychological disorder
DOCP
Di-ortho cresyl phosphate
FAR
U.S. Federal Aviation Regulation
JAR
Joint Aviation Regulation
MOCP
Mono-ortho cresyl phosphate
MSDB
Material Safety Data Bulletin
NTE
Neurotoxic esterases
OHS
Occupational Health and Safety
OP
Organophosphorus
OPICN
Organophosphorus ester-induced chronic neurotoxicity
OPIDN Organophosphorus ester-induced delayed neurotoxicity
PAN
Phenyl-alpha naphthylamine
TCP
Tricresyl phosphate
TOCP
Tri-ortho cresyl phosphatekp
1
Introduction
As already noted in Chapters 10 and 11, the oils and hydraulics used in
aircraft engines can be toxic, and specic ingredients of oils can be irritating, sensitising (such as phenyl-alpha-naphthylamine) or neurotoxic (for
example, ortho-containing triaryl phosphates such as tri-orthocresyl phosphate) [1, 2]. If oil or hydraulic uid leaks occur, this contamination may be in
the form of unchanged material, degraded material from long use, combusted
or pyrolised materials. These materials can contaminate aircraft cabin air in
the form of gases, vapours, mists and aerosols.
Notwithstanding emergency situations, a range of other situations can
arise whereby aircraft cabin air can be contaminated [3]. These include:
uptake of exhaust from other aircraft or on ground contamination sources,

application of de-icing uids,
hydraulic uid leaks from landing gear and other hydraulic systems,
excessive use of lubricants and preservative compounds in the cargo hold,
preservatives on the inside of aircraft skin,
large accumulations of dirt and brake dust may build up on inlet ducts
where auxilliary power units extract air from near the aircraft belly,
ingestion of oil and hydraulic uid at sealing interfaces, around oil cooling
fan gaskets and in worn transmissions,
oil contamination from synthetic turbine oil,
engine combustion products (for example, defective fuel manifolds, seal
failures, engine leaks).
Signicant contaminants include: aldehydes; aromatic hydrocarbons; aliphatic
hydrocarbons; chlorinated, uorinated, methylated, phosphate or nitrogen
231
compounds; esters; and oxides [46]. One additional problem is the lower
partial pressure of oxygen that is present in the cabins of planes ying at
altitude [7].
To date, most studies that have been carried out to measure atmospheric
contamination in aircraft by engine oil leaks or hydraulic uids are sufciently awed on procedural and methodological grounds as to render their
conclusions invalid. Further, no monitoring has occurred during an oil leak.
International aviation legislation such as the US Federal Aviation Regulations (FAR) and airworthiness standards for aircraft air quality state crew
and passenger compartment air must be free from harmful and hazardous
concentrations of gases or vapors [8]. Where contamination of air in the
ight deck and passenger cabin occurs that is sufcient to cause symptoms of
discomfort, fatigue, irritation or toxicity, this contravenes such standards and
legislation.
Inhalation is an important route of exposure, with exposure to uncovered
skin being a second, less signicant route (for example, following exposure to
oil mists or vapours). Ingestion is unlikely.
Occasionally, such exposures may be of a magnitude to induce symptoms
of toxicity. In terms of toxicity a growing number of aircrew are developing
symptoms following both short-term and long-term repeated exposures, including dizziness, fatigue respiratory problems, nausea, disorientation, confusion, blurred vision and tremors [911]. Neurotoxicity is a major ight
safety concern especially where exposures are intense [12].
2
Toxic Ingredients of Jet Oils
The engine oils that are used in jet engines are precision oils that need to operate in extreme conditions. Some commercial jet oils have been in use as
engine oils in aviation for decades. For example, Mobil USA note that Mobil
Jet Oil II (a jet oil with close to half the market share) has been essentially
unchanged since its development in the early 1960s and most changes have
involved slight revisions of the ester base stock due to changes in raw material
availability [13].
Chemical exposures in aircraft are not unheard of. In 1953, the US
Aeromedical Association rst expressed their concerns about the toxicity
risks of cabin air contamination by hydraulics and lubricants [14]. Other
risks have been identied more recently, either as part of the chemicals routinely used in maintaining aircraft [15], or as toxicological factors in aviation
accidents [16, 17].
A complex approval process exists for ensuring that materials used in
aviation are manufactured to relevant standards, and the jet engine oil specication of the US Navy MIL-PRF-23699 is used for jet oils. This process of
232
approval and re-approval for new product formulations has meant that there
is some resistance to modifying formulations (for example, for health and
safety reasons).
Consequently, changing approved formulations is not conducted without
signicant justication. In the case of the additive tricresyl phosphate (TCP),
manufacturers have been reluctant to modify product formulations by substituting toxic TCP additives that perform well in critical applications. This has
meant that potentially toxic products have continued to be available and used
long after their toxicity was recognised [18].
It is not known if an approved formulation containing, for example 3%
tricresyl phosphate, is considered a change in formulation if the proportion
of individual isomers in the TCP mixture is altered, but the 3% remains unchanged. However, as Mobil indicate, only the base stock esters have been
modied over the past thirty or so years, suggesting that the mixture of isomers in TCP stock has not been changed.
Using a typical commercial Jet Oil (Mobil Jet Oil II), various sources, such
as the suppliers label on the cardboard box the cans are shipped in, the
product Material Safety Data Bulletin (MSDB), and information from the
manufacturer, list the following ingredients [6]:
synthetic esters based in a mixture of 95% C5 -C10 fatty acid esters of pentaerythritol and dipentaerythritol;
3% tricresyl phosphate (Phoshoric acid, tris(methylphenyl) ester, CAS No
1330-78-5);
1% phenyl-alpha-naphthylamine (PAN) (1-Naphthalenamine, N-phenyl,
CAS No 90-30-2);
a substituted diphenylamine;
a last entry ingredients partially unknown is also noted on some documentation.
Of these ingredients, the most toxicologically signicant components are the
substituted diphenylamine, phenyl-alpha-naphthylamine (PAN) and tricresyl
phosphate (TCP).
2.1
The Substituted Diphenylamine
The substituted diphenylamine is variously reported as benzamine, 4-octylN-(4-octylphenyl), (CAS No 101-67-7) or 0.11% N-phenyl-benzeneamine,
reaction product with 2,4,4-trimethylpentene (CAS No 68411-46-1), and used
as an antioxidant, in concentrations not greater than 1% (see Fig. 1).
There is little toxicity data available for this ingredient, although it is not
believed to be toxic by single exposure (no data on long-term exposure). The
disclosure of this ingredient in hazard communication by identity probably
233
Fig. 1 Substituted diphenylamines
relates to its environmental effects, such as poor biodegradability and toxicity

to aquatic invertebrates [19].
2.2
N-Phenyl-alpha-naphthylamine
N-Phenyl-alpha-naphthylamine, (CAS No 90-30-2), also known as phenylalpha-naphthylamine (PAN), is a lipophilic solid as an antioxidant in lubrication oils and as a protective agent in rubber products (see Fig. 2). In these
products, the chemical acts as a radical scavenger in the auto-oxidation of
polymers or lubricants. It is generally used in these products at a concentration of about 1% (its concentration in jet oils). The commercial product has a typical purity of about 99%. Named impurities are: N-phenyl-2naphthylamine (CAS No 135-88-6, 500 to below 5000 ppm), 1-naphthylamine
(below 100500 ppm) and 2-naphthylamine (below 3 to 50 ppm), aniline (below 100 to 2500 ppm), 1-naphthol (below 5000 ppm), 1,1-dinaphthylamine
(below 1000 ppm).
PAN is readily absorbed by mammalian systems and rapidly biotransformed [20]. Both urine and faeces appear to be the main routes of excretion [21].
By single dosing, PAN has a short-term low toxicity, with LD50 s above
1 g/kg. The chemical has a similar mechanism of toxicity to many aromatic
amines, of methaemoglobin production. PAN is not irritating in primary skin
and eye irritation studies. However, in a guinea pig maximisation test, PAN
was shown to be a strong skin sensitiser [22]. This result is supported by case
studies in exposed workers [23, 24]. At the concentration used (1%), Mobil Jet
Fig. 2 N-Phenyl-1-naphthylamine
234
Oil II meets cut off criteria (1%) for classication as a hazardous substance in
Australia for sensitisation properties.
Most genotoxicity studies report negative results, suggesting little genotoxicity potential [21].
Most repeated dose toxicological studies focus on its potential carcinogenicity. An experimental study, using both PAN and the related compound
N-phenyl-2-naphthalenamine administered subcutaneously to mice found
a heightened incidence of lung and kidney cancers [25]. While the methodology used in this study makes evaluation of the results problematic (use of
one gender, small sample sizes, limited number of dose groups, subcutaneous
administration as an inappropriate route of exposure, and so on). A high incidence of various forms of cancer was also found among workers exposed
to antirust oil containing 0.5% PAN [26]. While these animal and human results offer only limited information, they are at least supportive of a mild
carcinogenic effect.
This must be contrasted with the results of long-term carcinogenicity
bioassays in rats and mice conducted by the US National Toxicology Program with the structurally related N-phenyl-2-naphthylamine (studies were
not carried out on PAN), which have not reported any carcinogenic potential
for this chemical [27].
2.3
Tricresyl Phosphate
Tricresyl phosphate (CAS No 1330-78-5), is also known as phosphoric acid,
tris(methylphenyl) ester or tritolyl phosphate. TCP is a blend of ten tricresyl phosphate isomer molecules, plus other structurally similar compounds,
including phenolic and xylenolic compounds. TCP is a molecule comprised
of three cresyl (methylphenyl) groups linked to a phosphate group. The location of the methyl group in the cresyl group is critical for the expression
of neurotoxicity, with ortho-, meta- or para- prexes that denote how far
apart the hydroxyl and methyl groups are on the cresol molecule. Technically, there are 27 (33 ) different combinations of meta, ortho and para cresyl
groups in TCP (see Fig. 3). Since the apparently different three-dimensional
structures of the molecule are not chemically locked in place, they are not
optical isomers. Therefore, structures with similar numbers of cresyl groups
(such as ppm, pmp and mpp) are considered the same molecules. This gets
the apparent 27 structures down to the real ten isomers conventionally described.
CAS number descriptors for tricresyl phosphate chemicals have been introduced to differentiate between ortho-cresyl and non-ortho-cresyl isomers:
CAS No 78-30-8 tricresyl phosphate (containing o-o-o, o-o-m, o-o-p, o-mm, o-m-p, o-p-p isomers);
235
Fig. 3 Structure of Tricresyl phosphate
CAS No 78-32-0 tricresyl phosphate (containing m-m-m, m-m-p, m-p-p,

p-p-p isomers).
TCP is a compound with a toxicity typical of the organophosphorus compounds. Human toxicity to organophosphorus (OP) compounds has been
known since at least 1899, when neurotoxicity to phosphocreosole (then used
in the treatment of tuberculosis) was reported [28]. The study of OP toxicity is extensive, and generally characterised by a toxicity of inhibition of the
esterase enzymes, most particularly cholinesterases [29] and neurotoxic esterases [30]. The mechanism of effect is phosphorylation [31].
Signs of low level intoxication include headache, vertigo, general weakness,
drowsiness, lethargy, difculty in concentration, slurred speech, confusion,
emotional lability and hypothermia [32]. The reversibility of such effects has
been questioned [33].
Signs of poisoning are usually foreshadowed by the development of early
symptoms related to acetylcholine overow and include salivation, lacrimation, conjunctivitis, visual impairment, nausea and vomiting, abdominal
pains and cramps, diarrhoea, parasympathomimetic effects on heart and circulation, fasciculations and muscle twitches [34]. This is the basic site of
inhibition for all OP molecules [35, 36].
A second reaction with certain OPs (including TCP) leads to further
neurotoxic and neuropathological changes. This is inhibition of neurotoxic
esterases (NTE) which produces a progressive distal symmetrical sensorimotor mixed peripheral neuropathy, called organophosphorus-induced delayed neurotoxicity (OPIDN) [36, 37]. The mechanism of toxicity is now fairly
well understood, as indeed are the organophosphorus structures which are
predicted to cause OPIDN [38].
OPIDN has a severe pathology. It is quite likely that such a severe condition
would be presaged with a range of clinical and pre-clinical signs and symptoms. These have been reported extensively, and an intermediate syndrome
was dened in 1987 [39].
More recently, chronic exposure to organophosphates has been associated
with a range of neurological and neuropsychological effects [4044]. Such
236
symptoms (mainly neurological and neurobehavioural symptoms) may also

be seen in exposed individuals who have been sufciently fortunate in not
having exposures that were excessive enough in intensity or duration to lead
to clinical disease.
A distinct condition chronic organophosphate-induced neuropsychological disorder (COPIND) has been described, of neurological and neuropsychological symptoms [45]. These include:
diffuse neuropsychological symptoms (headaches, mental fatigue, depression, anxiety, irritability);
reduced concentration and impaired vigilance;
reduced information processing and psychomotor speed;
memory decit and linguistic disturbances.
COPIND may be seen in exposed individuals either following single or
short-term exposures leading to signs of toxicity [46], or long-term low level
repeated exposure with (often) no apparent signs of exposure [43]. The basic
mechanism of effect is not known, although it is not believed to be related to
the esterase inhibition properties of organophosphorus compounds. It is also
not known if these symptoms are permanent.
In addition, since the introduction and extensive use of synthetic organophosphorus compounds in agriculture and industry half a century ago, many
studies have reported long-term, persistent, chronic neurotoxicity symptoms
in individuals as a result of acute exposure to high doses that cause acute
cholinergic toxicity, or from long-term, low-level, subclinical doses of these
chemicals [4749]. The neuronal disorder that results from organophosphorus ester-induced chronic neurotoxicity (OPICN), which leads to long-term
neurological and neurobehavioral decits and has recently been linked to
the effects being seen in aircrew despite OP levels being too low to cause
OPIDN [50].
Furthermore, OPICN induced by low-level inhalation of organophosphates
present in jet engine lubricating oils and the hydraulic uids of aircraft
could explain the long-term neurological decits consistently reported by
crewmembers and passengers, although organophosphate levels may have
been too low to produce OPIDN [50].
While the description above relate to the general toxicity of OPs, they
are characteristic of exposure to tricresyl phosphate. The ten isomers that
make up TCP are toxicologically different, and it is well established that the
ortho-containing isomers are the most toxic [5153]. Of the ten isomers of
TCP, six contain at least one ortho-cresyl group: three mono-ortho (MOCP)
isomers, two di-ortho (DOCP) isomers and one tri-ortho (TOCP) isomer, triorthocresyl phosphate (TOCP). Other, similar ortho- containing chemicals,
such as the xylenols and phenolics, are also present in commercial TCP formulations in small amounts. Manufacturers of TCP have reduced the levels of
ortho-cresyl and ortho-ethylphenyl isomers to reduce the potential for neu-
237
rotoxicity of products containing TCP [18]. How much these renements had
removed the toxic impurities outlined above is not known. Indeed, toxicity was still being detected in commercially available products in 1988 [18],
and questions have been raised about the lack of consistency between stated
ingredient data and actual amounts of toxic isomers present in commercial
formulations, and their impact on exposed individuals [6].
3
Effects of Aircraft Oil Leaks on Crew
Where exposure may be to high levels of airborne contaminants, it is not unreasonable for signs of irritancy and discomfort to be observed. Similarly,
it is not unreasonable to consider that a person exposed to a chemical that
contains 1% of a sensitiser and 3% of a neurotoxicant might show signs of irritancy and neurotoxicity. These symptoms are often reported in air crew who
may be exposed to aircraft uids.
The earliest case found in the literature of toxicity following jet oil exposure and adverse health problems in air crew was reported in 1977 [55]. A previously healthy member of an aircraft ight crew was acutely incapacitated
during ight with neurological impairment and gastrointestinal distress. His
clinical status returned to normal within a day. The aetiology of his symptoms
was related to an inhalation exposure to aerosolised or vapourised synthetic
lubricating oil arising from a jet engine of his aircraft.
Other studies of exposures in aircraft exist in the literature, including
a 1983 study of eighty nine cases of smoke/fumes in the cockpit in the US Air
Force [56], a 1983 study of Boeing 747 ight attendants in the USA (this paper
linked symptoms to ozone) [57], a 1990 study of aerospace workers [58], and
a 1998 study of BAe 146 ight crews in Canada over a four-month period [9].
A recent report of seven case studies considered representative of the common symptoms of irritancy and toxicity described similar symptoms [10],
and a follow up survey by the same research group reported similar ndings
in a larger group of fty crew respondents [59]. Two union-based studies in
pilots provide additional data [60, 61].
These studies investigated different exposures and situations, and the
range of symptoms in these studies was quite broad, affecting many body
systems. However, there are common themes in symptom clusters in these
studies, as shown in Table 1 overleaf.
While this Table shows a long list of symptoms, it is possible to characterise
many symptoms more consistently. For example, different papers report
dizziness or loss of balance or light-headededness or feeling faint or feeling
intoxicated or disorientation. It would be incorrect to regard such symptoms
as being entirely different from each other they point to a basic neuropsychological dysfunction affecting balance. But rather than dismissing such
4%
73%
3/7
2% 4/7
Neurotoxic symptoms
26% 23% 15%

26%
11% 13%
9%
8%
8% 6/7
3/7
1% 4/7
3% 3/7
2% 4/7
Nausea, vomiting, gastrointestinal symptoms

Abdominal spasms/cramps/diarrhoea
Blurred vision, loss of visual acuity
Shaking/tremors/tingling
Numbness (ngers, lips, limbs), loss of sensation
Gastrointestinal symptoms
Skin symptoms secondary

to irritation
Sinus congestion
35% 54%
5% 2/7
Nose bleed
17%
1/7
Throat irritation, burning throat, gagging and coughing 2% 64% 57% 43% 2/7
Cough
69%
2/7
Difculty in breathing, chest tightness
68%
3/7
Loss of voice
35%
1/7
Rashes, blisters (on uncovered body parts)
36%
4/7
7/7
35% 74% 57% 24% 4/7
4%
58%
20%
50%
40%
4%
10% 12%
5% 15%
5% 16%
5% 4%
8%
32% 37%
48% 16%
4%
76%
12%
62%
76%
14%
62% 26%
[10] [59] [60] [61]

7
50
21 106
Respiratory symptoms secondary

to irritation
Symptoms of direct irritation

to eye, airways or skin
Fainting/loss of consciousness/grey out

Respiratory distress, shortness of breath,
respiration requiring oxygen
Irritation of eyes, nose and throat
Eye irritation, eye pain
Loss of consciousness/
Inability to function
Reference [56] [57] [58] [9]

Number of cases/reports 89 248
53 112
Sign or symptom
Symptom cluster
Table 1 Studies reporting signs and symptoms in aircrew
238
26% 20% 60%

3% 6%
29%
32%
Disorientation
Dizziness/loss of balance
Light-headed, feeling faint or intoxicated
Chest pains
Severe headache, head pressure
Fatigue, exhaustion
Chemical sensitivity
Immune system effects
General increase in feeling unwell
Behaviour modied, depression, irritability
Change in urine
Joint pain, muscle weakness, muscle cramps
Cognitive/neuropsychological
symptoms related to higher
nervous system function
Nonspecic general symptoms
26%
47%
35% 54%
7% 81%
25% 52%
Trouble thinking or counting, word blindness, confusion, 26% 39% 42%

coordination problems
Memory loss, memory impairment, forgetfulness
42%
Neurological symptoms related

to basal nervous system function
15%
6%
32%
6%
26%
Reference [56] [57] [58] [9]

Number of cases/reports 89 248
53 112
Sign or symptom
Symptom cluster
Table 1 (continued)
2/7
4/7
16%
72% 16%
21%
22%
86% 21%
62% 21%
72% 26%
21%
21%
40%
4%
38% 5%
30%
33%
30%
10%
3%
27%
8%
3%
33%
66% 26% 11%
7/7
4/7
4/7
7/7
2/7
7/7
7/7
4/7
58% 21% 22%
6/7
[10] [59] [60] [61]

7
50
21 106

239
240
symptoms as being multitudinous and variable [62], it may be more appropriate to re-categorise symptoms with clearer denitions, so that the articial
distinctions between symptom reporting can be claried, and a shorter list of
symptom clusters be developed (as shown in the rst column of Table 1).
4
Other Factors of Importance to the Aviation Industry
The cockpit or cabin of an aircraft is a unique environment. It is a specialised working environment for the air crew that cannot (indeed, must not)
be equated with workplaces at sea level, or workplaces where specialised ventilation and escape are possible [63].
The process of aircraft pressurisation means that the working environment
is hypoxic. Flying crew are required to conduct complex operations requiring
high order cognitive skills and coordination expertise. Flight attendants may
be required to direct emergency procedures requiring composure and condence. Anything that may have an impact on the delivery of these tasks can
have serious consequences.
A lowered level of oxygen may in turn have an impact on the emergence of
adverse health problems to toxic exposures.
For these reasons, the application of conventional occupational health and
safety procedures to this specialised environment are inappropriate. Examples of these include:
permissible exposure standards for occupational exposures to airborne
contaminants extenuating circumstances on board aircraft (including
humidity and cabin pressure) have not been studied to the extent that
a suitable exposure standard can be identied that incorporates these factors or identies interactions between factors [64];
There is not agreement on a toxicological standard among aviation toxicologists to apply to aircraft. Exposure standards were developed by
the American Conference of Industrial Hygienists (ACGIH) for the average worker at or near sea level pressure in relatively good health. Flight
crew work in conditions where atmospheric pressure is reduced. [67] Most
chemicals do not have exposure standards and of those that do exist most
are still regulated by voluntary standards set before 1971, when adopted
uncritically and unchanged with new science having had no impact on
them. [68];
it is incorrect to assume the exposure standard for TOCP as being adequately protective for a TCP containing mixture of TCP isomers as other
ortho isomers (MOCPs, DOCPs) are more toxic than TOCP [65];
procedures for assessing the risks of exposures to more than one chemical,
that may act in synergy to produce toxicity (for example, carbon monoxide and lowered oxygen);
241
under circumstances of exposure to mixtures of contaminants, levels may

be well below recommended levels in currently accepted exposure standards yet still generate complaints or signs and symptoms, because they
act in synergy with other contaminants or because some standards may
be outdated and have not incorporated more recent scientic and medical
evidence [64];
ventilation rates for buildings.
Occupational exposure standards may be inadequate to protect nonworkers,
for example passengers.
Further, an oil leak from an engine at high pressure and temperature may
burn or pyrolise before it enters the cabin. This produces carbon-containing
materials which, in the presence of energy and oxygen, produce the two
oxides of carbon: carbon dioxide (CO2 ) and carbon monoxide (CO). The
rst of these (CO2 ) is produced in the presence of an abundance of oxygen,
the second (CO), where stoichiometric concentrations of oxygen are lacking
(usually in conditions of incomplete combustion). Both of these oxides are
gases, one (carbon monoxide) is quite toxic at low concentrations, causing
toxic asphyxiation. Single or short-term exposure to CO insufcient to cause
asphyxiation produces headache, dizziness, and nausea; long-term exposure
can cause memory defects and central nervous system damage, among other
effects [66].
Many combustion and pyrolysis products are toxic. The toxic asphyxiants,
such as carbon monoxide, have already been introduced above. Some thermal
degradation products, such as acrolein and formaldehyde are highly irritating. Others, such as oxides of nitrogen and phosgene, can produce delayed
effects. Still others, such as particulate matter (for example, soot) can carry
adsorbed gases deep into the respiratory tract where they may provoke a local
reaction or be absorbed to produce systemic effects.
A leak of such an oil from an engine operating at altitude would see
most of the oil pyrolise once it leaves the conned conditions of temperature and pressure operating in the engine. While it seems reasonable that any
ingredients with suitable autoignition or degradation properties that allow
such a transformation after release from the engine could be radically transformed, it is possible to speculate in only general terms about the cocktail of
chemicals that could form. Presumably it would include carbon dioxide, carbon monoxide, partially burnt hydrocarbons (including irritating and toxic
by-products, such as acrolein and other aldehydes, and TCP (which is stable at high temperatures). These contaminants will be in gas, vapour, mist
and particulate forms. These contaminants could not be classied as being of
low toxicity. The possible problems that might arise from exposure to such
a cocktail cannot be dismissed without proper consideration.
242
5
Conclusions
What emerges in the analysis of this data is a pattern of symptoms related to
local effects to exposure to an irritant, overlaid by development of systemic
symptoms in a number of body systems, including the nervous system, respiratory system, gastro-intestinal system, and possibly the immune system
and cardiovascular system. These symptoms may be expressed specically to
these systems, or may be seen more generally, such as headache, behavioural
change or chronic fatigue.
The symptoms reported by exposed individuals as shown in Table 1 are
sufciently consistent to indicate the development of a discrete occupational
health condition, and the term aerotoxic syndrome is introduced to describe
it (Etymology: aero refers to aviation, toxic to toxicity of exposure and associated symptoms). Features of this syndrome are that it is associated with
air crew exposure at altitude to atmospheric contaminants from engine oil or
other aircraft uids, temporarily juxtaposed by the development of a consistent symptomology including short-term skin, gastro-intestinal, respiratory
and nervous system effects, and long-term central nervous, respiratory and
immunological effects (see Table 2). This syndrome may be reversible following brief exposures, but features are emerging of a chronic syndrome
following signicant exposures [10, 11, 59].
The presence of contaminants in ight decks and passenger cabins of commercial jet aircraft should be considered an air safety, occupational health
and passenger health problem:
As shown in the section on leaks, incidents involving leaks or engine oil
and other aircraft materials into the passenger cabin of aircraft occur
frequently and are unofcially recognised through service bulletins, defect statistics reports and other sources. From the analysis in Chapter 11,
the rates of occurrence of incidents are higher than the aviation industry
admits, and for some models of aircraft are signicant. These need appropriate reporting, follow up investigations and health investigations for
those exposed.
The oils used in aircraft engines contain toxic ingredients which can cause
irritation, sensitisation and neurotoxicity. This does not present a risk to
crew or passengers as long as the oil stays in the engine. However, if the oil
leaks out of the engine, it may enter the air conditioning system and cabin
air. Where these leaks cause crew or passenger discomfort, irritation or
toxicity, this is a direct contravention of the US Federal Aviation Authoritys and the European Joint Aviation Authorities airworthiness standards
for aircraft ventilation (FAR/JAR 25.831).
As indicated by manufacturer information and industry documentation,
aviation materials such as jet oils and hydraulic uids are hazardous and
243
Table 2 Aerotoxic syndrome: short- and long-term symptoms

Short term exposure
Long term exposure
Neurotoxic symptoms: blurred or tunnel

vision, nystagmus, disorientation, shaking
and tremors, loss of balance and vertigo,
seizures, loss of consciousness, parathesias;
Neuropsychological or Psychotoxic
symptoms: memory impairment, headache,
light-headedness, dizziness, confusion and
feeling intoxicated;
Neurotoxic symptoms: numbness (ngers,

lips, limbs), parathesias;
Neuropsychological or Psychotoxic
symptoms: memory impairment
forgetfulness, lack of coordination, severe
headaches, dizziness balance, sleep
disorders;
Gastro-intestinal symptoms: nausea,

vomiting;
Respiratory symptoms: cough, breathing
difculties (shortness of breath), tightness
in chest, respiratory failure requiring
oxygen;
Cardiovascular symptoms: increased heart
rate and palpitations;
Gastro-intestinal symptoms: salivation,

nausea, vomiting, diarrhoea;
Respiratory symptoms: breathing
difculties (shortness of breath), tightness
in chest, respiratory failure, susceptibility
to upper respiratory tract infections;
Cardiovascular symptoms: chest pain,
increased heart rate and palpitations;
Skin symptoms: skin itching and rashes,
skin blisters (on uncovered body parts),
hair loss;
Irritation of eyes, nose and upper airways.
Irritation of eyes, nose and upper airways;

Sensitivity: signs of immunosuppression,
chemical sensitivity leading to acquired
or multiple chemical sensitivity
General: weakness and fatigue (leading to
chronic fatigue), exhaustion, hot
ashes, joint pain, muscle weakness and
pain.
contain toxic ingredients. If such uids leak into the air supply, cabin
and ight deck, toxic exposures are possible. Presently, the aircraft manufacturers, airline operators and the aviation regulators deny that this is
a signicant problem.
Leaks of oil and other uids into aircraft may be considered of a nuisance
type, but where they affect the health and performance of crew, or the
health of passengers, this is to be considered a ight safety and health
issue and must be given appropriate priority.
Pilots continue to y when experiencing discomfort or symptoms. There
is a lack of understanding by pilots regarding the toxicity of the oil leaks,
occupational health and safety (OHS) implications and the necessity to
use oxygen. This is further compounded by the airline health professionals
244
who, when confronted with a pilot who has been exposed in a fume event
and who is concerned about its consequences, have a poor understanding
of the short and long-term medical issues that may arise and tend to be
dismissive about their implications.
Attempts by the industry to minimise this issue, such as acceptance of
under-reporting of incidents, inadequate recognition of the extent of the
problem, inadequate adherence/interpretation of the regulations, inadequate monitoring, inappropriate use of exposure standards and care provided to crew reporting problems, have perpetuated this problem.
The health implications, both short and long-term, following exposure to
contaminants being reported by crew and passengers must be properly
addressed. A syndrome of symptoms is emerging, called aerotoxic syndrome, suggesting these exposures are common and a substantial group of
affected individuals exists.
Where contaminants impair the performance or affect the ability of pilots to y planes, as has been reported for a number of incidents, this is
a major safety problem. Where contaminants cause undue discomfort or
even transient health effects in staff and passengers, this is a breach of FAR
25.831 and other regulations.
Contaminants in the air of an occupational environment should, under normal circumstances, alert management to a potential problem [63]. Proper
medical and scientic research needs to be undertaken in order to help airline
management and crew to better understand both the short-term and longterm medical effects of being subjected to air contamination.
Over the past fty years, the concept of duty of care has emerged as one of
the most important legal responsibilities for employers. In the workplace, the
duty of care of an employer to its workers has been crystallised into OHS legislation. Aviation safety is something that a person outside the industry would
understand to cover all aspects of safety, including the health and safety of
its workers. However, this does not seem to be how all industry insiders see
it. Many in the industry see aviation safety as being about making sure the
planes keep ying. Both the aviation regulators and the airlines themselves
think that OHS is not their business which is strange, because if they do not
look after the health and safety of workers in the industry, then who will?
More scientic and medical research is needed on the short and longterm effects of exposure to contaminated air and, until this is completed, all
areas of the aviation industry should take fume exposure events seriously;
they should be seen as an important part of educating crew and the aviation
industry, thereby addressing the problem.
Many of the worlds leading experts who have seen aircrew from around
the globe or were familiar with the issue spoke at a recent conference held in
London by the British Airline Pilot Association (BALPA) looking at the issues
of contaminated air by engine oils and concluded:
245
There is a workplace problem resulting in chronic and acute illness

amongst ight crew (both pilots and cabin crew).
The workplace in wich these illnesses are being induced is the aircraft
cabin environment. This is the resulting in signicant ight safety issues, in
addition to unacceptable ight crew personnel healt implications.
Further, we are concerned the passengers may also be suffering from similar symptoms to those exhibited by ight crew.
References
1. AMA CAT (1953) Aviation Toxicology: an Introduction to the Subject and a Handbook of Data. AeroMedical Association Committee of Aviation Toxicology, Blakiston,
USA
2. Tupper CR (1989) Chemical hazards in aeromedical aircraft. Aviation, Space and Environmental Medicine 60:7375
3. ASHRAE (1999) Air Quality Within Commercial Aircraft: ASHRAE Standard 161.
American Society for Heating, Refrigeration, Airconditioning and Energy, Atlanta
4. Mattie DR, Hoeich TJ, Jones CE, Horton ML, Whitmire RE, Godin CS, Femming CD,
Andesen ME (1993) The comparative toxicity of operational air force hydraulic uids.
Toxicology and Industrial Health 9:9951016
5. Hewstone RK (1994) Environmental health aspects of lubricant additives. Science of
the Total Environment 156:243254
6. Winder C, Balouet J-C (2002) Toxic ingredients in commercial jet fuels. Environmental Research 89:146164
7. Balouet J-C, Kerguelen M, Winder C (2001) Toxicity and hypoxia: Hyperbaric pressure and LC50 s for Carbon monoxide or Hydrogn cyanide. Toxicology 164:164
8. FAA (1965) Federal Air Regulations, Transport Category Airplanes, Airworthiness
Standards, Section 25.831. US Federal Aviation Authority, Washington, DC
9. van Netten C (1998) Air quality and health effects associated with the operation of the
BAe 146-200 aircraft. Applied Occupational and Environmental Hygiene 13:733739
10. Balouet J-C, Winder C (1999) Aerotoxic syndrome in air crew as a result of exposure
to airborne contaminants in aircraft. Paper presented at the American Society of Testing and Materials (ASTM) Symposium on Air Quality and Comfort in Airliner Cabins
New Orleans, USA, 27-28 October
11. Winder C, Balouet J-C (2001) Symptoms of irritation and toxicity in aircrew as a result of exposure to airborne chemicals in aircraft. Journal of Occupational Health and
Safety Australian and New Zealand 17:471483
12. BASI (1997) British Aerospace plc BAE 146-300 10 July 1997: Preliminary Report
9702276. Canberra: Bureau of Air Safety Inspection, 23 October 1997
13. Buck WH (1999) Mobil Oil Corporation. Correspondence, 27 July 1999
14. CAT, AMA (1953) Aviation Toxicology: An Introduction to the Subject and a Handbook of Data. Committee of Aviation Toxicology, Aero Medical Association, Blakiston,
USA
15. Tupper CR (1989) Chemical hazards in aeromedical aircraft. Aviation, Space and Environmental Medicine 60:7375
16. Smith PW, Laceeld DJ, Crane CR (1970) Toxicological ndings in aircraft accident
investigation. Aerospace Medicine 41:760762
246
17. Trimble EJ (1996) The management of aircraft passenger survival in re. Toxicology
115:4161
18. Mackerer CR, Barth ML, Krueger AJ, Chawla B, Roy TA (1999) Comparison of neurotoxic effects from oral administration or ingestion of tricresyl phosphate and jet
engine oil containing tricresyl phosphate. J Toxicology and Environmental Health
57:293328
19. ACC (2001) High Production Volume Chemical Submission, Substituted Diphenylamines. American Chemistry Council, Arlington VA, 18 December 2001. At:
http://www.epa.gov.gov/chemrtk/subdiapha/c13378.pdf
20. Miyazaki K, Kawai S, Sasayama T, Iseki K, Arita T (1987) Absorption, metabolism and
excretion of N-phenyl-1-naphthylamine in rats. Yakuzaigaku (Archives of Practical
Pharmacology) 47:1722 (English abstract)
21. IPCS (1998) Concise International Chemical Assessment Document No 9: N-Phenyl1-naphthylamine. International Programme on Chemical Safety, Geneva
22. Boman A, Hagelthorn G, Jeansson I, Karlberg A-T, Rystedt I, Wahlberg JE (1980)
Phenyl-alpha-naphthylaminecase report and guinea pig studies. Contact Dermatitis
6:299300
23. Kalimo K, Jolanki R, Estlander T, Kanerva L (1989) Contact allergy to antioxidants in
industrial greases. Contact Dermatitis 20:151152
24. Carmichael AJ, Foulds IS (1990) Isolated naphthylamine allergy to phenyl-alphanapthylamine. Contact Dermatitis 22:298299
25. Wang H-W, Wang D, Dzeng R-W (1984) Carcinogenicity of n-phenyl-1-naphthylamine and n-phenyl-2-naphthylamine in mice. Cancer Research 44:30983100
26. Jarvholm B, Lavenius B (1981) A cohort study on cancer among workers exposed to
an antirust oil. Scandinavian J Work Environment and Health 7:179184
27. US NTP (1988) Toxicology and Carcinogenesis Studies of N-Phenyl-2-naphthylamine
in F344/N Rats and B6C3F1 Mice (feed studies). US National Toxicology Program,
http://ntp-server.niehs.nih.gov/htdocs/lt-studies/tr333.html
28. Echobion DJ (1993) Organophosphorus ester insecticides. In: Echobion DJ, Joy RM
(eds) Pesticides and Neurological Diseases, second edition. CRC Press, Boca Raton
29. Aldridge WN (1954) Tricresyl phosphate and cholinesterase. Biochemical J 56:185
189
30. Johnson MK (1975) Structure activity relationship for substrates and inhibitors of hen
brain neurotoxic esterase. Biochemical Pharmacology 24:797805
31. Earl CJ, Thompson RHS (1952) Cholinesterase levels in the nervous system in triortho-cresyl phosphate poisoning. British J Pharmacology 7:685694
32. Eyer P (1955) Neuropsychopathological changes by organophosphorus compounds
a review. Human and Experimental Toxicology 14:857864
33. Kilburn KH (1999) Evidence for chronic neurobehavioral impairment from chlorpyrifos and organophosphate insecticide (Dursban) used indoors. Environmental
Epidemiology and Toxicology 1:153162
34. Minton NA, Murray VSG (1988) A review of organophosphate poisoning. Medical
Toxicology 3:350375
35. Johnson MK (1975) Organophosphorus esters causing delayed neurotoxic effects.
Archives of Toxicology 34:259288
36. Metcalf RL (1982) Historical perspective of organophosphorus ester-induced delayed
neurotoxicity. Neurotoxicology 3:269284
37. Baron RL (1981) Delayed neurotoxicity and other consequences of organophosphate
esters. Annual Reviews of Entolomology 26:2948
247
38. Johnson MK (1990) Organophosphates and delayed neuropathyIs NTE alive and
well? Toxicology and Applied Pharmacology 102:385399
39. Senanayake N, Karalliede L (1987) Neurotoxic effects of organophosphorus esters. An
intermediate syndrome. New England J Medicine 316:761763
40. Dille JR, Smith PW (1964) Central nervous system effects of chronic exposure to
organophosphate insecticides. Aerospace Medicine 35:475478
41. Savage EP, Keefe TF, Mounce LM, Heaton JA, Burcar PJ (1988) Chronic neurological
sequelae of acute organophosphorus pesticide intoxication. Archives of Environmental Health 43:3845
42. Rosenstock L, Keifer M, Daniell WE, McConnell R, Claypoole K (1991) Chronic central nervous system effects of acute organophosphate pesticide intoxication. Lancet
338:225227
43. Steenland M (1996) Chronic neurological effects of organophosphate pesticides.
British Medical J 312:13111312
44. UK DoH (1999) Toxicology of OPs and the mechanisms involved. In: Organophosphates. Committee on Toxicity of Chemicals in Food, Consumer Products and the
Environment, Chap. 5. UK Department of Health, HMSO, London, pp 4958
45. Jamal GA (1997) Neurological syndromes of organophosphorus compounds. Adverse
Drug Reactions and Toxicology Reviews 16:133170
46. Savage EP, Keefe TF, Mounce LM, Heaton JA, Burcar PJ (1988) Chronic neurological
sequelae of acute organophosphorus pesticide intoxication. Archives of Environmental Health 43:3845
47. Gershon S, Shaw FB (1961) Psychiatric sequelae of chronic exposure to organophosphorous insecticides. Lancet 1:13711374
48. Metcalf DR, Holmes JH (1969) EEG, psychological and neurological alterations in
humans with organophosphorous exposure. Annals New York Academy Science
160:357365
49. Callender TJ, Morrow L, Subramanian K (1994) Evaluation of chronic neurological
sequelae after acute pesticide exposure using SPECT brain scans. J Toxicology and
Environmental Health 41:275284
50. Abou-Donia MB (2004) Organophosphorus ester-induced chronic neurotoxicity.
Archives of Environmental Health 58:484497
51. Smith MI, Elvove I, Valaer PJ, Frazier WH, Mallory GE (1930) Pharmacologic and
chemical studies of the cause of the so called ginger paralysis. US Public Health
Reports 45:17031716
52. Henschler D, Bayer HH (1958) Toxicologic studies of triphenyl phosphate, trixenyl
phosphates, and triaryl phosphates from mixtures of homogenous phenols. Archives
of Experimental Pathology and Pharmacology 233:512517
53. Henschler D (1958) Die Trikresylphosphatvergiftung. Experimentelle Klarung von
Problemen der tiologie und Pathogenese (Tricresyl phosphate poisoning. Experimental clarication of problems of etiology and pathogenesis). Klinische Wochenschriften 36:663674
54. Daughtrey W, Biles R, Jortner B, Erlich M (1996) Subchronic delayed neurotoxicity evaluation of jet engine lubricants containing phosphorus additives. Fundamental
and Applied Toxicology 32:244249
55. Montgomery MR, Wier GT, Zieve FJ, Anders MW (1977) Human intoxication following inhalation exposure to synthetic jet lubricating oil. Clinical Toxicology 11:423426
56. Rayman RB, McNaughton GB (1983) Smoke/fumes in the cockpit. Aviation, Space
and Environmental Medicine 67:738740
248
57. Tashkin DP, Coulson AH, Simmons MS, Spivey GH (1983) Respiratory symptoms of
ight attendants during high altitude ight: Possible relation to cabin ozone exposure.
Archives of Occupational and Environmental Health 52:117137
58. Sparks PS, Simon GE, Katon WJ, Altman LC, Ayars GH, Johnson RL (1990) An outbreak of illness among aerospace workers. Western J Medicine 158:2833
59. Winder C, Fonteyn P, Balouet J-C (2002) Aerotoxic syndrome: A descriptive epidemiological survey of aircrew exposed to in-cabin airborne contaminants. J Occupational
Health and Safety Australia and New Zealand 18:321338
60. Cox L, Michaelis S (2002) A survey of health symptoms in BAe 146 aircrew. J Occupational Health and Safety Australia and New Zealand 18:305312
61. Michaelis S (2003) A survey of health symptoms in BALPA Boeing 757 pilots. J Occupational Health and Safety Australia and New Zealand 19:253261
62. NAS (2001) The Airliner Cabin Environment and the Health of Passengers and Crew.
US National Academy of Science, National Academy Press, Washington, DC
63. Singh B (2004) In ight smoke and fumes. Aviation Safety Spotlight, March 2004
64. Fox R (1997) Air Quality Testing Aboard Ansett Airlines BAe 146 Aircraft: Final
Report. Allied Signal Aerospace, 25 November 1997
65. Craig P, Barth M (1999) Evaluation of the hazards of industrial exposure to tricresyl phosphate: A review and interpretation of the literature. J Toxicology and
Environmental Health Part B Critical Reviews 2:281300. Consultants, Mobil Business
resources Coorporation, NJ
66. WHO (1986) Diseases caused by asphyxiants: Carbon monoxide, Hydrogen cyanide
and its toxic derivatives, and Hydrogen sulde. Early Detection of Occupational Diseases. World Health Organization, Geneva, pp 154164
67. Fox RB (2000) Air quality and comfort measurement aboard a commuter aircraft and
solutions to improve perceived occupant comfort levels. In: Nagda NL (ed) Air quality and comfort in airliner cabins. ASTM STP 1393. American Society for Testing and
Materials, West Conshohocken, PA
68. Michaels D (2005) Doubt is their product. Scientic American, June 2005

DOI 10.1007/b107247
Published online:
Aircraft Air Quality Malfunction Incidents:

Design, Servicing, and Policy Measures
to Decrease Frequency and Severity of Toxic Events
Richard Best Susan Michaelis (u)
School of Safety Science, UNSW, 2052 Sydney, NSW, Australia
r.best@bigpond.com, s.michaelis@student.unsw.edu.au
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
250
Development of the Regulatory Framework . . . . . . . . . . . . . . . . .
251
Ventilation Regulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
252
4
4.1
4.2
4.3
4.4
4.5
4.6
Sources and Examples of Defects Covering Contaminated Air

From Lubricant and Fluid Leaks . . . . . . . . . . . . . . . . .
Service Bulletins . . . . . . . . . . . . . . . . . . . . . . . . . .
Airworthiness Directives . . . . . . . . . . . . . . . . . . . . .
Service Information Leaets or Letters . . . . . . . . . . . . .
All Operator Letters or Message . . . . . . . . . . . . . . . . .
Other Sources of Information . . . . . . . . . . . . . . . . . .
A Case Study in Reporting Documentation: The BAe 146 . . .
.
.
.
.
.
.
.
253
254
255
256
256
256
258
Regulatory Inconsistencies Indicating Health and Safety Issues . . . . . .
258
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
261
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
262
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Abstract Aircraft air supply contamination from leaking oil and hydraulic uids has
a long history in commercial aviation. There is a wide range of aviation legislation covering the required processes to be followed when this type of defect occurs, including
reporting, maintenance procedures, airworthiness requirements, crew tness for ight
and emergency procedures. A variety of evidence showing that contaminated air has an
extensive and well-documented history will be examined. It is clear that the regulations
are not being adhered to or enforced. A variety of issues emanating from these failures
will be reviewed as well as suggestions made as to what can be done to effectively resolve
them.
Keywords Aircraft air contamination Airworthiness standards Jet oil leaks Fumes
Aircraft defects Aircraft modications
Abbreviations
AD Airworthiness directive
AOM All operator message
APU Auxiliary power unit
250
ATA
BAe
CAA
CASA
ER
FAR
JAR
ICAO
MEL
OHS
SB
SIL
R. Best S. Michaelis
Air Transport Association
British Aerospace
Civil Aviation Authority (UK)
Civil Aviation Safety Authority (Australia)
Engineering release
Federal Aviation Regulation
Joint Aviation Regulation
International Civil Aviation Organisation
Minimum equipment list
Occupational health and safety
Service bulletin
Service information leaet or letter
1
Introduction
In assessing the design, servicing and repair policy measures to decrease contaminated air event frequency and severity, an appreciation of the regulatory
aspects and development of aircraft pressurisation and air conditioning systems is necessary.
The safety issues relative to the supply of clean air in the cabin of a modern jet aircraft should be viewed in terms of the requirements of aviation
legislation and the airworthiness standards that are required to enable the aircraft to be issued with a document called a Certicate of Airworthiness. This
certication allows for subsequent and continuing operation.
Pressurisation in aircraft refers to the sealing of the cabin, the supply of
air to it from an external source and the control of pressure. Pressurisation
is needed in commercial aircraft that y at high altitude because the human body requires a continual supply of oxygen, the quantity varying with
the amount of physical effort. Pressurisation permits cabin altitudes to remain below 8000 ft. (about 2400 m) whilst the aircraft ies generally between
30 000 and 40 000 ft. (910012 100 m). Associated with pressurisation, the
cabin atmosphere must be provided with an air conditioning system which
is necessary because of the extremes of temperature in which an aircraft may
be operating from day to day. An aircraft needs to be heated or cooled to
maintain a temperature of approximately 22 C, while in ight outside temperatures may be as low as 60 C.
Further, the creation and maintenance of a comfortable atmosphere requires a certain amount of fresh air to be provided for each person. A number
of methods of supply of air for pressurization and air conditioning have been
employed, such as engine-driven compressors, air-driven compressors and
bleed air, in which some air that could be used for combustion in the engine
is bled off for the cabin air supply. Additionally, since the 1970s, some air
251
has been recirculated. This involves the re-use of part of the used air from
the cabin mixed with the incoming fresh air from the compressors, therefore reducing the fuel usage by reducing the amount of air bled off from the
source.
2
Development of the Regulatory Framework
These design issues were recognised as components of an aviation safety system and so, to ensure international uniformity, like-minded countries met
and formed the International Civil Aviation Organisation (ICAO) in 1944.
ICAO developed certain protocols that contained the standards which all
signatories states are obliged to uphold. Legislative backing by each member state is required to ratify the Organisations protocols, with any noncompliances being notied to the ICAO Secretariat.
International airworthiness standards set down by ICAO detail what is required for an aircraft to be deemed t for ight or airworthy and each
Table 1 Development of airworthiness standards to enable safe operation of the aircraft

Type design
Specication of appropriate design specication
Type certication
Aircraft or product conforms

to the appropriate design document
Type data certicate sheet (aircraft)
Supplemental type certicate
(aviation product)
Technical service order
(equipment or component).
Parts manufacturing authority
(part or component)
Production approval
Manufacturing approvals
Operational approvals
Registration of aircraft in the national

register
Certicate of airworthiness
(certicated aircraft)
Special ight permits
(non-certicated aircraft)
Maintenance release/return to service
issuance (dependent on certicate of
airworthiness or ight permit)
252
country adopts these design aspects into its own legislation [1]. The national
regulations for many countries will parallel or be harmonised with the ICAO
requirements of the US Federal Aviation Regulations (FAR) and European
and UK Joint Aviation Regulations (JAR) [2]. Airworthiness standards cover,
among other things, aircraft design, aircraft materials, engines and auxiliary
power unit (APU) requirements, aircraft performance, fuel and oil systems,
and aircraft ventilation [3].
The international airworthiness standards specied by the ICAO have been
promulgated and distributed under ICAO Publication Annex 8 (Airworthiness of Aircraft) which expects the member state (country of registry) to
effect design standards through appropriate instruments to give legislative
backing to the standards and specications.
Leading countries where aircraft manufacture occurs generally issue publications providing guidance on a range of subjects from First-of-Type Aircraft Introduction Procedures through to Continuing Airworthiness so as
to ensure the ongoing safety of the product.
To assure the continuing airworthiness of any certicated product it was
expected the framework in each ICAO members own country procedures
would include a Service Difculty Reporting System whereby operational
or design problems could be brought to the attention of the appropriate authorities and the holder of the Type Certicate (or equivalent) so that the
proper consideration and resulting action or product improvement could be
implemented.
The development of airworthiness standards to enable safe operation of
the aircraft can be summarised by the steps shown in Table 1.
3
Ventilation Regulations
An aircraft must be maintained in an airworthy state in order to y. Many
countries adopt Part 25 of US FARs and/or Part 25 of the JARs as the airworthiness standards for transport category aeroplanes.
The airworthiness standard for aircraft ventilation, developed in 1965 [4]
with the latest amendment in 1997, is FAR/JAR regulation number 25.831,
which requires that:
a. Each passenger and crew compartment must be ventilated, and each crew
compartment must have enough fresh air (but not less than 10 cu. ft. per
minute per crewmember) to enable crewmembers to perform their duties
without undue discomfort or fatigue.
b. Crew and passenger compartment air must be free from harmful or hazardous concentrations of gases or vapours.
253
c. There must be provisions made to ensure that the conditions prescribed in

paragraph (b) of this section are met after reasonably probable failures or
malfunctioning of the ventilating, heating, pressurisation or other systems
or equipment.
The airworthiness ventilation regulation for transport aircraft, FAR/JAR
25.831 a/b, established in 1965 [4, 5] is the major ventilation regulation that
must be met for an aircraft to be considered t for ight. Where exposure
events occur as outlined in the previous chapter on frequency of events, part
25.831a is being ignored. With regard to part 25.831b, industry claims it is
not known if the contaminants are at or above harmful levels despite no or
inappropriate testing being carried out in ight to determine levels of contamination at the time of the leaks. Additionally, when the regulation came
into effect, part b was thought to only cover carbon monoxide, carbon dioxide
or ozone and all other contaminants were not considered [6]. Some regulators claim that aircraft toxicants are more of an occupational health and
safety (OHS) issue than an air safety issue. For example, the UK Civil Aviation Authority (CAA) has recently advised [7] that crew discomfort such as
headaches, nausea and irritation due to contamination is not its responsibility unless the safety of ight and landing are affected. This is not in the intent
of airworthiness regulation 25.831a, which implies that undue discomfort and
fatigue has the ability to affect crew performance and therefore could impair
ight safety.
4
Sources and Examples of Defects Covering Contaminated Air
From Lubricant and Fluid Leaks
Information about an aircrafts operation, defects and its continuing airworthiness is received from various sources, as information ows in both
directions between the aircraft manufacturer and the operator, based on inservice experience. In order for the information to be set out in a uniform
industry-wide standard, the Air Transport Association (ATA) has devised
various chapter codes. These codes relate to particular subject matters and
identify to all what the particular topic is. Some examples are ATA 21Air
conditioning; ATA 36 Pneumatics; ATA 49Auxiliary power unit (APU);
ATA 71, 72Power plant general, engines. This information takes various
formats explored later.
254
4.1
Service Bulletins
Service bulletins (SB) are based upon information gained from the eld and
are issued on a variety of compliance options. They are issued by the manufacturer identifying inspections or modications that have been issued. The
SB will list the title of the modication, effectivity, reason and in some cases
background for its release, description, compliance, man-hours, costs and so
on. In rare cases, these may be issued as an alert SB indicating a higher status.
SBs may be issued for information only, optional or recommended, often with
a statement as to when this might be undertaken. The manufacturer cannot
make the modications or inspections required by SBs mandatory.
The nonmandatory nature of such advice is problematic. Aircraft operators should take greater responsibility to assess whether a modication or
inspection requirement ought to be implemented, as currently although operators subscribe to the SBs, the requirement to assess them is only implied
and there is no requirement to act on advice in safety-related SBs [8]. On the
British Aerospace (BAE) 146 for example, SBs from 1983 up until 2000 were
all for information only, optional or recommended and usually at a time to
suit the operator, except for one mandated in 1985 [9]. As such it is unknown
how many will have been acted upon, but quite likely many will not have been
undertaken as these are not mandatory.
It appears that many in the aviation industry view modications and inspections for oil leaking into the air supply as part of its ongoing product
improvement and enhancement [10], rather than as a mandatory requirement
to meet the airworthiness regulations. Despite the BAe 146s long history of
fumes, oil and hydraulic uid leakage into the air supply and strong evidence
of crew discomfort going back to at least 1983, and despite the fact that contaminated air breached the ventilation airworthiness requirements, no other
SBs were made mandatory until 2000. Since 2000 only ve SBs on three aircraft types have been made mandatory by the regulators in selected countries
through their inclusion in airworthiness directives (ADs): the MD series aircraft [11]; the Rolls Royce 307 series engines [12]; and the BAe 146 [1318].
Examples of SBs include:
BAeSB 49-5-35040G: 24 October 1984. Title: APUIntroduce an improved compressor inlet duct seal. Reason: Inadequate sealing between
APU accessory drive gearbox oil sump and compressor inlet duct, and between top and bottom halves of inlet duct, allows fumes to be sucked from
the bay area through the APU and into passenger cabin. Description:
Improved silicone rubber seal conguration. This SB is for information
only. Retrospective embodiment is not intended because in service experience has shown that this modication is not a complete answer to the
problem. Compliance: Information [19].
255
Allied Signal, Garrett APU Division (BAe 146)SB GTCP36-49-5899:

November 1989. Title: APU: Replace compressor seal assembly. Reason:
The current compressor seal has shown an unacceptable rate of failure
which can result in smoke in the cabin. Background: The failure of the
compressor seal assembly allows gearbox oil to leak into the compressor
inlet, resulting in smoke in the cabin. The new seal has been redesigned
to improve sealing characteristics and reliability. Compliance: Recommended at operators convenience [20].
Other SBs dealing with the air quality issue on the BAe 146 include
Refs. [2124].
4.2
Airworthiness Directives
A national regulator such as the US Federal Aviation Authority or the UK
CAA can make a SB mandatory when it feels a signicant or real safety
issue exists or is likely to exist [10, 25], by issuing an AD. However, regulators have been reluctant to issue ADs in relation to contaminated air despite
evidence and acknowledgement that breathing oils and fumes is a potential threat to ight safety and failure to meet the airworthiness ventilation
when undue discomfort and fatigue is occurring associated with contaminated air [4, 13, 14, 23, 24]. This reluctance may be economically driven or
may be due to a lack of expertise in this eld as the regulators and manufacturers have admitted that toxicants in aircraft cabins are outside their eld of
expertise [26]: The regulatory bodies as admitted by CASA yesterday, are not
competent to rule on such a highly specialised area. Neither are the airlines or
the manufacturers. [10].
Compliance dates with mandatory ADs vary greatly, despite ADs being
issued where a safety risk exists or could exist. While some ADs require
inspections and possible maintenance before further ight or within 10 h provided the source of contamination can be identied and isolated following
suspected oil contamination [13], others allowing smoke and odours into the
cabin require modications 36 months or more after the AD was issued [11].
In two cases service information data indicating oil contamination of the
ducting in 1984 [27] and an optional SB in 1993 [22] indicating oil contamination of the APU inlet duct were made mandatory through ADs 18 and 10 years
later, respectively [1518]. An AD will list the details, background and a compliance date by which the requirements must be completed. In some cases
months or years are assigned for completion.
Examples of an SB made mandatory by the regulator and an actual AD
include:
BAeSB 21150: 20 March 2001. Title: Air conditioningInspect engine
oil seals, APU and environmental control system jet pump and air condi-
256
tioning pack for signs of oil contamination. Reason: Incidents have been
reported involving impaired performance of the ight crew ... In the past,
oil leaks and cabin/ight deck odours and fumes may have come to be regarded as a nuisance rather than a potential ight safety issue. However
whilst investigations are being carried out, oil leaks and cabin/ight deck
odours must be regarded as a potential threat to ight safety, they should
not be dismissed as a mere nuisance and should be addressed as soon as
possible. Compliance: Mandatory [23].
CAA AD 003-10-2002 (BAe 146) December 2002 BAe SB 21156. Title: Air
conditioningInspect air conditioning sound attenuating ducts for signs
of oil contamination. Reason: Action required due to incidents reported
of impaired performance of ight crew. Compliance: Mandatory [15].
4.3
Service Information Leaflets or Letters
Service information leaets (SIL) are information documents, usually issued
to disseminate information generally supporting a SB-related modication or
inspection.
Example of SILs include:
BAe 146 SIL 21/7: December 1984. Title: Oil Contamination of Air Conditioning System [27].
BAe 146 SIL 21/45: November 2000. Title: Cabin Air Quality Trouble
Shooting Advice and Relevant Modications (includes sources of contamination, modications, medical tests and crew health survey) [28].
Other SILs dealing with the air quality issue on the BAe 146 include
Refs. [2932].
4.4
All Operator Letters or Message
All operator letters or all operator messages (AOM) are information sent by
manufacturers to aircraft operators on a particular subject. An example of an
AOM is:
BAe AOM Ref 00/030V: January 2001. Title: Smoke and Fumes. Reason:
Measures to take when smoke or smell from air conditioning system is
sensed ... The air supply is protected from contamination by seals, which
achieve maximum efciency during steady state operation. However, they
may be less efcient during transients (engine acceleration or deceleration) or whilst the engine is still achieving an optimum operating temperature. Improvements in seal design continue to increase efciency, and
when available, modications are provided for the engines and APU [33].
19842001
1997 and 2001
19992001+
19921998
Various
Allied Signal/Honeywell
BAe
Ansett (BAe/Allied Signal)
BAe, Honeywell,
Normalair/Garrett,
NJS, Ansett
CASA, CAABAe 146
20012003
19842003
15+
19842001
Service information
leaet
Service information
Leaet
All operator message
Internal engineering
releases/orders
Various: EMM, CMM,
MM, engineering notice,
internal memo
Airworthiness directive
Total
27+
19842000
Allied Signal/
Honeywell
Allied Signal/Garrett
APU Division
Allied Signal
BAe
6+
202+
9+
6
32
20
85+
19842003
BAe
Number
Service bulletins
aircraft manufacturer
Service bulletins
engine
Service bulletins
APU
Year(s)
Source
Report type
Table 2 Selected modications and data available for review for the BAe 146 aircraft
[13, 14]
[38, 39]
[33]
[BA6 21-20-29]
[SIL ALF/LF-8]
[2732]
[20]
[21]
[19, 2224]
Sources
(see, for example)

257
258
4.5
Other Sources of Information
Other sources of information include defect reporting in the aircraft technical
log, defect reports sent from the aircraft operator to the regulatory authority, informal communications between aircraft operator and manufacturers
or defects reports to regulators from the Type Certicate holder.
Sources of information also include Manufacturers Operations Manual/
Notices to Aircrew/Operators, which is information provided to operators
and aircrew by the manufacturer highlighting operational information [35,
36]. Other records include engine maintenance manuals listing modication
details and procedures such as engineering maintenance manuals revising oil
leakage inspection procedures [37], notices to pilots regarding air conditioning contamination [38, 39], engineering updates or engineering releases (ER)
such as the following:
Ansett (Allied Signal) ER AR5 49-20-33: November 1993: Ejector System
to Reduce Gearbox Pressure which Prevents Oil Leaking Past Compressor
Carbon SealKnown Cause of Smells in Cabin: See SB GTCP36-49-6661
(work undertaken 11/9410/96)
4.6
A Case Study in Reporting Documentation: The BAe 146
Table 2 brings together the various documentation and reporting means described earlier and lists selected air quality/contaminated air related data
in various formats for the BAe 146 and 146RJ aircraft and its engines, the
ALF/LF 502/507. The table includes only revisions up to 2003 and includes
data collected by the authors from a range of sources.
A number of issues are raised by reviewing the information in Table 2.
Information is generally not available for review, difcult to source or the
history of the problem is generally not clearly identied. For example, improvements made to a number one bearing seal in 1995 for the BAE 146
engine were still causing problems in 2003 [21, 39, 40], yet many in the industry believed that the problem was xed. The various ADs raised between 2001
and 2003 did not clearly demonstrate the problems were well known many
years earlier.
5
Regulatory Inconsistencies Indicating Health and Safety Issues
In some cases, aircraft air contamination modications or inspections will relate to a maintenance or engineering issue. However, from rst principles, oil
259
leaking into the aircraft air supply is a design issue, with initial certication
clean air airworthiness requirements needing to be met as well as as on an
ongoing continuing airworthiness basis throughout the aircrafts operating
life [3]. BAe acknowledged this in 2000 when advising that engine or APU
seals may be less efcient during transients (engine acceleration or deceleration) and improvements in design when available would be provided [36].
However, a design fault that leads to ongoing engineering and operational
problems is difcult to rectify, appears to be accepted fatalistically as costly to
x, and ultimately, often ignored.
The substantial amount of evidence regarding oil leakage and crew and
passenger discomfort would indicate continuing airworthiness is not being maintained as oil and hydraulic fumes are not being viewed as part of
the continuing airworthiness requirements. As shown in Sect. 6.3 TSa , there
is a very large and varied volume of industry modication and defect data
available, directly related to oil leaks and contaminated air dating back over
20 years [34]. Moreover, industry information of effects of oil contamination
dates back to at least 1981 [41]. Impaired crew performance was rst documented in 1977 [42], with increasing reference to impaired crew performance
in recent years. It is clear that the aviation industry has shown great reluctance to openly and satisfactorily deal with the issue.
Fumes (predominantly Mobil Jet Oil II leaking into the bleed air supply [43]) identied as oil leaking past the engine and APU oil seals [10, 44]
have been seen as a nuisance rather than a potential threat to ight
safety [23, 24]. Oil contamination and noxious fumes are not being regarded
as a major defect or equivalent and are often not regarded as needing to be
reported as required by the various regulations and therefore the subsequent
requirements are not fullled, as the following examples show:
Prior to the issue of the AD by CASA, there was no specic requirement
for National Jet Systems (NJS) to report to CASA on incidents of air contamination. [45].
We dont regard fumes as an immediate threat to aviation safety ... Obviously if we did we would have to ground ights. [46].
Oil fumes are more of a health problem than an aircraft technical defect
as not all pilots affected and there is no mandate to look at health. [47].
Toxins in cabin air are an OHS issue and not responsibility of the Aviation Regulator which is responsible for short and medium term effects on
safety. [26].
In Australia, for example, the aircraft operators and the Civil Aviation Authority (CASA) have failed to view fumes and oil contamination as a major defect
as required in its own legislation, manuals and advisories [4850]. These issues are not being reported in all cases, as there is a failure to view the issue as
an industry responsibility [26]. This is evidenced by the previous comments
showing that prior to the issue of the rst BAe 146 air contamination related
a In the submission by Winder & Michaelis is neither
TS
Chap. 6.2 nor 6.3.
Editors or typesetters annotations (will be removed before the nal TEX run)
260
AD in March 2001, CASA did not view fumes as reportable or a safety issue.
Comments by a major aircraft operator suggesting they are different as they
require fumes to be reported [43] shows the industry indifference to the reporting system, which is clearly not working. Industry accepts that all engine
oil seals can leak and that it is an inevitable feature of the design of air conditioning systems [26, 43]. However, it is recognised that no one modication
is a complete x: The modications will not solve the problem completely
they are to reduce the number of events. and The modications that have
been developed are really around the reliability of the seals and making sure
they dont fail as frequently. So they are improvements to the reliability, rather
than improving the quality of the sealing. [10].
A recent 2004 comment by CASA very importantly completely contradicts
the industry perception that fumes are not reportable occurrences and turns
around the long-held CASA position that such defects are not major defects
and therefore not reportable: All instances of smoke or fumes in the aircraft
cabin that adversely affect the quality of cabin air on Australian registered
aircraft ...are categorised by the CASA as a Major Defect. [51].
Failures of oil seals are seen as the common factor in the majority of fume
incidents [5256] and are often hard to identify [5257]. Inspections for oil
leaks and fumes often take place between ights with engineering comments
including not safety of ight, no fault found, report further or repair at
company convenience with reports of fumes sometimes ongoing over days,
weeks or months [52, 58, 59]. Factors involved include the difculty in precisely locating the oil leak, cases where more than one engine or APU oil
leak combination occurs and residual contamination of the air conditioning
packs [52].
After air contamination by oil and hydraulic uids on an aircraft it is
common practice to continue to operate the aircraft with part of the bleed
air supply or one aircraft air conditioning pack deselected under the minimum equipment list (MEL) system. However this does not take into account
several important factors. There may be difculty in accurately determining the exact source of the contamination; additionally When an oil leak
from an engine or APU is repaired, the system downstream must also be
thoroughly cleaned to eliminate unintentional introduction of contaminants
into the cabin. [60, 61]; or there is no effective way to adequately clean
bleed air ducts in situ once they have become contaminated with oil breakdown products. Adequate cleaning requires removal of the ductwork to wash
out oil products with cleaning typically reserved for major maintenance
checks [60, 61].
Crews appear to be signicantly under-reporting contaminated air events.
Therefore, the true scale of the problem remains unknown and the issue is
continually downgraded. Some examples include fumes not being seen as
major defects or similar with compulsory reporting and under-reporting,
which is known to be occurring (as discussed in the chapter on rates of inci-
261
dents) [53, 6165]. Reasons for under-reporting include fumes being seen as
a highly repetitive occurrence and almost as a normal part of ight, fear of
reporting ongoing problems, lack of understanding of effects of fumes on the
individual and regulatory requirements and crews being advised by company
doctors that there are no adverse health implications. In addition to the reporting, airworthiness, defect and maintenance investigation regulations not
being met, contaminated air affects crew tness with regard to ight regulations and emergency procedures, such as the ability of the cabin crew to
evacuate the cabin in 90 s.
Also, the use of emergency oxygen is not being seen as a serious incident, as
established by the ICAO [66]. Fume contamination is clearly a safety deciency
by denition but as an example of the downgrading of the issue, the Australian
Transport Safety Bureau stated it was a possible safety deciency [52, 67, 68].
While the use of oxygen when contamination events occur is required for
ight crew, it was only recently added to the emergency and abnormal procedures checklist. Previously oxygen was only required when smoke or re
occurred [33, 35]. Emergency 100% oxygen is not generally being used in shortterm transient fume events which are part of the design problem, but are seen
as a normal part of ight by pilots. At the same time, cabin crew and passengers
are not provided with any effective protection against contaminated air. The UK
CAA requires mandatory occurrence report to be made where oxygen is used
in fume incidents, yet this was generally not occuring. It only recently advised
crews to use oxygen in all fume events [69]; however, this is still not occurring
in all cases and the reports are often not made. Additionally the subtle incapacitation effects of odourless gases or fumes such as carbon monoxide may not
alert the pilots to the need to use oxygen.
6
Conclusion
There is a wide variety of legislation and data that support that engine oils and
hydraulic uids are leaking into the cabin air supply. There is increasing evidence to show this is affecting crew performance and health both short and
long term. There is evidence going back many years clearly showing that oil
and hydraulic leakage is a major ongoing problem and that the regulatory requirements are not being adhered to or enforced. These defects are in fact part
of the ventilation airworthiness requirements and must continue to be met for
an aircraft to be considered t for ight. However, industry attitudes towards
contaminated air have been complacent and irresponsible. This is alarming,
especially when it is apparent that commercial aircraft have no immediate or
effective back-up system to protect crews and passengers should the bleed air
become contaminated, despite the technology existing that could address this
262
problem. This complacency has allowed the obvious safety implications of operating crew experiencing discomfort when exposed to oil fumes and the shortand long-term health implications continue for many years.
For a safety system to work effectively all the components of the system
need to operate as designed. The lack of recognition of the full implications
of the contaminated air problem with the BAe 146 and other aircraft models
highlights the breakdown in the inter-relationships on which a properly functioning safety management system is dependent.
Steps to be taken to reduce this problem include the need to review clean
air airworthiness requirements to cover all contaminants and view undue
discomfort and fatigue as contrary to the legislation. It is necessary to view
clean air under FAR/JAR 25.831a/b as part of ongoing aircraft certication
requirements. Also, correct reporting and under-reporting problems must be
resolved. All regulations including airworthiness ventilation regulation, defect reporting and maintenance procedures, tness for duty and emergency
procedures must also be met. Appropriate monitoring of aircraft air (during
contaminated air events) for all hazardous compounds must take place and
independent, appropriate testing of oils and their pyrolysis breakdown products must be undertaken. Less toxic oils ought to be used and a review of the
toxicity of oils in terms of human inhalation in aircraft and the applicability
of exposure standards should be undertaken. Continued ight under the MEL
system when the air supply is suspected to be contaminated and crews show
signs of discomfort should not occur. Modications relating to contaminated
air should be made mandatory as distinct from merely optional, for information, or recommended at operator convenience. All crew should use oxygen
when air contamination occurs. The collation of data worldwide should take
place so as to review major issues and trends.
There is a need to educate all within the industry, including crews, operators, manufacturers and regulators, that contaminated air must be reported
and addressed immediately.
Better designed engine/APU bearing oil seals must be made a priority.
Bleed air lters that are effective in removing applicable contaminants must
be tted and maintained to protect crews and passengers from contaminated
air. Finally a review of bleed air systems and possible bleed-free systems
such as being used on the Boeing 7E7 Dreamliner should be undertaken.
References
1. FAA/ASA (1997) Pilots handbook of aeronautical knowledge. US Federal Aviation
Authority)/Aviation Supplies and Academics, Washington
2. CETS (1998) Improving the continued airworthiness of civil aircraft: a strategy for
the FAAs aircraft certication standard. Commission on Aircraft Certication Safety
Management. National Research Council, Washington
263
3. Hoy B (2000) What makes an aircraft airworthy? Flight Safety Australia, Civil Aviation
Safety Authority of Australia, Canberra
4. FAA (1965) Federal aviation regulations, transport category airplanes, airworthiness
standardsVentilation section 25.831, February 1965 (amendment 2587effective
7 May 1996). US Federal Aviation Authority, Washington
5. JAA (2003) JAR airworthiness ventilation regulation 25.831a/b. Amendment 16 May
2003. http://www.jaa.nl/section1/jars/430782.pdf
6. Best R (2001) Certication of aircraft in Australia. Aviation air quality. In: Winder C,
Michaelis S, Weber RO (eds) Proceedings of the Aviation Air Quality Symposium,
Australian Defence Force Academy/University of New South Wales, 7 December 2000,
University of New South Wales, Sydney, pp 7882
7. Hamilton T (2004) Letter to British Airline Pilots Association (BALPA) Aircraft Environment Task Group. UK Civil Aviation Authority, London
8. Villiers D (2004) Watching brief. Flight Safety Australia, Civil Aviation Safety Authority of Australia, Canberra
9. BAe (1985) Service bulletin SB 21-2400543A. British Aerospace Systems, Hateld
10. BAe (2000) British Aerospace submission and evidence by British Aerospace Systems to the Australian Senate Inquiry into Air Safety (19992000) BAe 146 Cabin Air
Quality. Parliament of Australia, Canberra
11. FAA (2000) Airworthiness directive (AD) 2000-15-17, MD series aircraft. US Federal
Aviation Authority, Washington
12. FAA (2001) Airworthiness directive AD 2001-08-15: Rolls Royce 3007 series engines.
US Federal Aviation Authority, Washington
13. CAA (2001) Airworthiness directive (AD) 002-03-2001: Air conditioningTo inspect
engine oil seals, APU and ECS jet pump and air conditioning pack for signs of oil. UK
14. CASA (2001) Airworthiness directive AD/BAe 146/86: Environmental control
systemInspection for contamination. Civil Aviation Safety Authority of Australia,
Canberra
15. CAA (2002) Airworthiness directive (AD) 003-10-2002: Air conditioningTo inspect
air conditioning sound-attenuating ducts for signs of oil contamination. UK Civil
Aviation Authority, London
16. CASA (2003) Airworthiness directive AD/BAe 146/102: Air-conditioning duct
Inspection. Civil Aviation Safety Authority of Australia, Canberra
17. CASA (2003) Airworthiness directive AD/BAe 146/105: APUAir-inlet duct
Modication. Civil Aviation Safety Authority of Australia, Canberra
18. CAA (2003) Airworthiness directive (AD) 007-04-2003: Airborne auxiliary power
(APU)Introduction of improved APU inlet exible duct part no DXA07175. UK
19. BAe (1984) Service bulletin SB 49-5-35040G: APUPower plantIntroduce an improved compressor inlet duct seal (Garrett change 13). British Aerospace Systems,
Hateld
20. Allied Signal (1989) Service bulletin SB GTCP36-49-5899: APU: Replace compressor
seal assembly. Allied Signal Aerospace, Phoenix
21. Allied Signal (1995) Service bulletin SB ALF502R 72-342. Allied Signal Aerospace,
Phoenix
22. BAe (2003) Service bulletin SB 4936: Airborne auxiliary power (APU)Introduction
of improved APU inlet exible duct part no DXA07175. British Aerospace Systems,
Hateld (fourth revision). Original issue 1993
264
23. BAe (2001) Service bulletin SB 21150: Air conditioningTo inspect engine oil seals,
APU and ECS jet pump and air conditioning pack for signs of oil contamination.
British Aerospace Systems, Hateld
24. BAe (2002) Service bulletin SB 21156: Air conditioningTo inspect air conditioning sound attenuating ducts for signs of oil contamination. British Aerospace Systems,
Hateld
25. CASA Australian civil aviation regulation 37AAirworthiness directives. Civil Aviation Safety Authority of Australia, Canberra
26. CASA (2000) submission and evidence of the Civil Aviation Safety Authority of Australia to the Australian Senate Inquiry into Air Safety (19992000) BAe 146 Cabin Air
Quality. Parliament of Australia, Canberra
27. BAe (1984) Service information leaet SIL 217: Oil contamination of air conditioning
system. British Aerospace Systems, Hateld
28. BAe (2000) Service information leaet SIL 2145: Cabin air quality trouble shooting
advice and relevant modications. British Aerospace Systems, Hateld
29. BAe (1990) Service information leaet SIL 36/9: Installation of catalytic converters in
bleed air system. British Aerospace Systems, Hateld
30. BAe (1990) Service information leaet SIL 21/27: Oil Contamination of air conditioning system. British Aerospace Systems, Hateld
31. BAe (1990) Service information leaet SIL 36/11: Bleed hogging trouble shooting
procedures. British Aerospace Systems, Hateld
32. BAe (1991) Service information leaet SIL 21/30: Cabin and ight deck malodors
trouble shooting procedures. British Aerospace Systems, Hateld
33. BAe (2001) All operator message (AOM) Ref 00/030V: Smoke and fumes. British
Aerospace Systems, Hateld
34. Rolls Royce (1983) Notice to operators: RB21122B, RB 211524. Cabin odour
known causes. Rolls Royce
35. BAe (1990) BAe 146 Manufacturers Operations Manual: Emergency and Abnormal
Procedures Check List/Smoke or Fire Protection. British Aerospace Systems, Hateld
36. BAe (2000) BAe 146 manufacturers operations manual: notice to aircrew, operational
notice: no OP 16 and 43 (issue 1). British Aerospace Systems, Hateld
37. Honeywell (2001) Engine manual ALF502RTemporary revision no 72-888: revised
oil leakage inspection procedures. Honeywell (Allied Signal) Phoenix
38. NJS (2001) Operations manual: notice to BAe 146 pilots 37/97: BAe 146/Avro RJair
conditioning contamination. National Jet Systems, Adelaide
39. Honeywell (2002) ALF/LF 502/507 engineering activity update. Honeywell (Allied
Signal) Phoenix
40. Allied Signal (2003 service bulletin SB ALF/LF 72-1082. Allied Signal Aerospace,
Phoenix
41. SAE (1981) Aerospace information report AIR 1539, 30/1/81, environmental control system contamination: sources of vaporous contamination. Society of Automotive
Engineers, Warrendale, PA
42. Montgomery MR, Wier GT, Zieve FJ, Anders MW (1977) Human intoxication following inhalation exposure to synthetic jet lubricating oil. Clin Toxicol 11:423426
43. (2000) Ansett Australia submission and evidence by Ansett Australia to the Australian
Senate Inquiry into Air Safety (19992000) BAe 146 Cabin Air Quality. Parliament of
Australia, Canberra
44. Mobil (1983) Correspondence: Mobil Jet Oil II. Mobil Oil Corporation, Environmental
Affairs and Toxicology Department, New York
265
45. OBrien K (2002) BAe 146 aircraft: Q398. Hansard, Australian Parliament House, Canberra
46. CASA spokesperson (2003) News in sciencePossible solution to toxic aircraft cabin
air. Australian Broadcasting Commission Radio News, Sydney
47. Borger H (2003) High on the ight deck. Occupational health and safety magazine.
CCH, Sydney
48. CASA (2000) Civil aviation regulation 2. Civil Aviation Safety Authority of Australia,
Canberra
49. CASA (2000) Major defect reporting manual. Civil Aviation Safety Authority of Australia, Canberra
50. CASA (2001) Defect reports, Civil Aviation Advisory Publication 511(1). Civil Aviation Safety Authority of Australia, Canberra
51. Vamvakinou M (2004) Air safetyCabin air quality: Q3011. Hansard, Australian Parliament House
52. ATSB (1999) Occurrence brief 199702276, BAe 146, VH NJF, Fumes in cabin from
oil leak in APU, 10 July 1997. Australian Transport Safety Bureau, Canberra, http://
53. ATSB (2000) Occurrence brief 200002431, BAe 146, VH NJL, Fumes in cabin from oil
Leak in APU, 30 April 2000. Australian Transport Safety Bureau, Canberra, http://
54. ATSB (2003) Occurrence brief 200203243, BAe 146, VH NJR, Fumes in cabin from
oil leak in APU, 22 July 2002. Australian Transport Safety Bureau, Canberra, http://
55. ATSB (2003) Occurrence brief 200204912, BAe 146, VH NJL, Fumes in cabin from oil
leak in APU, 20 October, 2002. Australian Transport Safety Bureau, Canberra, http://
56. ATSB (2003) Occurrence brief 200205865, BAe 146, VH NJX, Fumes in cabin from
oil leak in APU, 2 December, 2002. Australian Transport Safety Bureau, Canberra,
http://www.atsb.gov.au/aviation/occurs/occurs_detail.cfm?ID=542
57. NJS submission and evidence by National Jet Systems, Adelaide, to the Australian
Senate Inquiry into Air Safety (19992000) BAe 146 Cabin Air Quality. Parliament of
Australia, Canberra
58. Pavlinovich N (2003) Witness statement to Australian Industrial Relations Commission. U2002/6475 N. Industrial Relations Commission, Perth
59. CAA mandatory occurrence reporting database. UK Civil Aviation Authority, London
60. ASHRAE (1999) standard 161Air quality within commercial aircraftCommittee
review draft
61. AFAP submission and evidence by Australian Federation of Air Pilots, Melbourne, to
the Australian Senate Inquiry into Air Safety (19992000) BAe 146 Cabin Air Quality.
62. Senate of Australia (2000) Air safety and cabin air quality in the BAe 146 aircraft. Senate Rural and Regional Affairs and Transport References Committee. Parliament of
Australia. Final report, Canberra
63. Michaelis S (2003) A survey of health symptoms in BALPA Boeing 757 pilots. J Occup
Health Saf Aust N Z 19:253261
64. BALPA (2004) B757 and BAe 146 fume event reports (2004). Unpublished data. British
Airline Pilots Association, London
65. Singh B (2004) In ight smoke and fumes. Australia Aviation Safety Spotlight 0304
66. ICAO (2001) Aircraft accident and incident investigation, 9th edn, annex 13. International Civil Aviation Organisation, Paris
266
67. ATSB air navigation act 1920. Australian Transport Safety Bureau, Canberra
68. ATSB Editorial: New air safety accident or incident report. Australian Transport
Safety Bureau, Canberra, http://www.atsb.gov.au/aviation/editorial/forms_article.cfm
69. UK CAA (2002) ight operations department communication (FODCOM) 21/2002:
smoke/fume occurrences. UK Civil Aviation Authority, London

DOI 10.1007/b107248
Aircraft Cabin Air Filtration and Related Technologies:

Requirements, Present Practice and Prospects
S. Michaelis1 (u) T. Loraine2
1 School
Of Safety Sciences, UNSW, 2052 Sydney, Australia

s.michaelis@student.unsw.edu.au
2 The British Air Line Pilots Association (BALPA), 81 New Road,
Harlington, Hayes UB3 5BG, UK
tristanloraine@balpa.org
Susan Michaelis is a former BAE 146 pilot and now a part time masters student at UNSW.
Tristan Loraine is Chairman of the BALPA Cabin Air Quality Task Group which was set
up by BALPA to explore all issues of the cabin environment and to make recommendations. His input to this article represents emerging thinking from the task group. Echo-Air
diagrams courtesy of Indoor Air Technologies Inc, Canada and USA.
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
268
Recirculated Air Composition . . . . . . . . . . . . . . . . . . . . . . . . .
269
3
3.1
3.2
3.3
Recirculated Air Filtration Methods

Direct Interception . . . . . . . . .
Diffusional Interception . . . . . . .
Inertial Impaction . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
269
270
271
271
4
4.1
4.2
4.3
High Efciency Particulate Air Filters . . . . . . . . . .

Number of Recirculated Air Filters per Aircraft Type .
Cost Saving of Recirculated Air . . . . . . . . . . . . .
Operational Effect of Unserviceable Recirculation Fans
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
271
273
274
275
5
5.1
5.2
5.2.1
5.2.2
5.2.3
5.3
5.4
5.5
5.6
5.6.1
5.6.2
Bleed Air Filtration . . . . . . . . . . . . . . .

Non Regenerative Chemical Filtration System .
Regenerative Chemical Filtration Systems . . .
Temperature Swing Adsorption . . . . . . . .
Pressure Swing Adsorption . . . . . . . . . . .
Pressure Temperature Swing Adsorption . . .
Plasma . . . . . . . . . . . . . . . . . . . . . .
Ultraviolet Light . . . . . . . . . . . . . . . . .
Nanocrystalline Materials . . . . . . . . . . . .
Catalytic Converters . . . . . . . . . . . . . . .
The Reduction Catalyst . . . . . . . . . . . . .
The Oxidation Catalyst . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
275
276
277
277
278
278
278
279
279
280
280
280
ECHO-Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
281
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
285
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
287
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
268
S. Michaelis T. Loraine
Abstract Aircraft cabin air being supplied from the engines or APU is known to occasionally be contaminated with hydraulic uids, engine oils, and pyrolysis products of these
which need to be removed to ensure that the crew and passengers are not exposed to any
contaminants. One way of achieving this is to lter these contaminants out of the outside
air before it reaches the crew and passengers. Additionally, some aircraft cabin air is recirculated and this also needs to be ltered to remove bacteria and viruses. This chapter
reviews a number of catalytic, physical, and ventilation system alternatives to simple ltration that could help to eliminate the risk of contaminated outside air or recirculated
air from entering the passenger cabin.
Keywords Air quality HEPA Bleed air Cabin air TCP Bacteria Virus
Aircraft cabin fumes Contaminated air Engine oils Hydraulic uids
Abbreviations
APU
ASHRAE American Society of Heating, Refrigeration and Air-conditioning Engineers
ASTM
American Society for Testing and Materials
FAR
HEPA
High Efciency Particulate Air Filter
JAR
SIL
Service Information Leaet
TCP
tri-cresyl-phosphate
TOCP
tri-ortho-cresyl-phosphate
DOCP
di-ortho-cresyl-phosphate
MOCP
mono-ortho-cresyl-phosphate
VOC
volatile organic compound
1
Introduction
Over the last few decades, aircraft manufacturers have sought ways to make
aircraft engines more efcient and burn less fuel to make them more economical. An aircraft engine as well as producing thrust to propel the aircraft has
other demands put on it. These include hydraulic pumps, electrical generators and provision of cabin air. Some of the air which could be used for thrust
is bled off and passed through air-conditioning packs where it is cooled and
supplied to the aircraft cabin. This air pressurizes the aircraft cabin and provides air to allow crews and passengers to survive whilst the aircraft often
ies in extreme conditions such as pressures as low as 0.2 atmospheres and
temperatures as low as 60 C.
In an effort to reduce operating costs many aircraft and engine manufacturers reduce the bleed air requirement by recirculating some of the cabin
air and therefore put less demand on the engine for bleed air. Recirculation
was commonplace before the jet age began. For example, the Boeing Stra-
Aircraft Cabin Air Filtration and Related Technologies
269
tocruiser of the late 1940s was equipped with an air recirculation system. In
jet aircraft, ltered/recirculated air combined with outside air came into use
principally with the introduction of high-bypass-ratio fan engines. At Boeing,
this began with the 747 in 1970 [1]. This idea has evolved over thirty years until today when about 50% of the cabin air is recirculated air and 50% is fresh
bleed air [2]. It is estimated that in so doing, an average airliner will save over
US$ 60 000 per annum [3] compared to airliners not recirculating cabin air.
Aircraft cabin air may contain numerous bacteria and viruses and these need
to be ltered out if any cabin air is to be recirculated. Additionally, the air
coming from the engines and APU is known to occasionally be contaminated
with hydraulic uids, engine oils, and pyrolysis products of these which also
need to be removed. This chapter looks at aircraft ltration options and provides an introduction into their capabilities and the part they play or could
play in commercial aviation.
2
Recirculated Air Composition
Bleed air is outside air and should be fresh and clean, unless it becomes contaminated with aerosol droplets or vapors of engine oil, hydraulic uids, other
organic vapors, carbon dioxide or other by-products of combustion as they
enter the aircraft. Recirculated air is not fresh air and given the large numbers of passengers on an aircraft, there are high concentrations of particulates
(bres, dust, skin particles), bacteria (up to 30 000 bacteria per minute per
passenger can be released into the cabin environment from skin scales) [4],
other micro-organisms as well as odors. These contaminants are all a potential risk to passengers and crews. Bacteria thrive in high humidity, and
viruses in low humidity. Both conditions are found on commercial aircraft.
In addition passengers will be more vulnerable to infection during a ight
compared with normal non ight conditions. This is due to the closed conditions of the aircraft cabin environment, the small amount of available airspace
per passenger, air continually being blown over the head area, and contact
with people from diverse backgrounds.
3
Recirculated Air Filtration Methods
Recirculated air-ltration systems have been designed to enhance passenger and crew health and comfort by controlling bacteria and viruses. Bleed
air from the engine is cooled in an Air Conditioning Pack before going
to a Mix Manifold where it is mixed with air from the cabin that is be-
270

Fig. 1 Flow Chart of Recirculated Air
ing recirculated to provide crews and passengers with a mix of fresh air and
recirculated air.
The recirculated air-ltration systems are placed beside the Mix Manifold so that the cabin air that is to be recirculated passes through the airltration system as it enters the Mix Manifold.
Filters used in this application should be able to remove particles down to
the size of viruses (0.01 m (micron) in diameter and below [4]), as well as
bacteria and other particulate matter up to 10 m in diameter. The 0.3 micron
benchmark is used in efciency ratings because it approximates the most difcult particle size for a lter to capture [5].
Table 1 Some typical dimensional comparisons
Item
Diameter (m)
Typical associated illness
Human hair
Red blood cell
Mycobacterium bacteria
Pneumococci bacteria
Inuenza virus
Rhinovirus virus
30 50
8.0
0.2 1.0
0.5
0.1
0.03
Tuberculosis
Pneumonia
Flu, Croup, Pneumonia
Common Cold
To enable lters to remove particles efciently over a large dimensional

window, they use several mechanisms of ltration which combine all of these
mechanisms to a varying degree. The mechanisms used are:
3.1
Direct Interception
Filters are made up of matrices with a well-dened pore size. If the particles
are larger than the pores, they are captured by direct interception on the sur-
271
face of the lter element. This mechanism can be likened to the wire screens
used to separate gravel from sand.
3.2
Diffusional Interception
Very small particles like viruses could pass through the empty spaces of
a lter but they are inuenced by Brownian motion. This is caused by the collision of rapidly moving gas molecules with the aerosol size range particles
and droplets. A simple example of the three dimensional Brownian motion
could be described in two dimensions as a drunken man wandering around
the square [4]. The zig-zag movement of the microscopic particles caused by
these collisions substantially increases the probability of collision with a bre
within a thick lter element, such as is normally employed for HEPA levels
of efciency. The Brownian motion causes small particles to be collected on
the individual bres and pore walls of the lter. Particles of about 0.1 m
diameter and below are captured using this principle.
3.3
Inertial Impaction
Particles that have a higher density than air deviate from the air ow as it
passes through the lter and impact on the surfaces or walls of the pores
where they are captured. Inertial impaction works best for particles in the
range 0.3 to 10 m.
4
High Efficiency Particulate Air Filters
The rst HEPA lter was designed in the 1940s by the research and development rm Arthur D. Little under a classied government contract as part
of the Manhattan Project, where the rst atomic bomb was developed during World War II. A major advancement in air ltration technology, the lter
solved a critical need to control very small particles which had become contaminated by nuclear radioactive sources [6].
Considering the condensation nuclei of radioactive iodine to be most
harmful, researchers focused on the ability to capture solid particles that were
created through the condensation of gases and liquid aerosols into solid matter. Having identied 0.3 micron particles as the most penetrating size and
representative of the particle of concern, 0.3 microns was established as the
particle size fraction at which to determine lter efciency performance [6].
HEPA lters used in the aerospace industry are made of micro glass bres and
272
are similar to those used in hospitals. However, in the critical areas of hospitals where these are used, they lter outside air for removal of particulates
and aerosols, not to recirculate potentially infectious air. Their performance
or efciency is normally reported as the capture percentage for 0.3 micron
particles and to be meaningful, the reported efciency must relate to particle
size and ow velocity.
According to the European air lter efciency classication, a HEPA lter
can be any lter element rated between 85% and 99.995% removal efciency
for 0.3 micron particles. However, for aircraft cabin air recirculation systems,
this denition has been tightened and the current aerospace industry standard is 99.99% minimum removal efciency by sodium ame test to British
Standard BS.3928 or 99.97% minimum removal efciency by di-octyl phthalate (DOP) test according to ASTM publication D 2986-95 [7]. This is the
efciency standard now specied by Airbus and Boeing for their new generation aircraft.
The sodium chloride test consists of challenging the lter with an
aerosol mist of sodium chloride (NaCl) particles, with a mean particle
size of 0.58 micrometers. The DOP test consists of challenging the lter
with an aerosol mist of di-octyl phthalate oil droplets, with a mean size of
0.3 micrometers. The removal efciency, or penetration, is calculated as a percentage by measuring the aerosol concentrations upstream and downstream
of the lter element under test.
The American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) in their ASHRAE STANDARD 161 - Air Quality Within
Commercial Aircraft Committee Review Draft, state in relationship to HEPA
lters and recirculated air:
All air that is recirculated through the aircraft systems shall pass through
a high efciency particulate air (HEPA) lter before it is supplied to the cabin.
HEPA lters used for this purpose shall meet or exceed the requirements of
Institute of Environmental Science and Technology (IEST) Filter Type B,
MERV 17 or H13 according to EN 1822-1 and shall provide 99.97% collection efciency for 0.3 micron particles. The leak tests conducted to meet this
requirement shall be on the standard aircraft holding frame used for the lter in accordance with IEST-RP-CC0034.1 or EN1822-4. These lters and their
mountings shall be designed, installed and maintained as per manufacturer
recommendations to prevent bypassing of unltered air due to media failure,
improper installation, or other causes. Air used for recirculation should be extracted from the cabin at locations where the air is expected to be the least
contaminated. [8]
Signicant differences between microbes and chemical compounds (such
as DOP and NaCl) makes the use of chemicals unsuitable for rating the microbial removal efciency of air lters.
Hence, HEPA lters on transport-category aircraft remove particles with
an efciency higher than 99.97% at 0.3 micron, signicantly reducing the level
273
of airborne-particulate contamination. Using the denition of a HEPA lter as

either 99.99% sodium ame or 99.97% DOP, the rst Airbus HEPA lters were
introduced in 1994. The Airbus SIL, (ref. Airbus SIL21-065) [5] states the existing lter element of 98% efciency with a mean particle size of 8 microns
(when tested with gravimetric method) has been replaced in production by
a new lter element with a 99.99% efciency with a particle size of about
0.3 microns (when tested by sodium ame test).
The rst HEPA lters (99.97% DOP) for Boeing aircraft were introduced
on the B747-400 in 1998 (ref. Boeing SIL 747-SL-21-52-A) [5]. In 1999, United
Airlines became the rst major airline to install HEPA lters throughout the
airlines eet [9]. Other airlines also have been specifying HEPA lters for new
aircraft in recent years and retrotting some aircraft.
HEPA lters typically are disposable and roughly about 3 feet by 2 feet by
8 inches thick. This thickness accommodates the pleated structure of the lter medium which substantially increases the available lter area within the 3
by 2 foot frame, which reduces the pressure required to force the air through
the lter. They have about a 4500 hour service life (although no regulation
currently enforces change of used lters) and weigh about 5.5 lbs each [10].
Additionally, combined particulate and odor removal cabin lters are being offered by lter manufacturers such as Pall Aerospace in conjunction with
Airbus [11]. These have an odor absorber tted in series with the HEPA lter to remove odors and Volatile Organic Compounds (VOCs). The combined
lters are directly interchangeable with the existing particulate cabin air lter elements. They weigh about twice the weight of a HEPA lter and Design
Service Life is given as: To be determined on in-service experience [11].
Airlines so far have generally selected the HEPA only lter due to the nancial
penalties associated with the lower service life of the combined lters which
need replacing more frequently. However, combined lters allow the operator to produce air that is closer to the quality of the outside air, which it is
supposed to replace, so perhaps this should be regulated.
4.1
Number of Recirculated Air Filters per Aircraft Type
The number of lters present on a commercial aircraft to lter recirculated air
depends on aircraft types. Some examples are given in Table 2. Newer aircraft
appear to have an increasing number of lters in an attempt to improve the
quality of the recirculated air.
In a drive to make engines more efcient and use less fuel, industry argues that by removing contaminants to such high levels there is justication
to reduce further the amount of fresh air supplied from the engines from current industry practices. This would potentially enable industry greater cost
savings but this is a much debated issue currently being evaluated by organi-
274
Table 2 Relationship of lter numbers to maximum passenger carrying capability

Aircraft type
Maximum number of passengers

[12, 13]
Airbus
A300-600/A310
Airbus
A318/A319/
A320/A321
Airbus
A330/A340
298/247
117/134/164/199
Boeing
737 -300, -500,
-600, -700
Boeing
737 -400,
-800, -900
Boeing
747-400
Boeing
757
Boeing
767
Boeing
777
DC-10/
MD-11
From 293 for A330-200

up to 419 for A340-600
Up to 149
Up to 189
Up to 524
in 2 class layout
228 on the -200
280 on the -300
Up to 375
No of lters [14]
4
(8 on extended
range ones)
1
10
2
2
Up to 550
Up to 410
zations such as the ASHRAE in the USA and by the European Association of
Aerospace Industries (AECMA) in Europe.
4.2
Cost Saving of Recirculated Air
In the early days all passengers were supplied with 100% unrecirculated air.
An important question is whether the nancial savings made by the airline
industry by recirculating air are really what the consumer wants, and whether
this additional exposure risk is warranted. It has been estimated that the introduction of air recirculation saves airlines an average of US$ 60 000 per
aircraft [4], per year. For 300 trips per year of a 200-seat aircraft this amounts
to one dollar saved (by the airline) per passenger trip. This saving rises only
to about two dollars per passenger trip for half the size of the aircraft, or for a
50% load factor for the 200-seat aircraft. The 15-cent per passenger hour fuel
275
cost estimate for a 50% increase in the fresh air ventilation rate for a DC10-10
given in another survey by Nagda et al. is consistent with this. For a 6-hour
ight, this would calculate to a cost of about 90-cents per passenger trip. If
only one more passenger became ill on a ight that used air recirculation
compared with a ight without it, the cost to that individual and to society
would far exceed the saving by the airline to use recirculation [15].
4.3
Operational Effect of Unserviceable Recirculation Fans
A typical aircraft Minimum Equipment List (Dispatch Deviation Manual)
which tells pilots and engineers what minimum equipment needs to be functioning prior to departure makes the following statements in relation to
Recirculation Fans:
Boeing 747-400
Number of
recirculation fans
tted = 4
With
one fan failed, increase

ight planning fuel by 0.3%
With
two or more fans failed,

increase ight planning fuel
by 0.8% [16]
A Boeing 747-400 doing a typical long range ight from London to Singapore will burn 150 000 kg of fuel. Therefore, 0.3% will be an extra 450 kg, 0.8%
will be 1200 kg of extra fuel needed.
This shows that the amount of extra fuel that would be used in exchange
for not having recycled air, a cost that perhaps passengers would be prepared
to pay if they were asked to make an informed choice.
5
Bleed Air Filtration
So far, we have only looked at ltering recirculated air which is very different
from ltering the engine bleed air or APU ventilation air, which sometimes
become contaminated with hydraulic uids, engine oils or pyrolysis products of these not intended to ever be in the cabin air. Hydraulic uids and
engine oils usually contain a selection of toxic ingredients which include
N-phenyl-1-naphthylamine (a skin sensitiser) and the organophosphate Tricresyl phosphate (TCP) [17]. The bleed air is known to become contaminated
sometimes [18, 19], which prompts the question of whether this can be ltered to remove these contaminants, and what system could be used to do it.
If the lters were placed in the correct location, i.e. at the pack outlet, then
perhaps some of the issues could have been addressed many years ago. In particular Donaldson Company, Inc., have chemical adsorptive lters which been
276
produced for various Boeing commercial airplanes for over 20 years and these
lters are claimed to remove VOCs including organophosphates such as TCP
from the recirculated cabin air [34].
Bleed air coming off the engines is very hot pressurized air and if ltered at
source it would need a different ltration device from those designed for the
ltration of microscopic particles and droplets from recirculated air which
is at room temperature and cabin pressure. The next section explores the
techniques to better lter bleed air which include ltering hot bleed air or
ltering the bleed air at less extreme temperatures and pressures as it leaves
the air conditioning packs after it has been cooled, before it reaches the mix
manifold or aircraft cabin.
5.1
Non Regenerative Chemical Filtration System
Commercial aircraft could use cold bleed air ltration to remove engine oils
and hydraulic uid contaminants that could be based around a popular adsorbent such as activated carbon. This was highlighted at a Cabin Air Quality
Seminar in 1991 [20].
AMETEK Aircontrol Technologies in Middlesex, England have produced
activated carbon Odor Removal Filters for the BAe 146 for some 10
years [21]. The activated carbon is a different product from the carbon
absorbent beds offered by other lter manufacturers to remove odors. AMETEK Aircontrol Technologies use an activated carbon cloth about which the
suppliers state that: The activated carbon cloth (ACC) used in the Ametek ight deck and cabin lter are of the same specication to those currently being supplied for some military NBC applications. Our ACC was originally developed in conjunction with CDE Porton Down. Extensive testing
has shown the size and distribution of the pore structure in these materials
make them very good at adsorption of nerve gases which are organophosphorus compounds. Tri cresyl phosphate (TCP) is also an organophosphorus
compound and therefore we see no reason why it should not be equally well
adsorbed [22]. As with all lters, it must be remembered that they are only
effective if regularly replaced before they become saturated.
Activated carbon is the main component found in standard NBC (Nuclear,
Chemical & Biological) gas masks and protective suits. Coal becomes activated charcoal when it has been heated with steam or carbon dioxide, and
in the absence of air. This process opens up millions of very small pores
between the carbon atoms, resulting in highly porous charcoals that have
surface areas of 3002000 square meters per gram. These so-called active, or
activated, charcoals are widely used to adsorb odorous or colored substances
from gases or liquids, that attach to it by physical attraction. The huge surface
area of activated charcoal gives it a large number of bonding sites. When certain chemicals pass next to the carbon surface, they attach to the surface and
277
are captured. Activated charcoal is good at trapping carbon-based impurities (organic chemicals), as well as gases like chlorine. The more absorption
sites become saturated, the less the lter will work. Therefore, the lter life
will be dependent on the concentration of contaminants and the amount of
contaminated air it is exposed to, as any particulate lter eventually clogs.
This is also true, of course, for the HEPA lters used for recirculation of cabin
air, which must be changed regularly to remain effective.
Air purication technologies used on NASA spacecraft are worth looking at as they detail how far back technology was available to protect crew
members when this was given a priority over cost issues. The early space programs, Mercury, Gemini, and Apollo, employed equipment that relied heavily
upon physical and chemical adsorption and coarse particulate matter ltration to address these challenges. These used activated carbon to remove trace
contaminants. Trace chemical contamination control still relied upon expendable adsorption beds. Little change was realized with the development of the
Space Shuttle. Air purication systems used on board the Shuttle Orbiter actually reverted to systems similar to those used before Skylab. Expendable
chemical and physical adsorption systems have been the rule. As a result, mission duration is limited to 15 days or less [23]. Skylab, Americas rst space
station, employed a similar approach for cabin air purication with the exception that carbon dioxide partial pressure control was provided by a pressure
swing adsorption system.
5.2
Regenerative Chemical Filtration Systems
To increase mission life or time between lter replacements a Regenerative
Chemical Filtration System would be more advantageous. Sorbtion technologies capable of being regenerated can be classied by the method used to
drive off the gases adsorbed, these are:
Temperature Swing Adsorption (TSA)
Pressure Swing Adsorption (PSA)
Pressure and Temperature Adsorption (PTSA)
All these systems usually use a minimum of two lter beds, one adsorbing
while the other is regenerating by the removal of previously adsorbed challenges. Following regeneration, the incoming air is diverted to the cleaned
bed and the bed previously on line commences its regeneration cycle.
5.2.1
Temperature Swing Adsorption
A TSA system is similar to current charcoal systems in that it removes gases
at low (ambient) pressure, but is then regenerated by heating the regenerat-
278
ing bed to temperatures in the order of 170 C (338 F). The regenerated bed
must be fully cooled before it can be used on line and, unlike current systems,
the adsorbent material cannot be treated to improve the range of chemicals
adsorbed as such treatments would be destroyed during the regenerative heat
cycle.
5.2.2
Pressure Swing Adsorption
In the case of a PSA system, a technology which was rst developed in
1956 [24], the on line bed operates at an elevated pressure and the off line bed
regenerates at low pressure. The beds are designed to adsorb gases at elevated
pressure, and release the gases when the pressure is removed.
5.2.3
Pressure Temperature Swing Adsorption
PTSA systems combine the characteristics of both TSA and PSA systems in
that adsorption takes place at elevated pressure and desorption is achieved by
removal of the pressure and heating of the bed.
PSA systems appear to offer the most solutions to commercial aviation
if a will existed to do so. Pall Aerospace has integrated PSA protection into
numerous military systems, including the Apache, Cobra and Comanche helicopters, and an advanced armored test bed vehicle for the army. A PSA system
has even been installed in the personal limousine of the Head of State of
a friendly foreign nation [24]. PSA uses in commercial aircraft ltration have
been discussed since at least 1991 [20]. Experience gained by NASA with the
Skylab project of the 1970s demonstrates how long the technology has been
available.
5.3
Plasma
Plasma technology is likened to cold combustion. Instead of using heat to
break up contaminants, the plasma cells destroy molecules using highly reactive strongly oxidizing free radicals atoms or molecules that have unpaired
electrons. From the chemical point of view, destruction reaction rates normally associated with temperatures of 10 000 to 100 000 K can be realized with
the gas at near ambient temperature [25].
Nonthermal Plasma Systems in combination with a particulate capture
lter may be able to remove particulates and decompose chemicals and biocontaminants with lower than current energy and maintenance costs.
The questions that need to be addressed before such technology could be
adapted for use in commercial aviation would be:
279
What are the effects of air ow rate/residency time on efcacy?

What are the effects of plasma depth/throw (i.e. what is the relationship
between the plasma size and the volume of air that can be ltered?)
What is the range of contaminants that can be decomposed?
When bleed air is contaminated with engine oil, hydraulic uid, or pyrolysis products of these, what reaction products are produced and how safe
are they?
5.4
Ultraviolet Light
The key to achieving the cleanest indoor air is a multistage air purier. Ultraviolet light systems can be used to supplement other ltration techniques.
There are four principal wavelengths in the ultraviolet spectrum that
lend themselves to specic applications: Photochemical UV-A, Erythemal
UV-B, Germicidal UV-C, and extreme shortwave UV Energy, which generates
ozone [26]. The destruction of germs and bacteria by germicidal ultraviolet
light is accomplished quickly and effectively. The UV-C rays strike the various
microorganisms whether they are bacteria, virus, yeast, mold or algae, and
they break through the outer membrane. The radiation reaches the heart of
the organisms (commonly known as the DNA) where it causes abrupt modications. The modied DNA transmits incorrect codes or messages, which
brings about destruction of the microorganisms.
5.5
Nanocrystalline Materials
Nanotechnology refers to the emerging set of tools, techniques and unique
applications involving the structure and composition of materials at the nearatomic, or nanometer level. A nanometer (nm) is a billionth of a meter,
or a millionth of a millimeter. Although the original suggestions date back
to about 1970, systematic large scale research on nanocrystalline materials
only began around 1990 [27]. Nanocrystalline materials exhibit a wide array
of remarkable chemical and physical properties, and can be considered as
new materials that bridge molecular and condensed matter. One of their remarkable properties is enhanced surface chemical reactivity (normalized for
surface area) toward incoming adsorbates [28].
As a result of their high surface areas and their enhanced surface reactivity,
nanocrystalline MgO, CaO and Al2 O3 have shown remarkably high capacities
to chemically adsorb organic compounds and substantially outperform the
activated carbon samples that are normally employed for such purposes [29].
The literature suggests that such materials are good candidates for adsorbents of toxic industrial chemicals including metals [30]. Toxic chemicals
that have been demonstrated to be chemically decomposed by nanoactive
280
metal oxides at room temperature include but are not limited to the chemical
warfare agents GD, VX and HD [31] and the organophosphates dimethylmethyl phosphonate (DMMP), paraoxon, parathion, diisopropyluorophosphate (DFP) [32].
Nanocrystalline materials are now part of the ltration world and are worthy of further investigation to see how they could be used in commercial
aircraft ltration systems.
5.6
Catalytic Converters
To prevent oil breakdown products from entering the cabin air, catalytic converters have been used to clean the cabin air [33]. Donaldson was part of the
development team for a catalytic bleed air purication unit for a military airplane cabin air application in the 1950s and is still producing that product
today [34]. Catalytic converters used in cars have been around since the early
1970s and are now included in virtually every car sold in the United States.
They usually use two different types of catalysts, a reduction catalyst and an
oxidation catalyst. Both types consist usually of a ceramic structure coated
with a metal catalyst, usually platinum, rhodium and/or palladium. The idea
is to create a structure that exposes the maximum surface area of catalyst to
the exhaust stream, while also minimizing the amount of catalyst required (to
reduce cost). There are two main types of structures used in catalytic converters honeycomb and ceramic beads. Most cars today use a honeycomb
structure.
5.6.1
The Reduction Catalyst
The reduction catalyst is the rst stage of the catalytic converter. It typically uses platinum and rhodium to help reduce the NOx emissions. When an
NO or NO2 molecule contacts the catalyst, the catalyst rips the nitrogen atom
out of the molecule and holds on to it, freeing the oxygen in the form of O2 .
The nitrogen atoms bond with other nitrogen atoms that are also stuck to the
catalyst, forming N2 . For example:
2NO N2 + O2
or 2NO2 N2 + 2O2
5.6.2
The Oxidation Catalyst
The oxidation catalyst is the second stage of the catalytic converter. It reduces
the unburned hydrocarbons and carbon monoxide by burning (oxidizing)
them over a platinum and palladium catalyst. This catalyst aids the reaction
281
of the CO and hydrocarbons with the remaining oxygen in the exhaust gas.
For example:
2CO + O2 2CO2
The efciency of a catalytic converter depends on the catalyst used, pressure, temperature and residency time. A catalytic converter may be set up to
function with a preset amount of contamination in mind but if this changes
and the converter becomes overloaded they may then produce signicant
amounts of contaminants. These could potentially include elevated carbon
monoxide and carbon dioxide levels, as well as the presence of unconverted,
or semi-converted, hydrocarbon oil constituents and a reduction in the oxygen concentration in the air being delivered to the cabin [33]. Catalytic converters introduced on the BAE 146 did not achieve the desired efciency and
were removed by some operators for this reason [35].
As of December 2004 ltration companies such as Pall Aerospace were
re-examining catalytic converters in the search for economical solutions to
ongoing contaminated bleed air problems. Additionally Englelhard, who were
pioneers in automobile catalytic converters, recently commercialized a combined VOC and ozone converter for use in the Airbus A 320 [36].
6
ECHO-Air
The current ventilation system in commercial aircraft is based on an idea
which is many decades old and widely used. A newly developed alternative
ventilation/ltration system called ECHO-Air has been designed to try to improve overall system performance [37]. A prototype of the concept involved in
this system has just been tested in a Boeing 737 [38].
The ECHO Air system is designed to eliminate or signicantly reduce aircraft envelope condensation which can promote undesirable and potentially
hazardous effects such as rain in the plane, microbial growth, electrical
system deterioration, fuselage corrosion, and dead weight accumulation. By
pressurizing the envelope with a portion of the dry ventilation air prior to
its entry into the cabin, the system provides a dynamic barrier that prevents
cabin air inltration into the envelope through thermal-gradient induced
stack pressures. Coincidentally, passing ventilation air through the envelope improves cabin air quality through absorption and ltering of such
contaminants as ozone, oil aerosols and combustion VOCs without the high
pressure drops associated with standard ltration systems.
An airow controller driven by an electronic control unit is used to control the envelope ventilation either positively or negatively with respect to the
cabin, in relation to the phase of the ight (ground, take-off, ascent, cruise,
descent, landing and taxiing). The ECU also monitors the system operation.
282
On demand from a cockpit switch, the envelope pressure can be set positive
or negative with respect to the cabin, giving the pilot full control to act in the
case of smoke in the cabin [39].
The main idea is to divide the air circulation into two separate ows of air:
one that goes in to the cabin, and one that goes in to the envelope (the space
between the cabin liner and the fuselage) (Fig. 2).
Envelope tubing (plastic, or metal for the re suppressant injection version) is used to allow a controlled ow of air inside the envelope, both to and
from it. Flow blockers are used to reduce stack pressures, and to control the
air distribution within the envelope. Stack pressure is the pressure differential that exists across the liner due to the buoyancy effect of the air inside
the envelope and the difference in air pressures caused by the extreme cold
temperature of the envelope air near the fuselage during ight in comparison
with the air temperature inside the cabin. The system uses an airow controller to pressurize the envelope, either positively or negatively with respect
to the cabin, so as to offset at least stack pressures and upstream molecular
diffusion. Upstream molecular diffusion occurs across the liner coming from
the lower pressure air stream.
Methods of sealing and openings in the cabin liner, as well as dened pressure differences across the liner, are used to control the ow of air through the
liner. Leakage dimensions (areas and thickness) in the liner are set together
with air velocities to limit upstream molecular diffusion of specic gases and
vapors of concern, and to control cabin air circulation and exhaust.
When depressurizing the envelope relative to the cabin, a secondary air return path (additional to the cabin oor openings) is provided across the liner
through the envelope. This depressurization is used to provide a more direct
Fig. 2 Illustration of the main features of the echo-air system
283
Fig. 3 Exhaust ows during the exhaust mode of operation of the echo-air system. (Illustrations courtesy of Indoor Air Technologies Inc, Canada and USA)
284
and effective path for venting cabin air contaminants, to exhaust envelope air
pollutants including smoke and re suppressants in the event of a re, and to
exhaust envelope moisture (Fig. 2).
When pressurizing the envelope positively relative to the cabin, the air supply is of low humidity and thermally conditioned to assist with cabin thermal
conditioning and maintain a dry envelope preventing moist cabin air entry to
the envelope. On the ground this dry air is supplied by the aircraft APU or by
a ground-based air conditioning unit. In the air, its source is a mixture of the
bleed trim and pack air and is also used to ventilate the cabin. Bleed air is dry
at altitude and the use of it for envelope supply keeps the envelope dry by preventing moist cabin air from entering the envelope, while entering the cabin
itself to mix with cabin circulation air.
The ECHO-Air system has several advantages over current cabin ventilation designs used in the industry, as summarized below:
When pressurizing the envelope negatively with respect to the cabin:
It allows reduction in pathogen spread within the cabin by providing more
direct exhausting of air contaminated by the passengers or as a result of
a terrorist gas or aerosol release, and taking advantage of thermal plumes
that normally rise to the ceiling;
It facilitates the reduction of volatile organic compounds in the cabin air
by exhausting directly through the envelope of gases formed when the
envelope is warmed during taxiing or on ground;
It accelerates the clearing of smoke in the cabin in case of re in the cabin
through additional exhausting from the envelope;
It allows suppression of re or pyrolysis in the envelope by the direct injection of a re suppressant without impacting cabin air, and it prevents
envelope smoke from entering the cabin by venting it directly to the outdoors. If the envelope is initially under positive pressure (see below), such
an event will be detected as normal by smoke in the cabin. At such a time,
the pilot switches ECHO Air to envelope depressurization mode for an
envelope re hazard. If the envelope is already under negative pressure
at the time of such an event, a smoke sensor in the envelope air exhaust
will detect this. During such a hazard mitigation period, ECHO Air envelope ventilation is set at minimum depressurization exhaust rate, so as to
maintain the envelope under a negative pressure while minimizing both
air supply to the re and dilution of the re suppressant being injected.
When pressurizing the envelope positively with respect to the cabin:
It lters all or a portion of the cabin ventilation air before it enters the
cabin. This ltration reduces incidents of passenger exposure to combustion products ingested when taxiing behind other planes, for example, and
to bleed air oil aerosols if an upstream engine lubrication bearing seal
fails;
285
It provides an additional airway path thereby enabling an increase in the

total ventilation/recirculation rate of the cabin air through envelope ventilation, thereby reducing pathogen spread through higher rates of air
dilution and air circulation through pathogen lters and sterilizers;
It prevents entry of humid cabin air both on the ground using ground air
conditioners, and in the air using engine bleed air, thus reducing moisture
accumulation in the envelope, improving thermal and acoustic performance of the insulation material in the envelope, and reducing fuselage
corrosion inside the aircraft;
It allows humidication of the cabin air without the drawbacks of increased accumulation of moisture inside the envelope;
It enables cabin heating and cooling via the envelope, thereby providing
a more comfortable cabin temperature with smaller thermal gradients and
cold drafts to passengers.
7
Conclusions
There are currently no airworthiness standards or regulations which specify
the level of ltration removal efciency which must be used on board aircraft.
The quality of the air, if regulations are enforced, should be regulated to some
extent by FAR/JAR 25.831 which states:
Each crew compartment must have enough fresh air (but not less than 10
cubic feet per minute per crewmember) to enable crewmembers to perform
their duties without undue discomfort or fatigue.
Crew and passenger compartment air must be free from harmful or hazardous concentrations of gases or vapors.
The majority of modern, large, commercial aircraft, which use a recirculation type of cabin air system, utilize ne HEPA ltration, (99.99% minimum
sodium ame efciency). A small number of aircraft types have lters with
lower efciencies. Some older aircraft have either total outside air ventilation,
or a small amount of unltered recirculation combined with the outside air.
HEPA lters will only work properly if properly installed (HEPA lters
shall be installed in order to minimize the recirculation of bacteria and viruses
in the air distribution system but there is no denitive time interval for replacing a cabin air lter element. HEPA lters shall be maintained according to
best practices manufacturers specications [40]). The time interval varies
between aircraft types. Manufacturers recommend that airlines follow the
guidelines provided by the manufacturers in the aircraft maintenance manuals. Pall Aerospace states that: It is often the case that airlines will choose
to replace cabin air lter elements at regular hard time intervals to t in
with routine scheduled maintenance periods, such as a C-Check. The denition
286
of a C-Check varies between aircraft models and operators [5]. Donaldson

Company also states: Recommended service interval is one C-Check for most
operators [41].
The AMETEK Aircontrol Technologies BAE 146 lters previously discussed
show that technology exists to help protect passengers and crews but regulators have to date not made these mandatory, simply optional or for information. Activated carbon technology has existed for many years to protect
military personnel from potential exposures to organophosphate based nerve
gases, such as Sarin. The failure of regulators to enforce protective measures
may be driven by economics, as lters must be regularly replaced before they
become saturated and ineffective [42].
To ensure that crews and passengers are safe from the organophosphates
and other contaminants present in the contaminated engine bleed air, appropriate prevention measures by way of ltration safety options should be
mandated by the regulators.
In an industry so focused on ight safety with numerous back up systems
it seems inconceivable that no protection systems are in place to protect crew
and passengers from bleed air malfunctions.
The potentially serious health problems to exposed passengers and crew
from this source deserve to be given a higher priority than seems to be directed to this problem at the moment. (One particular aircraft type has over
200 Service Bulletins, Service Information Leaets, etc. referring to contaminated air on this aircraft model since its introduction in the 1980s [42]). It
seems wiser and safer to have in place safety systems that cater for all mechanical failures, poor designs or malfunctions in relationship to bleed air
contamination rather than argue that bleed air contamination events are statistically a rare event and not worthy of the proper independent research and
investment.
If appropriate prevention measures cannot effectively protect crews and
passengers then the only option is to return to the original bleed air philosophy of early jet aircraft and not allow the bleed air from the engine to directly
enter the aircraft cabin. Interestingly, the latest Boeing 787 has been designed
to use bleed free engines, where the air supply for the aircraft is not directly taken from the engine air in the traditional manner but reported to be
supplied from separate compressors. The engines in development for the Boeing 787, The General Electric GENX (General Electric Next-generation) and the
Rolls-Royce Trent 1000 will eliminate the use of bleed air [43]. Some might
argue that perhaps bleed free engines are the safest option. The British
Airline Pilots Association (BALPA) invited every leading airline, aircraft manufacturer, engine manufacturer, lubricant manufacturer, regulators and every
person who has ever had an input into the issues of contaminated air, whether
independent or from industry to make a presentation at the Contaminated
Air Protection Conference in April 2005 and the conclusions were very clear,
alarming, simple and included:
287
There is a workplace problem resulting in chronic and acute illness

amongst ight crew (both pilots and cabin crew).
Further, we are concerned the passengers may also be suffering from similar symptoms to those exhibited by ight crew.
An absurd situation often exists where those who regulate the industry
often have no direct responsibility towards passenger and crew health and
seemingly no urgency to enforce such protective measures. CAAs prime responsibilities for passengers are to regulate for their safety. It has no direct
responsibilities for passenger health or comfort [44] Thus, HSE has no active
responsibilities in relation to the health of airline passengers or crew [45].
Filtration companies have had the technology for many years to decrease
the effect of contaminated bleed air. Regrettably, their main clients, the airlines and manufacturers, have not had the will, nor see the need to invest in
the technology.
References
1. Boeing (2005) Cabin air quality. http://www.boeing.com/commercial/cabinair/facts.
html. Cited 2005
2. Finnair (2005) http://www.nnair.com/web/nnair/scripts/template_2level_white.
jsp?pageid=-13038. Cited 2005
3. Pall Aerospace (2005) Boeing and McDonnell Douglas data. In: Cabin air ltration.
Pall Aerospace, East Hills, NY
4. Pall Aerospace (2005) Cabin air ltration. APME 500a COD/2m/5/92
5. Available at: http://www.donaldson.com/en/aircraft/cabinair/index.html
6. Available at: http://www.airtesters.com/HEPA_lters.cfm
7. Pall Aerospace FAQs for BALPA AETG website campaign August 2004. Available at:
http://www.balpa.org
8. ASHRAE Standard 161 (2004) Air quality within commercial aircraft committee review craft, January 2004. In: Sections: 6.3.1 Recirculated Air Quality
9. INDA (2000) INDA e-FILTER newsletter. http://www.inda.org/period/enews/jan00.
html. Cited 3 Jan 2000
10. Pall Aerospace Technical Data Sheet APME P/N QA06423-01. Not dated.
11. Airbus Publicity Card by Pall Aerospace Combined Particulate & Odour Removal
Cabin Air Filters APM528/BP/2M/0402 April 2001
12. Available at http://www.airbus.com assuming 32 inch seat pitch. A330200 assumes
30B at 40 in + 263Y at 32 in pitch and A340-600 assumes 36B at 40 in + 383Y at 32 in
pitch
13. Available at: http://www.boeing.com
14. Data taken from applicable aircraft specic maintenance manuals
15. Hocking MB (1998) Indoor Air Quality: Recommendations Relevant to Aircraft Passenger Cabins. Am Ind Hyg Assoc J 59:446454
16. Boeing 747-400 Minimum Equipment List. Last updated 2004
17. Winder C, Balouet JC (2002) The Toxicity of Commercial Jet Oils. Environ Res Section
A 89:146164
288
18. Michaelis S (2003) A Survey of Health Symptoms in BALPA Boeing 757 Pilots. J Occupational Health & Safety, Aust & NZ 19(3):253-261
19. Cox L, Michaelis S (2002) A survey of symptoms in BAe 146 aircrew. J Occupational
Health and Safety Australia and New Zealand 18:305312
20. Needelman WM (1991) New Technologies for Airliner Cabin Air Contamination Control. Adsorption & Chemical Methods For Gaseous Pollutants. Needelman WM
Associate Director, Scientic and Laboratory Services Department, Pall Corporation.
Presentation at the International Seminar on Cabin Air Quality in Commercial Airliners, Paris, 19 June 1991
21. Available at: http://www.aircontroltechnologies.co.uk
22. Letter from Giles M of Vapour Management Systems Ltd to Loraine T of British Airline Pilots Association AETG, Cited 4 October 2004
23. Perry JL, LeVan D Air Purication In Closed Environments: Overview Of Spacecraft
Systems. Available at: http://www.natick.army.mil/soldier/jocotas/ColPro_Papers/
Perry-LeVan.pdf
24. Available at: http://www.pall.com/Aerospace_2947.asp
25. Golkoski C, Hedge A (2003) Nonthermal plasma air ltration technology. Super Pulse
& Dept. Design & Environmental Analysis, Ithaca, NY
26. Available at: http://www.purennatural.com/fs.php?center=airpuriers%2
Fultravioletairpuriers.php
27. Available at: http://physics.umbc.edu/ takacs/nano.html
28. Klabunde KJ, Stark JV, Koper O, Mohs C, Park DG, Decker S, Jiang Y, Lagadic I,
Zhang D (1996) Nanocrystals as stoichiometric reagents with unique surface chemistry. J Phys Chem B 100:1214212153
29. Khaleel A, Kapoor PN, Klabunde KJ (1999) Nanocrystalline metal oxides as new absorbents for air purication. Nanostructured Mater 11(4):459468
30. Decker SP, Klabunde JS, Khaleel A, Klabunde KJ (2002) Catalyzed destructive adsorption of environmental toxins with nanocrystalline metal oxides: uoro-, chloro-, bromocarbons, sulfur, and organophosphorous compounds. Env Sci Technol 36(4):762
768
31. Wagner GW, Procell LR, OConnor RJ, Munavali CL, Carnes CL, Kapoor P, Klabunde KJ (2001) Reactions of VX, GB, GD, and HD with nanosize Al2 O3 : formation
of aluminophosphonates. J Am Chem Soc 123:16361644
32. Rajagopalan S, Koper O, Decker S, Klabunde KJ (2000) Nanocrystalline metal oxides
as destructive adsorbents for organophosphorus compounds at ambient temperatures. Chem Eur J 8:26022607
33. Van Netten C, Leung V (2000) Comparison of the constituents of two jet engine lubricating oils and their volatile pyrolytic degradation products. Appl Occupational
Environ Hyg 15(3):277283
34. Verbrugge K (2004) Sales Engineer, Donaldson Europe personal communication.
September 2004
35. BA6 36-10-11: Ansett internal engineering release on work undertaken on BAe modications. Remove engine/APU catalytic converters. Ansett work completed 7/95
36. Air Transportation Center of Excellence (2004) Proposal to FAA by air transportation
center of excellence for airliner cabin environment research, vol 1
37. Walkinshaw DS, Mitalas GP, McNeil CS, US Patent 6,491,254 (Dec 10, 2002)
38. Indoor Air Technologies Inc. Media release available at: http://www.cyberus.ca/dsw/
iat/echoairboing.html
39. Interview with leader of Echo Air Project, Mr Doug Walkinshaw. http://www.
indoorair.ca/iat/echoairfaq.html
289
40. ASHRAE Standard 161 (2004) Air quality within commercial aircraft committee review draft, January 2004. Section: B. Maintenance
41. Donaldson (2005) BIOAdvantage aircraft cabin air HEPA lter with antimicrobial
protection, product datasheet. http://www.donaldson.com
42. Available at: http://www.aopis.org
43. Available at: http://encyclopedia.thefreedictionary.com/Bleed%20air and http://www.
balpa.org/intranet/BALPA-Camp/The-Aircra/The-Aircraft-Environment.pdf
44. House of Lords Select Committee on Science and Technology Air Travel and HealthHL Paper 121-I. (2000) 3.19 CAA lays down aviation safety standards in areas broadly
similar to those of ICAO and JAA, and sets them out in regulations made under the
Air Navigation Order (ANO)
45. House of Lords Select Committee on Science and Technology Air Travel and HealthHL Paper 121-I. (2000) 3.21 The Health and Safety at Work etc. Act 1974 applies to
aircraft in and over Great Britain but has no role outside the airspace above Great
Britain. The Executive (HSE) set up under the Act seeks to avoid duplicating the activities of other regulatory bodies associated with health and safety. Its interface with
CAA is the subject of a Memorandum of Understanding. Aircraft have been exempted
from many regulations made under the governing Act (p 1)

DOI 10.1007/b107249
Aircraft Cabin Air Quality Trends

Relative to Ground Level Standards
James E. Cone
Environmental and Occupational Disease Epidemiology,
New York City Department of Health and Mental Hygiene, New York, NY 10007, USA
jcone@health.nyc.gov
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
Candidate Ground Level Standards and Standard-setting Agencies

US Occupational Safety and Health Administration . . . . . . . . .
American Conference of Governmental Industrial Hygienists . . . .
US Environmental Protection Agency . . . . . . . . . . . . . . . . .
American Society for Heating, Refrigeration
and Air Conditioning Engineers . . . . . . . . . . . . . . . . . . . .
Society of Automotive Engineers . . . . . . . . . . . . . . . . . . . .
European Community . . . . . . . . . . . . . . . . . . . . . . . . . .
Spacecraft Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . .
Submarine Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . .
3
3.1
3.2
3.2.1
3.2.2
3.2.3
3.2.4
3.2.5
3.2.6
3.2.7
3.2.8
3.2.9
Types of Contaminants and their Regulation

Other Specic Contaminants of Concern . .
Specic Classes of Compounds, by Use Type
Pesticides . . . . . . . . . . . . . . . . . . . .
Jet Fuels . . . . . . . . . . . . . . . . . . . .
Jet Oils . . . . . . . . . . . . . . . . . . . . .
Hydraulic Fluids . . . . . . . . . . . . . . . .
Carbonyl-containing Compounds . . . . . .
Dusts and Particulates . . . . . . . . . . . .
Physical Parameters . . . . . . . . . . . . . .
Infectious Disease Transmission . . . . . . .
Tobacco Smoke . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
4
4.1
4.2
4.3
Effects of Aircraft Environmental Systems . . . . . . . . .

Environmental Control Systems . . . . . . . . . . . . . . .
Filtration Systems . . . . . . . . . . . . . . . . . . . . . . .
Distribution of Air and Temperature Control in the Cabin
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Conclusions and Recommendations
294
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
297
297
298
299
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
299
300
300
300
300
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
304
304
305
305
305
305
309
309
309
310
310
310
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
311
311
312
312
. . . . . . . . . . . . . . . . . . . . .
312
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
313
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Abstract Aircraft cabin air quality has attracted much attention, summarized recently by
a detailed examination and commentary by a U.S. National Academy of Sciences Committee. Ventilation of aircraft has several signicant variables that require control measures
that are seldom of concern for occupied space at ground level. The principal of these special requirements are the need to compensate for the substantial difference between cabin
294
J.E. Cone
and outside pressures, the much lower available space per occupant in aircraft cabins,
and the need for coping with more extreme external temperatures than are common at
ground level. The breadth of these concerns is of interest in the policies and regulatory
aspects of a number of agencies which are briey described, and their roles and areas of
potential interest outlined. Types of possible contaminants are listed, and the limits which
have been set by several of these agencies for many of these potential contaminants are
tabulated. In addition recent measured aircraft cabin concentrations of several key contaminants are listed. This chapter provides an overview of the general air quality variables
affecting enclosed space to enable these to be related to the special needs of some of the
less common enclosed spaces described in the following chapters.
Keywords Air contaminants Gases Vapors Aerosols Particulates
Physical parameters Infective agents
Abbreviations
ACGIH American conference of governmental industrial hygienists
ASHRAE American society for heating, refrigeration and air conditioning engineers
CFU
colony-forming unit
ECS
environmental control system
EPA
U.S. environmental protection agency
FAA
federal aviation act
FAR
federal aviation regulation
HEPA
high efciency particulate air [lter]
MSDS
manufacturer(s) safety data sheet
NAAQS national ambient air quality standard
NAS
national academy of sciences
NACOSH national advisory committee on occupational safety and health
NASA
national aeronautics and space administration
NRC
national research council
OSHA
US occupational safety and health act
PEL
permissible exposure limits
SARS
severe acute respiratory syndrome
SEALS submarine escape action levels
SMACs spacecraft maximum allowable concentrations
STEL
short term exposure level
TLV
threshold limit value
TWA
time weighted average
VOC
volatile organic compounds
1
Introduction
Signicant concerns have been raised regarding the impact of the cabin air
environment on the health and safety of passengers and crew. The combination of high occupant density, relatively low ventilation rates, and varying but
potentially long occupancy periods up to 18 hours on some intercontinen-
Aircraft Cabin Air Quality Trends Relative to Ground Level Standards
295
tal ights create the potential for adverse health impacts. Aircraft cabins are
a uniquely challenging environment. As Hocking has noted, aircraft cabins
have the smallest available airspace per person of any current social environment, and occupants of a fully loaded aircraft typically have about 3570 ft3
(12 m3 ) of available airspace per person, approximately 1/10th that of a typical ofce worker or a spectator in an auditorium [1].
In 2003, the US National Academy of Sciences [2] (NAS) released its latest
report, funded by the Federal Aviation Administration. The NAS report, most
importantly, recommends that air quality in commercial aircraft be monitored with routine surveillance of air-quality characteristics such as ozone,
carbon monoxide, carbon dioxide, ne particulate matter, cabin pressure,
relative humidity, and temperature. In addition, it called for a detailed research program to be launched to investigate specic questions about the possible association between air contaminants and observed or reported health
effects.
As noted by the NAS report, since passage of the Federal Aviation Act in
1958, the Federal Aviation Agency maintains authority over the regulations
related to operation and safety of civil aircraft (Public Law 85726). The Occupational Safety and Health Act (OSHA) was adopted in 1970 to regulate
health and safety provisions for workers (Public Law 91516). Exemptions
from OSHA coverage included workers in industries regulated by other agencies such as the Airlines (FAA), Railroads (Federal Railway Administration),
maritime workers, and federal, state and local government workers. The FAA
exercised its option to regulate the safety and health of airline cabin workers beginning in 1975 (40 FR 29114, DOT 1975). Federal Aviation Regulations
(FARs) that have been subsequently promulgated by the FAA to govern air
quality in commercial aircraft so far include O3 , CO, carbon dioxide (CO2 ),
ventilation, and cabin pressure (14 CFR 21, 14 CFR 25, 14 CFR 121, and 14
CFR 125). Similarly, the European Joint Airworthiness Authority (JAA) regulates European cabin air through Joint Aviation Regulations.
The National Academy of Sciences (NAS) report [2] concluded that the current design standard for the minimum amount of outside air circulated into
cabins is about half the ventilation rate often required for building environments. Reduced ventilation rates in buildings have been linked to increased
reports of health symptoms and sick leave, but whether building ventilation
standards are appropriate for airplanes has not been determined. Studies
of transmission of infectious airborne diseases such as tuberculosis during
ights suggest that the spread of infectious agents during ights does not
appear to be facilitated by aircraft ventilation systems, but rather by the
high density of people, the committee concluded. An aircrafts environmental
control system itself can be a source of contamination during abnormal operations when engine oil, hydraulic uids, or de-icing solutions enter the cabin
through the air-supply system in what is called bleed air. Many crews and
passengers have reported air quality incidents involving smoke or odors
296
J.E. Cone
within cabins. The NAS committee said FAA should study the need for and
feasibility of installing equipment to remove vapors and particles from air
supplied by the environmental control system on all ights.
Other countries have also convened reviews of airline cabin air quality,
including the British House of Lords [3] and the Australian Parliament [4].
Rep. John Mica (R-FL), the Chairman of the U.S. House Subcommittee on
Aviation, stated in a June 5, 2003 hearing, Flight crews and passengers have
continued to raise concerns about the cabin air quality in commercial aircraft.
There have also been questions about the possible transmission of contagious
diseases in-ight. Most recently, the focus has been severe acute respiratory
syndrome, or SARS. Often those who y complain of headaches, fatigue,
fever, and respiratory difculties. The unanswered question is whether these
complaints are due to poor cabin air quality or to other factors inherent when
ying for a long period of time in a conned space with other people.
Patricia Friend, President, Association of Flight Attendants, noted in the
same hearing the following issues related to airline cabin air quality that are
currently unresolved:
Inadequate ventilation and standards for aircraft;
Polluted air supply on the ground from exhaust fumes and heated deicing
uids;
Exposure to heated oils and hydraulic uids that can leak or spill into air
supply systems;
Reduced oxygen in the ambient air during ights which is generally
equivalent to altitudes of 6000 to 8000 feet;
Inadequate attention to the thermal environment;
Exposure to ozone gas which can result in respiratory distress and increase
susceptibility to infection;
Exposure to potentially high concentrations of pesticides that are sprayed
in planes on some international ights.
Unless adequate solutions are found to the above problems, these concerns
will tend to increase with the growth of air trafc, the tendency for airlines to
seek greater fuel efciency, and the trend toward future generations of aircraft
providing less fresh air ventilation and more recirculation of air in aircraft
cabins. Despite the above concerns, and the work of several active committees
tasked with such standard development for over the past 10 years, there are
currently no accepted International, North American or European standards
for the air/environmental quality within aircraft cabins. In light of these developments, and in particular, the worldwide SARS epidemic of 2003, U.S.
Senator Dianne Feinstein (D-Calif.) has called for a national standard for airplane cabin air quality [5]. The absence of US regulations addressing a wider
range of additional cabin air contaminants and environmental factors is the
source of signicant concern on the part of airline industry workers, unions
and the ying public. Particularly in light of current economic stressors on
297
major sectors of the airline industry, these problems may be exacerbated in

the future.
This chapter will address the following questions:
1. What are the existing ground level air quality standards of potential relevance to the airline cabin environment?
2. What are relevant measured levels of airborne contaminants of concern
aboard aircraft?
3. What are potential actions and policy options for development of airline
cabin air quality standards in the future?
Standards are reference values to which something can be compared. If properly developed and maintained they represent the combined knowledge of
designers, manufacturers, and consumers and are useful as benchmarks for
industry, providing a way to insure compatibility, comparability, enabling
mass production and a means of measuring and testing of products [6]. Common weaknesses of standards include failure to prevent contamination due
to conict of interest in standards setting organizations, failure of standards
setting bodies to update standards with new information or changes in technology, and failure of enforcement.
Federal regulatory agencies other than the US FAA (with its limited set of
airline cabin environmental standards), such as the U.S. Occupational Safety
and Health Administration (OSHA) and the U.S. Environmental Protection
Agency (EPA) have not established exposure limits for the unique environment of aircraft, nor are their existing standards necessarily appropriate for
this environment. What are the agency standards and standard-setting processes that currently exist that might be appropriate as a starting point for
considering development of a more comprehensive set of airline cabin air
standards?
2
Candidate Ground Level Standards and Standard-setting Agencies
2.1
US Occupational Safety and Health Administration
The US Occupational Safety and Health Administration (OSHA) was established in 1970 as an administrative agency in the US Department of Labor.
OSHA is the primary federal agency responsible for maintaining minimal
standards for indoor air of workplaces. The setting of standards for workers
would have spill-over benets in terms of potentially protecting passengers
and the general public, although workplace standards are often signicantly
less stringent compared with standards for the environment of the general
public, including potentially more susceptible populations. The National Ad-
298
J.E. Cone
visory Committee on Occupational Safety and Health (NACOSH) has noted

since consensus standards were rst adopted in the two years after the passage of the Occupational Safety and Health Act, a relatively small number of
standards have been promulgated. Further, standards such as Permissible Exposure Limits (PEL) have not been successfully updated. The average time
to develop and promulgate a standard is ten years .... During the time these
important standards were under consideration, hundreds of workers continued to be killed or seriously injured annually by these hazards. NACOSH
made recommendations for streamlining and speeding up the standard setting process. However, underlying problems causing the ineffectiveness of
OSHA standard setting, including lack of political will in administrations
fundamentally opposed to regulation and the increasing weakness of unions
compared with corporate interests, were not addressed. Some states that opt
to have state OSHA plans have had more success in updating and adopting
standards. This is permitted as long as they adopt standards that are at least
as effective as the federal standards. Even in the best states, however, there
is a relative lack of systematic scientic or worker input into the standard
setting process, resulting in the tendency toward wholesale and uncritical
adoption of industry consensus standards such as the American Conference
of Governmental Industrial Hygienists Threshold Limit Values.
2.2
American Conference of Governmental Industrial Hygienists
The American Conference of Governmental Industrial Hygienists (ACGIH)
establishes industry standards through a Threshold Limit Values (TLV)
Committee. This effort started in 1946. The membership of ACGIH represents industry, government, academia and to a diminishing extent, labor
organizations and was drawn from four disciplines: industrial hygiene, toxicology, occupational medicine and occupational epidemiology. The TLV
Committee sets guidelines and recommendations, not regulations, and they
publish an annual booklet of recommended limits for chemical substances
and physical agents, primarily for use by industrial hygienists. According to
the preface of the TLV booklet, TLVs are health-based recommendations derived from assessment of the available published scientic information from
studies in exposed humans and from studies in experimental animals [7].
Criticism of the TLV development process has focused on the lack of adequate documentation of the committees decision-making processes, lack of
an effective means of preventing conicts of interest from tainting committee decisions [8], and lack of explicit scientic rationale or health basis [9]
for many of the TLVs that have been developed. Nevertheless, the TLVs tend
to be more protective than current Federal OSHA regulations, and are more
frequently updated. As Peter Montague has noted, During the 20 years that
OSHA spent setting 12 new PELs, the ACGIH TLV Committee revised 234
299
TLVs downward, making them more protective (and stricter than the corresponding PELs which had been adopted in 1970 but never revised), and
adopted 168 new TLVs for which there were no PELs [10].
2.3
US Environmental Protection Agency
The Environmental Protection Agency is responsible for setting ambient air
quality standards for the general public, including sectors of the population
that may be most vulnerable. Since airlines serve the same general public, and
include passengers that are in the vulnerable groups, EPAs ambient air quality
standards could serve as a starting point for aircraft cabin air standards. The
EPA national ambient air quality standards mandated by the 1991 Clean Air
Act cover criteria air pollutants with a focus on those associated with smog:
ground level ozone, carbon monoxide and particulates, volatile organic compounds and 189 specic hazardous air pollutants. Many, but not all of these
overlap with chemical contaminants of concern in aircraft cabins.
2.4
American Society for Heating, Refrigeration and Air Conditioning Engineers
The American Society for Heating, Refrigeration and Air Conditioning Engineers (ASHRAE) establishes standards for many aspects of building ventilation. One standard in particular is most applicable to ventilation of airline
cabins: Standard 622001. The purpose of ASHRAE Standard 62, as dened
in Sect. 1, is to specify minimum ventilation rates and indoor air quality that
will be acceptable to human occupants and are intended to minimize the potential for adverse health effects. The scope of Standard 62 applies to all
indoor or enclosed spaces that people may occupy, except where other applicable standards and requirements dictate larger amounts of ventilation than
this standard. This guideline is voluntary, and has not been adopted by the
FAA. Most useful for aircraft environments may be the standard for comfort
(odor) criteria with respect to human bioefuents. In addition, ASHRAE recommends that indoor CO2 concentrations be maintained less than 700 ppm
above the outdoor air concentration [11], Table 4. As the NAS 2002 report
has pointed out, ASHRAE Standard 621999 is also generally more restrictive than FAAs FAR 25 with respect to both O3 and CO [12]. In addition,
it states that [13] Assuming that the cabin temperature and pressure apply,
and not the outside conditions, it is seen that ASHRAE Standard 621999
would require 50100% more outside air than the current requirement in
FAR 25.
Temperature and humidity guidelines are also provided by another
ASHRAE Standard (551992, Thermal Environmental Conditions for Human
Occupancy ASHRAE 1992), that proposes voluntary ranges of temperature
300
J.E. Cone
and humidity that are generally found comfortable related to activity level
and clothing.
2.5
Society of Automotive Engineers
The Society of Automotive Engineers has published a recommended practice guideline, Procedure for Sampling and Measurement of Engine Generated
Contaminants in Bleed Air Supplies from Aircraft Engines Under Normal Operating Conditions, ARP4418 (SAE 1995), that includes a table from AIR4766,
Air Quality for Aircraft Cabins that species the maximal concentrations of
contaminants in engine bleed air. Multiple other guidelines relevant to airline
cabins have been prepared by SAE, for example, Testing of Airplane Installed
Environmental Control Systems (ECS) ARP217 March 1999.
2.6
European Community
The European Commission Directive 2000/39/EC of 8 June 2000 established
a rst list of indicative occupational exposure limit values in implementation
of Council Directive 98/24/EC on the protection of the health and safety of
workers from the risks related to chemical agents at work [14]. Unfortunately,
the amount of overlap between the regulated occupational exposures listed
and those likely to be of concern in airline cabins is small (Table 1).
2.7
Spacecraft Guidelines
The National Aeronautics and Space Administration (NASA) requested that
the National Research Council (NRC) develop spacecraft maximum allowable
concentrations (SMACs) for airborne contaminants. A subcommittee was established and four [17] reports of its ndings have been published to date,
including recommendations by NASA scientists and contractors on 35 substances of concern. Recognizing that differences exist regarding conditions
aboard spacecraft compared with airline cabins, most notably duration of
trips, weightless conditions, and unique contaminants that each environment
may face, nevertheless, the spacecraft air guidelines may be a useful starting
point for considering possible similar standards for airlines. The 1 hour and
24 hour SMACs, however, are clearly intended for emergencies, and may result in some mild mucosal irritation symptoms. They are not intended for
protecting the general public, especially susceptible populations. These are
further discussed in other sections of this volume.
5000 ppm
Carbon
dioxide
Carbon
monoxide
Nitrogen
dioxide
700 ppm
above ambientg
9 ppm (8 h)
9 ppm (8 h)
0.055 ppm
(ann avg)
0.05 ppm
ASHRAEa
50 ppm
5 ppm
0.05 ppm
(annual average)
5000 ppm
0.1 ppm
OSHA PELc
35 ppm (1 h)
0.12 ppm (1 h)
0.08 ppm (8 h)
EPA NAAQSb
3 ppm (TWA),
5 ppm (STEL)
0.05 ppm (TWA)

(heavy work),
0.08 ppm
(moderate work),
0.1 ppm
(light work)
5000 ppm (TWA),
30 000 ppm (STEL)
25 ppm (TWA)
ACGIH TLVd
NAS SMACs
b EPA
621999.
NAAQS, 40 CFR 50.
c PEL = OSHA permissible exposure limit.
d TWA = time-weighted average concentration in a normal 8-h workday and a 40-h workweek, to which nearly all workers may be repeatedly
exposed, day after day, without adverse effect (ACGIH 1999). STEL = short-term exposure level is a 15-min TWA exposure that should not be
exceeded at any time during the workday (ACGIH 1999).
e FAA airworthiness standards (14 CFR 25) for ozone: 0.25 parts per million by volume, sea level equivalent, at any time above 32 000 ft; and
0.1 parts per million by volume, sea level equivalent, time-weighted average during any 3-h interval.
f National Institute for Occupational Safety and Health (NIOSH) recommended exposure limit (REL) not to be exceeded at any time for O is
3
0.10 ppm (NIOSH 1997); California Air Resources Board California ambient air-quality standard (CAAQS) for O3 is 0.09 ppm for 1-h exposure
(CARB 1999); and World Health Organization guideline for O3 is 0.06 ppm for 8-h exposure (WHO 2000).
g Applies to use of carbon dioxide as a proxy for odors from bioefuents; not a limit on exposure to carbon dioxide.
a ASHRAE
0.1 ppm
0.25 ppm
Ozonee,f
50 ppm
35 ppm (1 h)
FAA
Contaminants
Table 1 Limits on contaminants that may be found in aircraft cabin air (adapted from NAS 2002, reprinted with permission) [37]

301
PMh10
PMh2.5
Formaldehyde
Acetone
Acetylaldehyde
Acrolein
Benzene
Temperature
ASHRAE_55-1996i
ASHRAEa
1 ppm
0.1 ppm
200 ppm (TWA)
10 ppm
0.75 ppm (TWA)

2 ppm (STEL)
1000 ppm
150 g/m3 (24 h)

65 g/m3 (24 h)
OSHA PELc
EPA NAAQSb
0.5 ppm (TWA)

2.5 ppm (STEL)
0.1 ppm (ceiling)
10 ppm (TWA)
15 ppm (STEL)
500 ppm (TWA),
750 ppm (STEL)
25 ppm (ceiling)
1000 ppm
0.3 ppm (ceiling)
ACGIH TLVd
1 hour 10 ppm,
24 hour 6 ppm
1 hour 0.075 ppm,
24 hour 0.035 ppm
1 hour 10 ppm,
24 hour 3 ppm
1 hour 50 ppm
24 hour 50 ppm
NAS SMACs
b EPA
621999.
NAAQS, 40 CFR 50.
d TWA = time-weighted average concentration in a normal 8-h workday and a 40-h workweek, to which nearly all workers may be
repeatedly exposed, day after day, without adverse effect (ACGIH 1999).
STEL = short-term exposure level is a 15-min TWA exposure that should not be exceeded at any time during the workday (ACGIH 1999).
h PM = particulate matter less than 10 microns in diameter; PM
10
2.5 = particulate matter less than 2.5 microns in diameter.
i Thermal Environmental Conditions for Human Occupancy.
a ASHRAE
Acetic acid
Freon 113
FAA
Contaminants
Table 1 (continued)
302
J.E. Cone
Methylene chloride
Hydrogen sulde
Toluene
Xylene
Pyrethrum
Vinyl chloride
1 ppm
1000 ppm
50 ppm (ceiling)
200 ppm
OSHA PELc
5 mg/m3
1 ppm
200 ppm
100 ppm
1 hour 0.1 ppm

50 ppm
24 hour 0.005 ppm
(PA) [15]
500 ppm
EPA NAAQSb
NAS SMACs
50 ppm (TWA)
100 ppm (TWA)
150 ppm (STEL)
5 mg/m3
5 ppm
50 ppm
1 hour 130 ppm

24 hour 30 ppm
1 hour 100 ppm

24 hour 35 ppm
1000 ppm (TWA)

39.4 ppm (ceiling)
5 ppm
1 hour 10 ppm
24 hour 10 ppm
0.01 ppm
1 hour4ppm
24 hour 0.3 ppm
10 ppm
15 ppm
(Submarines [16])
ACGIH TLVd
b EPA
621999.
NAAQS, 40 CFR 50.
d TWA = time-weighted average concentration in a normal 8-h workday and a 40-h workweek, to which nearly all workers may be repeatedly
exposed, day after day, without adverse effect (ACGIH 1999). STEL = short-term exposure level is a 15-min TWA exposure that should not be
exceeded at any time during the workday (ACGIH 1999).
a ASHRAE
Hydrazine
Ethanol
Ethylene glycol
2-ethoxy ethanol
ASHRAEa
FAA
Contaminants
Table 1 (continued)

303
304
J.E. Cone
2.8
Submarine Guidelines
The US Navy has proposed Submarine Escape Action Levels (SEALS) that are
intended to be protective of the health of personnel in a disabled submarine.
U.S. Navy Health Research Centers Toxicology Detachment proposed two exposure levels, called submarine escape action level (SEAL) 1 and SEAL 2, for
each gas. SEAL 1 was dened as the maximum concentration of a gas in
a disabled submarine below which healthy submariners can be exposed for
up to 10 days without experiencing irreversible health effects. SEAL 2 is dened as the maximum concentration of a gas in a disabled submarine below
which healthy submariners can be exposed for up to 24 hours without experiencing irreversible health effects. These were reviewed by the NAS and
found to be adequate with the exception of chlorine [16]. They are similarly
not intended for protecting the general public, especially susceptible populations and thus are not likely to be useful with regard to development of airline
cabin air quality standards. These are further discussed in other sections of
this volume.
3
Types of Contaminants and their Regulation
Specic standards for pollutants and physical environmental characteristics
potentially found in aircraft cabins are listed in Table 1. Some are commonly
found in other indoor or transportation environments, and others are relatively unique to the aircraft cabin. Chapters 10 to 12 of this volume address
these contaminants in detail. Measurement of levels of contaminants is particularly a problem in difcult to predict intermittent cabin air quality incidents.
The US FAA ventilation standard species that the air of the cockpit and
cabin must be free of harmful or hazardous concentrations of gases or vapors (14 CFR 25, Section 831). For example, according to the standard, CO
concentrations in excess of 1 part in 20 000 parts of air (50 ppm) are considered hazardous, and CO2 concentrations during ight may not exceed 0.5%
by volume (sea-level) or 5000 ppm.
3.1
Other Specific Contaminants of Concern
Spengler [18] reported a wide range of sources of potential contaminants
in cabin air: Volatile Organic Compounds (VOCs), including; fuel exhaust
(toluene, xylenes, benzene, decane, undecane, hexane pentadiene), distilled
spirits and human bioefuents (propan-2-ol, ethanol, acetone), air fresheners
305
and cosmetics (limonene, toluene), dry cleaning agent (tetrachloroethene),

refrigerants (dichlorodiuoromethane), solvents (butan-2-one, toluene, 1,1,1trichloroethane, xylenes) and plastic resin (vinyl acetate).
Table 2 lists existing data on measured air contaminants aboard aircraft
from recent studies. Additional studies in the future will expand this information, particularly if the NAS recommendations for additional sampling are
followed.
3.2
Specific Classes of Compounds, by Use Type
3.2.1
Pesticides
Pesticides used on aircraft include 2% phenothrin aerosol or residual spraying using a permethrin emulsion. In a study analyzing contents of aerosol
sprays, VOCs were found in all preparations including ethyl benzene and
xylene isomers along with phenothrin. Residual sprays contained cis-, and
trans-permethrins, palmidrol, and occasionally naphthalene [19]. There are
currently no US or European standards designed to control pesticide exposures aboard aircraft. US OSHA standards exist for some of the inert
ingredients such as xylene and ethyl benzene.
3.2.2
Jet Fuels
Jet fuels are complex mixtures of hydrocarbon components and performance additives. JP-8 is one common military jet fuel containing naphthalenes
while Jet A and A-1 are among the most common sources of nonmilitary occupational chemical exposure. Jet fuel varies by airplane and engine type.
Combustion of jet fuel results in CO2 , H2 O, CO, various carbon-containing
particles, NOx , and a large number of complex organic compounds. OSHA
as well as EPA standards exist for various components of jet fuels and their
combustion byproducts.
3.2.3
Jet Oils
A recent review summarized the hazards of jet oils: Jet oils are specialized
synthetic oils used in high-performance jet engines. They have an appreciable
hazard due to toxic ingredients, but are safe in use provided that maintenance personnel follow appropriate safety precautions and the oil stays in the
engine. Aircraft engines that leak oil may expose others to the oils through
uncontrolled exposure. Airplanes that use engines as a source of air for cabin
2223
na
78
Nagda
et al. 1989a
92
min
max
na
199e
Particulate matter, g/m3

mean
37 (PM3.5 )
176
(PM10 )
140
200
na (total
particles)
3
10
36
23
60
1400
1200
1800
2
10
na
na
< 0.1
7
1469 225
924
1959
51 15
< 20
122
3861091c
293
2013
Study and number of ights

Spengler
ASHRAE/
Haghighat
et al. 1997
CSS 1999
et al. 1999
6
8
43
Nitrogen oxides, ppb

mean
min
max
Dechow
1996
x
0.7
0.8
1.3
1162
na
na
CSS
1994
35
Carbon monoxide, ppm

mean
0.6
min
na
max
1.3
Carbon dioxide, ppm

mean
1756 660
min
765
max
3157
Ozone, ppb
mean
min
max
Contaminants
or
Characteristic
na
1980
117c,d
4.549.6f
na
na
1.92.39c
1.0
4.0
6831557c
423
2900
0
90
na
Lee et al.
1999a
16
Table 2 Contaminant concentrations reported in published studies (from NAS, 2002, reprinted with permission)
30
380
na (PM10 )
580 700
< 200
3100
0.87 0.65
< 0.2
9.4
1387 351
664
4238
200 180
< 50
1000
Waters
et al. 2001
37
< 10 (PM2.5
& PM10 )
0.2
na
0.8
1380
na
1755
Nagda
et al. 2001b
10
306
J.E. Cone
Nagda
et al. 1989a
92
Formaldehyde, ppb
mean
min
max
Bacteria, CFU/m3
mean
min
max
Fungi, CFU/m3
mean
min
max
na
0
360
na
0
110
9.0 12.7
na
61
na
na
2200
(ppb)
CSS
1994
35
131.1 123.4
na
642
VOC, g/m3 , with ethanol

mean
min
max
Contaminants
or
Characteristic
Table 2 (continued)
na
20
1700
7
3
26
na
na
2200e
(ppb)
Dechow
1996
x
201f
na
na
3171e
608
1805e
na
<1
37
na
39
244
2.9 1.7
< 0.6
4.9
900 450
380
1500

Spengler
ASHRAE/
Haghighat
et al. 1997
CSS 1999
et al. 1999
6
8
43
na
17
107
na
44
93
Lee et al.
1999a
16
na
0
< 0.07
Waters
et al. 2001
37
7.2 (g/m3 )
na
13 (g/m3 )
Nagda
et al. 2001b
10

307
x
1.8
nba
14 3.2
8.8
27.8
na
na
6950
18
17
19
na
5500
6900
16.8
na
na
20.323.8c
19
27
23 1.7
17.8
26.1

Spengler
ASHRAE/
Haghighat
et al. 1997
CSS 1999
et al. 1999
6
8
43
23.0
22.2
25.6
Dechow
1996
x
24.4
na
na
CSS
1994
35
10.042.6c
4.9
55.5
21.325.3c
17.8
26.3
Lee et al.
1999a
16
Waters
et al. 2001
37
5500
8000g
10.5
na
34.3
23
26
Nagda
et al. 2001b
10
b Values
from non-smoking ights.

represent those in cabin during cruise.
c Range of means.
d Particle size range measured was not specied.
e Values from Space DR, Johnson RA, Rankin WL, Nagda NL (2000) The airplane cabin environment: past, present and future research. pp. 189
214 in Air Quality and Comfort in Air Cabins, Nagda NL (ed), ASTM STP 1393. West Conshohochen: American Society for Testing Materials.
f Geometric mean.
g Range varied depending on aircraft type. For B767 and B747, cabin-pressure altitude was 5,5006,500 ft. during cruise, for B737, approximately
8,000 ft.
a Data
Temperature, C
mean 24.1 1.6
min 21.0
max 27.2
Relative humidity, %
mean 21.5 5.1
min 9.9
max 30.8
Cabin-pressure altitude, ft
mean 4344
min 2.415
max 7212
Contaminants
or
Nagda
Characteristic et al. 1989a
92
Table 2 (continued)
308
J.E. Cone
309
pressurization may have this source contaminated by the oil if an engine

leaks. Examination of the ingredients of the oil indicates that at least two ingredients are hazardous: N-phenyl-1-naphthylamine (a skin sensitizer) and
tricresyl phosphate (a neurotoxicant, if ortho-cresyl isomers are present).
Publicly available information such as labels and MSDS understates the hazards of such ingredients and in the case of ortho-cresyl phosphates by several
orders of magnitude. [20] Applicable standards: Airborne Exposure Limit
for Tri-orthocresyl Phosphate (TOCP): OSHA Permissible Exposure Limit
(PEL) is 0.1 mg/m3 (TWA) and the ACGIH Threshold Limit Value (TLV) is
0.1 mg/m3 (TWA).
3.2.4
Hydraulic Fluids
Hydraulic uids used in commercial aircraft include, for example, Skydrol
500B-4. This is a re resistant hydraulic uid including a proprietary phosphate ester mixture composed principally of dibutyl phenyl phosphate and
tributyl phosphate [21]. Hydraulic uids may similarly enter the airline cabin
if air used for ventilation is contaminated. The existing standards for tributyl
phosphate: OSHA General Industry PEL 5.0 mg/m3 and the ACGIH TLV:
0.2 ppm, 2.2 mg/m3 TWA.
3.2.5
Carbonyl-containing Compounds
Carbonyl compounds (e.g. formaldehyde, acetaldehyde, acetone, and acrolein)
may be found in airline cabin environments at low levels [22]. OSHA standards exist for each of these compounds.
3.2.6
Dusts and Particulates
Existing standards for dusts and particulates of concern in airline cabin air
such as dander and cat allergens, food particles such as peanuts, and surface dusts and dust mite allergens consist primarily of OSHAs nuisance dust
standard of 5 micrograms/m3 , a level that is not appropriate for indoor transportation environments.
The US EPA has established standards for ultrane particles (PM10 m and
PM2.5 m ) in ambient air. Particles in the ultrane, and more generally, in the
submicron ranges are produced mainly from combustion, gas to particle conversion, nucleation processes or photochemical processes, with some of them
emitted directly by the source and some formed in the air from the precursors
emitted by the sources [23]. These constitute the largest number of particles
in ambient air. Cigarette smoke was a major source of respirable suspended
310
J.E. Cone
particulates in cabin air until smoking was progressively controlled and then
nally banned from all US domestic and international ights in the year 2000
(see Sect. 3.2.9 below).
3.2.7
Physical Parameters
Standards for physical parameters of cabin environment affecting cabin air
quality including temperature, relative humidity and cabin air pressure have
been set by the US FAA. However, existing standards will require extensive
review: For example, the FAA requirement for cabin air pressure was set in
1964 without any rationale, according to Eileen Abt, a National Academy
of Science staff director who worked on their recent panel on cabin air quality. It has never been revisited. Cabin air velocity/mixing is a more recent
concern. Air ow patterns and air velocity have signicant effects on thermal
environment and air quality around passengers [24]. Air distribution inside
aircraft cabins is a key factor affecting comfort as well as potentially affecting
disease transmission. Cabin altitude average during cruise, maximum rate
of change during ascent, descent, tilt of take-off and landing all are potential
concerns.
3.2.8
Infectious Disease Transmission
Infectious diseases of concern in airline cabin environments include bacteria, e.g., Mycobacterium tuberculosis (M. tb) and bacterial byproducts such as
endotoxin, viruses, such as inuenza, measles, mumps and chicken pox and
corona viruses such as is associated with Severe Acute Respiratory Syndrome
(SARS). The risk of transmission of infectious agents that are aerosolized
such as M. tb from an infectious person to passengers or crew on an aircraft
appears to be highest on long ights (8 hours or more) [25] and among those
working or sitting closest to the infectious person. There are no currently
applicable standards for microorganisms in ground environments, although
there have been attempts to set an OSHA standard for tuberculosis that have
been unsuccessful to date. Guidelines from the US Centers for Disease Control and Prevention [26] have been used by US state OSHA enforcement units
for control of tuberculosis in health care workplaces but not aircraft cabin environments. The World Health Organization has also developed guidelines to
prevent tuberculosis transmission aboard aircraft [27].
311
3.2.9
Tobacco Smoke
The difculties in regulating indoor air aboard aircraft are illustrated by the
case of tobacco smoke. A petition was rst submitted in 1969 to the US FAA
to request federal action to control tobacco smoking aboard aircraft [28]. The
FAA did not act on the petition, taking the position that tobacco smoke was
not likely to be a health problem for those exposed to second-hand smoke
aboard aircraft. The US Civil Aeronautics Board was more receptive and beginning in 1973 began regulating smoking aboard aircraft primarily through
establishing no-smoking zones in the cabin and separating cigar and pipe
smokers. A series of federal reports were published in 1986 summarizing the
mounting scientic evidence for harmful health effects of passive smoking,
including The health consequences of involuntary smoking, by the US Surgeon
General [29], and the US National Academy of Sciences report, Environmental tobacco smoke: measuring exposures and assessing health effects [30]. In
addition, the US National Academy of Sciences published, in the same year,
The airliner cabin environment, its rst report on Cabin Air Quality [31] that
proposed banning smoking from all ights within the US. Despite the efforts
of the tobacco industry, in 1988, a law to ban smoking aboard US domestic
ights of less than two hours went into effect. In 1990, Congress made the
ban on smoking on domestic ights permanent and expanded it to include all
domestic ights of six hours or less [32]. Most international airlines banned
smoking in the 1990s. Finally, smoking was banned from all domestic and international ights to and from the US in 2000. A few airlines internationally
continue to allow smoking on ights [33].
4
Effects of Aircraft Environmental Systems
4.1
Environmental Control Systems
The primary purpose of the ECS is to maintain cabin pressure in a range from
a maximum of 101 kPa (14.7 psi) on the ground at sea level to a minimum of
75 kPa (10.9 psi) in ight regardless of the altitude at which the aircraft ies.
A Congressional Aviation Subcommittee memorandum has summarized air
supplied to various aircraft types: Older model airplanes, such as the DC-9,
the B-727, and half of the DC-10s, provide 100% fresh air to the aircraft cabin.
Newer models of jet aircraft, including the MD-80, DC-10, B-737, 747, 757,
and A-300, 320 and 310, provide up to 50% re-circulated air. The recycled
air system allows newer model aircraft to conserve fuel. The effectiveness of
these ltration systems is often the focus of debate on cabin air quality [34].
312
J.E. Cone
4.2
Filtration Systems
Recirculated air in aircraft cabins is used to reduce the cost of compressing
outside air and maintain air circulation. Current practice is to lter recirculated air with particle lters. Filter efciency varies from 40% on MD-80
aircraft to 99.97% (High Efciency Particulate Air HEPA for 0.3 micron
particles) on most recently manufactured aircraft. [35] Filter changes occur
most often at scheduled maintenance checks, usually at 4000 to 12 000 ight
hours [36]. Although these lters remove various sized particles from recirculated air depending on the lter efciency, including bacteria and viruses
when HEPA lters are used, they do nothing to remove chemical contaminants in the form of gases. Optional activated charcoal lters may be used
on some aircraft to remove organic chemical contaminants, but these are
uncommon.
4.3
Distribution of Air and Temperature Control in the Cabin
In a typical aircraft, air is supplied in amounts sufcient to maintain thermal uniformity throughout the cabin and exhausted along the whole length
of the cabin. Air distribution is accomplished by single (narrow body aircraft)
or multiple diffusers (wide body aircraft) located in the middle of the ceiling
in the aisles, above the windows, or near the overhead baggage compartments. The NAS has noted that Adequate temperature control in the cabin
requires that conditioned air be supplied to the cabin at about 0.65 kg/min
(1.4 lb/min) per person to maintain a comfortable temperature. This requirement is more than twice the FAR 25 requirement of 0.25 kg/min per person
for outside air.
5
Conclusions and Recommendations
Existing ground level standards may be used as a starting point for development of airline cabin air quality standards, but as they exist currently, none
can be simply adopted as appropriate without careful review by a panel of independent experts. Possible policy and action options for the future include:
1. Implementation of NAS recommendations for additional sampling and research. This will provide a better scientic basis for future standard setting
activities.
2. It is the role of FAA to promulgate appropriate cabin environmental standards for the wide range of contaminants and other environmental quality
parameters potentially impacting crew and passengers.
313
3. If the FAA fails to exercise its preemption and promulgate such standards
in a timely fashion, US Federal OSHA could develop appropriate emergency temporary cabin air quality standards covering cabin crew.
4. Similar efforts are needed at the level of the European Union and United
Nations to establish worldwide cabin air quality standards.
A new sense of urgency is needed among the responsible agencies to increase
the likelihood that appropriate airline cabin air standards will be established
in the near future.
References
1. Hocking MB (2000) Passenger aircraft cabin air quality: trends, effects, societal costs,
proposals. Chemosphere 41:603615
2. Committee on Environmental Studies and Toxicology, National Research Council.
(2002). The airliner cabin environment and the health of passengers and crew.
http://books.nap.edu/openbook/0309082897/html/index.html. Cited 4 Oct 2005
3. House of Lords (2000) Air travel and health. House of Lords, Session 19992000, 5th
Report/HL Paper 121. Select Committee on Science and Technology, House of Lords,
Parliament, Great Britain
4. Parliament of the Commonwealth of Australia (2000) Air safety and cabin air quality
in the BAe 146 Aircraft: report by the Senate Rural and Regional Affairs and Transport
References Committee, Parliament House, Canberra
5. http://feinstein.senate.gov/03Releases/r-sars1.htm. Cited Oct 2004
6. Nondestructive Testing Resource Center. Standards.
http://www.ndt-ed.org/GeneralResources/Standards/standards.htm. Cited Sept 2004
7. Mattromatteo E (1988) TLVs: changes in philosophy. Appl Ind Hyg 3(3):F12F16
8. Senn Tarlau E (1990) Guest editorial; industrial hygiene with no limits. Am Ind Hyg
Assoc J 51(1):A9A10
9. Castleman BI, Ziem GE (1994) American conference of governmental industrial hygienists: low threshold of credibility. Am J Ind Med 26(1):133143
10. Montague P (1994) The scientic basis of chemical safetypart 2: standards that kill.
http://www.monitor.net/rachel/r416.html. Cited Sept 2004
11. Boschi N (2005) Aircraft Cabin Indoor Air Environment Requirements. In: Hocking MB, Hocking D (eds) Air Quality in Airplane Cabins and Similar Enclosed Spaces.
Handbook of Environmental Chemistry, vol. 4H. Springer, Berlin Heidelberg
12. Committee on air quality in passenger cabins of commercial aircraft (2004) The airliner cabin environment and the health of passengers and Crew, p 67
13. Committee on air quality in passenger cabins of commercial aircraft (2004) The airline cabin environment and the health of passengers and crew, p 66, http://www.
nap.edu/books/ 0309082897/html/66.html. Cited May 2005
14. European Commission Directive 2000/39/EC of 8 June (2000) Establishing a rst list
of indicative occupational exposure limit values in implementation of Council Directive 98/24/EC on the protection of the health and safety of workers from the risks
related to chemical agents at work. http://europa.eu.int/eur-lex/pri/en/oj/dat/2000/
l_142/l_14220000616en00470050.pdf. Cited Sept 2004
15. Pennsylvania Ambient Air Quality Standards. http://www.dep.state.pa.us/dep/
deputate/airwaste/aq/standards/standard.htm accessed 11/04
314
J.E. Cone
16. National Academy of Sciences, Board on Environmental Studies and Toxicology Review of Submarine Escape Action Levels for Selected Chemicals (2002)
http://books.nap.edu/books/0309082943/html/211.html#pagetop. Cited Sept 2004
17. Subcommittee on Guidelines for Developing Spacecraft Maximum Allowable Concentrations for Space Station Contaminants (19921996) Spacecraft maximum allowable concentrations for selected airborne contaminants. http://books.nap.edu/books/
0309054788/html/222.html#pagetop
18. Spengler JD, Burge H, Dumyahn TS, Muilenburg M, Forester D (1997) Environmental survey on aircraft and ground-based commercial transportation vehicles. Harvard
School of Public Health, Harvard University, Cambridge, MA
19. van Netten C (2002) Analysis and implications of aircraft disinfectants. Sci Total Environ 293(13):25762
20. Winder C, Balouet JC (2002) The toxicity of commercial jet oils. Environ Res
89(2):14664
21. Healy CE, Nair RS, Ribelin WE, Bechtel CL (1992) Subchronic rat inhalation study
with Skydrol 500B-4 re resistant hydraulic uid. Am Ind Hyg Assoc J 53(3):17580
22. Building Research Establishment (2000) Air quality in the aircraft cabin http://www.
dft.gov.uk/stellent/groups/dft_aviation/documents/page/dft_aviation_027561.hcsp.
Cited 22 Nov 2000
23. http://www.deh.gov.au/atmosphere/airquality/health-impacts/pubs/health-impacts.
pdf
24. Yuanhui Z (2004) University of Illinois, Urbana, Ill. Experimental Characterization of
Airows in Aircraft Cabins. Proceedings of the American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) Annual Meeting, Nashville, TN,
26-30 June 2004
25. http://www.cdc.gov/travel/diseases/tb.htm accessed 1/05
26. Centers for Disease Control (1994) Guidelines for preventing the transmission of
mycobacterium tuberculosis in health-care facilities. MMWR 43(RR13):1132
27. World Health Organization. Communicable Diseases Cluster (1998) Tuberculosis
and air travel: guidelines for prevention and control. http://www.who.int/docstore/
gtb/publications/aircraft/PDF/98_256.pdf. Cited Jan 2005
28. Neilsen K, Glantz SA (2004) A tobacco industry study of airline cabin air quality:
dropping inconvenient ndings. Tobacco Control 13(1):i20i29
29. US Department of Health and Human Services (1986) The health consequences of involuntary smoking: a report of the Surgeon General.: Public Health Service, Centers
for Disease Control (DHHS Publication No (CDC) 87-8398), Rockville, MD
30. Committee on Passive Smoking NRC (1986) Environmental tobacco smoke: measuring exposures and assessing health effects. National Academy Press, Washington,
DC
31. National Research Council (US) Committee on Airliner Cabin Air Quality (1986). The
airliner cabin environment: air quality and safety. National Academy Press, Washington, DC
32. Holm AL, Davis RM (2004) Clearing the airways: advocacy and regulation for smokefree airlines. Tobacco Control 13(I):i30i36
33. A list of airline smoking policies is maintained at http://hem.passagen.se/fungus/
airlines.html accessed 1/05, last updated 7/15/2003
34. http://www.house.gov/transportation/aviation/06-05-03/06-05-03memo.html
accessed 10/04
315
35. Committee on air quality in passenger cabins of commercial aircraft (2004) The airline cabin environment and the health of passengers and crew, p 50 http://books.nap.
edu/books/0309082897/html/50.html#pagetop. Cited Oct 2004.
36. Lundquist J (2002) Pall Aerospace, personal communication, July 5, 2001, cited in
NAS, Board on Environmental Studies and Toxicology. Pall Aerospace, East Hills, NY
37. Committee on air quality in passenger cabins of commercial aircraft (2004) The airline cabin environment and the health of passengers and crew, p 44, http://www.nap.
edu/books/ 0309082897/html/44.html. Cited May 2005

DOI 10.1007/b107250
Automobile, Bus, and Rail Passenger Air Quality

Clifford P. Weisel
Environmental and Occupational Health Sciences Institute, University of Medicine &
Dentistry of New Jersey Robert Wood Johnson Medical School, 170 Frelinghuysen
Road, Piscataway, NJ 08854, USA
weisel@eohsi.rutgers.edu
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
318
Principles of Exposure and Contribution Due to Proximity to a Source . .
318
Time Activity Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
320
4
4.1
4.2
4.3
4.4
Concentration of Pollutants Within Motor Vehicles

Carbon Monoxide . . . . . . . . . . . . . . . . . . .
Volatile Organic Compounds . . . . . . . . . . . . .
Carbonyl Compounds . . . . . . . . . . . . . . . . .
Particulate Matter . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
323
323
327
330
330
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
331
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
332
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Abstract Concentrations of pollutants emitted in the exhaust of gasoline and diesel engines and from evaporation of fuels (e.g. CO, CO2 , NOx , particulate matter, and volatile
organic compounds) have been measured at elevated levels within the enclosed spaces of
automobile, bus, and train cabins in urban centers throughout the world and compared
to ambient air levels. The magnitude of the elevation has been linked to vehicle type, fuel
composition and trafc congestion. The amount of time spent traveling and the mode
of transportation used varies across the population and by location. Controls on mobile
emissions have resulted in declines in the exposure to pollutants within vehicles in countries in which they have been implemented. The exposure received within cars, buses and
trains is a signicant portion of the total daily exposure to pollutants emitted by mobile
sources.
Keywords Bus Car Exposure In-cabin automobile Mobile sources
Abbreviations
BEADS
Benzene Exposure and Absorbed Dose Simulation
BEAM
Benzene Exposure Assessment Model
CO
carbon monoxide
HAPEM-MS Hazardous Air Pollutant Exposure Model for Mobile Sources
NAAQS
National Ambient Air Quality Standard
NEM
NAAQS Exposure Model
NHAPS
National Human Activity Pattern Survey
318
NO2
SHAPE
C.P. Weisel
nitrogen dioxide
Simulation of Human Activity and Pollutant Exposure
1
Introduction
Automobiles, buses and trains have enclosed spaces that people spend varying amounts of time in on most days. Due to the proximity of these enclosures
to the emission from gasoline and diesel engines and fuel tanks, elevated
levels of compounds emitted from mobile sources are encountered in these
locations leading to potential exposures. This chapter reviews the literature
of exposures associated with driving and riding in private and public transportation.
2
Principles of Exposure and Contribution Due to Proximity to a Source
The exposure to environmental contaminants occurs at the juxtaposition of
an individual with a location or activity that causes contact with that contaminant. The contact can be through breathing (inhalation exposure), touching
(dermal exposure) or consuming (ingestion exposure) the contaminant. Since
the concentration of environmental pollutants varies in space and time and
the amount of time spent in different locations, and activities are different for
a single individual from day to day as well as between individuals, a distribution of exposures will exist for a population. To estimate population-based
exposures for pollutants emitted from mobile sources, such as combustion
products (e.g. carbon monoxide, diesel particulates) and the fuel components themselves, it is important to determine the distribution of time spent
in those vehicles (cars, buses, trucks, trains), and the distribution of air
concentrations within the interior of vehicles. The concentration present in
the interior of vehicles is often elevated for pollutants emitted from mobile
sources compared to other locations where people spend time (indoors at
home and work, outdoors), due to the proximity of those vehicles to the emission sources. The concentrations of pollutants emitted from cars, buses and
trucks in roadway air are much greater than background ambient air concentrations and roadway air can enter into the cabin of automobiles and buses.
Automobiles and buses are adjacent to the tailpipe of the vehicle immediately in front of them when trafc is stopped or moving slowly, which can be
a source of pollutants for the vehicles interior. In addition, evaporation from
the fuel tank and leaks in the engine or exhaust system from the vehicle itself can enter the interior of cars and buses, and periodic fueling results in
319
the vehicle being in close proximity to a large source of volatile hydrocarbons,

which may cause pollutants to penetrate into the interior of the vehicle. The
enclosed nature of vehicles can result in a spike in concentration from small
emissions.
Pollutants of concern from mobile sources include carbon monoxide, nitrogen oxides, volatile organic compounds (predominantly hydrocarbons and
fuel additives), aldehydes and particulate matter. Gasoline and diesel fuel are
petroleum distillates containing hydrocarbons between C4 and C12 , and C10
and C19 , respectively. Within the United States, The Clean Air Act Amendments of 1990 specify the allowable emission rates of volatile organic compound from gasoline, gasolines benzene content and the minimum oxygen
content of fuel as well as the allowable air concentrations of carbon monoxide and nitrogen oxide [1]. The European Parliaments have issued a number
of directives to reduce evaporative and exhaust emission from mobile sources
starting in the early 1990s [2]. The World Health Organization has provided
air pollution guidelines for many pollutants with mobile sources that are used
by countries throughout the globe as a basis for trying to establish regulations
on emissions [3].
To reduce the levels of pollutants emitted and the amount of ozone formed
in the atmosphere during the summer, regulations have been implemented
in the US to alter the formulation of gasoline. Ozone is an air pollutant that
is formed photochemically from nitrogen oxides and hydrocarbons, and has
known adverse health effects for individuals with respiratory problems. The
changes in gasoline formulation have resulted in the fuel containing oxygenated compounds as a major component of the fuel, thereby reducing
emission and subsequent exposure to carbon monoxide and hydrocarbons.
However, these modications in fuel content have resulted in emissions of
a different suite of compounds, in particular ethers and aldehydes. Globally
the composition of fuel varies dependent upon the regulations in the country
and what fuel/raw materials are available at different prices. Considerations
for formulation of fuels are the octane rating and how much power the fuel
can provide the engine. Different additives and fuel compositions have been
used to obtain a sufciently high octane rating to provide a smooth operating engine. One of the early additives was tetraethyl lead, which caused
widespread distribution and exposure to lead, a neurotoxicant that affected
developing fetuses and young children. The use of tetraethyl lead has been
banned in many countries with the introduction of the catalytic converter,
a device to reduce hydrocarbon and nitrogen oxide emissions that was poisoned by lead. Without lead in gasoline, some countries have increased the
amount of aromatic hydrocarbons to obtain the higher octane ratings necessary for smoother running engines. The specic composition of fuel will
control the proportions and types of compounds emitted from automobiles
and trucks and thus the concentrations of compounds within the cabins of
vehicles.
320
C.P. Weisel
3
Time Activity Patterns
To determine how much time is spent by people within the cabins of automobiles, buses and trains, time activity patterns are generated based on
questionnaire and survey response data. The most comprehensive time activity pattern database collected within the United States which includes the
amount of time spent in or near vehicles by geographic location, age and gender, is the National Human Activity Pattern Survey (NHAPS) [4, 5]. NHAPS
is a probabilistically-based population sample, which implies it is representative of the larger population from which a randomly selected sub-population
was queried. Time activity data were collected as part of the EXPOLIS study
in Europe, which measured pollutant concentration and administrated questionnaires in six cities [6] and in Canada as part of the Canadian Human Activity Pattern Survey (CHAPS) [7]. Time activity pattern data have also been
collected in support of individual studies in different countries to construct
databases for evaluating specic exposures (e.g. India: [8]; Mexico: [7, 9, 10]).
These studies do not always have specic information that distinguishes
modes of trafc, such as private car and types of public transportation. Information on transportation can sometimes be obtained from statistical or
census information collected by governmental agencies which can be linked
to for many countries [11].
Since the largest compilation of activity pattern data has been collected
in the US, those data are presented in this manuscript, though comparisons
to other countries are included. The average percentage of time spent in the
ve major microenvironments or distinct locations where a uniform or modeled exposure can be determine for the total US population is: 5.5% within
a vehicle, 68.7% in a residence, 5.4% in an ofce-factory, 12.8% in other indoor locations and 7.6% outdoors [12] with a similar distribution of time for
the working European population [13] (Fig. 1). These proportions will vary
with location, age and season. A portion of the population does not use vehicles at all. For the portion of the US population that travels in a vehicle, the
overall mean daily amount of time spent traveling was 95 minutes or 6.6% of
a day with an additional 78 minutes or 5.4% [4] of the day spent near a vehicle
(Fig. 2).
Differences in the amount of time spent in and near vehicles have been reported by gender, with males spending more time per day in and near vehicles
(mean 103 minutes in and 111 minutes near vehicles) than females (mean 92
minutes in and 50 minutes near vehicles) (Fig. 2).
Adults spend more time in vehicles than children (ages 14: 68 75 min,
511: 71 77 min, 1217: 82 80 min, 1864: 104 111 min, > 64: 91
94 min) (Fig. 2). The differences in the amount of time spent in vehicles: geographically within the US (Northeast 99 107 min, Midwest 101
115 min, South 96 98 min, West 94 104 min), by day of week (weekday
321
Fig. 1 Percent of time for different microenvironments for the US (total population) and
Europe (working population) (Data from: Klepeis et al (12) and De Bruin et al (13))
Fig. 2 (a) Time spent in vehicle by gender and age (b) Amount of time spent in or near
vehicles per day for US population (Data from: Klepeis et al (4))
94 101 min, weekend 103 112 min) and by season (winter 94 101 min,
spring 100 110 min, summer 98 104 min, fall 97 104 min), are not large.
The fraction of time spent traveling is similar in cities in Mexico (10%) [9]
and Europe (7.5%) [14]. The utilization and ownership of cars and other vehicles in many Asian cites are available (e.g. Korea: [15]). Traveling by a car
was more prevalent than in a truck or van, bus, train or subway, or airplane,
322
C.P. Weisel
Fig. 3 Percent of time for different modes of transportation for the US and Europe (Milan)
based on the NHAPS and Expolis data (Fig. 3). Additional activities associated with exposure to vehicle emissions because of close proximity, though
not within an enclosed space, included walking, riding a motorcycle, riding
a bicycle, being on a sidewalk, parking lot, and at a service station (Fig. 4).
The greatest amount of time spent in vehicles was from approximately
6 am through 11 pm, with weekdays displaying a trimodal pattern, with peaks
at 7:30 am, 12:30 pm and 5:00 pm, times when people commute to and from
work or school or go to lunch. On the weekend, the time of day when people
were in a car was more evenly distributed during the daylight hours, though
some skewness toward the latter part of the day was evident. In Europe, maximal amounts of time spent traveling was between 7 am and 7 pm.
The use of public transportation is dependent upon its availability and the
accessibility of automobiles to the residents. Public transportation is most efcient in large urban centers, while greater access to automobiles occurs in
the countries or regions of countries with higher socio-economic strata. As
indicated above, while the automobile is the preferred mode of transportation
Fig. 4 Percent of time spent near roadways by different activities in the US
323
in the US overall, residents of some cities rely upon public transportation.

Within New York City more than 50% of the commuters use the train or bus
to get to and from work as do between 20 and 40% of commuters in six
other cities with populations in excess of 250 000 (Source: U.S. Census Bureau,
2002 American Community Survey ). However, many other large cities in the
US, such as Los Angeles, rely more on private automobiles. Highly-populated
cities in Europe, Asia and Central/South America also have large numbers
of commuters using public transportation or shared vehicles operated by independent owners. The motorcycle and bicycle are also popular modes of
transportation in Asian urban settings. For example in Taiwan, more than
twice as many motorcycles as cars are registered, approximately one for every
two people [16]. While motorcycles and bicycles are not enclosed environments, their proximity to trafc results in potential high exposure during
commuting. In rural areas of developing countries the percentage of time
spent indoors and outdoors and in vehicles differs from what was determined
for the US and Europe, with time spent outdoors much greater and in vehicles
much lower (e.g. [8, 10]), due to the low number of automobiles available on
a population basis.
4
Concentration of Pollutants Within Motor Vehicles
4.1
Carbon Monoxide
The rst set of studies to determine the concentrations of pollutants within
the cabins of motor vehicles was done for carbon monoxide on automobiles
in the US [1720]. Carbon monoxide (CO) was studied due to the prevalence of automobiles as the preferred mode of transportation in the US and
the availability of real time monitors for carbon monoxide that could measure levels present in roadways. Researchers in Europe, Mexico and Asia have
since focused on exposure within public transport vehicles and other pollutants. Since cars drive near other cars, trucks and buses, the contaminants
in the interior air of automobiles is primarily associated with penetration of
contaminants present in roadway air.
In the initial studies, carbon monoxide was used as a marker of automotive exhaust emissions to determine the population-based exposure in
a number of urban settings [17, 18, 2022]. The frequency distribution of
Colorado exposures representing nonsmoking urban populations was determined in Washington, D.C. and Denver, Colorado using personal exposure
monitors [17]. Since these measurements were made more than thirty years
ago, emission controls and changes in fuel composition have been implemented in the US that have decreased the emissions of CO and other pol-
324
C.P. Weisel
lutants from the tailpipes of cars. Thus, the CO levels that individuals are
exposed to today in the US are expected to be lower. However, in countries
where many older cars or cars with less stringent emission controls are in use,
the CO concentrations reported in these initial studies may still be applicable. More than 10% of Denver residents and 4% of Washington, D.C. residents
were exposed to CO concentrations exceeding 9 ppm for 8 hours during the
winter of 19821983. The mean CO exposure levels measured with personal
monitors were similar to the mean concentrations measured at the monitoring site, with the highest CO values in both Washington, D.C. and Denver, but
the 90th and 98th percentile concentrations measured with personal monitors
for the exposed population were higher than the concentrations measured
at the ambient monitoring sites [17]. This implies that peak exposures to
individuals occurred within the cabin of the automobile and that ambient
monitoring stations would underestimate the concentrations associated with
those exposures.
A second approach used to understand CO levels within the cabin of a vehicle was to measure CO concentrations within test vehicles driven by staff
along predetermined routes, rather than measuring the concentration of the
general population while they traveled on their usual commute. In Washington, D.C. eight prescribed automotive commuter routes, four bus routes and
three rail routes were evaluated using a microenvironmental study designed
to assess commuter exposures [20]. The routes represented major commuter
routes of the city. A series of different parameters: street trafc density, parking garage use, presence of roadway tunnels, street canyons and expressways
were ascertained to determine which could affect roadway CO levels. Parking in indoor garages resulted in higher CO levels on the return trip than the
morning trip. This was explained by the residual CO that penetrated the car
while it was parked in an indoor parking garage, where concentrations build
up during the day due to exhaust emissions into an enclosed area, and remained in the cars interior during a portion of the trip home. An inverse
relationship between driving speed and CO concentration was also identied.
Driving speed is a controller of the air exchange rate between the cars interior and surrounding roadway air. Driving speed could also be an indicator
of trafc density. As the automobile left the garage and the city center, it entered less congested areas that would have lower roadway CO concentrations
whereby higher air exchange rates would result in the cars interior CO concentrations decreasing as the trip progressed. The average CO concentration
for automotive commuters in Washington, D.C. was 9 to 14 ppm during a 40
to 60 minute period, twice a day. A similar sampling approach was recently
used in Athens, Greece for a variety of transportation modes and for cars
traveling in a range of trafc densities [23]. The mean CO cabin concentrations for cars measured in that study was 21 ppm (range 14.640 ppm).
Multiple modes of transportation have been examined in a number of
locations. The mean CO concentrations measured on two days in the four
325
microenvironments in US cities designated as transportation modes were:

automobiles (7.6/4.4 ppm), buses (8.2/10 ppm), trucks (6.8/7.7 ppm) and
motorcycles (13.2/11.1 ppm) [17]. These concentrations exceeded the values
measured in the other 18 microenvironments examined, except for four high
exposure microenvironments which were locations frequented by automobiles (parking garages, service stations, within 10 feet of a road) and a manufacturing facility. Differential CO concentrations were measured in Athens
by type of vehicle and season (winter/summer) [23]. The mean concentrations were highest in the winter when the combustion efciency of motors is
lower. The mean and peak winter CO levels for the different transportation
modes were: automobile 21/50 ppm, bus 10/26 ppm, trolley 9.6/18 ppm, rail
4/5.7 ppm and pedestrian 12/19 ppm. The exposures for these routes, when
the time spent traveling among the different transportation modes along
with the time spent waiting for the public vehicles to arrive were incorporated, were: car 10 ppm-hours, bus 5 ppm-hours, trolley 5 ppm-hours, train
1.5 ppm-hours; and pedestrian 5.5 ppm-hours. Comparison of the CO concentration across ve different cities during the 1990s showed differentials of
nearly an order of magnitude (Fig. 5) [13, 23].
Ott and co-workers (1981, 1993) evaluated CO levels inside automobiles
driven along a single prescribed route along a Californian Arterial Highway during two different decades. The original measurements were made
in 19801981 and a second set in 19911992 to evaluate long-term temporal variations [24]. Detailed information was collected about the driving
conditions, trafc and the condition of the car being driven to assess fac-
Fig. 5 Carbon monoxide concentration for ve different microenvironments associated

with mobile sources (car interior, bus interior, trolley interior, train interior, pedestrian
near roadway) in ve cities during the 1990s. (Source: Duci et al. 2003 Summary of Four
Cities, Bruiner de Bruin 2004 Milan)
326
C.P. Weisel
tors that could control CO concentration in cars. The Californian in-car

CO concentrations were regressed against a series of roadway, automotive
and meteorological parameters. The regression model indicated that the surrounding vehicle counts and season were predictive of the automotive cabin
CO concentration. The known decline in CO exhaust emissions from automobiles due to regulations implemented between 1981 and 1991, when the two
sets of measurements were made, resulted in a corresponding decline in the
magnitude of CO exposure that occurred while driving in cars to a greater
extent than had been observed in the ambient CO air concentrations. One
explanation for the stronger association observed between declining emission rates and cabin CO air concentrations is the strong inuence of near
source emissions, the tailpipes of the surrounding automobiles, on the cabin
concentrations and therefore driver exposures to compounds in automotive
emissions. Gomez-Perales et al [25] also found a decline of a factor of three
in CO concentrations in vehicles in Mexico City between 1991 and 2002. Chan
and Chung [26] studied the CO, nitrogen oxide (NO) and nitrogen dioxide
(NO2 ) concentrations inside cars under four driving environments (highway, countryside, urban street and tunnel) and three ventilation conditions
(air-conditioning with air recirculation, air conditioning with fresh air intake and natural ventilation) in Hong Kong. They also determined that the
number of cars surrounding the vehicle was an important determinant of the
concentration both as it related to the roadway type and within a particular roadway. The roadway air was able to penetrate into the cabin under all
ventilation conditions, thereby controlling the concentration. However, there
was a time lag in the CO concentrations between the roadway air and the
in-vehicle air when the ventilation conditions were set to recirculation compared to natural conditions both for increases due to trafc jams and for
decreases when the trafc was owing more freely. As expected, the high air
exchange rate associated with open windows resulted in any rapid change in
the air concentration outside the vehicle also occurring inside the vehicle.
The CO concentration was typically 1 ppm when driving in areas not surrounded by other cars and 5 to 7 ppm when driving within a tunnel and trafc
jams.
While measurements can directly indicate what the concentration and exposure within an automotive cabin is, the results are limited to the exact
set of conditions under which the measurements were made. To extrapolate the concentrations to population exposures and for use in estimating
how proposed changes in emissions may affect concentrations and exposures,
a variety of exposure models have been developed. These models can estimate general population exposures to different environmental contaminants,
including CO and other automobile emissions. Several of the models that include a component associated with commuting and compounds important in
automotive emissions are: Simulation of Human Activity and Pollutant Exposure (SHAPE) [21], Benzene Exposure Assessment Model (BEAM) [27], Na-
327
tional Ambient Air Quality Standard (NAAQS) Exposure Model (NEM) [28],
a probabilistic version of NEM (pNEM) [29], Hazardous Air Pollutant Exposure Model for Mobile Sources (HAPEM-MS) [29], and Benzene Exposure
and Absorbed Dose Simulation (BEADS) [30].
Exposures models have been run and evaluated for CO exposures within
automobiles and/or population distribution for populations in Germany [31],
Massachusetts, US [32], California, US [24, 33], Taipei, Taiwan [34], New
Jersey, [35], Paris, France [36], Mexico City, Mexico [37], Amsterdam, the
Netherlands [38] and Nottingham, United Kingdom [39]. A review of these
studies and other published reports have shown a decline in ambient and
in-vehicle CO levels over the last two decades [40]. Overall, in-vehicle CO
concentrations were elevated compared to ambient air. The values measured
in the US during the 1990s are lower than in other countries, with median
(and 90th percentile) CO concentrations in automobiles driven on suburban and urban commutes of 2.3 (5.9) ppm and 1.9 (6.9) ppm, respectively,
while the average CO concentrations in the urban settings of other countries
are between 7 to 20 ppm for the European cities and Taipei, and 40 ppm for
Mexico City. In-vehicle CO air concentrations in rural regions which have
limited trafc were lower in all studies. Linear regression models of CO exposure concentrations within automobiles based on central monitoring stations
have shown only moderate predictive power [37] with the ambient monitoring stations typically underestimating the exposure levels [34]. Therefore,
extrapolation of ambient air CO concentrations underestimates the exposure
of commuters to automotive derived compounds.
4.2
Automobiles emit volatile organic compounds (VOCs), in particular hydrocarbons, a major component of gasoline and diesel fuels, the two most common fuels for mobile vehicles. Thus, a major focus of recent papers measuring
exposures and concentrations within vehicles have examined these compounds. There are a number of health concerns associated with exposure to
VOCs and the methodologies to sample and analyze or directly measure them
have improved during the past decade. Both evaporative and exhaust emissions contain VOCs. VOC emissions have declined as emission controls for
both evaporative and exhaust emissions have improved and engine efciency
has advanced from carburetor-based to fuel injection engines. The most commonly analyzed VOCs have been benzene and other aromatic compounds,
though some studies have measured alkanes and fuel additives. In general the
vehicles interior concentrations are ve to ten times higher than the ambient
atmosphere for many of these compounds and two to three times higher than
indoor values, though some individual cities can have much higher levels in
the interior of their cars, hundreds of g/m3 (Fig. 6).
328
C.P. Weisel
Initial studies on VOCs were conducted in the US to identify the major factors that inuence the VOC contaminant levels within the automobile. Route
location (urban, suburban or rural) and time of day had the largest inuence on the concentration [35, 41, 42]. These factors were highly related to the
trafc density, that is, the number of cars immediately adjacent to the vehicle in which the sampling was being conducted. This result is similar to that
observed for CO levels in cars [21]. Within a single city, the highest concentrations were observed when cars were driven through a tunnel [35]. VOC air
concentrations in and near tunnels have been shown to be greatly elevated
compared to ambient levels and are a source of automotive-related emissions for the surrounding area in a number of countries [4348]. Seasonality
was found to inuence the relative concentration of the various VOCs within
a tunnel in Korea [48]. The same factors that controlled the CO concentration within automobiles (window position, ventilation and meteorological
conditions) had smaller but determinable inuences on the in-vehicle VOC
concentrations.
Most of the US studies examining VOC concentrations were done by researchers (< 10 different cars per study) driving cars over preselected commuting routes. One large study done in Los Angeles, CA, collected air samples
in the automobiles of 140 employees of a Californian state agency during
their normal commute over two seasons [49]. A large amount of variability
was observed in the VOC concentrations, probably the result of the different
routes followed and the variety of automobiles that were sampled. The latter
Fig. 6 Mean benzene and toluene (examples of aromatic VOC emitted from mobile
sources) concentrations in automobile cabins in cities throughout the world. Ambient
1990 are typical levels measured in the US and Europe. Los Angeles 1989 [49], Boston
1991 [32], North Carolina 1991 [41], Taiwan 1994 [54], New Jersey 1995 [35], Paris
1995 [36], Amsterdam 1995 [38], Korea 1995 [55], Korea 1998 [56], Los Angeles 1998 [50],
Sacramento 1998 [50], Korea 1999 [42], Korea 2002 [57], Hong Kong 2003 [51], China 2003
[58], Sydney 2004 [53]
329
could be very important as poorly maintained or older vehicles could have

problems in the engines increasing engine emissions, or poor body integrity
which could increase the penetration of exhaust emissions into the automobile. Carburetor-powered vehicles and cars with malfunctions could result in
self VOC contamination of an automobiles interior from evaporative emissions of gasoline within the engine compartment. This was observed in New
Jersey where higher levels were observed in a carburetor engine powered vehicle compared to a car with a fuel injected engine [35]. Further, there were
malfunctions of one car in New Jersey (a valve which did not open fully and
a pin hole leak in a hose) which resulted in increased gasoline evaporation
within the engine, signicantly elevating the VOC concentrations in that cars
interior. A second study conducted in California [50] observed lower in cabin
concentrations of selected VOCs and particle mass (PM2.5 ) than reported during the previous decade in Los Angeles [49]. These results indicate that the
emission controls on automobiles and on gasoline have reduced commuters
exposures to VOCs in Los Angeles, similar to what had been reported for CO.
Several studies have examined the VOC concentrations, mainly aromatic
compounds, in automobiles and other forms of transportation in Asian countries and Australia [25, 42, 5153]. As was found in the US studies, driving
route and engine type (carburetor vs. fuel injection) were major factors affecting the concentration in the interior of automobiles. Season, which is associated with fuel composition, and ventilation rate in rural settings, but not
car model nor driving duration, inuenced the VOC concentrations within
the car [42]. The concentrations within automobiles of VOC vary across location and time and are higher than background concentrations in urban
settings.
Individuals who walk or bicycle near roadways or ride motorbikes were exposed to higher VOC levels than measured at the ambient monitoring station
but lower values than measured in automobiles. Concern has also been expressed as to exposure to 1, 3 butadiene, which is present in auto exhaust,
since 1, 3 butadiene has the highest cancer potency of any single compound
present in automotive exhaust [59]. It has been shown to be present in the air
near tunnel exhaust which, like automobile cabins, is highly impacted by automotive emissions [60], but no measurements inside the automobile cabin
have been reported.
One recent VOC gasoline additive used to reduce CO emissions in several
countries, methyl tert butyl ether (MTBE), can be the single largest component (up to 15% volume basis) in many fuel blends. MTBE is mainly released
through evaporative emissions, though it can be in exhaust emissions when
incomplete combustion occurs. MTBE was found to be elevated inside automobiles and buses relative to other microenvironments [61, 62] and can be
particularly high (mg/m3 rather than tens of g/m3 ) during and just after
lling the automobile fuel tank [63]. The VOC air concentrations, particularly benzene and toluene within the compartments of public transportation
330
C.P. Weisel
vehicles, are generally lower than the levels in cars driven along the same
roadways. Values measured in buses were typically one half of the values
measured in cars or taxis, with that measured in trains one half the value
measured in buses [5153, 55, 60, 62, 64, 65], although in some studies the
concentrations had smaller differences across the transportation modes.
4.3
Carbonyl Compounds
The combustion of fuel, which for petroleum is predominantly hydrocarbons, can result in the emissions of partially oxygenated compounds, such
as carbonyl compounds. The addition of ethers to reduce carbon monoxide
emissions or the use of methanol or ethanol based fuels also causes emissions of carbonyl compounds, particularly formaldehyde and acetaldehyde.
Formaldehyde concentrations measured at several cities in the US (Boston,
MA, Los Angeles, CA and Sacramento, CA) in the cabin of automobiles, in
subways, while walking and bicycling, have all been similar to each other (4 to
20 g/ m3 ) [32, 49, 50], and to ambient air levels (19 g/m3 ), but lower than
measured indoors (60 g/m3 ) [66]. Median formaldehyde and acetaldehyde
concentrations in cars in Korea were 24 and 13 g/m3 , respectively, while in
public buses they were 25 and 14 g/m3 , respectively [57].
4.4
Particulate Matter
Mobile sources of particle matter include diesel engines and to a smaller extent gasoline engines, abrasion of engine parts, brakes and tires. Automobiles
are primarily gasoline powered, whereas buses and trucks generally justify
the greater economy and power of diesel engines. Diesel particles are primarily in the respirable size range (PM3.5 -particles with diameters of less than
3.5 micron). Abrasion processes produce larger size particles.
In recent studies, particle mass loadings in the cabins of vehicles
driven along urban routes in Madison, WI were found to be 105
30 g/m3 [67], in Sacramento, CA were 622 g/m3 , and Los Angeles, CA
were 29107 g/m3 [50]. These levels were lower than roadway air concentrations due to removal of particles by the ventilation system of the cars.
The particle levels increased with trafc density. Real-time measurements
of PAHs on particles smaller than 1 m in cars driven in the US (North
Carolina and California) were elevated compared to ambient levels, with baseline levels of PAHs in automobiles of < 200 ng/m3 and short term spikes of
> 1000 ng/m3 [68]. Adams et al. [69] summarized the literature on exposure values of particulate matter in a series of microenvironments related to
mobile sources. They reported that different studies used particle size cutoff
varying from 2.5 to 10 microns, with some studies not providing information
331
on size cutoff but rather that the particles were analyzed by UV absorbance or
as black smoke. These analysis methods primarily measure elemental carbon
whose principal source is from combustion which generates small particles.
Large variations were observed among the results collected in the cities summarized by Adams et al. [69], with samples collected in Europe and the US
being in the tens of g/m3 while Delhi, India had levels of 389 249 g/m3 .
PM concentrations in buses were similar to or higher than those found in cars
in the same cities. Underground trains appeared to have concentrations in the
hundreds of g/m3 in several studies. A recent study of PM2.5 and PM10 levels
in a variety of public transportation modes in Hong Kong observed concentrations of tens to almost 200 g/m3 in buses and trams [58]. Lower PM levels
were identied in Hong Kong in vehicles that were air conditioned (ac) vs.
non-air conditioned (non-ac) vehicles which used natural ventilation (i.e. had
their windows opened). The higher ventilation rates in the non-ac vehicles
resulted in more of the PM emitted from surrounded vehicles entering the
vehicles. Lower PM levels were also identied on the upper deck of double
decker trams.
One group of individuals who may be exposed to PM from buses are
children who take school buses [70]. The buses frequently stop to pick up
and discharge passengers which may result in self entrainment of PM in the
buss interior. Further, the large number of buses idling at single pickup and
drop-off points, e.g. the school, may result in school children having elevated
exposures to diesel emissions while they congregate around the buses.
Specic fuel additives can also be present in PM emitted in the exhaust.
Lead associated with gasoline emissions, where tetraethyl lead is used as an
octane enhancer and anti-knock agent, is a major source of lead exposure.
This exposure is not limited to the enclosed space of the automobile cabin.
While lead is no longer added to gasoline in many countries, its use in any
country is a major potential health concern to children. A second major additive, methylcyclopentadienyl manganese tricarbonyl (MMT) can mobilize
manganese into the air. Elevated levels of particulate manganese were measured for taxi cab drivers compared to ofce workers in Toronto, Canada,
indicating that in-cabin levels of particulate manganese are elevated compared to ambient and indoor air when MMT is used as an additive in fuel [71].
Platinum, which could originate from degradation of the catalytic converter
used to reduced exhaust emissions from cars, was present in air samples collected in buses in Munich, Germany, with the highest value of 33 pg/m3 on
the route with the greatest trafc density [72].
332
C.P. Weisel
5
Conclusion
Elevated exposures to air pollutants (CO, particulate matter and volatile organic compounds) emitted from mobile sources occur within the enclosed
spaces of private and public transportation compared to other locations. The
magnitude of those exposures is a function of the trafc density, the utilization of pollution controls and the operating conditions of the surrounding
vehicles, as well as the vehicle being driven. The duration and manner of the
commute varies by location and individual resulting in large differences in exposure with major differences in the primary mode of transportation used
and the levels observed between developed and developing countries. Estimates of total exposure to pollutants emitted from mobile sources need to
consider time spent within vehicles.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Sawyer RF (1993) Environ Health Perspect 101(6):5

Bureau USC. http://wwwcensusgov/acs/www/
World Health Organization (2000) WHO Reg Publ Eur Ser:V
Klepeis N, Tsang AM, Behar JV (1996) Analysis of the National Human Activity Pattern Survey (NHAPS) Respondents from a Standpoint of Exposure Assessment. US
EPA, Las Vegas, NV
Tsang AM, Klepeis NE (1996) Descriptive Statistics Tables from a Detailed Analysis of
the Nation Human Activity Pattern Survey (NHAPS) Data. US EPA, Las Vegas, NV
Jantunen MJ, Hanninen O, Katsouyanni K, Knoppel H, Kuenzli N, Lebret E, Maroni M,
Saarela K, Sram R, Zmirou D (1998) J Expo Anal Environ Epidemiol 8:495
Leech JA, Wilby K, McMullen E (1999) Can J Public Health 90:244
Balakrishnan K, Sambandam S, Ramaswamy P, Mehta S, Smith KR (2004) J Expo Anal
Environ Epidemiol 14:S14
Ramirez-Aguilar M, Cicero-Fernandez P, Winer AM, Romieu I, Meneses-Gonzalez F,
Hernandez-Avila M (2002) J Air Waste Manage 52:50
Muller E, Diab RD, Binedell M, Hounsome R (2003) Atmos Environ 37:2015
Netherlands S http://wwwcbsnl/en/service/links/defaultasp
Klepeis NE, Nelson WC, Ott WR, Robinson JP, Tsang AM, Switzer P, Behar JV, Hern SC,
Engelmann WH (2001) J Expo Anal Environ Epidemiol 11:231
De Bruin YB, Carrer P, Jantunen M, Hanninen O, Di Marco GS, Kephalopoulos S, Cavallo D, Maroni M (2004) J Expo Anal Environ Epidemiol 14:312
Kousa A, Kukkonen J, Karppinen A, Aarnio P, Koskentalo T (2002) Atmos Environ
36:2109
Figures T: http://wwwinfogovhk/td/eng/transport/tf_menu_indexhtml
Statistics BA. http://wwwdgbasgovtw/english/dgbas-e0htm
Akland G, Hartwell T, Johnson T, Whitmore R (1985) Environ Sci Tech 19:911
Cortese A, Spengler JD (1976) J Air Pollution Control Assoc 26:1144
Spengler J, Burge H, Dumyahn T, Muilenber M, Forester D (1997) Environmental Survey on Aircraft and Ground-Based Commercial Transportation Vehicles. Harvard
School of Public Health for the Commercial Airplane Group, Boston, MA
333
20. Flachsbart PG, Mack GA, Howes JE, Rodes CE (1987) Japca 37:135
21. Ott W, Thomas J, Mage D, Wallace L (1988) Atmos Environ 22:2102
22. Ott W, Willits N (1981) CO exposure of Occupants of Motor Vehicles: Modeling the
Dynamic Response of the Vehicle Use. University UES, Stanford, CA
23. Duci A, Chaloulakou A, Spyrellis N (2003) Sci Total Environ 309:47
24. Ott W, Switzer P, Willits N, Hildemann L, Yu Y (1993) Trends of in-vehicle CO exposures on a California arterial highway over one decade. In: Proc 86th Annual Meeting
of the Air and Waste Management Association, Denver, CO
25. Gomez-Perales JE, Colvile RN, Nieuwenhuijsen MJ, Fernandez-Bremauntz A, Gutierrez- Avedoy VJ, Paramo-Figueroa VH, Blanco-Jimenez S, Bueno-Lopez E, Mandujano F, Bernabe-Cabanillas R, Ortiz-Segovia E (2004) Atmos Environ 38:1219
26. Chan AT, Chung MW (2003) Atmos Environ 37:3795
27. BEAM (1992) BEAM Users Guide. Environmental Monitorying System Laboratory,
Ofce of Research and Development, US EPA, Las Vegas, NV
28. McCurdy T (1995) J Expo Anal Environ Epidemiol 5:533
29. Johnson TR (1995) J Expo Anal Environ Epidemiol 5:551
30. Macintosh DL, Xue JP, Ozkaynak H, Spengler JD, Ryan PB (1995) J Expo Anal Environ
Epidemiol 5:375
31. Rudolf W (1990) Sci Total Environ 93:263
32. Chan CC, Spengler JD, Ozkaynak H, Lefkopoulou M (1991) J Air Waste Manage Assoc
41:1594
33. Ott W, Switzer P, Willits N (1994) Air Waste 44:1010
34. Liu JJ, Chan CC, Jeng FT (1994) Atmos Environ 28:2361
35. Lawryk NJ, Lioy PJ, Weisel CP (1995) J Expo Anal Environ Epidemiol 5:511
36. Dor F, Lemoullec Y, Festy B (1995) J Air Waste Manage 45:103
37. Fernandez-Bremauntz AA, Ashmore MR (1995) J Exposure Anal Environ Epidemiol
5:497
38. Vanwijnen JH, Verhoeff AP, Jans HWA, Vanbruggen M (1995) Int Arch Occ Env Hea
67:187
39. Clifford MJ, Clarke R, Riffat SB (1997) Atmos Environ 31:1003
40. Flachsbart PG (1995) J Expo Anal Env Epidemiol 5:473
41. Chan C, Ozkaynak H, Spengler J, Sheldon L (1991) Environ Sci Technol 25:964
42. Jo WK, Park KH (1999) Atmos Environ 33:409
43. De Fre R, Bruynseraede P, Kretzschmar JG (1994) Environ Health Perspect 102(4):31
44. Tsai JH, Liang CP, Lee DZ, Sheu YC, Lin SJ (1997) J Environ Eng 123:406
45. Fraser MP, Cass GR, Simoneit BRT (1998) Environ Sci Technol 32:2051
46. Touaty M, Bonsang B (2000) Atmos Environ 34:985
47. Hsu YC, Tsai JH, Chen HW, Lin WY (2001) Chemosphere 42:227
48. Na K, Kim YP, Moon KC (2002) Atmos Environ 36:1969
49. SCAQMD (1989) In-vehicle Characterization Study in the South Coast Air Basin.
South Coast Air Quality Management District, Los Angeles, CA
50. Rodes C, Sheldon L, Whitaker D, Clayton A, Fitzgerald K, Flannagan J (1998) Measuring Concentrations of Selected Air Pollutants Inside California Vehicles. California Air
Resources Board, Sacramento, CA, pp 95339
51. Lau WL, Chan LY (2003) Sci Total Environ 308:143
52. Chan LY, Lau WL, Wang XM, Tang JH (2003) Environ Int 29:429
53. Chertok M, Voukelatos A, Sheppeard V, Rissel C (2004) Comparison of BTEX and NO2
exposure levels for commuting microenvironment in Sydney car, train, bus, bicycle
and walking. National Clean Air Conference, Clean Air Society of Australia and New
Zealand, pp 2327
334
C.P. Weisel
54.
55.
56.
57.
58.
59.
60.
61.
Chan CC, Lin SH, Her GR (1994) Atmos Environ 28:2351

Jo WK, Choi SJ (1996) J Air Waste Manage Assoc 46:749
Jo WK, Park KH (1998) Environ Int 24:259
Jo WK, Lee JW (2002) Environ Res 88:44
Chan LY, Lau WL, Lee SC, Chan CY (2002) Atmos Environ 36:3363
Ye Y, Galbally IE, Weeks IA (1997) Atmos Environ 31:1157
Duffy BL, Nelson PF (1996) Atmos Environ 30:2759
Lioy PJ, Weisel CP, Jo WK, Pellizzari E, Raymer JH (1994) J Expo Anal Env Epidemiol
4:427
Lee JW, Jo WK (2002) Sci Total Environ 291:219
Vayghani SA, Weisel C (1999) J Expo Anal Environ Epidemiol 9:261
Barrefors G, Petersson G (1996) Environ Technol 17:643
Jo WK, Yu CH (2001) Environ Res 86:66
Zhang JF, He QC, Lioy PJ (1994) Environ Sci Technol 28:146
Ptak TJ, Fallon SL (1994) Part Sci Technol 12:313
Buckley T, Ott W (1996) Demonstration of real-time measurements of PAH and CO
to estimate in-vehicle exposure and identify sources. In: Proc Int Specialty Conference on Measurement of Toxic and Related Air Pollutants, Research Triangle Park,
NC, USA. Air & Waste Management Association and the US EPA, pp 803810
Adams HS, Nieuwenhuijsen MJ, Colvile RN, McMullen MAS, Khandelwal P (2001) Sci
Total Environ 279:29
Wu CC, Suarez AE, Lin ZB, Kidwell CB, Borgoul PV, Caffrey PF, Ondov JM, Sattler B
(1998) Atmos Environ 32:1911
Zayed J, Mikhail M, Loranger S, Kennedy G, LEsperance G (1996) Am Ind Hyg Assoc
J 57:376
Schierl R, Fruhmann G (1996) Sci Total Environ 182:21
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.

DOI 10.1007/b107251
Indoor Air Quality on Passenger Ships

A.D. Webster1 (u) G.L. Reynolds2
1 Environment
Health & Safety Department, Amerada Hess Corporation,

1185 Avenue of the Americas, New York, NY 10128, USA
awebster@hess.com
2 Lloyds Register, 71 Fenchurch Street, London EC3M 4BS, UK
gill.reynolds@lr.org
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
336
Indoor Air Pollution and Health Effects . . . . . . . . . . . . . . . . . . .
337
3
3.1
3.2
Indoor Air Quality on Ships . . . . . . . . . . . . . . . . . . . . . . . . . .

Shipboard Air Quality Assessments . . . . . . . . . . . . . . . . . . . . . .
Air Quality and Comfort Criteria . . . . . . . . . . . . . . . . . . . . . . .
337
339
345
Remedial Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
346
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
348
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
348
Abstract Generally, shipboard air quality problems are not severe in comparison to those
in many homes and ofces. However, a number of problems have been identied which
warrant attention in order to safeguard passenger comfort and health. The HVAC systems
of some ships, particularly older vessels, are prone to microbial contamination. A lack of
awareness of the potential problems at the design stage and subsequent lack of maintenance has allowed moulds and fungi to inltrate the HVAC system where these collect
and grow. These microbes not only pose an immediate risk of respiratory illness and allergic responses, they are also a nuisance with respect to the general maintenance and
cleaning of vessels and hence are often a cause of complaint. Airborne microbial sampling has been used extensively to determine the potential for passenger exposure to this
type of contamination. Due to the complexities of shipboard HVAC design, proper ltration offers the best method of keeping the system clean and preventing the accumulation
of microbial contamination. Ventilation efciency is a problem on some vessels, especially in smaller cabins or densely occupied communal areas. Thermal comfort is also
a source of complaint which requires attention on many ships. Proactive monitoring to
evaluate indoor air quality and identify remedial measures reduces the likelihood of problems developing. Cost-effective improvements can be made to the design and operation
of ventilation systems that reduce contamination and improve air quality. It is apparent
that many of the problems encountered could have been designed out. In particular, attention should be paid to the prevention of ingress of water into the supply and exhaust
systems, thereby restricting the potential for microbial proliferation. Provision of better
access for inspection and cleaning of ductwork would also be benecial.
Keywords Indoor air quality Ships HVAC Microbial contamination
336
A.D. Webster G.L. Reynolds
Abbreviations
ASHRAE American Society of Heating, Refrigerating, and Air Conditioning Engineers
CEC
Commission of the European Communities
CO
Carbon monoxide
Carbon dioxide
CO2
ETS
HVAC Heating, ventilation, and air conditioning
Oxides of nitrogen
NOx
PAH
Polycyclic aromatic hydrocarbons
SOLAS Safety of Life at Sea Convention
SO2
Sulphur dioxide
VOC
Volatile organic compound
1
Introduction
Ill health, allergenic reaction, discomfort and nuisance may result from poor
air quality and badly maintained ventilation systems. Poor indoor air quality
may also affect the revenue potential of a passenger ship. Operators need to be
aware of the importance of proper ltration and ductwork cleanliness as well
as the benets of regular indoor air quality investigations in order to assist in
early identication of potential problems.
There are three overriding considerations for the ship owner and shipyard and the heating, ventilation, and air conditioning (HVAC) manufacturer
regarding the design and operation of an HVAC system:
Comfort,
Safety,
Economy.
The HVAC system needs to provide good-quality air which is dust and odour
free with a minimum of noise and draught and which is able to maintain thermal comfort under a wide range of operating conditions. These requirements
have to be balanced against SOLAS re safety considerations and the operators desire for low power consumption, low equipment space and weight, low
maintenance costs, and simplicity in design and operation.
A major priority for passenger ship designers is to gain space for revenue earning by increasing the number of cabins or space in public rooms;
thus fan rooms and HVAC systems often have a low standing. However, it
has been suggested that a well-functioning HVAC plant producing a good
indoor climate will positively inuence revenue from passengers, indirectly
through increased spending induced by a pleasant atmosphere and directly
by a higher repeat booking rate. In addition, increased operational costs resulting from poor maintenance procedures will also affect revenue in the long
term.
337
2
Indoor Air Pollution and Health Effects
Indoor air quality problems have always existed, for example in the form of
mould, condensation, and combustion by-products. However, in recent years
awareness of indoor pollution problems has increased [1]. The recent apparent deterioration in indoor air quality is largely attributed to:
The implementation of energy conservation measures leading to a reduction in ventilation;
The introduction of an increasing number of synthetic materials from
which a release or evaporation of VOCs may occur.
The range of substances which may pollute the internal environment is diverse. Sources may be of biological as well as man-made origin and may
originate in the outdoor environment or be produced internally. The invisible
nature and lack of identiable odour of many substances commonly regarded
as indoor air pollutants has led to a lack of public awareness of their existence.
The more important sources of these pollutants are: outdoor air, building materials, interior furnishings, human activity, and biological contamination by
living organisms (Table 1). Many individual pollutants, in particular those
associated with combustion and the myriad of volatile organic compounds
(VOCs), derive from a range of sources. Pollutants with specic sources such
as asbestos are comparatively rare.
Similar types of health effects are attributed to many indoor pollutants.
However, the severity of the attributed health risks will vary. Effects range
from mild irritation of nasal and mucous membranes to toxic and carcinogenic effects, although in only a few cases, such as acute allergenic reactions
or carbon monoxide poisoning, is there a clear-cut relation between a pollutant and an associated health effect.
Concentrations, and consequently exposure to airborne pollutants, vary
widely. Furthermore, the levels of many pollutants are likely to change constantly due to the intermittent nature of their production, for example those
associated with environmental tobacco smoke (ETS), gas cooking, and vehicular emissions. Concentrations of pollutants indoors will depend primarily
upon the rate of entry or emission of the pollutant, the degree of ventilation, and the existence of a variety of decay processes including absorption
of gases and vapours by furnishings, interaction with other airborne species,
and deposition of particulate matter. Pollutants measured in the highest concentrations are usually those originating indoors. Air exchange with outdoor
air will subsequently serve to dilute indoor sources provided the outdoor air
is not more highly contaminated.
338
Table 1 Summary of common pollutants and their sources in the indoor environment
(After [1])
Sources
Biological
Metabolic processes
Pollen, moulds, fungi,
algae, insects, animal dander,
house dust mites
Micro-organisms
Building materials & interior furnishings
Chipboard, plywood
Adhesives
Fire retardent materials,
insulation
Furniture, fabrics, wallpaper,
carpeting
Paints
Human activity
Cooking, heating
Washing, cleaning,
personal hygiene
Walking, dusting,
vacuum cleaning
Tobacco smoking
Pest control
HVAC cleaning
Outside air
Motor vehicles
Industry, commerce,
power stations
Pollutants
CO2 , water vapour, particulate matter,

odorous organic compounds
Allergenic compounds
Pathogenic viruses, bacteria, fungi

Aldehydes (i.e. formaldehyde)
Solvents, aldehydes
Mineral bres (e.g. asbestos)
Organic compounds
Formaldehyde
Solvents, heavy metals
NOx , CO, CO2 , SO2 , particulate matter,
formaldehyde, hydrocarbons, PAH
Water vapour, VOC
Particulate matter (resuspension of)
ETS (CO, NOx , ne particles, VOC)
Chlorinated organic compounds
Biocide
NOx , CO, particulate matter, VOC, PAH
NOx , CO, SO2 , particulate matter,
VOC, PAH
3
Indoor Air Quality on Ships
Although a signicant body of data and expertise is available for land-based
environments, ships pose unique problems with respect to indoor air quality.
The type and scale of problems encountered are dependent upon the:
Type of vessel,
Age of vessel,
339
Area of operation climate, ambient air quality, etc.,

Period and density of passenger occupation, and
HVAC system design and maintenance.
3.1
Shipboard Air Quality Assessments
To address the lack of data concerning shipboard air quality, LR [2] conducted a major assessment of indoor air quality onboard a wide cross-section
of vessel types including cruise ships, passenger ferries, and passengercarrying cargo vessels. Ofce buildings and residential premises were also
investigated for comparative purposes. An extensive range of pollutants were
initially monitored (Table 2); however, as investigations evolved assessments
concentrated on a much narrower range of key pollutants utilising more
accurate and discrete samplers to determine pollutant concentrations during typical periods of occupancy. Wherever possible, standard investigation
methodologies were employed but modied to allow monitoring to be carried out without needlessly attracting the attention of passengers. Inspection
of HVAC systems and their operational and maintenance procedures was an
integral part of the assessment since lter efciency and ductwork cleanliness
are key factors in determining indoor air quality.
Overall, these assessments found that pollutant sources are generally more
limited on board passenger ships as compared to land-based locations such
as homes and ofces. The lack of combustion sources in passenger spaces
limits the range of pollutants considerably. Concentrations of the products of
combustion oxides of nitrogen (NOx ), and carbon monoxide (CO) are usually very low except in those cases on vehicle ferries where the emissions from
cars and lorries during loading or disembarkation can disperse up and down
stairways from vehicle decks.
Carbon dioxide (CO2 ) concentration is a key parameter for assessing indoor air quality and ventilation efciency. CO2 is a component of air with outdoor concentrations typically ranging from 360 to around 500 ppm. Indoor
concentrations range from 360 ppm to 3000 ppm and above, with these high
CO2 concentrations attributed to the respiration of human occupants [3]. Although not harmful itself at these concentrations, CO2 acts as a surrogate
indicator of ventilation efciency as well as particulates, bioaerosols, and
other pollutants resulting from human occupancy. Above 1000 ppm symptoms such as drowsiness, perceived stufness, inability to concentrate, and
odours increase. This, however, is due to the accumulation of a range of indoor pollutants rather than CO2 itself. More recent studies by the U.S. Air
Force suggest that effects may begin at concentrations as low as 600 ppm [20].
CO2 is now regularly used in investigations as a basic indicator of indoor
air quality and to assess ventilation rates. Often occupant-generated CO2 can
be used as a tracer to determine the air exchange rate (Fig. 1).
340
Table 2 Monitoring methodologies in Lloyds Registers shipboard monitoring programme [2]

Parameter
Instrument/Methodology
Carbon dioxide
Carbon monoxide
Nitrogen dioxide
Volatile organic
compounds
Formaldehyde
Non-dispersive infrared or electrochemical

Continuous Non-dispersive infrared
Diffusion tube
Photoacoustic instrumentation or
adsorption tubes, gas chromatography
2,4-DNPH sampler high-performance
liquid chromatography
Adsorption tubes
Gravimetric analysis
Impaction/laboratory enumeration and
identication
Filter collection/phase microscopy
Filter collection/adsorption tube
Random water samples/specialist
Microbiological analysis
Thermometer, hygrometer, anemometer
CO2 or peruorocarbon tracer/
Gas chromatography/
Electron-capture detection
Photography, endoscopy
Odours:
Particulate matter
Airborne micro-organisms
Airborne asbestos
polycyclic aromatic hydrocarbons
Legionella
Thermal comfort:
Air exchange rate
Ventilation rate
ductwork cleanliness
Assessments by Lloyds Register [2] found high CO2 concentrations and

poor air exchange rates in the passenger cabins of several vessels, especially
where local supply and/or exhaust units were switched off or were blocked.
Generally, cabin air exchange rates were found to be worse on passenger ferries. Conversely, air exchange rates in the public rooms of ferries were better
than those on cruise ships.
Air exchange rates have also been determined by Lloyds Register using the
peruorocarbon tracer method [21]. This method is used when CO2 concentrations exceed 1000 ppm for extended periods, suggesting that there may be
a problem with ventilation efciency. However, it can only be used in those
situations where air is not recirculated. This method has been used to verify
air exchange rates on a number of vessels and has proved to be very sensitive.
Elevated CO2 concentrations are also indicative of potential odour problems. However, major odour problems are usually attributed to a specic
source which is often difcult to determine.
VOCs were initially envisaged as a major problem on board passenger
ships due to the relatively high amount of cleaning, redecorating, and refur-
341
Fig. 1 Simultaneous monitoring of carbon dioxide concentrations, temperature and relative humidity at three locations on a passenger ship [2]
bishing that occurs. However, comparisons of shipboard monitoring [2] with

VOC concentrations in homes and ofces [4] suggest that ships are relatively
free of this type of pollution (Fig. 2). This is believed to be due to removal
of these vapours by the mechanical ventilation systems aboard ships. Many
onshore indoor environments such as homes and ofces do not have mechanical ventilation systems, and often there is greater emphasis on saving
342
Fig. 2 Total volatile organic compound concentrations on four vessels, an ofce building
and in the 100 Homes survey [2, 4]
energy, thus reducing air exchange rates. These reduced air exchange rates
allow pollutant concentrations to build up, increasing exposure.
High VOC concentrations were typically associated with localised external
sources such as reneries or passenger activity (e.g. perfume use). In addition, a few peak concentrations were attributed to redecoration; however,
these elevated concentrations appeared to decay over a period of a few days.
The results of extensive formaldehyde measurements reveal a similar story. As
Fig. 3 demonstrates, apart from isolated peaks, concentrations were low when
compared to homes and ofces.
Particulate concentrations on ships are relatively high when compared to
homes and ofces. However, high concentrations are generally attributed to
passengers smoking or to areas where resuspension of settled dust occurs (for
example heavily used passageways). It seems that high air exchange rates and
the associated high airow in some spaces cause particulates to remain airborne for much longer, resulting in higher measured suspended particulate
concentrations. However, the high air exchange rate in these locations means
that the concentrations of other pollutant parameters were generally much
lower than in land-based areas.
It is likely that a signicant proportion of particulate matter is of biological
origin and consists of airborne spores, moulds, and fungi. Indeed shipboard
HVAC systems seem to be extremely prone to contamination by moulds and
fungi. This mould typically appears in the form of black/brown deposits
which line the walls of the ductwork and accumulate in low-pressure areas
343
Fig. 3 Formaldehyde concentrations on four vessels, an ofce building and a residential

survey [2, 4]
(Fig. 4). On certain ships this contamination has been found to completely
clog ducts and reheat units on a regular basis.
Spores of moulds and fungi gain entry to ships either directly via open
doors and windows or alternatively through the inlets to the HVAC system.
The inlet air handling units of passenger ships usually incorporate some form
of ltration system which prevents the ingress of particulate matter, which
may include spores. Unfortunately, very often, dirty and ill-tting lters allow
dust and spores to pass into the HVAC ductwork unimpeded (Fig. 5).
Most microbes require moisture, organic nutrients, and warm temperatures in order to grow. Once the moulds have gained access to the ship, it
is water that is the main factor limiting growth. Therefore, microbial growth
occurs in those parts of the system prone to moisture incursion or condensation. Following a period of growth, spores are distributed further into the
ductwork and eventually may spread throughout the ship. Once this microbial
growth has established in the HVAC system it is very difcult to remove.
An indication of lter efciency and the level of contamination can be
gained by sampling the air for viable micro-organisms. A count of the viable airborne particles from samples taken both inside and outside a ship
can then be made. Where the mould and fungi colony counts from inside
the ship are more than three times greater than outside, further investigation
is prompted, as this suggests that the source of the contamination is present
within the ship. High counts both inside and outside the ship suggest that
344
Fig. 4 Dirt and mould contamination removed from HVAC ductwork
Fig. 5 Broken air handling unit lter frames and gaps in lter material
lters are poorly maintained and that spores are passing through the HVAC
system unimpeded. Inspection of the air handling units and ductwork usually
conrms these ndings [2].
Many of the most prolic types of mould found to be present, such as Cladosporium and Penicillium, are not particularly harmful to human health.
However, these species may play host to other more virulent species of fungi,
bacteria, and even viruses which may be of much greater risk to health [5].
For instance Aspergillus colonies are found in samples containing high Cladosporium counts. Aspergillus spores may cause a variety of health effects
345
which include invasive aspergillosis, allergic bronchopulmonary aspergillosis, and hypersensitivity pneumonitis [6, 7].
Such problems are not conned to passenger ships and have been identied in other types of inspected vessels. This suggests that people are not
the principal source of fungal contamination. Clearly microbial contamination and ventilation system hygiene are important issues for ship operators,
and future ship design needs to take account of these operational problems.
The main considerations are prevention of ingress of water into the supply
and exhaust systems and provision for inspection and cleaning of ductwork
throughout the life of the ship.
Bacteria from external sources are much less of a problem on ships due to
the sterilising effect of sea water and sea spray. Within a ship people tend to be
the main source of bacteria, and therefore the level of contamination is largely
dependent upon the hygiene and behaviour of the occupants.
In recent years there has been much concern regarding the risk of legionella aboard passenger ships. Lloyds Register [2] conducted extensive
sampling of water systems onboard passenger ships, and no legionella contamination was detected. However, the design of ship HVAC systems often
precludes sampling due to an insufcient volume of available water.
Opinion varies as to whether legionella contamination can occur in a ships
air conditioning system. Of more immediate concern is legionella infection
from spa pools or the potable water supply. Many authorities [8] recommend
rigorous cleaning and maintenance aimed at prevention. As a precautionary
measure it is prudent to conduct regular sampling of the potable water system
as well as bathing waters and any standing water in the HVAC system. However, it is important to remember that the absence of legionella in random
samples does not guarantee its absence from other parts of the system. Indeed detailed microbiological analysis suggests that conditions in the supply
system are conducive to legionella proliferation.
3.2
Air Quality and Comfort Criteria
The key guidance concerning indoor air quality is that of ASHRAE relating to
thermal comfort [9] and Ventilation for Acceptable Indoor Air Quality [3]. In
addition, the international standard on air conditioning and ventilation design in the passenger accommodation of ships makes recommendations with
respect to temperature, relative humidity, supply, and exhaust air [10].
These and additional criteria have contributed to the guidelines used in the
evaluation of air quality aboard ships (Table 3). The criteria are designed both
to protect human health and indicate the anticipated level of passenger comfort. The comfort parameters, although not directly associated with ill health,
may exacerbate or raise susceptibility to other health problems.
346
Table 3 Summary of air quality and comfort criteria
Parameter
Criteria
Basis/Source
Temperature
Relative humidity
Air ow
Ventilation
CO2
1927 C
3070%
< 0.15 m/s
[9, 11, 12]
1000 ppm 90% of locations

1250 ppm 100% of locations
Cabins 8 L/s person
Dining-room >10 L/s person
Bar >16 L/s person
Smoking lounge >36 L/s person
100 g/m3
Indoor:outdoor ratio > 3 : 1
Must not be detected
0.01 bres/ml
[3]
Air exchange
Formaldehyde
Airborne fungi
Legionella
Asbestos
[3]
[13]
[14]
[5]
[15]
[16]
4
Remedial Measures
Problems relating to thermal comfort and air exchange rate can usually be
identied and solved relatively easily, provided the system has the required
design capacity. Monitoring and calibration of control systems should be sufcient to improve matters. However, the monitoring tools required to assess
thermal comfort and/or air exchange rate are rarely in evidence on board
ships.
With respect to microbial contamination and particulate matter there are
three general approaches to remedial action:
1. Design spaces and systems to prevent indoor contamination.
2. Maintain indoor conditions so that contamination does not occur or recur.
3. Clean up existing contamination.
The rst two remedial measures rely to a large extent on proper ltration. In
most circumstances, lters should be sufcient to remove microbial agents
larger than 1 or 2 m in diameter. Some ships use prelters (low efciency,
high arrestance), which is a good way to prolong the life of the main, moderately efcient lters. Filters prevent the ingress not only of micro-organisms
and spores but also the dirt and debris which act as nutrients. Typically, in
inspected ships, the cleanliness of the ductwork is directly related to the condition of the lters.
There are no statutory requirements with respect to lter material in marine applications. European and American guidelines have been published
347
which aim to improve air quality and prevent microbial contamination in

buildings. European guidance [1719] recommends that air lters with an
average arrestance of at least 90% should be placed at the intake. A second
lter with an average arrestance of > 99% (average efciency of 0.4 m particles, 8090%) should be located on the supply side of the air handling unit
behind a fan or any aerosol-producing device. Further necessary lters, for
example high-efciency particulate air lters, should be installed close to the
room [6, 17]. The American Conference of Governmental Industrial Hygienists make similar recommendations [5].
Problems regarding ltration are often directly associated with poor maintenance. Gaps or holes in lter material are the main route by which particulates, micro-organisms, and, in particular, fungal spores enter the HVAC
system (Fig. 5). Once ingrained, these fungi are recalcitrant, and therefore
every effort should be made to prevent their entry from the earliest stage.
Mould growth in the HVAC ductwork also presents problems with respect
to day-to-day operation and maintenance. Complaints associated with dust
and debris falling from supply grills can keep crew members occupied almost on a permanent basis. The reduction in ow which is often associated
with mould physically blocking ventilation ducts or recirculation units may
present a whole range of knock-on problems related to thermal comfort, air
exchange rates, and a concomitant build-up of pollutants.
Where microbial contamination does occur, it may become necessary to
clean the HVAC system. There are four generally recognised processes that
may be used to clean HVAC ductwork [18]:
1. Vacuum
2. Steam
3. Compressed air
4. Chemical/disinfection
Physical cleaning methods (1 to 3) are generally regarded as the most effective. However, problems associated with access to ductwork and the complexity of some ship HVAC design means that often chemical cleaning and/or
disinfection is the only method available. Chemical biocides are often advocated to kill moulds and other microbial growth from HVAC systems, and
extensive claims are often made regarding the residual properties of these
biocides. Investigations into the use of biocides for cleaning ductwork have
revealed severe shortcomings with respect to their efcacy. In general terms,
the use of biocides for ductwork disinfection is not recommended [18].
348
5
Conclusions
Generally, shipboard air quality problems are not severe in comparison to
problems in many homes and ofces. However, the potential nancial implications are believed to be great. The HVAC systems of some ships, particularly
older vessels, are prone to microbial contamination. A number of problems
have been identied which warrant attention in order to safeguard passenger comfort and health. Ventilation efciency is a problem on some vessels,
especially in smaller cabins or densely occupied communal areas. Thermal
comfort is also a source of complaint that requires attention on many ships.
Proactive monitoring to evaluate indoor air quality and identify remedial
measures reduces the likelihood of problems developing.
There is a growing awareness that cost-effective improvements can be
made to the design and operation of ventilation systems which reduce contamination and improve air quality. It is apparent that many of the problems
encountered can be designed out. In particular, attention should be paid
to the prevention of ingress of water into the supply and exhaust systems,
thereby restricting the potential for microbial proliferation. Provision of better access for inspection and cleaning of ductwork would also be benecial.
Disclaimer
The views expressed in this article are those of the authors and not those of
Lloyds Register. Neither the authors nor Lloyds Register assume any responsibility or liability for any loss, damage, or expense caused by reliance on the
information or advice in this article.
References
1. Reynolds GL, Lester JN, Perry R (1991) Environmental tobacco smoke and other pollutants in the home. In: Armitage AK (ed) Other peoples tobacco smoke. Galen Press,
Tucson, AZ, pp 133148
2. Webster AD (1997) The Contribution of Ventilation System Design and Maintenance
to Air Quality on Passenger Ships. The Institute of Marine Engineers, London
3. ASHRAE (2001) ASHRAE Standard Ventilation for Acceptable Indoor Air Quality.
ASHRAE 622001. American Society of Heating, Refrigerating, and Air Conditioning
Engineers, Atlanta
4. Brown VM, Cockram AH, Crump DH, Gavin MA (1993) Indoor air assessment in the
UK carried out by the Building Research Establishment Advisory Service. In: Proceedings of the 6th International Conference on Indoor Air Quality and Climate,
Helsinki, 2:1519
5. ACGIH (1989) Guidelines for the Assessment of Bioaerosols in the Indoor Environment, American Conference of Governmental Industrial Hygienists, Cincinnati
6. CEC (1993) Biological Particles in Indoor Environments: Indoor Air Quality and Its
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
349
impact on Man. Report no. 12. Commission of the European Communities, Luxembourg
World Health Organization (1988) Indoor air quality: biological contaminants. WHO
Regional Publications, European series no. 31, Copenhagen
HSE (1993) The control of legionellosis including legionnaires disease, HS(G)70
Health and Safety Executive. http://www.hse.gov.uk/. Cited 2005
ASHRAE (1992) ASHRAE Standard thermal environmental cnditions for human
occupancy. ASHRAE 55-1992. American Society of Heating, Refrigerating, and AirConditioning Engineers, Atlanta, GA
International Organization for Standardization (2002) Air-conditioning and ventilation of accommodation spaces on board ships design conditions and basis of
calculations. ISO 7547
International Organization for Standardization (1992) Moderate thermal environments determination of the PMV and PPD indices and specication of the conditions for thermal comfort. ISO/DIS 7730, Draft International Standard
International Organization for Standardization (1985) Thermal environments instruments and methods for measuring physical quantities. ISO 7726
CIBSE (2000) Guide A, Environmental Design. Chartered Institution of Building Services Engineers, London
World Health Organization (2000) Air quality guidelines for Europe, 2nd edn. WHO
Regional Publications, European series no. 91, Copenhagen
British Standards Institution (1992) Methods for sampling for legionella organisms in
water and related materials, BS 7592
DoE (1991) Asbestos materials in buildings. Department of the Environment, HMSO,
http://www.doeni.gov.uk/. Cited 2005
Commission of the European Communities (1992) Guidelines for ventilation requirements in buildings: indoor air quality and its impact on man. Report No. 11,
Luxembourg
Loyd S (1993) Ventilation system hygiene: A review, 5th edn. The Building Services
Research and Information Association, Bracknell
CEN (2003) Particulate air lters for general ventilation determination of ltration
performance. EN 779
Stonier RT (1995) CO2: powerful IAQ diagnostic tool, Heating Piping & Air Conditioning, Cleveland, Ohio
Winberry WT, Forehand L, Murphy NT, Ceroli A, Phinney B, Evans A (1993) Methods
for the determination of indoor air pollutants USEPA Methods. Noyes Data Corporation, New Jersey

DOI 10.1007/b107252
Submarine Atmospheres
Waldemar Mazurek
Maritime Platforms Division, Defence Science and Technology Organisation,
Department of Defence, 506 Lorimer St., Fishermans Bend, 3207 Victoria, Australia
wally.mazurek@dsto.defence.gov.au
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
352
Early Submarines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
353
3
3.1
3.2
3.3
World War II Generation Submarines

Snorting . . . . . . . . . . . . . . . .
Air Purication . . . . . . . . . . . .
Air Monitoring . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
353
354
356
359
4
4.1
4.2
4.2.1
4.2.2
4.2.3
4.2.4
4.2.5
4.3
4.4
Nuclear-Powered Submarines . . . . . . .
Air Quality Standards . . . . . . . . . . . .
Air Purication . . . . . . . . . . . . . . .
Carbon dioxide removal . . . . . . . . . .
Oxygen Generation . . . . . . . . . . . . .
Carbon Monoxide and Hydrogen Removal
Volatile Organic Compounds . . . . . . . .
Aerosols . . . . . . . . . . . . . . . . . . .
Early Air Quality Problems . . . . . . . . .
Air Monitoring . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
360
361
363
363
364
365
365
365
366
367
5
5.1
5.2
Post-War Conventional Submarines . . . . . . . . . . . . . . . . . . . . . .

Air Purication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Air Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
369
369
370
6
6.1
6.2
New Generation of Conventional AIP Submarines . . . . . . . . . . . . . .

Air Purication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Air Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
374
374
376
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
378
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
379
.
.
.
.
.
.
.
.
Abstract Atmosphere control in submarines has developed to meet the operational requirements. Until the end of WWII submarines were primarily semi-submersibles spending most of their time on the surface and submerged for periods of 12 h or less. However,
rudimentary control of oxygen and carbon dioxide was available in some WWI boats. In
the latter years of WWII, the requirement for longer dive times increased the demand
for atmosphere control and the development of atmosphere monitoring instrumentation.
The introduction of nuclear-powered submarines eliminated the need for air-dependent
propulsion, and initially their dive times were limited only by air quality problems.
352
W. Mazurek
The solution of these problems led to long-term (3 months) atmosphere control techniques, real-time air monitoring capabilities and the establishment of toxicological data
for a large number of air contaminants. These developments have also impacted on atmosphere control in conventional diesel-electric submarines. More recently a new generation
of submarines with non-nuclear air-independent propulsion has emerged. Although their
dive times are limited to 23 weeks, this capability can be best exploited with the development of new energy efcient air purication technology.
Keywords Submarine atmosphere Air quality Air contaminants Air purication
Air monitoring Exposure levels
Abbreviations
AIP
Air-independent propulsion
amu
Atomic mass unit
atm
Atmospheres
CAMS
Central atmosphere monitoring system
CFC
Chlorouorocarbons
HCFC
Hydrochlorouorocarbons
HEPA
High efciency particulate air lter
IMS
Ion mobility spectrometer
MEA
Monoethanolamine
MESMA Module denergie sous-marine autonome
MPC90d Maximum permissible concentration for 90 days
MPC24h Maximum permissible concentration for 24 h
MPC60m Maximum permissible concentration for 60 min
SMAC
Spacecraft maximum allowable concentration
1
Introduction
The evolution of submarines over the past 150 years has also been an experiment in the development of an articial enclosed environment for man.
In order to full the strategic functions of submarines, conditions are required where humans can perform demanding tasks. There are many facets
of this environment but a primary requirement is human life-support with
particular emphasis on a physiologically acceptable atmosphere that is not
detrimental to health or cognitive functions. The management of that atmosphere has been an incremental process commensurate with the changing
strategic requirements of submarines and new developments in propulsion
technology. During this period, submarines have evolved from comical curiosities to 20 000 tonne leviathans with cataclysmic potential.
This evolution was made possible by a combination of unlimited airindependent power and a sustainable enclosed atmosphere. However, the
genesis of submarine atmosphere control lay in the early submarine designs.
353
2
Early Submarines
The need for a physiologically acceptable air quality was recognised in the
early man-powered submarines. These were small vessels rst used in the
American Civil War. Notable examples were the Hunley and the Alligator [1],
the latter being the rst submarine purchased by the US Navy (1862). It was
14 m long and powered by a hand-cranked screw and a crew of 16 (or possibly
more) [2]. Although human power provided limited speed it had the advantage of silent running, a principal requirement of submarines. The low air
volume necessitated internal atmosphere control even for the relatively short
periods spent underwater. This was achieved by passing air over lime with the
aid of bellows to remove carbon dioxide. Oxygen may have been generated on
board through a chemical reaction [3], but this is uncertain.
Subsequent generations of successful submarine designs incorporated various forms of mechanical propulsion including steam engines [4]. Once sufcient steam was generated in the boiler, the pressure could propel the submarine under water for several kilometres. The introduction of electric motors
in the late 19th century allowed the use of electric energy, stored in batteries,
to drive the submerged submarine [1, 4]. The batteries were charged on the
surface by combustion engine powered generators.
Originally, submarine internal combustion engines were gasoline powered,
rendering the submarines susceptible to fuel res. Caged mice were kept in
the engine room of the British submarines. A picture postcard from 1904
showed three white mice with a caption White mice are carried on every
British submarine and are entered on the payroll as part of the crew. Being
extremely sensitive they notify the slightest escape of gasoline, by squeaking [5, 6]. Despite the postcard caption it is thought that they may have also
been used for the detection of carbon monoxide, as mice are more sensitive
to carbon monoxide than humans. This was probably the rst record of air
quality monitoring in submarines. Interestingly, these hazards were of greater
concern than carbon dioxide and oxygen concentrations. Later, during WWI,
the introduction of diesel engines reduced the risk of fuel res in submarines
and they became the forerunners of the modern diesel-electric (conventional)
submarines.
Of the WWI submarines, only the German-built boats were equipped with
basic atmosphere control through the use of a carbon dioxide sorbent (sodalime) and cylinders of compressed oxygen. This enabled a dive time of up
to 72 h [7]. At the time other navies did not see the need for prolonged dive
times, the advantage of which only became apparent in the latter parts of
WWII.
354
W. Mazurek
3
World War II Generation Submarines
WWII submarines were, like their predecessors, semi-submersibles. They
were principally designed for surface operations and were tted with a keel
like any other surface vessel and had deck-mounted guns. They achieved
maximum speed on the surface and were slow when submerged. Because they
were low in the water they were difcult to detect visually.
In general, these submarines would stay submerged during the day to
avoid visual detection, surfacing at night to recharge the batteries and to
launch attacks on surface shipping [8]. For most operations a dive time of
15 h was adequate. The length of the dive was potentially governed by the battery power. For example, WWII Royal Navy submarines could operate at full
speed, underwater, for 1 h or at 2 knots for 36 h [9]. The strategy of diving
deep and staying silent was often used to avoid detection by sonar as well
as avoiding depth charges. In many cases this necessitated prolonged dive
periods.
Generally, the air volume per man was about 10 m3 [10, 11]. This allowed
a boat, without a carbon dioxide removal system, to stay submerged for
a period of 15 h [11] before the carbon dioxide concentration rose to 3% with
the normal crew complement. Beyond this period, carbon dioxide concentrations would build up to the point when physical and mental functions were
severely affected [12, 13]. In the same period the oxygen concentration would
fall to a tolerable 18% in the absence of a supplementary oxygen supply.
These concentrations represented the acceptable limits laid down by Royal
Navy Admiralty instructions for operating air purication equipment in submarines (Book of Reference BR 1326/45) [14]. The revised versions of these
instructions form the basis of current air quality specications for Royal Navy
submarines.
3.1
Snorting
Until the mid-1940s submarines relied largely on surfacing for air revitalisation and battery charging. In the concluding stages of WWII, German and
Japanese submarines were tted with a schnorkel(or schnorchel) [15, 16].
This concept dates back to just after WWI and was used by the Netherlands
Submarine Service from the beginning of WWII [17]. The schnorkel was
a tube that allowed the intake of air from the atmosphere to the submarine
while the boat was just below the surface. Fresh air was drawn into the submarine by a reduction in internal air pressure caused by the operation of the
diesel engines. Schnorkelling overcame the need for the submarine to surface
and thus reduced the risk of detection. Using the schnorkel, the submarine
355
could maintain a speed of 6 knots in calm water. It had been reported that
one German submarine was able to operate for 70 days without surfacing [17].
After the war, the German term schnorkel was anglicised to snort and is it
is commonly referred to as the snorting mast in English- speaking countries.
This process is now an integral part of modern submarine design.
The rst German experimental snorkels were installed in 1942. They consisted of a telescopic tube or, where there was insufcient space, a folding
(pivoted) tube was tted which was hinged at the deck level. The tube housed
separate ducting for the air intake (induction mast) and engine exhaust. Both
were approximately 18 cm in diameter [18]. Although the snorkel had been
under discussion for 15 years in British submarine circles it had never been
introduced [19] until after WWII. It was then that the Royal Navy evaluated
snorting masts on operational submarines but there were some misgivings
concerning their use, particularly regarding the pressure changes experienced in the submarine.
The early designs involved a folding (collapsible) snort mast with the air
intake and exhaust outlet in close proximity to each other (Fig. 1). When not
in use the mast rested on the deck (casing). A photograph of a Royal Navy
submarine HMS Truant (1945) showed such a mast with the inlet to the induction trunking approximately 1.5 m above the exhaust outlet, which was
on the aft side of the mast. The induction mast was equipped with a oatactuated valve to prevent the inadvertent intake of water. The valve could close
for 1 or 2 min with the engines still operating using the internal submarine air
volume as a buffer [10]. It appears, from photographs, that during snorting
the exhaust outlet was just below the surface [19]. Information from German
submariners indicated intake of the engine exhaust emissions through the
induction mast in the presence of tail winds.
A trial was conducted aboard HMS Taciturn in order to determine the
submarine air quality after prolonged continuous snorting (400 h) and the
effects on the crew. Some of the crew were reluctant to participate, as there
were German rumours that sterility was caused by snorting and that U-boat
crews suffered from mysterious ill effects [17]. During the trial the engines
Fig. 1 A schematic diagram of a folding snorting mast of the type tted to Royal Navy
submarines in 1945
356
W. Mazurek
were operated 1620 h/day instead of the usual 212 h/day. Under these
exceptional conditions oxygen levels were frequently very low (1516%),
probably as result of the low pressure while carbon dioxide concentrations
were at acceptable levels (0.150.2%). Carbon monoxide concentrations were
commonly 200 ppm in the engine room. These results, together with the
low pressures observed during snorting (0.6 atm) suggested insufcient airow through the induction trunking. Interestingly, German submariners
were instructed to avoid pressures less than 0.85 atm [17]. The low oxygen partial pressures would have exacerbated the toxic effects of carbon
monoxide [20].
Lethal concentrations of carbon monoxide (500010 000 ppm) have been
reported in the engine rooms of U-boats [17]. If true, the claims reect a poor
state of efciency of the U-boat diesel engines possibly due to an insufcient
supply of oxygen as a result of restricted air ow through the induction mast.
These conditions also tend to produce higher concentrations of partially combusted organic compounds such as acrolein and other irritants [21].
A study of the toxic effects of diesel exhausts, conducted by the UK Ministry of Defence in 1956, showed that the carbon monoxide concentrations
from diesel engines varied from 380 ppm to 1700 ppm depending on the
air/fuel ratio [21]. When animals were exposed to diesel exhaust emissions
with an average carbon monoxide concentration of 560 ppm, within 7 h there
was 100% mortality for guinea pigs (10/10) and 20% mortality for mice
(10/50). The results highlighted the hazards faced by the submarine crews
during snorting and vindicated some of their concerns.
Despite the potential health hazards associated with snorting, the US Navy
also tted snorting masts to its new post-war diesel-electric submarines. The
USS Pickerel had a telescopic mast with a deection plate immediately above
the exhaust outlet to diffuse the exhaust gases [15]. This measure was probably aimed at reducing the visible and infrared signature of the exhaust rather
than reducing the likelihood of taking in the exhaust gases through the induction mast. The Royal Navy tted telescopic induction masts in the early 1960s
which were separated from the exhaust trunking (HMS Truncheon) [22].
Some versions involved a separate induction mast with the exhaust emerging
at the top of the conning tower [19]. It is not known if this modication was
motivated by air quality concerns.
Snorting has become an integral part of modern diesel-electric submarine operations. Ever mindful of being detected visually or by radar, even in
peace time submarines generally avoid surfacing in their own territorial waters but rely heavily on snorting to recharge their batteries and ventilate the
submarine.
357
3.2
Air Purification
As the resolution of radar improved during WWII, even snorting became hazardous. In order to avoid detection in enemy waters and to elude destruction
by depth charges from aircraft and warships, increasingly long dive times
were required. Allied submarines were poorly prepared for this, having little or no air purication measures. As a result, the efciency of the crew
was severely affected during long dives [12]. For example, it was not unusual
for carbon dioxide levels to exceed 3% with crews experiencing difculties
in performing physical duties or even holding a conversation or lighting
a match [12, 23]. German and Italian submarines were better served being
equipped with a carbon dioxide removal system, an oxygen supply and rudimentary air monitoring.
In exceptional cases when additional crew were taken aboard, such as
boarding parties, carbon dioxide removal was made available to British submarines during the early war years. This was in the form of sodalime contained in trays and distributed throughout the submarine [13]. On the basis of Royal Navy medical advice at the time, compressed oxygen was not
carried.
In the nal years of WWII, the Royal Navy had both the means of removing carbon dioxide and generating oxygen in submarines but for some
reason they were more reluctant to use these resources than the Germans and
Italians, preferring to tolerate poor air quality [14]. Because of the lack of
space in submarines, there is always a reluctance to carry additional stores
or equipment unless absolutely necessary. Apart from this, the German and
Italian navies had a long-established culture of air purication practices in
submarines. There may have also been a greater need for these submarines
to stay deep for longer periods compared to the British boats as the Allied
anti-submarine operations improved. However, after an increasing number of
reports of the incapacitating health effects (such as headaches and inability to
think clearly) experienced in British submarines during prolonged dive times,
the Royal Navys attitude changed [13].
Both lithium hydroxide and sodalime are used in submarines for the removal of carbon dioxide. Sodalime is less hazardous in terms of skin exposure
and inhalation but lacks the higher rate of reaction of lithium hydroxide. It is
a mixture of calcium hydroxide and sodium hydroxide (approx. 35%).
A lithium hydroxide monohydrate intermediate is involved in the reaction
of carbon dioxide with anhydrous lithium hydroxide [24]:
LiOH(s) + H2 O(g) LiOH H2 O
2LiOH H2 O(s) + CO2 (g) Li2 CO3 (s) + 3H2 O(g)
(1)
(2)
358
W. Mazurek
A more complex reaction occurs with sodalime. In the presence of gaseous

carbon dioxide the overall reaction is [25]:
CO2 (g) + Ca(OH)2 (s)
H2 O
/
CaCO3 (s) + H2 O(l)
(Slow)
(3)
The reaction can be broken down into a number of steps with the following
relative reaction rates:
CO2 (g) + H2 O CO2 (in solution)
CO2 (soln) + NaOH NaHCO3
NaHCO3 + Ca(OH)2 CaCO3 + NaOH + H2 O
CO2 (soln) + Ca(OH)2 (s) CaCO3 + H2 O
(slow)
(medium fast)
(fast)
(very slow)
(4)
(5)
(6)
(7)
In 1944 the Royal Navy rst evaluated oxygen candles aboard HMS Thule.
These were electrically ignited containers of sodium chlorate mixed with iron
lings that gave off almost pure oxygen in a highly exothermic reaction [14].
They had the advantage of a higher oxygen density than the compressed gas
(and comparable to liquid oxygen) but once ignited they could not be extinguished.
Oxygen candles rely on the thermal decomposition of sodium chlorate to
produce oxygen. Iron powder is added to sustain the reaction and barium
peroxide is present to absorb any chlorine produced from the reaction of
sodium chlorate in the presence of moisture [26, 27]. The reactions are:
2NaClO3 2NaCl + 3O2
2Fe + O2 2FeO
(slightly exothermic)
(exothermic)
(8)
(9)
A number of side reactions occur in the presence of water that generate traces
of chlorine and chlorine-containing compounds which are removed by reaction with barium peroxide, for example:
BaO2 + Cl2 BaCl2 + O2
(10)
Oxygen candles continue to remain an integral part of submarine atmosphere

control on a routine basis and for use in emergency situations.
Although the above measures addressed the respiratory gases the general
air quality remained poor. After a day on patrol, the odours had been described as a combination of the smell of unwashed bodies, the slops drain
tank in the galley, the nauseating smell of long past meals, cooking, fuel
oil, and damp clothing [9]. These problems were later overcome, to a large
extent, by better housekeeping practices and the addition of large capacity
charcoal lters to the air conditioning system; however, some diesel-electric
submarines continue to operate without charcoal lters.
In addition to the problems of air composition, British submarine crews
operating in tropical waters were severely affected by high temperatures
359
resulting in heat stroke, swollen ankles and septic and fungal skin infections. It had been estimated that this reduced human efciency to 60% after
a 13 h dive. According to one account the internal temperature reached 63 C
in the waters of the Indonesian archipelago with debilitating consequences
for the crew [28]. This problem was largely solved by the middle of 1944
when all Royal Navy submarines operating in these waters were tted with
air conditioning. The result was a dramatic improvement in heat-related
symptoms [14].
3.3
Air Monitoring
Efforts were made to introduce carbon dioxide monitors aboard Royal Navy
submarines during the latter part of WWII but such instruments were considered to be too fragile [14]. A number of American Pauling (and later Beckman
model D-2) oxygen meters were evaluated. They were based on the paramagnetic properties of oxygen (compared with the diamagnetic properties
of nitrogen) and were found to be accurate and direct reading but they were
easily damaged. There was also an ofcial view that the appearance of air
monitoring equipment aboard submarines would adversely affect the morale
of the crew.
In a 1945 British report, 13 carbon dioxide monitors were evaluated, some
of which were taken from captured German submarines. They included instruments based on infrared absorption and thermal conductivity. However,
many of the carbon dioxide meters at the time relied on absorption of the carbon dioxide by a reagent (e.g. sodalime) within a small chamber followed by
measurement of the resultant decrease in pressure with a manometer. This
was the principle of a Drgerwerk instrument tested in German submarines
in 1942 [29]. When evaluated by the British, it was found to be a robust and
accurate instrument but cumbersome to use and requiring a skilled operator [14]. By the end of WWII the German and Italian submarines had instruments for measuring carbon dioxide, oxygen and humidity, thus potentially
avoiding ill effects from high carbon dioxide and low oxygen concentrations.
However, they were apparently not used regularly. The reason for this is not
known but the complexity of the procedures alone would have discouraged
their use in all but the most pressing situations.
Although the instrumentation of the time may not have been suitable
for wartime operational purposes it was sufciently robust for sea trials. In
1947 a successful air quality trial was conducted aboard a British submarine
using oxygen, carbon dioxide and carbon monoxide measuring instruments.
Oxygen concentrations were measured with a paramagnetic Pauling oxygen
analyser. Carbon dioxide was measured by thermal conductivity with an instrument from Cambridge Instruments. It represented a signicant advance
on the instruments available during the war. Carbon monoxide was meas-
360
W. Mazurek
ured with a Chemical Carbon Monoxide Indicator Mark III from the Royal
Aircraft Establishment, Farnborough. The operating principle of this instrument was not described [17]. The trials were signicant in that they indicated
a change in the culture of indifference towards air quality in Royal Navy submarines.
By the end of WWII submarines were spending more time submerged than
ever before and they were beginning to evolve as True Submersibles. The ultimate goal of being independent of atmospheric air for propulsion and life
support had been reached in 1939 when the Germans evaluated a small experimental submarine developed by Helmuth Walter, which used hydrogen
peroxide as a source of oxygen to run a diesel-fuelled turbine [10]. The hydrogen peroxide was catalytically decomposed to oxygen but was unstable and
was prone to explode unexpectedly. Despite this problem some operational
boats were constructed, although none went to war. After the war two British
submarines were built (Explorer and Excalibur) based on the German design.
They achieved an underwater speed of 26 knots for a short time and could
out-run many surface vessels [6, 19, 30].
4
Nuclear-Powered Submarines
The success of the US nuclear-powered submarines ended the brief irtation
with the hydrogen peroxide system that was instrumental in highlighting the
strategic advantages of air-independent propulsion (AIP). The nuclear reactor
provided the means for air independency and almost unlimited power. It was
a logical extension of the concept of AIP as conceived by Helmuth Walter.
The advances in air purication and air monitoring in the diesel-electric
submarines provided the initial technology for the nuclear-powered submarines. However, it was inappropriate as it relied on non-regenerative means
of carbon dioxide removal and oxygen generation. Since these submarines
were to spend months at sea and submerged for most of the time, the vast
quantities of sodalime (or lithium hydroxide) and oxygen candles (or compressed oxygen) required would make this approach impractical.
The limitations became apparent when one of the early nuclear-powered
submarines, USS Nautilus, put to sea in 1955. As a result of the poor air
quality, more comprehensive air purication equipment was gradually installed [31] to facilitate extended dive times. However, the 1958 historic transit of USS Nautilus under the polar ice cap was achieved with the installation
and use of reticulated compressed air supply (emergency breathing air) and
connecting face masks to avoid reliance on the submarine atmosphere when
air contamination was high.
The introduction of the missile-carrying submarines in the 1960s further
hastened the development of air purication [31]. The stealth requirements
361
of these submarines were pivotal to Cold War strategy. It was important that
they remained submerged on station for up to 3 months.
4.1
Air Quality Standards
With the potential of nuclear-powered submarines to remain submerged for
months at a time and the lessons learned from the WWII experiences with
conventional submarines, it was clear that the 1945 Royal Navy submarine air
quality requirements (BR 1326/45) were inadequate. A major reappraisal of
air quality standards for submarines was necessary to maintain a habitable
environment free of detrimental health effects.
Both the Royal Navy and the US Navy produced similar comprehensive
submarine atmosphere control manuals, which listed a substantial number of
potential air contaminants. The purpose of the manuals was to ensure the
health and safety of the crew and to prevent damage to the ships machinery
from atmospheric contaminants. The US Navy manual produced in 1979 described the procedures for atmosphere monitoring and exposure for over 35
compounds [32] some of which are listed below with the most likely source:
Oxygen (depletion)
Carbon dioxide
Carbon monoxide
Nitrogen dioxide
Hydrogen
Hydrogen chloride
Chlorine
Hydrogen uoride
Hydrogen Sulde
Sulfur dioxide
Ammonia
Refrigerant gases
Chlorinated hydrocarbons
Methanol
Isopropanol
Benzene
Toluene
Ethyl benzene
Xylene
Arsine
Stibine
Respiration
Respiration
Cooking, smoking, combustion engine
exhaust
Catalytic burners
Battery charging
Refrigerant decomposition
Batteries (electrolysis of seawater)
Refrigerant decomposition
Sewage
Fire
CO2 scrubber
Refrigeration equipment
Solvents
Inks
Solvents
Solvents, fuels
Solvents, fuels
Solvents, fuels
Solvents, fuels
Battery charging
Battery charging
362
W. Mazurek
The majority of these compounds would be typical of many industrial environments. The exceptions are arsine and stibine, which could be formed
during the charging of lead acid batteries due to the presence of arsenic and
antimony impurities in lead electrodes. Initially this was perceived as a problem but improvements in lead purity have largely eliminated this hazard.
The Royal Navy has classied submarine air contaminants into various categories [33].
Category I includes the life gases and requiring continuous measurements
in real-time.
Category II includes compounds that may be released into the air during
an unusual event such as a leak, mechanical failure or a re. Monitoring
of these compounds is in real-time but periodic during the course of the
event.
Category III compounds are those which may be present on occasions and
present a potential for chronic health effects. Such compounds are routinely and retrospectively analysed.
Category IV substances are those which do not pose a health hazard but
may indicate a malfunction in equipment. Such substances are monitored
retrospectively as required or during special trials.
Category V substances are those which have been removed from the monitoring lists because they are no longer considered a health threat, they may
however, represent a minuscule health risk. These are not monitored but
are periodically reviewed.
Unlike industrial workers, submariners are continuously exposed to submarine atmospheric contaminants for the duration of a patrol. In the case of
the missile submarines this was 90 days. Hence, the baseline maximum permissible concentration (MPC90d) was set for this period and it represents an
action level rather than a time-weighted average. In addition, it was acknowledged that there would be occasions when these concentrations would be
exceeded due to some unforeseen event. As a result, maximum permissible
levels were also set for 24 h (MPC24h) and 60 min (MPC60m). If the MPC90d
is exceeded then the atmospheric concentrations need to be restored within
24 h. If the MPC24h is exceeded then they need to be restored within 1 h and if
the MPC60m is exceeded then the crew need to don respirators. All exposures
above the MPC90d are expected to be rare events [34].
These concepts have also been embraced by the French and US navies and
they are similar to those used by the NASA space program [35] where they are
referred to as spacecraft maximum allowable concentrations (SMACs). Despite the differences of the two environments, spacecraft and submarines face
similar problems of air purication and there has been a good deal of collaboration between the two communities in air monitoring, air purication and
toxicology.
363
At the time when the original submarine air quality standards were developed for nuclear-powered submarines (1960s), there was some collaboration between the US, UK and French navies where industrial occupational
toxicological data were unavailable [34, 36]. It is obvious from the similarity
of the US and UK documents that there was a considerable degree of collaboration between the two countries. Unfortunately, the original deliberations
on these exposure levels were never properly documented and the rationale
for the MPCs remains a matter of conjecture. As a consequence, a comprehensive review was undertaken by the Royal Navy in the mid 1990s in the form of
a mini-toxicological review for each air contaminant [36].
Independent bodies periodically review the submarine toxicological data.
In the UK this responsibility falls on the UK Medical Research Council [34]
through the Royal Navy. In addition, permanent committees comprised of
submariners, equipment suppliers, navy medical personnel and analytical
chemists are charged with the duty of ensuring that current standards of
air quality are maintained [37]. In the USA the National Research Council periodically conducts independent reviews of submarine atmosphere
control.
4.2
Air Purification
4.2.1
Carbon dioxide removal
Nuclear-powered submarines have used regenerative carbon dioxide removal
systems based on zeolite molecular sieves and amines [26]. At rst molecular sieves were used. Sorption/desorption of carbon dioxide from molecular
sieves can be achieved by either pressure change (pressure swing) or temperature change (temperature swing). Two beds are used with one in the sorption
mode while the other is being desorbed. An additional advantage of this system is the removal of refrigerant gases from the submarine atmosphere. The
disadvantages are noise, large size, high power requirement and the removal
of moisture from the air. The latter problem is dealt with by passing the air
over pre-dryer beds then desorbing the water and returning the water vapour
to the submarine atmosphere.
The French have continued using molecular sieves while both the Royal
Navy and the US Navy have abandoned this system and turned to amines (in
aqueous solution) instead [26]. The amine plants are more efcient [38], quieter and smaller than the molecular sieve plants. The principle of operation is
based on the reversible reaction of primary and secondary amines with carbon dioxide to form carbamates [39] and bicarbonates [40] with the overall
364
W. Mazurek
chemical reaction being:

2RNH2 + CO2 RNHCO2 + RNH3 +
Heat
H2 O
RNH2 + HCO3 + RNH3 +
(11)
Sorption of carbon dioxide is achieved at ambient temperatures and desorption occurs at reux temperature (approximately 135 C). In the amine plants,
carbon dioxide laden air is passed through a packed absorber tower where
the carbon dioxide reacts with the amine. The reaction mixture is then passed
into a boiler where the pure carbon dioxide is stripped, compressed and discharged overboard, while the amine is recirculated back into the absorber
tower in a continuous process.
Monoethanolamine (MEA) is the most commonly used amine because of
its water solubility and relatively low volatility. The process requires monitoring of the amine concentration by acid-base titrations in the non-ideal
conditions of the submarine environment. On occasions there are spillages
and leaks of amine solution which require attention and are complicated by
the toxic nature of MEA. To reduce fugitive amine emissions from the scrubber, the discharged air is passed through an ion exchange resin lter bag
prior to entering the ventilation system. Atmospheric concentrations of carbon dioxide of 0.5% can be achieved with this type of scrubber and unlike
the molecular sieve system, there is no loss of nitrogen or water into the waste
carbon dioxide stream.
A solid amino acid potassium salt, N-methyl alanine (Alkazid M), has also
been evaluated and considered as an alternative to MEA [41]. It has the advantage of the low volatility of a salt but it has a tendency to precipitate as the
bicarbonate under some conditions [40] and appears to have lower efciencies, under the same conditions as MEA.
HOCH2 CH2 NH2
monoethanolamine (MEA)
KO2 CCH(CH3 )NHCH3
potassium N-methyl alanine (Alkazid M)
4.2.2
Oxygen Generation
Initially nuclear-powered submarines used compressed oxygen for air revitalisation and oxygen candles for emergencies. The system was quickly (1953)
replaced with electrolytic oxygen generators based on industrial equipment [42]. This is an energy intensive process, which can be afforded only
through the availability of a power supply such as a nuclear reactor. The by-
365
product, hydrogen, is discharged overboard.

2H2 O
electrolysis
/
O2 + 2H2
(12)
There are two types of electrolysers, low pressure and high pressure
(200 atm). The earlier high-pressure electrolysers used a liquid caustic electrolyte and the higher pressure facilitated the discharge of hydrogen overboard. The more recent low-pressure electrolysers use a solid polymer electrolyte. They have fewer components and are easier to operate [43].
4.2.3
Carbon Monoxide and Hydrogen Removal
Nuclear-powered submarines have fewer batteries than the diesel-electric
submarines and only a small diesel engine for emergency propulsion. Nevertheless, sufcient carbon monoxide and hydrogen is generated to warrant
a high temperature catalytic oxidiser to remove these gases from the atmosphere [44]. The catalyst (Hopcalite), a mixture of copper oxide and manganese dioxide, has been used for some time in coal mining for removal of
airborne carbon monoxide. Air is passed through a heat exchanger and then
through the catalyst bed which is operated at 315 C. The system is also effective in removing many organic air contaminants by oxidation to carbon
dioxide and water [45, 46] and there is also anecdotal evidence to suggest that
this process destroys biological pathogens in the submarines.
4.2.4
The submarine environment is contaminated by volatile organic compounds
(VOCs) emitted from machinery, electronics, construction materials, paints,
lubricating oils, hydraulic uids and human habitation activity. In addition to
the removal of VOCs by the catalytic burner, a large activated charcoal lter is
used in the air ventilation system. The charcoal can absorb up to 2025% of
its weight in VOCs [47] and it is effective in reducing odours and removing all
but the most volatile compounds.
4.2.5
Aerosols
There are many sources of aerosols in submarines. The major source in conventional diesel-electric submarines is the diesel engine exhaust. Analyses
of aerosols collected on lter papers in the machinery area of Royal Navy
nuclear-powered submarines showed that average concentrations were approximately 0.2 mg m3 , of which approximately 65% were aliphatic organic
366
W. Mazurek
compounds probably from lubricating oils. US submarines on the other hand

appear to have a greater contribution from cigarette smoking [48].
In order to reduce the emissions of aerosols, vent precipitators are used to
remove oil mists from the vents of machinery oil sumps while atmospheric
aerosols are removed by two-stage electrostatic precipitators. Some of these
are modular and can be removed for easy cleaning while the others are selfcleaning [45]. The latter use a cleaning solution, which is automatically and
periodically sprayed over the electrodes to remove deposits.
4.3
Early Air Quality Problems
The installation of the air purication equipment in nuclear-powered submarines was a revolutionary development in submarine technology even
though the processes were not new. Despite the use of proven technology, severe problems arose with this equipment in the early years (USS Nautilus, USS
Seawolf ) [31].
The air quality in the submarines was very poor and threatened to jeopardise the advantages of air-independent propulsion. The carbon dioxide
scrubbers were not sufciently effective [44] in removing the atmospheric
carbon dioxide and there were leaks of amine solution through the seals
around the pump shafts. The catalytic burners would periodically explode
and burst into ame as a result of build up of fat and oil emissions from the
galley. Refrigerant gases leaked from equipment and were decomposed into
acid gases (hydrogen uoride, hydrogen chloride) resulting in brass ttings
turning green and submariners suffering from throat and eye irritation [44].
The acid gases were allegedly causing loose tooth llings [31]. The extent of
the refrigerant leaks was reduced through better seals. Lithium hydroxide was
added to the Hopcalite catalyst bed and a lithium carbonate lter was tted
to the outlet. This partially solved the problem by removing the acid gases as
they were being generated [45].
The VOC concentrations were so high that the charcoal lter in the ventilation system was saturated within 24 h to 48 h, corresponding to 10 kg of VOCs
in 24 h from an atmosphere concentration of 200 mg m3 [47]. VOC concentrations in new submarines tend to be high due to off-gassing of paints,
electrical components and construction materials. By increasing the size of
the charcoal bed to 250 kg, increasing the capacity of the catalytic burner,
banning painting and improving housekeeping practices the VOC concentrations were reduced to below 30 mg m3 [47]. The fact that painting was
carried out during patrols showed an alarming degree of ignorance in submarine atmosphere issues in the US Navy at the time.
Unfortunately, the practices of the diesel-electric submarines were carried
over into the new class of submarines. To ensure good air quality the internal
environment needed to be managed not just in terms of air purication and
367
air monitoring but primarily by ensuring that unnecessary pollutants were

not introduced into the submarine. Some control was needed on the consumable and structural materials used on board. In addition to controls on
housekeeping practices, materials needed to be assessed for their off-gassing
properties as a condition of use in the submarines.
Although the severity of the early problems were reduced, many of the
issues persisted for a number of years. The US Naval Research Laboratory
provided annual progress reports on research into submarine atmosphere
purication and control over the period 19601967 [49]. Like any other environmental control problem, there needed to be an understanding of the many
factors contributing to air pollution within the submarine.
Anticipation of problems and resolution of unforeseen problems are part
of air quality management. One example of an anticipated problem was the
introduction of hydrochlorouorocarbons (HCFCs) as replacements for chlorouorocarbons (CFCs) in the early 1990s [50]. By their very nature HCFCs
are more susceptible to decomposition than CFCs. The catalytic burners were
already modied to remove the acid gases from the decomposition of the
CFCs. It was envisaged that the equipment would require further modication when HCFCs were present. From experiments, it was found that if the
operating temperature was lowered to 260 C the decomposition of HCFCs
(principally R-134a) would be reduced to the level where there would be no
acid gases emerging from the catalytic burner while maintaining an acceptable level of hydrocarbon oxidation [51]. As an additional measure, a greater
emphasis was placed on minimising refrigerant gas leaks. The high sensitivity
to CFCs and HCFCs (2 ppm) and the reliability of the central air monitoring system (vide infra) enabled refrigerant leaks to be identied at an early
stage [50].
In an example of an unforeseen problem, during the early 1990s, a yellow powder was found to be forming on the internal surfaces of submarines.
It was subsequently identied as 2,6-di-tert-butyl-4-nitrophenol. The source
of this compound emerged as being 2,6-di-tert-butylphenol, an antioxidant
additive used in lubricating oil and hydraulic uids. It turned out that the
oil aerosols in the submarine atmosphere were being nitrated when passing through the electrostatic precipitators. Subsequent toxicological studies indicated a relatively high toxicity to rats (80 mg/kg) [52]. Fortunately,
the solution to this problem was simply to choose a different oil, free of
2,6-di-tert-butylphenol. The unsettling aspect of this case is the possibility
of other similar reactions taking place in the submarine atmosphere despite constant and careful air monitoring. It is doubtful whether this phenomenon would have been discovered had it not been for the colour of the
product.
368
W. Mazurek
4.4
Air Monitoring
In 1954 a prototype atmosphere analyser was installed aboard the USS Nautilus but never operated. A subsequent working version (Mark II) measured
carbon dioxide, carbon monoxide, oxygen, hydrogen and hydrocarbons and
was about the size of a 3-drawer ling cabinet. Air samples were drawn from
eight locations throughout the submarine. Carbon dioxide, carbon monoxide
and hydrocarbons were measured by infrared absorption-based technology
developed in pre-war Germany [53]. As expected, sensitivity to hydrocarbons
was relatively low with a full scale range of 6500 ppm. Oxygen was measured
with a paramagnetic detector and hydrogen by thermal conductivity.
Later versions (Mark IIIIV) were tted with infrared Freon detectors.
They were insensitive to small refrigerant leaks. The cells were pressurised
(6 atm) to improve sensitivity, but both the cells and infrared detectors were
unreliable. The pressurised cells were replaced with atmospheric pressure
cells 1.2 m long to provide the necessary sensitivity. In an attempt to further
improve the sensitivity, the Mark V used an automated gas chromatograph to
analyse all the gases.
Although these instruments worked well in the laboratory they were unsuitable for submarines. The analyser was too complex for the crew and was
plagued with reliability problems. In the 1980s the hydrocarbon monitor was
replaced with a simpler portable photo-ionisation detector [54]. Eventually
the entire analyser was replaced with a mass spectrometric (magnetic analyser) central atmosphere monitoring system (CAMS) which over the next
25 years provided the much sought after reliability. The system involved no
prior concentration or separation of the air contaminants. The air sample
was introduced directly into the mass spectrometer and the air contaminants
were separated in the magnetic analyser according the mass/charge ratio
using a xed collector for each preset mass. Thus the selection of the target
compounds was built into the instrument hardware. Carbon monoxide was
measured by a separate instrument based on infrared absorption.
In the mid 1980s a more advanced version of the CAMS-I became available, the CAMS-II, which was able to scan over the desired mass range
(2300 amu) as determined by the software. This provided a greater degree of
exibility, brought about by the use of microprocessors which were previously
unavailable.
The early Royal Navy nuclear-powered submarines were tted with an
air monitoring system consisting of a gas chromatograph with four separate
packed columns for monitoring hydrogen, oxygen, carbon dioxide and carbon monoxide [55]. Like the early monitors in the US submarines, it also
suffered from reliability problems and in 1980 it was replaced with a British
version of the US Navy CAMS. However, the British CAMS used a quadrupole
analyser rather than a magnetic analyser. The quadrupole analysers were
369
more susceptible to drift off calibration [56]. For various reasons the British
CAMS proved to be far less reliable than the US equivalent.
The reliability problems associated with these air monitoring systems were
not necessarily due to the technology. Often it is the installation and maintenance procedures that can lead to these problems. The complex and often
hostile environment is a challenge to any air monitoring technology. There
is also a requirement for continuous operation for 90 days without factory
support or calibration.
In addition to real-time monitoring, both the US Navy and the Royal
Navy undertake retrospective air monitoring for gases and particulates.
In the Royal Navy, air samples are taken using sorbent tubes lled with
Tenax [57]. In excess of 30 organic compounds are quantitatively determined by thermal desorption gas chromatography/mass spectrometry. They
include chlorinated compounds, aromatic hydrocarbons, low molecular
weight alcohols and ketones. A further 30 compounds are determined semiquantitatively [48].
Airborne particulates are collected on glass bre lters, over a period of
24 h, and analysed for metals retrospectively by inductively coupled plasma
spectroscopy (ICP). The lter samples are also extracted with carbon dioxide (super critical uid) and analysed for polynuclear aromatic hydrocarbons
(PAHs) and the water-soluble ions; uoride, chloride, chlorate, nitrate, nitrite,
bromide, sulfate and phosphate are analysed by ion chromatography [48].
5
Post-War Conventional Submarines
The post-war diesel-electric submarines that were designed and built in the
1950s and up to the 1970s (e.g. Oberon class) were largely based on the WWII
concepts. In the period that followed, for the rst time, operational submarines were built without keels and the designs optimised for underwater
running. However, advances in the air purication system have been slow to
match the capabilities of the nuclear-powered submarines.
5.1
Air Purification
In addition to the management of carbon dioxide and oxygen, hydrogen and
carbon monoxide controls were implemented in the post-war submarines. In
order to overcome some of the problems of snorting, the British Oberon class
submarines were tted with 12 small catalytic oxidizers for carbon monoxide
and hydrogen [26, 58]. They contained alumina pellets coated with palladium
and enclosed in nickel wire mesh envelopes. The catalyst was electrically
heated and the system relied on convection for circulation of air through the
370
W. Mazurek
catalyst bed, the temperature of which would vary (60120 C) depending on

the atmospheric hydrogen concentration (the catalytic oxidation of hydrogen
being highly exothermic). The popular Oberon class submarines were sold to
several navies during the 1960s and 1970s.
French submarines were equipped with charcoal lters distributed throughout the submarine using a total of 80 kg (minimum), which was replaced
every 90 days [59]. On a trial basis, one of the Italian Sauro class submarines
has been recently retrotted with 19 lter units, each containing 1 kg charcoal, through which fan-driven air is passed. An ionisation precipitator and
a UV sterilising lamp (for airborne pathogens) are also included in each
of the units [60]. The submarine is also tted with a CAMS, similar to the
instrument used in the Royal Navy nuclear-powered submarines.
The air purication system on the Australian Collins class submarines,
which were commissioned over the period 1996 2003, is more advanced
than other conventional submarines. In addition to the sodalime and oxygen
candles, the submarines have a wet amine carbon dioxide scrubber similar
to that of the nuclear-powered submarines. A charcoal lter (100 kg) is tted to the air ventilation system in the forward compartment, which contains
the crew accommodation. As a result, VOC concentrations of 12 ppm can be
achieved in the accommodation areas [61].
A low temperature CO/H2 catalytic burner is installed in the engine
room. The catalyst consists of platinum and palladium-coated tin oxide,
which is more active and can operate at lower temperatures than the
palladium-coated alumina used in the Oberon class submarines [58]. Unlike
the high temperature Hopcalite catalytic burner used in the nuclear-powered
submarines, this system has little effect on hydrocarbons and minimal effect
on CFCs, although some of the more reactive VOCs are partially oxidised by
the catalyst. For example, ethanol and 1,1,1-trichloroethylene are oxidised to
acetaldehyde and vinylidene chloride, respectively [62], which are more toxic
than the parent compounds.
In an effort to cater for the growing demand of atmosphere control in conventional submarines, Drger Sicherheitechnik has produced a small-scale
modular air purication system with a particulate lter, a hydrogen/carbon
monoxide oxidation catalyst operated at ambient temperatures, and a high
temperature catalyst to remove methane. An activated charcoal lter is used
to remove higher molecular weight VOCs, and carbon dioxide is removed
with sodalime. The system is designed for a crew of 1230 and the relatively
short dive times typical of diesel-electric submarines [63]. To date this has not
been installed on any operational submarine.
371
5.2
Air Monitoring
Because the nuclear-powered submarines were the rst to adopt comprehensive air quality standards, they have set the benchmark for future generations
of conventional submarines, even though they have shorter underwater endurance. The challenge for conventional submarines in adopting this benchmark involves dealing with air contamination from fuel vapours and diesel
engine exhaust emissions, which are largely absent in the nuclear-powered
submarines.
Atmosphere monitoring in the Collins class submarines is achieved with
a suite of 49 gas sensors distributed throughout the boat and connected to
a central computer. They monitor the following airborne components: oxygen, carbon dioxide, carbon monoxide, hydrogen, Freon, Halons, chlorine,
nitrogen dioxide, hydrogen cyanide, hydrogen sulde, torpedo (Otto) fuel,
hydrocarbons and aerosols. The data from the sensors are logged for retrospective analysis and alarms are triggered if concentrations exceed the
MPC24h [64]. In addition, portable electrochemical sensors and colorimetric
tubes are used for emergency air monitoring.
A CAMS, based on an ion mobility spectrometer (IMS), has been proposed
for use in submarines [65]. This approach has also been adopted by Drger
in conjunction with their air purication suite. The instrument is lightweight
(6 kg) and has a low power consumption (6 W). It is reputedly able to monitor
a large variety of gases as well as selected VOCs similar to those targeted by
the US Navy CAMS. However, it is unable to measure hydrogen, oxygen, carbon dioxide and carbon monoxide [63]. Unfortunately, there is no published
information on the performance of the system in the submarine environment.
Currently IMS is used to detect torpedo (Otto) fuel leaks in the Australian
Collins class submarines and this technology has been found particularly
useful for real-time monitoring of MEA [66], which is otherwise difcult to
measure either in real-time or retrospectively. Its sensitivity to amines has
been enhanced by the use of 4-heptanone as the reagent in place of water [67],
providing a dynamic range of 0.053 ppm for the handheld military version
of the instrument [66].
Generally, the diesel-electric submarines that have received most attention
in terms of air quality have been those operated by the navies with nuclearpowered submarines. This was based on the premise that there could not
be two standards of air quality in their submarines. Currently only France
has both conventional and nuclear-powered submarines while the Royal Navy
withdrew the relatively new Upholder class diesel-electric submarines in the
1990s. Other navies have been motivated into a closer scrutiny of submarine air quality by the increasingly stringent occupational health and safety
regulations pertaining to air contaminants. Both the Canadian [68, 69] and
372
W. Mazurek
Australian navies have largely adopted the US and British air quality standards for nuclear-powered submarines.
Air quality surveys of conventional submarines, for example the British
Oberon class, have shown the major source of air contamination to be hydrocarbon vapours and aerosols [70]. The concentrations of VOCs in the
engine room, consisting largely of diesel fuel (hydrocarbons), were in the
range 250 ppm (Fig. 2). The highest concentrations are reached immediately
after the engines have been closed down. Under these conditions the engine
room is poorly ventilated and at an elevated temperature due to the hot engines, which leads to the volatilisation of fuel from various sources. These
concentrations of fuel vapour also occur in the more modern diesel-electric
submarines although the diesel vapours may be largely conned to the engine room. Those submarines with charcoal lters, such as the Australian
Collins class submarines, generally do not include the engine room in the air
ltration system.
Maintaining good air quality in the accommodation areas is important in
submarines as it offers the opportunity for the body to eliminate some of the
pollutants that may have been absorbed in more contaminated areas such as
the engine and machine rooms. It is mainly the engineering maintenance personnel who are exposed to these areas. The remainder of the crew often enjoy
almost indoor air quality.
The VOCs and semi-volatile organic compounds (SVOCs) are responsible
for a lingering odour on the clothes and hair of submariners which persists
for up to 2 days after leaving the submarine. The exhaled breath analyses of
submariners have shown a decay in VOCs, in agreement with this anecdotal evidence. In addition, the exhaled breath concentrations of submariners
tend to be as high as those of industrial (e.g. airport) workers exposed to
Fig. 2 VOC concentrations in the engine room of an Oberon class submarine, under
conditions of surface running and diving
373
signicantly higher concentrations of air contaminants [71]. This is probably

a reection of the continuous exposure in submarines compared to intermittent exposure in industry. Air contaminants are not necessarily the major
source of human exposure in submarines. Skin absorption of liquids may lead
to a much higher intake of contaminants. It has been found, however, that
there are signicant reductions in the exhaled breath concentrations of submariners from the more modern submarines such as the Collins compared
with the older Oberon class. This largely reects the improvement in air quality in the accommodation areas of the submarines where elimination of VOCs
from the body can occur.
In diesel-electric submarines, high levels of aerosols can occur during
snorting as a result of the intake of diesel engine exhaust emissions through
the induction mast. This regular phenomenon is known as getting your own
back in submariner parlance. Computational uid dynamics modelling and
salt water tank experiments with scaled models, have shown that an average
of 25% of the exhaust plume can be drawn down the induction mast, in the
worst case situation with a relative tail wind of 7 km/h. At higher relative
wind speeds (22 km/h) this is reduced to 5% [72]. The greatest impact of this
event occurs in the engine room and adjoining areas. Under these conditions
aerosol concentrations greater than 10 mg m3 do occur for short periods
(0.5 h). In practice, to avoid prolonged exposure at these concentrations, remedial action may be taken by changing course, but this may not always be
possible. However, under normal operating conditions, average aerosol concentrations (for the patrol period) are rarely greater than 0.2 mg 3 [73]. This
gure includes a substantial component of sea salt spray (approx. 30%). Unlike the WWII submarines, modern diesel-electric submarines do not require
the engine room to be manned while the engines are operating. However,
adjoining areas are manned.
Many navies continue to allow cigarette smoking aboard submarines even
though the practice is prohibited in most workplace environments. In some
cases this is allowed only in the engine room when the engines are running. However, limiting exposures to diesel exhaust emissions and allowing
cigarette smoking in the same place is paradoxical bearing in mind that when
unrestricted, cigarette smoking can be a major source of aerosols and carbon monoxide. For example, carbon monoxide concentrations can be maintained below 10 ppm [70] in the engine room, but concentrations between
3080 ppm have been reported in Canadian Oberon class submarines when
engines were not operating [68]. This was probably due to cigarette smoking
and indicates the signicant contribution smoking makes to carbon monoxide levels in submarines.
Despite the regular snorting, carbon dioxide levels up to 2.5% have been
reported in an Oberon class submarine [68], well above the maximum operating conditions for nuclear-powered submarines, but below the Royal Navy
limit (3%) set during WWII. These levels were reached despite the available
374
W. Mazurek
capacity to maintain carbon dioxide concentrations below 2% by the use of

sodalime. Some of submarines have the capacity to maintain carbon dioxide concentrations below 1%. The German navy requires that carbon dioxide
concentrations do not exceed 0.5% and this is achieved with lithium hydroxide rather than sodalime.
Refrigerant gas leaks are fairly common in most applications and because
of the enclosed environment in submarines substantial air concentrations
can be reached with a major leak. Typical background levels can be up to
30 ppm [68] but concentrations of 200 ppm are not uncommon [73]. The elevated levels can be reduced either by regular replacement of seals in the
air conditioning compressors or frequent snorting (or surfacing) as charcoal
lters are not effective in retaining the volatile refrigerant gases.
With the environmental restrictions on discharging waste-water at sea, the
bilge tanks can become a source of hydrogen sulde through the action of
anaerobic sulfate-reducing bacteria on seawater in the presence of nutrients.
Emissions from the tanks frequently result in hydrogen sulde levels exceeding 10 ppm [73]. Because of the toxicity of hydrogen sulde, monitoring in
the vicinity of the tanks and the use of charcoal lters on the tank vents is
essential.
6
New Generation of Conventional AIP Submarines
The most signicant recent advance in submarine technology has been the
development of air-independent propulsion (AIP); a revival of the Helmuth
Walter submarines. They may be regarded as a poor mans nuclear-powered
submarine. A number of AIP propulsion systems have been developed and
currently both Stirling engine (Sweden) and fuel cells (Germany) have been
employed in operational submarines. Both technologies require liquid oxygen. The Stirling engine is an external combustion engine. It is vibration- and
noise-free and almost any fuel can be used. The principal source of energy
for the fuel cell is hydrogen, although hydrocarbon and alcohol fuels can be
reformed to produce hydrogen.
6.1
Air Purification
Modern AIP can allow a submarine to operate submerged for 2 weeks or
more. Oxygen replenishment can be easily achieved due to the availability of
liquid oxygen, however, the problem of carbon dioxide removal is similar to
that of the nuclear-powered submarines. The AIP submarines in current production, such as the German Type 212 and Swedish Gotland class Type A19,
use a non-regenerative system (sodalime or lithium hydroxide). Ironically,
375
the same problem was confronted in the early years of the nuclear-powered
submarines. To fully utilise the potential benet of AIP, a regenerative carbon dioxide removal system is required in addition to other air purication
and air quality control measures. Unlike the nuclear-powered submarines,
however, AIP submarines have limited power and limited space and therefore the direct transfer of air purication technology from nuclear-powered
submarines is not appropriate.
The current technology for regenerative carbon dioxide removal is the
liquid amine scrubber. A more recently developed system, by EADS Space
Transportation GmbH (formerly, Astrium GmbH), uses solid amine resins
somewhat similar to ion exchange resins. The principle of operation is analogous to liquid amines, whereby carbon dioxide is absorbed by the resins at
ambient temperatures and desorbed at elevated temperatures. Two beds are
used; while one bed is in the absorption phase the other is being regenerated [26]. The electric power demand can be reduced by utilising the heat
generated by the fuel cell system [74]. The main advantage of this system is
that it is free of liquid amine leaks and has potentially lower fugitive emissions of amine vapours. Although trials have been conducted in AIP (Swedish,
Canadian) submarines thus far, the system has not been installed in an operational submarine.
Another promising but less mature system is based on the MEA scrubber
with the absorber tower being replaced by a hollow bre hydrophilic membrane separating the carbon dioxide laden air from a liquid amine (or amino
acid) stream. Carbon dioxide diffuses through the membrane and reacts with
the amine, after which it is removed from the liquid amine by reuxing in
a separate chamber [75]. It has been estimated that the power requirement
for such a system in a Dutch submarine would be 4.1 kW, removing 2.5 kg
CO2 /h. This corresponds to a slightly lower energy efciency than existing
MEA scrubbers although a full scale unit has yet to be built.
Other less mature processes for the removal of carbon dioxide have also
been proposed based on carbonic anhydrase enzyme which accelerates the
absorption of carbon dioxide into an aqueous media.
CO2 + H2 O
carbonic anhydrase
/
HCO3 + H+
(13)
The carbonic anhydrase is immobilised on polymethylmethacrylate beads in

a packed column. Carbon dioxide rich air is passed through the immobilised
enzyme in an aqueous medium, forming dissolved bicarbonate. Flushing the
system with water removes the bicarbonate while leaving the immobilised
enzyme behind [76]. A variation of this process involves the use of a liquid
membrane containing carbonic anhydrase aqueous solution to facilitate the
removal of carbon dioxide from an air stream passing through a hollow bre
membrane. The carbon dioxide is liberated from the liquid membrane under
reduced pressure in a continuous process [77].
376
W. Mazurek
A totally different approach to carbon dioxide removal has been proposed

by the French (Direction des Constructions Navales) where the gas removal
is integrated with the liquid oxygen heat exchanger in their MESMA (module
denergie sous-marine autonome) system involving the use of a closed cycle
Rankine turbine powered by ethanol. Moisture is rst removed from the air
stream by condensation, then the carbon dioxide is condensed (together with
air contaminants) at 150 C [59]. It has been estimated that carbon dioxide concentrations of < 0.7% could be achieved by this process. By utilising
the liquid oxygen heat exchanger, the electrical power consumption could be
limited to 1 kW. It is not known if this system is scheduled for installation in
the new French AIP submarines.
In the case of a disabled submarine, where there may be minimal electrical power, sodalime and lithium hydroxide will no doubt be retained as the
principal carbon dioxide absorbers. In recent years there has been a growing acceptance of the use of sorbent-lled permeable fabric curtains for
static removal of carbon dioxide, which rely on the natural convection of air
within the submarine and diffusion of air through the fabric and sorbent. The
curtains are pleated bags approximately 0.5 m wide and 1.5 m long with a capacity of 56 L of sorbent [78, 79]. They offer the advantage of not requiring
power together with a high sorption efciency, albeit at a much slower rate
than the more common technique of passing fan-forced air through a sorbent
bed. A signicant disadvantage is the degree of obstruction caused by the
deployment of large quantities of these curtains throughout the submarine.
In terms of removing carbon monoxide and hydrogen, low temperature
catalysts, such as platinum/palladium on tin oxide already in use in submarines, perform adequately with little or no power requirement. If the range
of these catalysts is to be extended to include the complete oxidation of
hydrocarbons (as in the nuclear-powered submarines) photocatalysis may
be required. Such lters have been proposed for use in commercial aircraft
cabin air purication [80]. The problem of dealing with the decomposition of
HCFCs arises again and it may therefore be expedient to use activated charcoal lters for the removal of VOCs.
Respirable particulates can be removed by the use of electrostatic precipitators with the advantage of low power requirement but the disadvantage of
producing ozone, leading to such problems as the yellow powder (see above).
High efciency particulate air (HEPA) lters offer considerable resistance to
air ow and require a larger investment of energy than the electrostatic precipitators. A large portion of the aerosols appear to be semi-volatile liquids,
which cannot be permanently trapped on HEPA lters whereas they can be
adequately removed by self-cleaning electrostatic precipitators.
377
6.2
Air Monitoring
There are currently two main approaches to real-time air monitoring in submarines. The US Navy CAMS has a proven track recorded of 30 years of
submarine service. A smaller version (Mini-CAMS), based on the CAMS II
and specically designed for conventional submarines was developed in the
early 1990s and tested by Italian, Swedish and British navies [56]. Like the
CAMS II, the major disadvantage of the system is the initial cost, the need for
long lengths of tubing for air sampling at various locations within the submarine (which may lead to some losses in the reactive gases) and the reliance on
one analyser for all gases.
A Fourier transform infrared analyser, analysing interferometer for ambient air (ANITA) has been proposed for use in submarines. This has emerged
from the European space program where there is a requirement for trace
gas monitoring in manned spacecraft. The instrument is able to identify and
quantify 32 contaminants simultaneously in real-time [81]. It is envisaged
that it would assume the role of a CAMS.
A more conservative approach is the use of dedicated sensors distributed
throughout the submarine. This requires infrared-based sensors which may
be used for carbon dioxide, carbon monoxide, VOCs, Freons and Halons with
Fig. 3 A Mini-CAMS (Hamilton Sundstrand, Land and Sea, USA), central atmosphere
monitoring system based on the CAMS II and developed for conventional submarines
378
W. Mazurek
electrochemical sensors for most of the other air contaminants. The major advantages of this system are in the redundancy and relatively low initial costs.
The greatest problem in using electrochemical sensors is cross-sensitivity,
particularly to hydrogen which can be present in concentrations greater than
500 ppm and in extreme cases up to 20 000 ppm. Other disadvantages include
the need for regular calibration and short life (< 2 years) of the electrochemical sensors.
As in the case of the nuclear-powered submarines, the long dive times of
AIP submarines will require whole air sampling and retrospective monitoring
for the purposes of comprehensive air analysis and as an insurance against
unexpected increases in air contaminants or unexpected species. This will involve a further investment in scientic resources in support of this increased
operational capability.
With these new developments in submarines, the problems encountered
with atmosphere control appear to have increasing relevance to manned
space travel at a time when space stations are being designed for long periods
of habitation. As a result there is growing cooperation between the two scientic communities working in these areas.
7
Conclusions
Submarines provided the earliest example of an articial enclosed working
environment. The successful evolution of this environment has required development of air purication technologies, air quality management strategies,
reliable air monitoring systems and health risk assessments to continuous
exposures of a wide range of air contaminants.
The advent of the nuclear-powered submarine ushered in a new era of
atmosphere control for submarines. Driven by strategic imperatives, it was
achieved through a massive investment in research. The result was a comprehensive system of air monitoring and air purication based on a toxicological
rationale which was well in advance of its civilian contemporaries. This was
carried out against a background of military conservatism.
The continuous evolution of submarines, with an increasing requirement
for long dive times, restricted internal space and limited power availability,
has placed increasing demands on these technologies. It remains to be seen
whether the introduction of AIP submarines, with their limitations in size
and power, will result in the necessary quantum leap in atmosphere control
technology, as was the case with the nuclear-powered submarines.
379
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
Compton-Hall R (1999) The submarine pioneers. Sutton, Gloucestershire, UK

Gray E (1996) Few survived. Leo Cooper, London, p 29
http://www.chinfo.navy.mil/navpalib/cno/n87/history/tech.html
Bowers P (1990) The Garrett enigma. Airlife, Shrewsbury, England
Compton-Hall R (1983) Submarine boats. Conway Maritime, London, p 124125
Preston A (2001) The Royal Navy submarine service. Conway Maritime, London, p 33,
161
Hay MF (1918) Secrets of the submarines. Skefngton, London, p 31
Lanigan RJ (1998) In: The epic story of the submarine Growler. RJL Express, Laurel,
FL, p 70
Thomas DA (1961) Submarine victory. Kimber, London, p 22
Gabler U (1986) Submarine design. Bernard and Graefe, Koblenz
Evans PJH (1997) Second international conference on submarine air monitoring and
air purication (SAMAP), Portsmouth, England
Warren CET, Benson J (1961) Will not we fear. Harrap, London, p 118
Bryant B (1958) One man band. Kimber, London, p 83, 103
Ellis FP (1950) Personnel research in the Royal Navy 19391945. Report RNP 50/580
Schratz PA (1988) Submarine commander. University Press of Kentucky
Boyd C, Yoshida A (1995) The Japanese submarine force and World War II. Naval
Institute Press, Annapolis, Maryland, USA, chap 1
Ellis FP (1947) Environmental factors affecting the health and efciency of personnel during prolonged periods submerged in submarines tted with the snort. Report
RNP 47/402. Royal Naval Personnel Research Committee, Medical Research Council
Tall JJ, Kemp P (1996) HM Submarines in camera. Sutton, Gloucestershire, UK, p 154
Lipscomb FW (1975) The British submarine. Conway Maritime, Greenwich
Parmeggiani L (ed) (1983) Encyclopedia of Occupational Health and Safety, vol 1, 3rd
edn. International Labour Ofce, Geneva, p 396
Pattle RE, Stretch H, Burgess F, Sinclair K, Edgington JAG (1956) The toxic action of
fumes from a diesel engine under four different running conditions. Technical paper no 535. UK Ministry of Supply, Directorate of Chemical Defence Research and
Development, Chemical Defence Experimental Establishment, Porton
Kemp P (1990) The T-class submarine. Naval Institute Press, Maryland, p 134
Schratz PR (1988) Submarine commander. University Press of Kentucky, Lexington
Boryta DA, Maas AJ (1971) Ind Eng Chem Process Des Devel 10:489
Rendell DJ, Clarke M (2003) Fourth international conference on submarine air monitoring and air purication (SAMAP), Emden, Germany
Clark AN (1990) J Nav Eng 32:215
Weast RC (ed) (1978) Handbook of chemistry and physics, 58th edn. CRC, Florida
Mars A (1956) HMS Thule intercepts. Elk, London, p 120
Cowan SL (1945) Carbon dioxide indicators, with special reference to the requirements of submarines. Report RNPL no 5/45. Royal Naval Physiological Laboratory
Compton-Hall R (1988) Submarine versus submarine. David and Charles, London,
p 25
Wyatt J (2001) Undersea Warfare 3:8
Panel on monitoring and panel on hyperbarics and mixtures, Subcommittee on Submarine Air Quality, Committee on Toxicology, Committee on Life Sciences, National
Research Council (1988) Monitoring the air in submarines, part 1. In: Submarine air
quality. National Academy Press, Washington, DC, chap 4
380
W. Mazurek
33. Evans PJH (2000) Third international workshop on submarine air monitoring and air
purication (SAMAP), Toronto, Canada
34. Dibben PR, Nicholson G (2000) Third international workshop on submarine air monitoring and air purication (SAMAP), Toronto, Canada
35. James JT (2000) Third international workshop on submarine air monitoring and air
36. Holden HJ, Brittain AJ, Dean MR (1997) Second international conference on submarine air monitoring and air purication (SAMAP), Portsmouth, UK
37. Bollan HR (2003) Fourth international conference on submarine air monitoring and
air purication (SAMAP), Emden, Germany
38. Coverdale A, Cassidy S (1987) J Nav Eng 30:528
39. Caplow M (1968) J Am Chem Soc 90:6795
40. Hook RJ (1997) Ind Eng. Chem. Res 36:1779
41. Miller RR, Piatt VR (1960) (eds) The present status of chemical research in atmosphere purication and control on nuclear submarines. NRL Report 5465. Naval
Research Laboratory, Washington DC
42. White JC (1967) Review of the electrolytic production of oxygen in nuclear-powered
submarines. In: Alexander AL, Piatt VR (eds) Fifth annual progress report: the
present status of chemical research in atmosphere purication and control on
nuclear-powered submarines. NRL Report 6491. Naval Research Laboratory, Washington DC
43. Dams R (1997) Second international conference on submarine air monitoring and air
purication (SAMAP), Portsmouth, UK
44. Rockwell T (1995) The Rickover effect. Wiley, New York, p 218
45. Shadle T, Daley T (1991) 21st International conference on environmental systems, San
Francisco, CA
46. Johnson JE, Christian JG (1960) A laboratory study of the efciency of Hopcalite in
catalysing hydrocarbon combustion. In: Miller RR, Piatt VR (eds) The present status of chemical research in atmosphere purication and control on nuclear-powered
submarines. NRL Report 5465. Naval Research Laboratory, Washington DC
47. Johnson JE (1967) Activated carbon-atmosphere purication and analysis. In: Alexander AL, Piatt VR (eds) Fifth annual progress report: the present status of chemical
research in atmosphere purication and control on nuclear-powered submarines.
NRL Report 6491. Naval Research Laboratory, Washington, DC
48. Goodall A, Moore S (2000) Third international workshop on submarine air monitoring and air purication (SAMAP), Toronto, Canada
49. Alexander AL, Platt VR (eds) (1967) Fifth annual progress report: the present status
of chemical research in atmosphere purication and control on nuclear submarines.
NRL Report 6491. Naval Research Laboratory, Washington, DC
50. Wyatt JR, Daley TJ, Wright KA (1997) 27th international conference on environmental
systems, Lake Tahoe, Nevada, USA
51. Callahan JH, Wyatt JR (2000) Third international workshop on submarine air monitoring and air purication (SAMAP), Toronto, Canada
52. Alexander WK, Still KR, Briggs GB (2000) Third international workshop on submarine air monitoring and air purication (SAMAP), Toronto, Canada
53. Thomas FS, Andreatech AJ, White JC (1956) Atmosphere analyser for atomic-powered
submarines. In: Report of progress PB 121.201. Naval Research Laboratory, Washington, DC
54. Wyatt JR (2000) Third international workshop on submarine air monitoring and air
purication, Toronto (SAMAP), Canada
381
55. Riches J, Beardmore D, Thompson P (1999) J Def Sci 4:70

56. Koslin ME, Cason RM (1994) Submarine air monitoring and air purication conference, Adelaide, Australia.
57. Kearn R, Bollan HR, Carp AD, Goodall AC, Ford J (2004) Humans in submarines,
Stockholm, Sweden, 1820 August 2004
58. Dibben P (1993) J Nav Sci 19:170
59. Bry-Peel AB, Kerros PK, Rouzies PR (1994) Submarine air monitoring and air purication conference/workshop (SAMAP), Adelaide, Australia
60. Ranieri B, Bettini A, Menghini A (2003) Fourth international conference on submarine air monitoring and air purication (SAMAP), Emden, Germany
61. Mazurek W, Gan TH, Hanhela PJ (2004) Manuscript in preparation
62. Mazurek W (1999) The oxidation of submarine air contaminants on the CO/H2 oxidation catalysts used in RAN submarines. Report DSTO-TR-0781. Defence Science and
Technology Organisation
63. Kah S, Fernando R (2000) Third international workshop on submarine air monitoring
and air purication (SAMAP), Toronto, Canada
64. Hanhela PJ, Gan TH, Mazurek W (2000) Third international workshop on submarine
air monitoring and air purication (SAMAP), Toronto, Canada
65. Bollan HR (1997) Submarine air monitoring and air purication conference,
Portsmouth, UK
66. Bollan HR, Brokenshire JL, Lunn MH (2000) Third international workshop on submarine air monitoring and air purication (SAMAP), Toronto, Canada
67. Gan TH, Corino G (2000) Anal Chem 72:807
68. Kane DM, Shergold DJ (1987) An air quality assessment onboard an Oberon class
submarine-HMCS Okanagan. Report DCIEM no 87-RR-28. Defence and Civil Institute of Environmental Medicine, Department of National Defence, Canada
69. Severs YD, Sabiston BH (2000) Third international workshop on submarine air monitoring and air purication (SAMAP), Toronto, Canada
70. Loncar M, Hanhela PJ, Gan TH, Mazurek W (1996) Air monitoring trials on HMAS
Ovens. Report DSTO-RR-0085. Defence Science and Technology, Australia
71. Earwicker S, Hanhela PJ, Denola G, Mazurek W (1998) 14th international clean air
and environment conference, Melbourne, Australia
72. Mazurek W, Gan TH, Gamble GI, Kennett SR (2004) Undersea defence technology
conference and exhibition (UDT Hawaii), Hawaii
73. Mazurek W, Gan TH, Hanhela PJ (2003) Fourth international conference on submarine air monitoring and air purication (SAMAP), Emden, Germany
74. Eaton D (2000) Third international workshop on submarine air monitoring and air
75. van der Gijp S, Albers J, van Es H, Feron P (2000) Third international workshop on
submarine air monitoring and air purication (SAMAP), Toronto, Canada
76. Johnson KG, Fradette S (2000) Third international workshop on submarine air monitoring and air purication, Toronto, Canada
77. Trachtenberg MC, Cowan RM, Goldman SL (2003) 33rd International conference on
environmental systems, Vancouver, BC, Canada
78. Horn WG, Benton PJ (2003) Fourth international conference on submarine air monitoring and air purication, Emden, Germany
79. Risberg J, Svensson T, Juvik T (2004) Humans in submarines, Stockholm, Sweden, 18
20 August 2004
80. Bull K (2003) Air quality in passenger aircraft conference, London, UK
382
W. Mazurek
81. Mosebach H, Kampf D, Honne A, Tan G (2003) Fourth international conference on

submarine air monitoring and air purication, Emden, Germany

DOI 10.1007/b107253
Air Quality Systems for Related Enclosed Spaces:

Spacecraft Air
L.M. Mulloth1 (u) J.E. Finn2
1 Science
Applications International Corporation, M/S 239-15,

NASA Ames Research Center, Moffett Field, CA 94035, USA
lmulloth@mail.arc.nasa.gov
2 Ion America Corporation, PO Box 97 NASA Research Park, Moffett Field, CA 94035,
USA
jnn@ionamerica.com
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
384
Environmental Control and Life-Support Systems for Spacecraft . . . . .
385
3
3.1
3.1.1
3.1.2
3.1.3
3.1.4
3.2
3.2.1
3.2.2
3.3
3.4
3.5
3.6
Elements of Air Quality Control in a Spacecraft

Carbon Dioxide Removal . . . . . . . . . . . . .
Physical Adsorption . . . . . . . . . . . . . . . .
Chemical Absorption . . . . . . . . . . . . . . . .
Biological Processes . . . . . . . . . . . . . . . .
Electrochemical Processes . . . . . . . . . . . . .
Carbon Dioxide Reduction . . . . . . . . . . . .
Sabatier Process . . . . . . . . . . . . . . . . . .
Bosch Process . . . . . . . . . . . . . . . . . . . .
Oxygen and Nitrogen Storage and Supply . . . .
Humidity and Temperature Control . . . . . . . .
Atmosphere Monitoring . . . . . . . . . . . . . .
Trace Contaminant Control . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
388
389
389
390
392
393
394
395
396
396
397
398
398
4
4.1
4.2
4.3
4.4
Current Methods and Systems . .

Carbon Dioxide Removal Assembly
Trace Contaminant Control System
CO2 Reduction . . . . . . . . . . .
Oxygen Generation . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
398
399
400
401
401
Air Revitalization Systems of the Future . . . . . . . . . . . . . . . . . . .
401
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
403
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Abstract Air quality in the small, closed environment of a spacecraft cabin is always
a critical matter for the safety, health, and comfort of the crew. The technologies used
to keep air breathable in spacecraft have a unique set of requirements because of several
constraints that become more important as the duration and distance of space missions
lengthen. Technologies must be extremely robust, as supplies and spare parts are few and
resupply may be impossible. They must be well coordinated and function in a tightly integrated life-support system. Mass, volume, and power consumption must be minimal due
to the high cost of launch mass and limited solar/battery energy.
384
L.M. Mulloth J.E. Finn
This article examines some of the issues associated with spacecraft air revitalization
and briey reviews some of the technologies developed to maintain quality and minimize waste through recycling of air. We emphasize approaches for long-duration missions
(i.e., more than one month), in which technologies need to be regenerable and the oxygen cycle needs to approach closure. We also discuss air revitalization systems for the
International Space Station and needs for long-distance missions such as Mars transit.
Keywords Life support Air revitalization CO2 removal Trace contaminant control
Abbreviations
4BMS Four-bed molecular sieve
CDRA Carbon dioxide removal assembly
CFU Colony forming units
ECLSS Environmental control and life support system
EDC Electrochemical depolarized cell
ISS
International Space Station
NASA National Aeronautics and Space Administration
SMAC Spacecraft maximum allowable concentration
TCCS Trace contaminant control system
1
Introduction
Modern manned spacecraft support a combination of habitat and laboratory
environments and provide comfortable living space as well as the resources
needed by the astronauts to perform scientic and engineering tasks. In addition to human support, the spacecraft must also be able to sustain plants
and animals for scientic experiments. The systems needed to make this life
support possible must not only be extremely reliable, but must have minimal mass, volume, power consumption, and heat rejection requirements due
to the cost associated with launch mass. Recovery of valuable consumables
from waste products becomes extremely critical as the duration of space missions increases. Among the subsystems present in a life-support system, the
air quality subsystem is the most critical, due to the sensitivity of the inhabitants and the small volume and nearly complete closure of the enclosed
space.
In a closed spacecraft cabin environment, the sources of air pollution
are numerous. Regardless of the length or nature of the mission, the major source of pollution inside a space cabin is the byproducts of human or
animal metabolism. Continuous operation of instruments or machinery for
scientic research and life-support systems cause additional loads on the
air revitalization system through chemical leaks into the cabin environment.
Potential sources of contaminants also include re suppressants, material out-
Air Quality Systems for Related Enclosed Spaces: Spacecraft Air
385
gassing, chemical spills, and products of scientic experiments and chemical

processes. The complexity and the amount of the equipment and instrumentation for air revitalization inside a spacecraft cabin vary with factors such
as the crew size, duration, and objectives of the mission. As the duration of
the mission increases, the need for storing and recycling of food and waste
increases and, correspondingly, the contaminant load increases due to added
processors, growth of microorganisms, etc.
Fundamental differences between the space and Earth environments include lack of gravity and atmosphere. Equipment and process selection for
air pollution control for the space cabin is made more complex because of
gravity-related issues. Terrestrial systems have the advantage of proximity
to basic consumables such as air and water. The low-pressure or vacuum
environment surrounding the spacecraft increases the potential of loss of
atmosphere through leakage. Oxygen and nitrogen have to be carried or produced onboard the spacecraft for making up the loss through respiration
and leakage. Unlike air quality issues in buildings and airplanes, the air inside the aircraft has to be scrubbed of airborne contaminants and recycled
back to the cabin. From this perspective, spacecraft air quality control processes share some similarities to that of submarines. However, submarines
have ready availability of water (which can be electrolyzed to produce oxygen)
and high-pressure surroundings that limit the potential leakage of air. Also,
power is usually more readily available than in spacecraft, which typically rely
on photovoltaic power.
This article presents a broad picture of air revitalization systems and issues
in spacecraft. We focus on technology choices for carbon dioxide removal,
the most critical and resource-consuming task, then discuss other air revitalization tasks including carbon dioxide reduction, oxygen generation, trace
contaminant control, and other subsystems. Finally, we briey describe systems for the International Space Station, and some directions that future air
revitalization systems can evolve.
2
Environmental Control and Life-Support Systems for Spacecraft
The basic objective of an environmental control and life-support system
(ECLSS) of a spacecraft or a space habitat is to provide a safe and comfortable
living environment for the crew. Subsystems of a spacecraft ECLSS include the
following [1, 2]:
Atmosphere revitalization
Atmosphere control and supply
Temperature and humidity control
Water recovery and management
386
Waste management
Fire detection and suppression
Food management
Human hygiene
Integrated system management and control
Approximate durations for which humans can live without oxygen, water,
and food are 4 min, 3 days, and 30 days, respectively [3]. Obviously, atmosphere revitalization and atmosphere control and supply are the most critical
ECLSS elements for a spacecraft, regardless of the type or duration of the mission. Air quality specications developed by NASA for astronauts are shown
in Table 1.
All elements of the ECLSS inuence the air quality control in a spacecraft. The interdependency of these elements becomes more prominent for
air quality control as the ECLSS becomes self-sustaining with closed-loop
technologies which recycle water and solid waste as well as air. For example,
incinerators may be used for recycling of solid waste. The byproducts of incineration will enter the air purication stream [5]. The types of gas-phase
contaminants produced by solid waste oxidation will depend on the process,
but will probably be far more concentrated than the contaminants present in
the breathing air.
Table 1 Typical air quality specications for US spacecraft [4]
Parameter
Operational
Specications
90-Day
degraded
28-Day
degraded
CO2 partial
pressure
Oxygen partial
pressure
Microorganisms
Dewpoint
0.4 kPa
(3 mm Hg)
1922 kPa
(2.73.2 psi)
500 CFU/m3
4.416 C
(4060 F)
1821 C
(6570 F)
70.3101 kPa
(10.214.7 psi)
1.0 kPa
(7.6 mm Hg)
1626 kPa
(2.43.8 psi)
750 CFU/m3
1.721 C
(3570 F)
1629 C
(6085 F)
70.3101 kPa
(10.214.7 psi)
1.6 kPa
(12 mm Hg)
1627 kPa
(2.33.9 psi)
1000 CFU/m3
1.721 C
(3570 F)
4.432 C
(6090 F)
70.3101 kPa
(10.214.7 psi)
Temperature
Pressure
Operational refers to an extended duration mission, length 90 days or more. Degraded refers to a reduced standard allowable for missions of shorter duration.
Fractional/English units (shown in parentheses) are the basis for the specications. CFU
colony-forming units, psi pounds per square inch (absolute)
387
Typically, for short-duration missions that last 15 days or less, only the
basic atmosphere revitalization functions are necessary to maintain life support. Some examples of short-duration missions are shown in Table 2. For
such missions, it is practical and cost-effective to carry onboard enough lifesupport consumables such as oxygen, nitrogen, and water to last for the entire
mission. For instance, the CO2 removal process design for air revitalization in
short-duration missions is generally open-loop: CO2 will be continuously removed from the cabin air using a chemical or physical-chemical method, but
the process may not be reversible to collect the waste CO2 to recover oxygen.
Recovery of consumables from the waste products is not critical since they
can be discarded to space or stored to bring back to the Earth. Control of trace
contaminants and microorganisms and water recovery are also less important
for such missions.
The scope of human spaceight has changed tremendously since the rst
human space ight of Gagarin in April 1961. The advances in ECLSS technologies have progressed from merely keeping one man alive in space for few
minutes to providing continuous life support for multiple crew-members for
several years, and not only to survive, but also to do complex experiments,
construction, and maintenance in space.
Ideally, a spacecraft ECLSS design should be reliable and require minimal
maintenance. It should also support scientic research and extravehicular
activities, and provide redundancy/backup capabilities to deal with unforeseen situations. The ECLSS of the International Space Station (ISS) represents
the most advanced and current scenario. The habitable volume of the ISS
is 425 m3 compared to 3 m3 for the Vostok that carried Yuri Gagarin to
space [6]. While capable of handling up to nine crewmembers at a time, the
ISS has been supporting two to three crewmembers since it was established
in 1998 [7, 8]. Air quality specications for the ISS are shown in Table 3.
Long-duration human missions of the future, to destinations such as Mars,
will make Earth-based resupply of consumables impractical due to the increased mass, volume, and storage requirement and corresponding increase
in launch cost. For such missions, it is imperative to implement regenerable,
Table 2 Examples of short-duration space missions. A detailed description of the history
of air revitalization systems used in human spaceight is available in reference [3]
Flight
Duration
Year
Cabin
volume
Vostok
Mercury
Gemini
Apollo
108 min
15 min to 34 h 20 min
4 h to 13 days
512 days
19601963
19601963
19641966
19681972
3 m3
1.56 m3
2.26 m3
5.9 m3
388
Table 3 Some basic air quality specications developed by NASA for the International
Space Station [9]
Parameter
Specication
Atmospheric pressure
Minimum allowable pressure
Nitrogen partial pressure
Oxygen partial pressure
Carbon dioxide partial pressure
Carbon dioxide peak partial pressure
92.9103 kPa (14.214.9 psi)

95.8 kPa (13.9 psi)
below 80.0 kPa (11.6 psi)
19.523.1 kPa (2.833.35 psi)
0.71 kPa (daily average) (5.3 mm Hg)
not to exceed 1.0 kPa (7.6 mm Hg)
Fractional/English units (shown in parentheses) are the basis for the specications. psi
pounds per square inch (absolute)
closed-loop air revitalization system technologies that recycle consumables

such as oxygen and water and maintain balance between the supplies and demands of air, water, and food. The ISS ECLSS has the reputation of supporting
humans for long durations, however, it does not have recycling capabilities
for its air and water purication technologies. Consumables such as oxygen,
water, and other accessories for life-support machinery are supplied from
Earth [10].
Potential scenarios of spacecraft or space habitats for the next few decades
include launch and transit vehicles, crew exploration vehicles (e.g., rovers),
and surface habitats on planetary surfaces [11]. The ECLSS system for each
scenario will be different. The ECLSS of a crew exploration vehicle to transport the crewmembers from the Earth to Moon may not require a closedloop conguration since the crew will have a short stay in the spacecraft.
On the other hand, for a space transportation vehicle such as a Mars transit vehicle, where the travel time may be more than a year, near closure of
life-support systems is essential. For surface habitats such as Mars or Moon
outposts, provisions for the utilization of local resources such as oxygen or
water generation from atmosphere and soil may be required in addition to the
closed-loop life support.
3
Elements of Air Quality Control in a Spacecraft
The atmosphere of a spacecraft is controlled by a set of cooperating equipment assemblies performing various tasks. As discussed previously, the list
of necessary tasks depend on the mission. Those that most directly affect air
quality include the following:
389
CO2 removal
CO2 reduction
Oxygen and nitrogen supply
Trace contaminant control
Temperature and humidity control
Atmosphere monitoring
Microorganism control
3.1
Carbon Dioxide Removal
Carbon dioxide is produced through human and animal metabolic activities.
The average rate of human CO2 production due to normal metabolic respiration is approximately 1 kg/day [3]. Because carbon dioxide is used as a re
suppressant, a sudden increase in the carbon dioxide concentration can result
from the response to a re.
Continuous removal of carbon dioxide is essential for maintaining the
air quality standards inside the cabin for both short and long-duration missions. CO2 removal processors are generally among the largest and most
power-consuming air revitalization processors. A variety of methods have
been developed to some degree for CO2 removal from breathing air in a space
cabin. This section focuses on some of the more prominent ones, including
physical adsorption, chemical absorption, electrochemical means, and biological conversion.
3.1.1
Physical Adsorption
Gas separation utilizing aluminosilicate zeolite adsorbents offers a fully reversible technique for CO2 removal. The polar zeolite selectively adsorbs CO2
from air. However, water vapor preferentially adsorbs on zeolites and drastically reduces their capacity for CO2 . Thus removal of water vapor from
the process air stream is necessary prior to CO2 removal, resulting in substantially increased power, mass, and volume requirements. Regeneration of
CO2 sorbents can be achieved by applying vacuum, heat, or a combination
of both. Thermal regeneration becomes very energy intensive, particularly
in a closed-loop system, where CO2 and water need to be recovered and
recycled.
Overall, physical adsorption is a reliable process and is highly suitable
for air revitalization in a spacecraft where separation of low-partial pressure
components are involved. Adsorption processors provide excellent performance with high reliability and long life due to little or no moving parts, and
have almost no microgravity related issues. Carbon dioxide can be produced
from such a processor at relatively high purity, allowing easy integration
390
with a carbon dioxide reduction system for oxygen recovery. The US carbon
dioxide removal assembly (CDRA) on the ISS is based on this CO2 removal
technique [3].
3.1.2
Chemical Absorption
Chemical absorption processes generally use a medium that reacts with CO2
in the air, often in the presence of water vapor. Two processes, solid amine
and metal oxide, are highly reversible and are the best candidates for longduration missions. Metal hydroxides and superoxides are also discussed because of their continued use for many spacecraft applications.
3.1.2.1
Solid Amines
Amines are excellent solvents for extraction of gaseous carbon dioxide in
many terrestrial applications. However, CO2 removal using liquid amines
is difcult in space due to the microgravity environment. Solid amines
address this limitation. Solid amines are microporous beads of polymeric
acrylic esters or similar materials whose internal surfaces are coated with an
amine such as polyethylenimine. In the presence of moisture, the resin absorbs CO2 , which can be easily reversed with the application of heat.
One of the more serious problems associated with this sorbent is degradation and ammonia production assumed to occur due to the cross-linking
of the polymer. Ammonia is also an irritant to the inhabitants of the cabin.
Solid amine systems with relatively low regeneration temperatures have been
demonstrated in recent studies and are a promising alternative to the current
zeolite adsorbent system for CO2 removal [12]. Astrium incorporated a solid
amine-based CO2 removal device in an Air Revitalization System Demonstrator tested in closed-chamber ground studies in the late 1990s [13]. NASA has
qualied a two-bed solid amine system for Extended Duration Orbiter (Space
Shuttle) ights [14].
3.1.2.2
Metal Oxides
Oxides and hydroxides of some metals are capable of absorbing carbon dioxide based on Eq. 1:
MO + CO2 MCO3
(1)
In many cases this reaction can be reversed thermally and thus regenerate
metal oxide, producing CO2 . The CO2 absorption and regeneration capacity
varies with each metal oxide. In a study conducted by McDonnell Douglas As-
391
tronautics, silver oxide was found to have the best rate of CO2 absorption and
regenerability compared to the other metal oxides [15].
Metal oxides absorb both water and carbon dioxide, so simultaneous removal of water vapor and CO2 are possible using this reagent. A study conducted by Allied Signal Aerospace demonstrated that a mixed metal oxide
formulation of silver and zinc oxide is a better CO2 absorbent than silver
oxide alone. Regeneration of the metal oxide and water was performed at
63 C. Regeneration of adsorbent from the carbonate to the oxide state is
achieved at 250 C [1618].
3.1.2.3
Alkali Metal Hydroxides
Developed for space applications in the 1950s, LiOH is the most widely used
alkali metal hydroxide. Many US and Soviet space ights have utilized LiOH
for CO2 removal in the space cabin and for extravehicular activities. Its high
capacity and simplicity of operation makes it attractive for spacecraft CO2 removal. Except for Skylab, all US space missions used lithium hydroxide [19].
CO2 removal by lithium hydroxide, however, is a non-regenerative process.
The reaction mechanism of lithium hydroxide with carbon dioxide is given
by Eq. 2:
2 LiOH H2 O + CO2 Li2 CO3 + 3 H2 O
(2)
Since intermediate formation of lithium hydroxide monohydrate is necessary

for the absorption of CO2 by LiOH, its capacity depends on the relative humidity of the feed air stream. For high absorption, the water content of the
inlet stream should be such that the rate of hydration does not exceed the rate
of carbonation. The effective moisture content at 25 C is found to be a water
partial pressure range of 0.933.33 kPa (725 mm Hg) [19].
Apart from its weight and non-regenerative nature, lithium hydroxide has
some additional limitations. The absorption process is strongly exothermic,
and generates 1422 kcal/mole of CO2 . This can produce temperatures high
enough to dissociate the lithium hydroxide monohydrate complex. In closed
cabin units, a moist air-carbon dioxide stream may be sufcient to cool it to
the desired level. Moisture load produced by the reaction (Eq. 2), probable
dust formation due to its granular nature, etc., are other undesirable features
of this reagent. Due to its high dependency on an optimum amount of moisture content in the feed, humidity control systems have to be included in case
of very high or very low feed streams [19].
A mixture of barium and calcium hydroxide, widely known as Baralyme,
is a popular carbon dioxide removal agent in submersibles and diving rebreathers, which has been studied for spacecraft CO2 removal. Sodasorb,
another popular CO2 absorbent consists of hydrated lime Ca(OH)2 , sodium
hydroxide (NaOH), potassium hydroxide (KOH), and 1419% moisture.
392
3.1.2.4
Alkali Metal Superoxides
Alkali metal superoxides serve the dual purpose of removing CO2 and providing oxygen to a life-support system. The reaction between carbon dioxide and
alkali metal super oxides in the presence of moisture generates oxygen. Thus,
a single bed of reagent can scrub CO2 and provide oxygen simultaneously.
The general reactions involved are given by Eqs. 35:
2 MO2 + H2 O 2 MOH + 1.5 O2
2 MOH(s) + CO2 M2 CO3 + H2 O
MOH + CO2 MHCO3
(3)
(4)
(5)
Many USSR spacecraft successfully used alkali metal super oxides for CO2
removal from the spacecraft atmosphere. Early Soviet missions such as Vostok, Voskhod, and Soyuz used a combination of potassium superoxide and
lithium hydroxide for CO2 removal. Since the oxygen formation and CO2 removal is dependent on the moisture content of air, these systems had some
difculty in controlling the partial pressure of cabin oxygen. Though lithium
superoxide is most effective in removing CO2 , it has the disadvantage of
being unstable at room temperature. The theoretical CO2 removal and O2
production capacity of the potassium superoxide is 0.309 kg and 0.388 kg respectively, per kg of sorbent. While being excellent CO2 removal agents and
oxygen producers, the CO2 scrubbing reactions of superoxides are exothermic, which may elevate the ambient temperature [20].
A large number of additional compounds have been studied for potential use in spacecraft, especially for simultaneous generation of oxygen
with removal of CO2 during extravehicular activity. Lithium peroxide reacts
with carbon dioxide in the presence of moisture to produce oxygen. Metal
ozonides (MO3 ) release even more oxygen on reaction with CO2 , but have low
stability.
3.1.3
Biological Processes
Plants, through photosynthesis, perform the opposite of the human metabolic
process. The human metabolic process can be represented roughly as in Eq. 6:
Cn H2n On + n O2 n CO2 + n H2 O + energy
(6)
Humans consume food (e.g., carbohydrate), use oxygen through respiration, and oxidize the food to generate metabolic energy. The photosynthesis
process of plants to produce carbohydrate is essentially the reverse reaction
393
as in Eq. 7:
n CO2 + 2n H2 O + energy (light) Cn H2n On + n O2 + n H2 O .
(7)
Equations 6 and 7 show a simple stoichiometry of how simultaneous CO2

removal and O2 replenishment can be achieved by plants in a closed lifesupport system.
The photosynthetic cycle of green algae can been exploited for concurrent oxygen production and carbon dioxide removal. An algae suspension
of Sorokin strain of Chlorella pyrenoidosa has been studied for this purpose [21]. An experimental column of algae suspension in water was found
to achieve complete removal of carbon dioxide from a 5% carbon dioxide-air
mixture and produced about 1500 cm3 /h of oxygen. Algae culture has an optimum temperature range between 3739 C (98102 F) for growth. The major disadvantage of the algae system is the high power requirement. Absorption of CO2 and O2 gas mixture into a Chlorella suspension and subsequent
stripping of the O2 gas generated by photosynthesis, using a hydrophobic
microporous hollow ber membrane made of polypropylene, has also been
investigated. Recent studies using membrane-photobioreactor using Chlorella
have shown promising results [21].
Among bioregenerative systems, plants have advantages over algae, since
plants are a more probable food source and less prone to water and nutrient
stress than algae. It is also difcult to achieve long-term, sustained production of algae systems. Algae systems must be kept in water, and managing
gas and liquid interfaces in microgravity conditions is a challenge. Biological processes in general require large volume and rely heavily on engineering
systems that require sensitive control systems for survival.
3.1.4
Electrochemical Processes
Electrochemical CO2 removal processes effectively concentrate CO2 by transforming it into the carbonate ion and pumping it through an electrolytic
membrane. The process can be performed either in the oxygen generation or
power generation modes (regenerative fuel cells). For the oxygen generation
mode, power is supplied and an inert gas or nitrogen is supplied to the anode,
and the cell works as a reverse fuel cell to produce oxygen. For the power generation mode, hydrogen is supplied to the anode as fuel while CO2 -containing
air is supplied to the cathode in the hydrogen mode operation, and power is
produced as a by-product.
A number of variations of the CO2 electrolysis cells have been investigated.
We focus on two, the electrochemical depolarized cell and the molten carbonate cell. Carbonation cells, solid state electrodialysis, and electroactive
organo-metal complexes have also been or are being investigated.
394
3.1.4.1
Electrochemical Depolarized Cell
An electrochemical depolarized cell (EDC) consists of a series of cells which
transfer carbon dioxide from the spent air to a stream of hydrogen. Each cell
consists of a cathode, anode, and a porous matrix. A suitable electrolyte such
as cesium carbonate (Cs2 CO3 ) held by the porous matrix acts as the ionic
carrier for CO2 from cathode to anode [22].
Air containing carbon dioxide is applied to the cathode, which initiates the
reactions shown in Eqs. 89:
O2 + 2 H2 O + 4 e 4 OH
2 CO2 + 4 OH 2 H2 O + 2 CO3 =
(8)
(9)
The anode reactions are shown in Eqs. 1011:

2 H2 + 4 OH 4 H2 O + 4 e
2 CO3 = + 2 H2 O 4 OH + 2 CO2
(10)
(11)
The overall reaction is given by Eq. 12, where the products are carbon dioxide and electricity:
2 CO2 + O2 + 2 H2 2 CO2 + 2 H2 O + 4 e
(12)
Electrochemical concentrators can be operated continuously without a desorbing cycle for a wide range of CO2 partial pressures and require less
equipment than adsorption and absorption processes. Chief disadvantages
include the use of ammable hydrogen and the possibility of electrolyte dryout.
3.1.4.2
Molten Carbonate CO2 Concentrator
This unit uses a molten salt as an electrolyte, compared to the EDC use
of an aqueous alkaline electrolyte. Molten lithium and potassium carbonate (at a ratio 62 : 38) has been used as the electrolyte, which is retained on
a porous and non-conductive matrix such as lithium aluminate. CO2 -laden
air enters the cathode and hydrogen or an inert gas is introduced into the
anode depending on the mode of operation. Molten carbonate concentrators
have disadvantages of a corrosive electrolyte and high operating temperature
(> 600 C) [19].
395
3.2
Carbon Dioxide Reduction
In order to minimize the oxygen loss, the waste CO2 that is removed from
breathing air can be treated in many ways to recover the oxygen. For a shortduration mission the economics may be in favor of carrying oxygen for
resupply and venting waste CO2 to space, considering the costs associated
with building, launching, and maintaining additional machinery for CO2 reduction. However, for long-duration missions, near-closed-loop systems and
hence O2 recovery from CO2 are essential. Even for missions to low Earth orbit where continuous human presence is established, such as in the ISS, O2
recovery can be quite worthwhile as it minimizes the amount of water that
needs to be resupplied.
Not all CO2 necessarily needs to be reduced. When a life support system is
open-loop with respect to food (as they are until bioregenerative systems become practical), then the waste product from the system should approximate
the elemental makeup of food. To illustrate, if food is very roughly considered
to be a carbohydrate (Cn H2n On ), the waste product might be a combination of
CO2 and CH4 (e.g., n/2CO2 + n/2CH4 is equivalent to Cn H2n On ) [23].
Practical and well-characterized methods of CO2 reduction include the
Sabatier and Bosch reactions. Both reactions produce water, which can be
electrolyzed to produce oxygen and hydrogen for recycle. These processes
function best when the CO2 content of the feed stream is high, which can
drive the choice of CO2 removal equipment. Some processes, such as physical
adsorption, can produce CO2 of the required purity directly, requiring only
gas compression as an interface to the CO2 reduction unit.
3.2.1
Sabatier Process
Carbon dioxide reacts with H2 in the presence of a catalyst such as ruthenium
to produce water and methane according to the exothermic Sabatier reaction,
Eq. 13:
CO2 + 4 H2 CH4 + 2 H2 O
(13)
Water is separated and pumped to an electrolyzer, from which hydrogen is recycled. Methane is often considered a waste gas; depending on the system it
might be vented or pyrolyzed to produce hydrogen.
396
3.2.2
Bosch Process
CO2 reacts with H2 in the presence of iron, nickel, or ruthenium catalysts by
the exothermic Bosch reaction, Eq. 14:
CO2 + 2 H2 2 H2 O + C (solid) + heat
(14)
Carbon formation is a disadvantage for the Bosch reactor for space systems
since the catalyst bed/cartridge lls with low-density carbon and has to be
replaced periodically.
3.3
Oxygen and Nitrogen Storage and Supply
Oxygen is consumed by humans and animals through respiration, and is also
lost from the cabin through leakage. Nitrogen is used as a buffer gas. It is necessary to replenish the atmosphere lost through leakage and airlock use to
maintain the appropriate partial pressures of oxygen and nitrogen inside the
spacecraft.
Sources of oxygen in a spacecraft include water electrolysis, carbon dioxide electrolysis, high pressure or cryogenic storage, and oxygen-generating
materials such as potassium superoxide (KO2 ), sodium chlorate candle, and
lithium perchlorate.
Water electrolysis is currently the most practical method for long duration
missions. Two methods have been developed to a high degree of technology
maturity, namely solid polymer electrolyte and static feed water electrolysis.
The solid polymer technique uses a proton exchange membrane such as
hydrated peruorinated sulfonic acid polymer as an electrolyte, whereas the
static feed technique uses an alkaline (KOH) electrolyte. The alkaline electrolyte is susceptible to contamination by soluble species such as CO2 . In
order to mitigate contamination of the electrolyte, the device evaporates water from a matrix and condenses it on the alkaline electrolyte surface.
Carbon dioxide can be electrolyzed directly to produce oxygen using
a solid oxide electrolyte such as yttria-stabilized zirconia. The process has
the advantage of not requiring a CO2 reduction step, and can also electrolyze
water simultaneously [24]. The process has not yet been developed for lifesupport applications because of some challenging design issues. It operates
at high temperature (typically 800 C or higher) and CO is generated as
a byproduct.
Sodium and lithium chlorate candles are commonly used for short-term
oxygen production. On decomposition (with heat) they product oxygen and
the chloride salt. For example (Eq. 15),
2 NaClO3 2 NaCl + 3 O2 + heat
(15)
397
Sodium chlorate is stable at room temperature and decomposes at about

500 C. Once the chlorate candle is heated to decomposition temperature, it
can sustain the reaction to completion. An appropriate housing must be implemented for addressing the ammability issues.
Potassium superoxide, described in a previous section, is another oxygenproducing salt. It generates oxygen on reaction with carbon dioxide.
3.4
Humidity and Temperature Control
Humidity control inside the cabin depends on a number of parameters. These
include the rate of water vapor release into the cabin from various sources,
operation patterns, mechanisms of water and condensate processing equipment, leak rate of cabin atmosphere, etc. Gas purication systems such as
pressure and temperature swing adsorption columns often capture small
amounts of water vapor which will be lost to space during the regeneration
process. The main sources of water vapor in the cabin are human and animal metabolic processes such as respiration and perspiration. The metabolic
process results in more water output than input. While most carbohydrates
contain oxygen and hydrogen in the right proportion to produce water directly, the excess hydrogen compared to oxygen in fats and proteins combines
with respiratory oxygen to produce excess water. Apart from metabolic processes, water vapor enters the cabin environment due to evaporation from
various sources including personal hygiene activities (washing, showering,
laundry, etc), exposed food and drinks, and spills. Water vapor removal is
essential for maintaining the appropriate temperature and humidity for the
crew.
Humidity control inside the cabin is achieved through the following
methods:
1. Condensation of moisture from the atmosphere followed by phase separation
2. Drying the atmosphere using a desiccant
The condensation method is generally preferred since recovery of the removed water vapor from the desiccants is an energy-intensive process. The
condensing heat exchangers inside space habitats have a specially designed
vapor-liquid separators that function in a microgravity environment. Modern space habitats use a hydrophilic coating on the walls of the vapor-liquid
separator, which allows water to adhere to the surface.
398
3.5
Atmosphere Monitoring
Monitoring of gases in the spacecraft atmosphere is a key part of air quality control in a spacecraft. Partial pressures of gases such as N2 and O2 must
be maintained through supply while partial pressures of gases like CO2 and
water vapor should be maintained through removal processes. It is also important to continuously monitor the presence of potentially hazardous gases
such as H2 , CH4 , and CO.
3.6
Trace Contaminant Control
Trace contaminants enter a space cabin through human, animal, and microbial metabolism, off-gassing of hardware and construction materials, accidental release of experimental chemicals, and leakage of vapors and gases
during equipment repair. The living area inside the spacecraft has a limited
volume and accumulation of these contaminants can reach hazardous levels. Trace contaminants must be removed to preserve the health and comfort
of the crew and the success of the mission, which may include life sciences
experiments whose results could be confounded by having too-large contaminant concentrations.
The spacecraft maximum allowable concentration (SMAC), for many potential contaminants is different from the industry standard since the crew
has a potential of continuous exposure to these chemicals. NASA has developed a list of contaminants based on likely sources, releases, and historical
data [25].
Adsorption on activated carbon is effective for high molecular weight compounds and is potentially regenerable by various means, including heat and
exposure to vacuum. It can be impregnated with phosphoric acid and other
additives to target specic compounds such as ammonia. Many important
contaminants such carbon monoxide, methane, and organic halides quickly
break through carbon beds; for these an oxidation catalyst (such as noble
metal on alumina) is effective. Acid gases produced in the catalytic oxidizer
can be removed by an alkaline scrubber, containing LiOH for example [22].
4
Current Methods and Systems
Although most of their life-support technologies were developed years ago,
the Space Shuttle and the ISS today represent state-of-the-art manned spacecraft (Table 4). The Space Shuttle air revitalization system design is based
on a short-term ight conguration. It typically uses open-loop technolo-
399
Table 4 Technologies for the main air revitalization tasks on Space Shuttle and ISS
AR task
Space Shuttle
ISS
CO2 removal
LiOH chemical
absorption
CO2 reduction
None
O2 generation
Trace contaminant
control
Cryogenic supply
Activated carbon
adsorption
Zeolite physisorption, other

regenerable sorption
processes
None at present,
Sabatier possible
in the future
Water electrolysis
Activated carbon adsorption,
catalytic oxidation
gies such as lithium hydroxide CO2 removal for air revitalization. Water and
CO2 reduction/oxygen generation is not critical for the Space Shuttle ights
because these are relatively short duration.
Technologies were selected for the ISS following a rigorous set of simulated
ight tests [3]. The ISS also uses proprietary Russian technology extensively.
Examples of technologies in use today for the ISS are presented below.
4.1
Carbon Dioxide Removal Assembly
The carbon dioxide removal assembly (CDRA) for the ISS US segment is
based on selective physisorption of carbon dioxide. Two packed beds containing CaA zeolite alternate in removing CO2 from the process air stream and
regenerating via CO2 desorption and venting to space vacuum [26]. Because
the capacity of CaA zeolite for CO2 is greatly diminished by adsorbed water
vapor, the sorbent beds are preceded by two packed desiccant beds containing
layers of NaX zeolite and silica gel. These beds also cycle between desiccation of the process stream and regeneration by warm processed air, in which
the desorbing water vapor is carried back into the cabin. The four beds in the
processor give rise to its name four-bed molecular sieve or 4BMS (Fig. 1).
The CDRA draws humid air from the cabin through a temperature and
humidity control system to bring down the temperature of the air stream to
about 4.5 C. The saturated air is dried to approximately 60 C dew point
in the desiccant section of the 4BMS assembly. The dry air is then cooled in
a heat exchanger before entering the CO2 removal section of the system.
The cool air entering the regenerated, hot bed cools the sorbent and promotes CO2 adsorption. The hot, dry, CO2 -free air leaving the sorbent bed is
directed through the desiccant beds, where it regenerates the desiccant beds
by sweeping the moisture back to the cabin environment.
400
Fig. 1 Flow diagram of the four-bed molecular sieve carbon dioxide removal assembly installed in the US segment of the ISS. In the second half of the cycle, adsorbing and
desorbing beds switch
In the current design, regeneration of the spent CO2 sorbent bed is

achieved by thermal and vacuum regeneration. The regenerated CO2 is
vented to space, which ultimately results in a loss of oxygen and must be made
up by through resupply of water amounting to nearly 1000 kg/year. NASA
may install a Sabatier reactor in the air revitalization loop for CO2 reduction,
water production, and subsequent oxygen recovery through electrolysis.
The Russian segments carbon dioxide removal system (Vozdukh) uses
a solid-regenerable sorbent that adsorbs carbon dioxide and water vapor, and
is regenerated by exposure to vacuum. Similar to CDRA, the water vapor is
removed from the air using a desiccant bed located upstream of the CO2 sorbent. The desiccant bed is regenerated by heat and dry air.
4.2
Trace Contaminant Control System
The current Trace Contaminant Control Subassembly (TCCS) for the US segment of the ISS is based on a design developed for Apollo in 1960 by Lockheed
Missiles and Space Company (Fig. 2) [27, 28]. There are three primary components: a phosphoric acid-impregnated activated carbon adsorbent canister,
a high temperature catalytic oxidizer, and a LiOH scrubbing bed. The cabin
air ows through the activated carbon column where contaminants such as
ammonia and high molecular weight hydrocarbons are removed. Low molecular weight hydrocarbons such as methane, hydrogen, carbon monoxide,
and formaldehyde that adsorb poorly or are displaced by water or other contaminants are removed in a thermal catalytic oxidizer downstream of the
carbon bed. Acid gases such as nitrogen oxides, sulfur oxides, and hydrogen
halides are removed by the LiOH scrubber [29, 30].
401
Fig. 2 Flow diagram of the trace contaminant control assembly installed in the US segment of the ISS
Although activated carbon can be regenerated thermally by exposure to

vacuum or by water vapor displacement [31], the carbon is not regenerated
on the ISS and is a resupply item.
The Russian segments trace contaminant system contains two regenerable
charcoal beds to remove low molecular weight contaminants, an expendable
charcol bed for the removal of larger, less volatile molecules, and a roomtemperature catalytic oxidizer to remove carbon monoxide and hydrogen.
4.3
CO2 Reduction
The ISS presently does not have CO2 reduction capabilities. However,
a Sabatier reactor is being developed for possible future installation [32].
4.4
Oxygen Generation
Oxygen for the ISS is generated via water electrolysis in the Russian segment.
Oxygen generation via solid polymer electrolysis may be provided in the US
segment at a future date. Water is provided by the water purication system
and from supplemental water [33].
5
Air Revitalization Systems of the Future
The capabilities of current air revitalization systems are inadequate for future
human space initiatives that involve long-duration space voyages and interplanetary missions, because of their reliance on resupply of consumables such
as water, stored gases, and expendable sorbents. Carrying consumables and
expendables aboard a long-duration mission is expensive from the standpoint
of launch costs and also may be less safe (if, for example, a critical supply
402
runs out). Future systems should recycle and reuse the consumables to the
largest extent possible and should minimize the mass, volume, and power
requirements to make long-duration human space ights a reality [34].
CO2 removal is one of the most energy-intensive processes in the ISS lifesupport scheme because of the thermal energy required for both CO2 and
water removal [35]. Alternate, lower power CO2 -removal technologies involving membranes and new sorbents are in development.
Current TCCS carbon beds are expendable. Regenerable adsorbents are
available and can be utilized for trace contaminant control. Many processors
of the current air revitalization system function independently.
Future systems will implement thermal integration of the processors
(Fig. 3). Current systems use the packed bed conguration for adsorption
beds, which has inherent disadvantages such as high pressure-drop, and low
heat and mass transfer characteristics. Adsorbents immobilized on lightweight, thermally conductive substrates have been developed and have been
widely utilized for terrestrial applications [36].
Fig. 3 Flow diagram of a proposed, integrated carbon dioxide and trace contaminant control system. The membrane dryer minimizes the energy consumption and size of the
desiccant beds associated with 4BMS technology. The size of the non-regenerable carbon
bed is signicantly smaller than current technology due to the use of thermal/vacuumregenerated contaminant removal columns
403
References
1. Bertoto D, Brambatti G, Valenzano G (2003) International conference on environmental systems, 2003-01-2515
2. Volosin J (1995) Development of recommended advanced life support program, requirements and r&d priorities, strategy and preliminary results. Lockheed Martin
3. Wieland O (1994) Designing human presence in space: an introduction to environmental control and life systems, NASA RP-1324. NASA Marshall Space Flight Center
4. Humphries R, Reuter L, Schunk RG (1986) Intersociety conference on environmental
systems, 860942
5. Patterson MT, Wignarajah K, Bubenheim DL (1996) Life Support Biosphere Sci 3:1
6. Williams DE, Lewis JF, Gentry G (2003) International conference on environmental
systems, 2003-01-2589
7. Carrasquillo RL, Bagdigian RM, Lewis JF, Perry JL (2004) International conference on
environmental systems, 2004-01-2385
8. Williams DE, Lewis JF, Gentry G (2003) International conference on environmental
systems, 2000-01-2593
9. Perry JL et al (1996) International conference on environmental systems, 961519
10. Thompson CD, Shkedi B (2004) International conference on environmental systems,
2004-01-2388
11. NASA (2004) Program overview. Ofce of Exploration Systems, NASA
12. Schladt M, Helble J, Filburn T (2004) International conference on environmental systems, 2004-01-2443
13. Raatschen W, Tan G, Witt J (2003) International conference on environmental systems, 2003-01-2623
14. Winkler HE, Ouelette FA, Smith GS (1990) International conference on environmental
systems, 901292
15. Colombo GV (1973) Study of CO2 sorbents for extravehicular activity. McDonnell
Douglas Astronautics, Huntington Beach, CA
16. Hart JM, Borghese JB et al (1992) International conference on environmental systems,
921299
17. Nacheff MS, Chang CH et al (1989) Intersociety conference on environmental systems,
891595
18. Nalette TA, Filburn TP (1992) International conference on environmental systems,
921298
19. Binot RA (1991) Atmosphere revitalization: life support and habitability manual, ESA
PSS-03-406 Issue 1. European Space Agency
20. Buban EE, Gray RE (1977) Intersociety conference on environmental systems, 77ENAs-58
21. Cheng LH et al. (2004) International conference on environmental systems, 2004-012350
22. Wydeven T (1988) NASA technical memorandum 101004
23. Genovese J (1999) International conference on environmental systems, 1999-01-2120
24. Iacomini CS (2004) PhD dissertation, The University of Arizona, Tucson, Arizona
25. NASA (1995) Spacecraft maximum allowable concentrations (SMAC) for airborne
contaminants JSC 20584
26. Mulloth LM, Finn JE (1998) NASA technical memorandum 1998-208752
27. Lamparter RA (1989) Final report under NASA contract NAS8-36406. Lockheed Missiles & Space Company
404
28. Smith JM, Olcott TM (1965) Report under NASA contract NAS 9-3415. Lockheed Missiles & Space Company
29. Perry JL (1998) NASA technical publication 1998-207978
30. Humphries WR, Reuter JL, Schunk RG (2003) NASA SSP 41000AP
31. LeVan MD, Appel WS, Finn JE, Finn CK (1998) International conference on environmental systems, 981760
32. Murdoch K, Smith F, Perry J, Green S (2004) International conference on environmental systems, 2004-01-2446
33. Samsonov NM et al. (2004) International conference on environmental systems, 200401-2494
34. Chiaramonte FP, Joshi JA (2004) NASA technical memorandum 2004-212940
35. Williams DE (2004) International conference on environmental systems, 2004-01-2386
36. Mulloth LM, Perry JL, LeVan MD (2004) International conference on environmental
systems, 2004-01-2373

Air Quality

Uploaded by

Copyright:

Available Formats

Air Quality

Uploaded by

Document Information

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

Air Quality

Uploaded by

Copyright:

Available Formats

What are the major design drivers for aircraft environmental control systems?

What are the major design drivers for aircraft environmental control systems?

What are some of the contaminants that aircraft environmental control systems aim to control?

What are some of the contaminants that aircraft environmental control systems aim to control?

Hdb Env Chem Vol.

4, Part H (2005): 324

Aircraft Environmental Control Systems

Regulatory Requirements and Guidelines . . . . . . . . . . . . . . . . . .

Environmental and Occupant-Related Constraints . . . . . . . . . . . . .

M. Dechow C.A.H. Nurcombe

Aircraft Environmental Control Systems

M. Dechow C.A.H. Nurcombe

Fig. 1 The environmental control system Airbus Library

Aircraft Environmental Control Systems

M. Dechow C.A.H. Nurcombe

Fig. 2 Typical pressure schedule SAE ARP1270

Aircraft Environmental Control Systems

actual oxygen saturation of the blood. The pH-value, occurrence of carbon

M. Dechow C.A.H. Nurcombe

Fig. 3 Typical circular ow pattern within cabin Airbus Library

Aircraft Environmental Control Systems

Fig. 4 Typical CFD calculation result for temperature Airbus Library

M. Dechow C.A.H. Nurcombe

Aircraft Environmental Control Systems

Fig. 6 Calculated decrease of relative humidity Airbus Library

M. Dechow C.A.H. Nurcombe

Aircraft Environmental Control Systems

M. Dechow C.A.H. Nurcombe

Aircraft Environmental Control Systems

M. Dechow C.A.H. Nurcombe

Aircraft Environmental Control Systems

M. Dechow C.A.H. Nurcombe

Aircraft Environmental Control Systems

care wards (Airbus A310). When peak concentrations of bacteria occurred,

M. Dechow C.A.H. Nurcombe

Fig. 7 Interdependency of factors Airbus Library

Aircraft Environmental Control Systems

M. Dechow C.A.H. Nurcombe

Hdb Env Chem Vol. 4, Part H (2005): 2551

Occupational and Public Health Risks

Elevated Levels of Bioefuents . . . . . . . . . . . . . . . . . . . . . . . . .

Extreme Temperatures (See also Chap. 3) . . . . . . . . . . . . . . . . . . .

Exhaust Fumes, Deicing Fluid, Fuel Fumes, and Cleaning Products . . . .

Reduced Oxygen Supply (See also Chap. 3, and 4) . . . . . . . . . . . . . .

Ozone Gas (See also Chap. 3) . . . . . . . . . . . . . . . . . . . . . . . . .

Insecticides (See also Chap. 9) . . . . . . . . . . . . . . . . . . . . . . . . .

Abstract Symptoms reported by passengers and crewmembers on commercial aircraft are

Occupational and Public Health Risks

Provide a sufcient amount of

Note: All concentrations provided as sea level equivalent

Carbon dioxide < 30 000 ppm

Each passenger and crew compartment must be ventilated, and each

Note: All concentrations provided as sea level equivalent

Provide an effective altitude (pressure)

Combustion heaters must be

Must demonstrate satisfactory

Occupational and Public Health Risks

0.25 ppm (ceiling)

dent incident reporting system for either crewmembers or passengers. In the

Occupational and Public Health Risks