645R78002
FINAL REPORT
LITERATURE REVIEW AND EVALUATION OF THE
HEALTH EFFECTS ASSOCIATED WITH
DIESEL EXHAUST EMISSIONS
October 1978
Prepared for:
Health Effects Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, N. C. 27711
EPA Project Officer: James R. Smith
EPA Contract No. 68 - 02 - 2800
SRC Contract L-1348
SYRACUSE RESEARCH CORPORATION
-------
Principal Authors
The Syracuse Research Corporation (SRC) and its consultants developed the
basic information for EPA and reviewers under Contract No. 68-02-2800.
James R. Smith served as the EPA Project Officer. The major authors of this
document are listed below.
SRC Staff:
Mr. Joseph Santodonato
Dr. Dipak Basu
Dr. Philip Howard
SRC Consultant:
Dr. Paul Sheehe
Department of Preventive Medicine
State University of New York
College of Medicine
Syracuse, New York 13210
-------
Preface
Engineering tests have shown a significant improvement in fuel economy
(25% or greater) in light duty vehicles equipped with diesel engines versus
those equipped with gasoline engines. Automobile manufacturers are considering
a major program for conversion to diesel engines in the automobile fleet by
1985. Available studies show rather large differences in emissions from diesel
engine.exhausts as opposed to gasoline engine exhaust. Conversion of a major
portion of. the automobile fleet to diesel engines may significantly change the
ambient concentrations of both regulated and unregulated pollutants, and hence
the potential human exposure pattern. Such changes may impact upon public
health, and consequently require changes in air quality standards, and/or new
emissions standards. An assessment of the current state of knowledge regarding
the health effects from diesel exhaust emissions, and the identification of
major research needs, are important factors which must be considered by the EPA
under the 1977 Amendment to the Clean Air Act.
In order to accomplish this objective, the following information on diesel
emissions has been reviewed in this document: physical and chemical character-
istics; biological effects in animals and man; epidemiologic studies; knowledge
gaps; and research needs.
ii
-------
TABLE OF CONTENTS
1.0 Summary and Conclusions 1
1.1 Biological Effects !
1.2 Physical and Chemical Characteristics 6
2.0 Introduction 9
3.0 Physical and Chemical Characteristics 11
3.1 Particulates 11
3.1.1 Physical Characteristics 12
3.1.2 Gasoline Exhaust 14
3.1.3 Emission Rate of Particulate Matter 17
3.1.4 Chemical Composition 22
3.1.5 Trace Metals 26
3.1.6 Inorganic Acids and Their Salts 27
3.1.7 Elemental Carbon and Unburned and Partially
Burned Fuel and Lubricant 33
3.1.8 Polycyclic Aromatic Compounds 39
3.1.8.1 Dependency of PNA Emission on Vehicle
Characteristics 46
3.1.8.2 Dependency of PNA Emission on Engine
Operation Mode 46
3.1.8.3 Variation of PNA Emission with Engine
Maintenance 48
3.1.8.4 Effect of Fuel Composition 53
3.1.8.5 Effect of Engine Mileage on PNA Emission 54
3.1.8.6 Effect of Exhaust Emission Control 54
3.2 Volatile Emissions 57
3.3 Fuel Economy 70
3.4 Smoke Results 71
3.5 Odor Rating 73
3.6 Noise 76
3.7 Engine Modification and Antipollution Devices for Diesel Cars 78
3.8 Effect of Irradiation of Automobile Exhaust 81
3.8.1 Photoreactivity of Gasoline Emissions 81
3.8.2 Photoreactivity of Diesel Emmissions 82
3.9 Research Gaps and Recommendations 85
3.9.1 Definition of Particulate Matter 85
3.9.2 Inadequate Particulate Sampling Procedure 85
3.9.3 Better Storage Method 86
3.9.4 Improvement of Analytical Methodology 86
iii
-------
TABLE OF CONTENTS (Cont'd)
Page
3.9.5 Identification and Quantification of New Components 86
3.9.6 Analysis of Sulfates 86
3.9.7 Effect of NO and 03 on Pollutant Formation 87
3.9.8 Quantification of Different PNA Levels 87
3.9.9 Uniformity in Data Reporting 87
3.9.10 Diesel Odor Characterization 87
3.9.11 Necessity for Using Additives 87
3.9.12 Regulation of Pollutants 88
References for Section 3.0 89
4.0 Biological Effects 95
4.1 In Vitro Studies 97
4.1.1 Mutagenicity in Bacterial Systems 97
4.2 In Vivo Studies 104
4.2.1 Absorption, Metabolism, and Excretion 104
4.2.2 Acute Toxicity 107
4.2.2.1 Inhalation Exposure 107
4.2.3 Subacute Toxicity 113
4.2.3.1 Inhalation Exposure 113
4.2.3.2 Dermal Exposure 122
4,2.3.3 Behavioral Effects 123
4.2.4 Chronic Toxicity 126
4.2.5 Bioassays for Carcinogenicity 127
4.3 Human Studies 131
4.3.1 Controlled Exposures 131
4.3.2 Epidemiologie Studies 131
4.3.2.1 Occupational Studies 131
4.3.2.2 Community Studies 140
References for Section 4.0 142
5.0 Identification of Knowledge Gaps 149
5.1 Biological Effects 149
6.0 Recommended Research 152
6.1 Biological Effects
iv
-------
1.0 Summary and Conclusions
1.1 Biological Effects
Human exposure to vehicular combustion products has been a matter of
public health concern for many years. Increasing utilization of internal com-
bustion engines in the development of industrial civilization has forced our
society to live with an increased burden of air pollutants. The prospect that
passenger cars equipped with diesel engines will soon represent a significant
proportion of new car production raises an important question concerning pos-
sible impacts on public health. Recognition of this situation allows the
opportunity for evaluation of a major environmental change caused by the intro-
duction of greater quantities and/or types of diesel-derived pollutants. It
must be recognized, however, that the diesel engine is by no means the only (or
even major) source of the pollutants of primary concern.
The components of automotive emissions which are generally regarded
to have greatest toxic potential include carbon monoxide, oxides of nitrogen,
aldehydes, hydrocarbons, sulfur dioxide, and particulates. In comparison to
the gasoline engine operating with or without a catalytic converter, the
diesel produces far greater quantities of carbonaceous particulate material.
These particles are of respirable size and have high surface areas, enabling
them to adsorb gaseous exhaust products. Among these products are small
amounts of irritant gases and, perhaps most significant, a large proportion of
the polycyclic organic matter (e.g., benzo[a]pyrene) produced during combus-
tion, some of which are carcinogens. This poses the potential risk of delivery
of adsorbed gases into the lung by carrier particles. In turn this may lead to
extensive localization of harmful materials in the lung, with the accompanying
threat of emphysema and cancer development. In addition, fibrotic changes may
occur leading to reduced lung compliance and/or obstruction.
1
-------
Experimental studies designed to establish whether diesel emissions
represent a significant threat to human health have been conducted only infre-
quently during the past 25 years. Moreover, these studies have not provided
definitive and comprehensive insight to the health effects of diesel emissions.
While it was recognized that the gaseous emissions from diesel engines are com-
parable to or lower than for the noncatalyst-equipped gasoline engine, the
toxicologic role of increased particulate production remains unclear. Just
recently, however, it was reported that organic extracts of diesel particulate
contained materials that were mutagenic to histidine-requiring strains of
Salmonella typhimurium in the Ames assay. This positive result raises a ques-
tion concerning the potential carcinogenlcity of this material, although it is
not known how the effects of widely dispersed diesel exhaust in air may be
related to the mutagenic effect of diesel extracts. These mutagens in diesel
exhaust included, but were not limited to, the well-known polycylic aromatic
carcinogens. Other positive compounds found included direct-acting (i.e., not
requiring metabolic activation) frameshift mutagens; probably composed of polar
compounds such as substituted polynuclear aromatics, phenols, ethers, and
ketones. The strong formal relationship between mutagenesis and carcinogenesis
thus may implicate extracts of diesel particulate as a carcinogenic material.
It is also known, however,.that extracts of airborne particulate
pollutants in urban atmospheres and gasoline engine exhaust are also mutagenic
in the Ames assay. Moreover, direct-acting mutagens are found in gasoline
engine exhaust (noncatalyst-treated) just as they are found in diesel emissions.
In addition, a high incidence of skin cancers has been produced in mice by
dermal administration with organic extracts of the particulate exhaust fraction
from diesel and gasoline engines, and from extracts of ambient particulate
-------
pollutants. However, quantitative differences in potentcy are likely to exist
among extracts from these various sources.
Investigators have for many years attempted to show that exposure to
automotive emissions may lead to the development of cancer. Indeed, it was
shown several decades ago that both diesel and gasoline engine exhaust contain
carcinogenic polycyclic aromatic hydrocarbons such as benzo[a]pyrene. Since
that time, chronic studies have been initiated which involve the inhalation of
total diesel emissions by rats and hamsters. Under the conditions of exposure
employed, these experiments have thus far failed to produce tumors of the
respiratory tract. On the other hand, evidence of serious damage to the
respiratory tissues has been obtained. Rats exposed to diesel exhaust for 20
months displayed extensive particulate accumulations in the lungs, accompanied
by vesicular emphysema and beginning interstitial fibrosis. Similar observa-
tions were made in hamsters, in addition to the presence of cuboidal meta-
plasia. It is not known whether this tissue damage in rats and hamsters is
reversible or if it may lead to significant shortening of life.
A series of in vivo studies with several animal species inhaling
diluted irradiated and non-irradiated diesel exhaust are being conducted by the
U.S. Environmental Protection Agency (EPA). The initial subacute exposure
studies using relatively high concentrations of diesel exhaust (1:12 dilution)
were designed to provide preliminary data on toxic effects and target organs.
Animals inhaling diesel exhaust for up to two months were found to have black
granular particles in alveolar macrophages, and black pigment in the bronchial
and carinal lymph nodes. These observations indicated the existence of clear-
ance mechanisms for diesel particulate.
-------
Among guinea pigs inhaling diesel exhaust several exposure-related
changes were seen including, increased pulmonary flow resistance, increased
lung weight to body weight ratios, and sinus bradycardia. Microscopic examina-
tion of the lungs revealed goblet cell hypertrophy and focal hyperplasia of
alveolar lining cells, possibly an early indication of damage to the alveolar
wall by diesel exhaust. Neither the reversibility of these lesions, nor the
degree of functional impairment which accompanies them has yet been determined.
Other changes which EPA investigators observed were biochemical alterations in
the lungs of rats, behavioral changes in rats, and increased susceptibility to
death by respiratory infection in mice.
The presently available data base does not allow for an accurate com-
parison to be made between the effects of environmentally realistic concentra-
tions of diesel emissions and catalyst or noncatalyst treated gasoline engine
exhaust. Nevertheless, diluted gasoline exhaust produced emphysematous lesions
in the lungs of dogs as well as a high incidence of bilateral renal sclerosis
in rats. Evidence which is available thus far indicates that the use of an
oxidation catalyst with gasoline engines will dramatically reduce the toxicity
of resulting emissions. This can be attributed to the substantial reductions
realized in the emission of most harmful gaseous components in the catalyst-
treated exhaust.
An overall assessment of the public health risks associated with
diesel exhaust exposure cannot be based solely on the results of available
animal studies. This is partly due to the fact that many areas of concern with
respect to the toxicity of diesel emissions remain to be explored. In addition,
parallel studies with diesel and gasoline engines have not been conducted which
-------
would allow direct comparison of results. The alteration of the environment by
increased utilization of diesel engines will probably be due to the relative
abundance of particulate matter which is emitted. Protection of the public
health will thus be best achieved by examining the potential adverse impact of
this component of diesel exhaust.
The ultimate goal in demonstrating that increased utilization of
dieselized vehicles is an environmentally acceptable substitute for gasoline
engines is to provide reliable epidemiologic evidence which supports this
claim. Unfortunately, the previous epidemiologic research which is often used
to support the safety of the diesel provides only a limited data base, which is
clearly inadequate for developing sound conclusions. It is this fact more than
anything else which prevents the formulation of a valid health risk assessment
for diesel emissions. Among the more recent occupational mortality and morbid-
ity studies which have been reported, it has not been possible to isolate
diesel emissions as a singularly important factor in contributing to the excess
deaths and adverse health effects occasionally observed. Several studies
involving populations of workers exposed to high levels of diesel emissions are
currently being conducted by NIOSH. Results of these investigations should be
forthcoming within the next several years. Since large populations that have
been exposed to ambient levels of light duty diesel emissions do not yet exist,
the possibility of conducting community studies at this time is remote, espec-
ially when the intervals necessary for the development of neoplasms in humans
after exposure are considered. Thus because of the lack of a broad-based
community study or well-controlled investigation of a worker population where
quantitative exposure data are available, no definitive judgement regarding
diesel emissions can now be made.
5
-------
There is presently no way to place into perspective the hazards of
diesel emissions relative to those of automotive emissions as a whole. In the
absolute sense, diesel exhaust is a noxious mixture with the potential to pro-
duce serious lung disease, behavioral alterations, biochemical changes, and
decrements in pulmonary function. Its risk as a human carcinogen, however, is
unquantified. When looked upon in light of what we know regarding the poten-
tial health effects of noncatalyst-treated gasoline exhaust, the impact of
diesel emissions remains unclear. Presently, there are no data which suggest
an increased carcinogenic threat from the substitution of diesel- for gasoline-
powered light duty vehicles. Further epidemiologic research must be pursued
and studies in laboratory animals conducted to characterize cause-and-effect
relationships and exposure-response parameters.
1.2 Physical and Chemical Characteristics
A comparison of well-maintained diesel cars (without emission control)
and gasoline cars (with emission control) for regulated vaporous emissions
shows that diesel cars emit more hydrocarbons than gasoline cars. With auto-
mobiles of comparable size, diesel cars emit twice as much hydrocarbons as
gasoline cars in the FTP mode. The difference between the two classes of cars
becomes even greater in the SET and FET mode, with diesel cars likely to emit
three times more hydrocarbons than gasoline cars. The 1977-79 Federal Standard
of 1.5 g/mile for hydrocarbon emissions, however, can be met by diesel cars.
With respect to CO, both gasoline and diesel cars have about the same emission
rate. In the FTP mode, gasoline cars have higher CO emission rates than diesel
cars, and the reverse is true in the FET and SET modes. The 1977-79 Federal
Standard o£ 15 g/mile is met by diesel cars. The NO emission for gasoline
A
-------
cars is higher than for diesel cars in all modes. Most diesel and gasoline
cars will meet the 1977-80 Standard for NO .of 2.0 g/mile. If a 0.5 g/mile
A
particulate emission standard is introduced for diesel cars in the future, most
cars will be able to meet this standard.
The fuel economy consideration is in favor of diesel cars. Based on
combined city/highway estimates, the fuel economy for diesel cars is 33 to 60%
better than in corresponding gasoline-powered cars.
In terms of unregulated emissions, diesel cars are higher sources of
carbonyls and substantially higher, sources of particulate emissions. Although
there is uncertainty about benzo[a]pyrene emission rates, it appears from the
work of Springer and Baines (1977) that diesel cars emit at least an order of
magnitude more benzo[a]pyrene than gasoline cars. The reliability of the BaP
data in this work, particularly for gasoline exhaust, can be doubted because of
the unreliable sample collection technique. It has been demonstrated by Gross
(1972) that an increase of 0.5% CO could cause a 45% increase of PNA emissions.
Based on this result one would not expect higher PNA emissions from diesel
cars. This aspect of research correlating PNA and CO emission rates should be
reinvestigated with newer cars equipped with catalytic converters.
Sulfur dioxide emission rates for diesel cars are substantially
higher than for gasoline-powered cars. This is, however, expected because the
national average diesel fuel contains 0.23 weight% sulfur compared to 0.03
weight% $ulfur for the national average gasoline fuels. Reduction of fuel
sulfur will reduce the S0? emission rate from diesel cars. The sulfate emis-
sion rate for gasoline-powered cars with no air pump in the catalytic system is
less than for diesel cars. However, with air pumps, the sulfate emission rates
become comparable.
7
-------
Diesel engines produce more visible smoke than gasoline-powered cars.
However, diesel cars are capable of operation within the EPA smoke visibility
limit, for the most part, with only brief excursions during rapid throttle
movement. .Diesel exhaust have more odor than gasoline exhausts and the odor
intensity of diesel exhaust may noticeably change during the transient cycle.
Interior noise levels are slightly higher with the diesel during acceleration
than in gasoline automobiles. The idle noise levels are also higher with
diesels compared to gasoline cars.
There is a substantial conflict in the available data base among
various authors due primarily to nonuniformity of experimental conditions and
uncertainty in the variable experimental parameters. In addition to these
conflicts in the data base, nonuniformity of data reporting sometimes makes it
difficult to compare results among various investigators.
A number of new suspected carcinogenic compounds, namely, methylene-
PNA's and nitro-PNA's, have been reported in gasoline exhausts. Their presence
can be expected in diesel exhaust and needs confirmation.
-------
2.0 Introduction
According to an EPA projection it has been estimated that automobiles
equipped with diesel engines are likely to increase at a rate of 5% per year
and capture up to 25% of the U.S. new car market by 1985. The major factor
behind the projected increase in diesel-powered automobiles is the considerable
i
fuel economy for this type of car. Of the total estimated auto fuel consump-
tion during the period 1976-2000, it is projected that gasoline will decrease
by about 2% and diesel will increase by about 37% (EHA, 1978). At the present
time, diesel-powered motor vehicles (mostly heavy duty) contribute about 1% of
all motor vehicle emissions (National Academy of Sciences, 1976). With the
steady increase of diesel-powered automobiles, the effect caused by the anti-
cipated change in the quantities of environmental pollutants emitted to the
atmosphere dictates the need for more thorough investigations. The present
report is prepared for the EPA to assess the environmental health impact as a
result of conversion of light duty vehicles from gasoline-powered to diesel-
powered engines. However, much of the information to date is on heavy duty
diesel engines and might thus lead to biassed extrapolations.
Although, diesel exhaust contains relatively low levels of CO, making it
suitable for use in mining operations, it is considered by some to be poten-
tially a major source of air pollution due to its propensity for emitting
visible smoke and obnoxious odor. Particular attention, therefore, should be
given to the latter parameters in order to evaluate their significance in
promoting any deleterious health effects. The exhaust emissions from gasoline-
powered automobiles, on the other hand, are steadily decreasing as a result of
introduction of emission control devices. With the rates expected to decrease
-------
even further when the statutory emission standard becomes mandatory, the
emissions from diesel-powered automobiles may become a significant factor in
air pollution. Likewise, recent decreases in particulate emissions from
stationary sources might make the effects of diesel emissions even more
evident. It is, therefore, important to make a comparative study between
diesel and gasoline exhaust with the objective of maintaining a comparable
level (in terms of health effects) of environmental pollution.
Although the presently available literature contains abundant information
in some aspects of exhaust emission rates from diesel- and gasoline-powered
automobiles, there is a lack of data in other areas. In many cases, emission
rate data determined with the objective of quantitating chemical species lack
accompanying details about the engine operating conditions and vice versa.
This makes the comparative study of emission rates between the two fuel-powered
automobiles very difficult and/or impossible. The following section undertakes
the task of presenting a comparative review of the physical and chemical sig-
nificance of different exhaust emission parameters from diesel- and gasoline-
powered automobiles, with particular emphasis on those with suspected injurious
health effects.
10
-------
3.0 Physical and Chemical Characteristics
The chemical compositions of diesel and gasoline emission have been broadly
divided into two sections, one section detailing the particulate matter and the
v
other vaporous emissions.
3.1 Particulates
The exhaust from both diesel and gasoline engines contain suspensions
of microsize solid particles and liquid droplets in gas or vapor. It is,
therefore, necessary to define which part of the exhaust can be considered as
particulate matter. There is no general agreement on this subject. According
to a commonly used definition anything other than condensed water that can be
collected on Type A glass files filtering media at a temperature not to exceed
125°F (51.78C) is considered as particulate matter. The choice of the type A
filter is based on the fact that it removes 98% of the particles larger than
0.05 ym diameter (Sampson and Springer, 1973) from the gas or vapor phase. The
selection of collection temperature of 125°F is on the basis of a compromise
between minimization of moisture condensation and maximization of particulate
collection. However, there are certain limitations to this definition. When
the hot exhaust from the automobiles is discharged into the atmosphere, inhala-
tion of particulate matter by humans occurs at ambient temperature after air
dilution and associated cooling of the exhaust stream. Far more serious
limitations of this definition may arise unless the retention efficiencies of
the filtering media can be demonstrated to be high for particles of size ranges
below 0.05 ym. Data regarding the efficiency of particulate collection from
automobile exhaust are very limited. Therefore, until a better method is
available, the particulate data developed by various authors should be inter-
preted in their proper perspective.
11
-------
3.1.1 Physical Characteristics
The physical characteristics of the particulate from both diesel
and gasoline exhaust are discussed individually.
Diesel Exhaust: X-ray spectroscopy shows soot to have a graphite
structure with hexagonal basic carbon units linked into platelets giving a
crystallite about 21 x 134° containing 10 to 30 mole percent of hydrogen
(Millington and French, 1966). The structure has a good resemblance to poly-
benzenoid substances, such as the polynuclear aromatic hydrocarbons (PNA's).
The basic crystallite units agglomerate into spheres with a diameter range of
100-800A" (Vuk and Johnson, 1975). These agglomerates containing as few as one
100A° spherical particle or as many as 4000 spheres combine to form particles
up to 30 vim diameter (Vuk et al., 1975). The other physical characteristics of
the particulates as measured by different authors are presented in Table 3.1.
Table 3.1 shows that the particulate matter has very large
surface area which make it a powerful adsorptive agent. The low still air
settling velocity will make it remain airborne for a long period after gener-
ation.
It should be recognized that several parameters, such as, fuel
composition, engine design and maintenance, operating conditions and emission
control devices may influence the physical characteristics of the emitted
particulates. The particle size is normally expressed in terms of either
aerodynamic diameter or Mass Median Equivalent Diameter (MMED: diameter of an
aerodynamically equivalent sphere of unit density). The effect of engine
operating parameters, such, as engine speed and load, on particle size was
studied by Vuk and Johnson (1975). From their work these authors concluded
12
-------
TABLE 3.1. PHYSICAL CHARACTERISTICS OF PARTICIPATE FROM DIESEL EXHAUST
Parameter Mass medium Particulate Surface area Settling velocity
diameter (ym) count no/cm3 m2/m3 mm/hr.
Value 0.1 - 0.3 107 2 0.25 - 40
Reference Vuk & Johnson, Frey & Corn, Frey & Corn, Frey & Corn,
1975, Dolan 1967 1967 1967
et al., 1975,
and Schreck,
1978
13
-------
that the particle diameter decreases slightly with increasing engine load and
temperature, Ddlan et_ &!_<, (1975), however, reported a shift in particle size
shown as in Figure 3.1 from the smaller "nuclei mode" to the large "accumula-
tion mode" with increasing engine load. The results of Laresgoiti e_t al.
(1977) and Schreck (1978) do not show any significant variation in particle
size either with engine speed or load. It can be concluded from these inves-
tigations that particle size may not be strongly dependent on engine operating
conditions. Although no definite explanation can be offered, it can be con-
jectured that the discrepancies between the various results are due to the
differences in engine design and fuel composition used in these experiments.
3.1.2 Gasoline Exhaust
The size of the particulate matter emitted from gasoline engine
exhaust was studied by Mueller et. aJL. (1962) and Moran and Manary (1970).
Their findings were similar to that of Sampson and Springer (1973). The latter
work is summarized below.
The particulate size from gasoline engine exhaust depends on the
fuel composition. Unleaded fuel results in particulates of larger aerodynamic
diameter. For leaded fuels, approximately 90 wt% of the emitted particulates
are below 0.35 urn diameter and over 98 wt% below 10.0 ym diameter. In case of
unleaded fuels, approximately 40 wt% of the total particulates are below
0.35 ym diameter and 88 wt% below 10.0 ym diameter. The particle size distri-
butions of the emitted particulates from leaded and unleaded fuels was studied
by Sampson and Springer (1973). They concluded that the weight of the smaller
particles (<0.35 ym diameter) was much higher with leaded than with unleaded
fuel,
14
-------
g
M
H
I
O
U
w
A TOTAL
D NUCLEI
o ACCUMULATION
FULL
LOAD
Figure 3.1. Variation in aerosol volume concentration with load
for the two size components of diesel exhaust:
the accumulation mode 0.08 ^ Dp <^ 2.0 ym and the
nuclei mode Dp <_ 0.08 ym (from Dolan et_ al_., 1975).
Dp: diameter of particles.
15
-------
It has been estimated by Mueller (1970) that the MMED of the
lead particles which can stay dispersed in the atmosphere fall in the range of
0.1 to 1.0 ym. Of the total lead particles emitted from leaded gasoline about
71-91% are respirable compared to 34-59% respirable particles in the total
particulate matter in the exhaust (Mueller, 1970).
The particle size of the exhaust is similarly dependent on the
fuel sulfur content. In gasoline vehicles equipped with a catalytic converter,
the sulfate emission rate is substantial compared to the total particulate
emission rate (see Section 3.1.6). More than 70% of the sulfate emitted by the
vehicles may be in the form of HLSO, with a geometric mean diameter of M).02 pm
(Wilson et^ al., 1977). The rest of the suifate is primarily in the form of
(NH,)2SO, and other refractory sulfates (Lee and Duffield, 1977). The mean
diameter for ammonium sulfate aerosol is 0.07 ym. Therefore, the introduction
of a catalytic converter to vehicles is bound to shift the emitted particulate
diameter to smaller size ranges.
When sulfuric acid aerosol is exhausted into the atmosphere,
most of the aerosol in the smaller nuclei mode undergoes growth into the larger
accumulation mode in the size range of 0.1 to 1.0 urn. When this aerosol is
inhaled, the high relative humidity in the pulmonary system causes the aerosol
droplets to grow further in size. As a result of water vapor absorption the
acid is also diluted. Thus Wilson e£ al. (1977) estimated that a 0.35 urn
diameter droplet at 50% relative humidity would grow to a 1.0 ym in diameter at
99% relative humidity and the concentration would decrease from approximately
10.5 N to less than 0.5 N. But particles in the smaller nuclei mode will
experience less growth and dilution due to the decrease in vapor pressure
16
-------
caused by the high curvature of small droplets. Both size ranges will deposit
appreciably in the bronchi and alveolar regions. However, the nuclei mode will
give greater total, deposition and more deposition in the alveolar region.
The diameter of the emitted particles not only depends on fuel
composition but also on engine operating conditions. Generally, it has been
found that for leaded fuels cyclic operations yield larger particles than
steady state operations (Habibi, 1970). The average size of the emitted lead
particles, also, increase significantly with mileage accumulation from a MMED
of 1.1 ym at 6350 average mileage to 4.7 ym at 21350 average mileage (Habibi,
1970).
The corresponding effects on the particle size for unleaded
fuels has not yet been studied.
3.1.3 Emission Rate of Particulate Matter
The emission rates of particulate matter from both diesel- and
gasoline-powered vehicles are presented in Table 3.2.
Several parameters affect the weight of particulates emitted
from vehicles operated by both fuels. These parameters include fuel composi-
tion, engine design and maintenance, operating conditions, engine mileage, and
the presence or absence of emission control devices.
In the case of diesel-powered vehicles, increase in fuel sulfur
and aromaticity has been shown to increase particulate emissions (Braddock and
Gabele, 1977). Similar increases in particulate emission with increase in fuel
aromaticity has been observed by Ter Haar et al. (1972) in case of gasoline
cars. The increase of particulate emission with increase in S-content of the
17
-------
TABLE 3.2. PARTICULATE EMISSIONS FROM DIESEL- AND GASOLINE-POWERED
PASSENGER CARS
Vehicle type
Diesel vehicles:
VW Rabbit1
2
Peugeot 504
3
Mercedes 240D
Mercedes 300D3
Oldsmobile 3501
Gasoline vehicles:
With emission control
VW Rabbit1
Oldsmobile 3501
Engine displacement
CID
90
129
146
183
350
90
350
Total particulates
in FTP mode, mg/mile
294.0
397.0
477.0
490.0
924.0
6.8
9.1
Without emission control
4
Leaded gasoline car a
4
Unleaded.gasoline car a
Advanced unleaded gasoline car c
246.0L
181.Ol
2.0
a. Various 1966-70 model cars.
b. Average of low and high mileage cars.
c. Data not available.
1. Ref. Springer and Baines, 1977
2. Ref. Braddock and Gabele, 1977
3. Ref. Springer and Stahman, 1977
4. Ref. TerHaar e£ al., 1972
5. EPA generated data cited in PEDCO Environ. Inc., 1978
18
-------
fuel has been observed by Stara ejt al_. (1974). Using a G.M. engine equipped
with a pelletized catalyst, these authors have reported an increase of particu-
late emission from 9.6 rag/mile with 0.05% S indolene fuel to a value of 30.2
mg/mile with 0.10% S indolene fuel. The effect of fuel additives with non
emission-controlled cars has been studied by Ter Haar et, a!L. (1972). Addition
of lead, carburator detergent, phosphorus and a commercial upper cylinder
lubricant all caused increases in particulate emission.
The effect of engine size and operating modes on particulate
emission rates has been studied by Springer and Stahman (1977) and Springer and
Baines (1977). In the case of diesel-powered automobiles, the increase in
particulate weight with the increase in engine size has been demonstrated by
Springer and Stahman (1977). The effect of diesel engine operating parameters
on the variation of particulate mass emission has been studied by Laresgoiti
et al. (1977) with a Mercedes Benz Model 240D car and the effect is shown in
Figure 3.2 and Figure 3.3.
From their work Laresgoiti et al. (1977) concluded that any
change in engine performance parameters which affects a change in combustion
temperature, combustion time, and fuel-to-air ratio in the combustion chamber
will cause variation in the emitted particulates.
For gasoline-powered vehicles, moderate variations in the air-
to-fuel ratio and spark timing has no significant effect on particulate emis-
sion rate (Ganley, 1973). However, an increase of the particulate weight by
about 300% was observed as the engine speed and load was increased from 40 MPH
to 70 MPH under road load conditions.
19
-------
3IOOrpm
2900 rpm
2150 rpm
25 50 75
PERCENT Of FULL LOAD
100
Figure 3.2. Exhaust particulate concentration as a function
of engine speed. (Laresgoiti et al., i 977)
20
-------
PERCENT OF
FULL LOAD:
1.000 2,000
REVOLUTIONS PER MINUTE
5,000
Figure 3.3. Exhaust particulate concentration as a function
of engine load. (Laresgoiti et_ al., 1977)
21
-------
The dependency of particulate emission rates for both diesel-
and gasoline-powered automobiles on engine size and operating modes is shown in
Table 3.3.
It can be seen from Table 3.3 that the particulate emission rate
for both catalyst equipped gasoline-powered cars and non-catalyst equipped
diesel-powered automobiles increases with increase in engine size and that the
emission rates are dependent on the engine operating modes. The emission rates
are higher during cold cycle than hot cycle. The particulate matter in exhaust
from a non-catalyst equipped diesel is about 47 to 102 times more than gasoline
engines fitted with a catalytic converter.
The use of a catalytic converter has the effect of increasing
the amount of particulate from both diesel and gasoline-powered automobiles.
Table 3.4 shows the effect of a catalytic converter on particulate emission
rates.
It is necessary to emphasize that the particulate emission rate
from non-catalyst cars operating with unleaded fuel has decreased substantially
(see Table 3.2) with the introduction of advanced lean-burn engines, probably
due to decreases in the sulfate emission rate. The primary reason for the
increase in particulate matter with catalyst equipped cars is due to conversion
of SCL to particulate sulfate (see Section 3.1.6).
The effect of different variables on the mode of particulate
emission is qualitatively summarized in Table 3.5.
3.1.4 Chemical Composition
The chemical composition of particulate matter in both gasoline
and diesel exhaust is complex. The characterization of diesel exhaust
22
-------
TABLE 3.3. VARIATION OF PARTICULATE EMISSION RATES WITH CAR SIZE
AND ENGINE OPERATING MODES3
Operating
mode
Particulate emission rate, mg/km
Size of gasoline-powered car
Size of' diesel-powered car
V.W. Rabbit
90 CID
Oldsmobile Cutlass
350 CID
V.W. Rabbit
90 CID
Oldsmobile Cutlass
350 CID
U>
1975 Federal Test 3.34
Procedure
Federal Test 4.95
Procedure cold
Federal Test 3.01
Procedure hot
Sulfur Emission Test 1.63
Fuel Economy Test 1.55
5.63
8.38
3.55
9.72
13.62
182.0
202.0
165.0
161.0
157.0
573.0
628.0
523.0
360.0
298.0
a. Ref. Springer and Baines, 1977.
-------
TABLE 3.4. EFFECT OF CATALYTIC CONVERTER ON PARTICULATE
EMISSION RATES
Vehicle category Particulate emission
rate, mg/mile
Light-duty gasoline-powered vehicles:
Catalyst3 6.0
Catalyst (excess air)a 15.0
Non-catalyst (lead fuel)a 250.0
Non-catalyst (unleaded fuel)a 2.0
Light-duty diesel-powered vehicles:
Non-Catalyst3 500.0
Catalyst Some small decrease
a. EPA generated data cited in FED Co. Environ. Inc., 1978.
b. Ref. Seizinger, 19/8
24
-------
TABLE 3.5. EFFECT OF VARIABLES ON PARTICULATE EMISSION FROM
GASOLINE- AND DIESEL-POWERED CARS
Variable
Engine type:
Indirect injection
Direct injection
Turbocharglng
Engine displace-
ment
Emission control:
Exhaust gas
recirculatlon (EGR)
Catalyst
Catalyst & water
scrubber
Thermal reactor
Engine speed
& load
Engine maintenance:
Engine deteriora-
tion
Combustion chamber
deposits
Fuel Composition
S-content
Aromatlclty
Fuel Additives
Lead
Halogen compounds
Nitromethane
Ba & Nl-additives
Methanol
Water injected
with fuel
Emission Parameter
Particulate mass
Particulate mass
Particulate mass
Particulate mass
Particulate mass
.Particulate mass
Particulate mass
Particulate mass
Particulate size
Particulate mass
Particulate mass
Particulate mass
Particulate mass
Particulate mass
Farticulate mass
Particulate mass
Particulate mass
Particulate mass
Particulate mass
Particulate mass
Effect
Diesel
increase
decrease
decrease
decrease with
decrease in
parameter
increase
increase
decrease
decrease
no change
increase with
increase in
variable
increase with
poor maintenance
increase with
increase in
variable
increase with
increase in
variable
increase
decrease
decrease but may
decrease engine
life
decrease
decrease
Gasoline
decrease with
decrease in
parameter
increase
decrease
increase with
increase in
variable
increase with
deposits
increase with
increase in
variable
increase with
increase in
variable
increase
increase
decrease
Reference
EEA, Inc., 1978
EEA, Inc., 1978
NIOSH, 1978
Springer & Balnes,
1977
NIOSH, 1978
Stewart et al., 1975
& Stara et at. , 1974
NIOSH, 1978
NIOSH, 1978
Laresgoiti et al . ,
1977 & Ganley,~T973
NIOSH, 1978
GI-OBS, 1972
NIOSH, 1978 &
Stara et al. , 1974
Braddock & Gabele,
1977 & TerHaar
et al., 1972
Sampson & Springer,
1973
Broome & Khan, 1971
Broome & Khan, 1971
Broome & Khan, 1971;
Apostolescu et al. , 1977
Broome & Khan, 1971
Greeves et. al. , 1977
Detergent
Particulate mass
Injector stays increase
clean but no hard
data on particu-
late emission
NIOSH, 1978 &
TerHaar, et al., 1972
25
-------
particulates has begun only recently and the available information is limited.
Complete combustion of fuel under perfect conditions should yield principally
carbon dioxide and water. Because the combustion process in an actual engine
is imperfect, several other products are produced. Both absolute and relative
concentrations of combustion products are influenced by numerous factors. Some
of the most prominent factors are: (1) air-to-fuel ratio, (2) ignition timing,
(3) inlet mixture density, (4) combustion chamber geometry, (5) the variable
parameters, such as speed, load, and engine temperatures, (6) fuel composition,
and (7) presence of emission control devices.
The particulate matter from both diesel and gasoline exhaust
contains a variety of products. Some of those which have been identified are:
(1) unburned carbon, (2) unburned and partially burned hydrocarbons originating
primarily from fuel and oil, (3) trace amounts of metals, (A) inorganic acids
and their salts, namely, sulfates and nitrates, and (5) polycyclic aromatic
hydrocarbons. The exhaust emissions from diesel and gasoline cars containing
each of these categories of compound are discussed individually.
3.1.5 Trace Metals
The origin of metals in automobile emissions is from two dis-
tinct sources, namely, fuel and lube oil, and engine and exhaust system wear.
When catalytic converters are used, the third possible source could be attri-
tion products from the catalyst. However, no trace metals from, the latter
source have been reported. When leaded gasoline is used, obviously the pre-
dominant metal content in the exhaust is lead (Campbell and Dartnell, 1973).
The lead emission rates from cars under cyclic operating conditions have been
studied by Ter Haar £t al. (1972); Habibi et^ al., (1970); Ninomiya e£ al. (1970)
26
-------
Typical emission factors for metals cannot be derived from
baseline characterization of auto exhaust by dynamometer tests as performed by
EPA, since attempts are made to keep variability of additives, oils, and lubri-
cants to a minimum. Emphasis is placed rather on the effect of emissions as a
function of variations in operating conditions. The data cited in Table 3.6
reflect this because the different test cycles differed significantly in the
average speed and variability of the operating mode. All the metal data,
except the precatalyst Cu and Fe data, given in Table 3.6 were obtained by an
X-ray fluorescence method. The precatalyst Fe and Cu data were obtained by
emission spectrometry.
A comparison of the results given in Table 3.6 with the results
obtained by Springer and Baines (1977) is interesting. With the exception of
iron, the latter authors have failed to detect any other metals listed in
Table 3.6 in all modes of operation with a 1976 Oldsmobile diesel Cutlass and a
1977 V.W. diesel Rabbit. Evidently, more research is needed in this field to
establish the possible emissions of trace metals from both diesel- and gasoline-
powered passenger cars.
3.1.6 Inorganic Acids and Their Salts
The sulfate emission rates for diesel- and gasoline-powered
passenger cars are .demonstrated .in Table 3.7,
Comparison of the average values shows that the diesel has
higher sulfur emission rates than non-catalyst gasoline cars. However, gaso-
line cars equipped with a catalyst have comparable sulfate emission rates to
diesel cars. These emission rates are substantially lower for advanced non-
catalyst and three-way catalyst gasoline-fueled vehicles. However, addition of
27
-------
TABLE 3.6. EMISSION RATES OF SELECTED METALS FROM A VARIETY OF
CARS UNDER DIFFERENT OPERATING CONDITIONS
Pre-catalyst cars
Metals
FTP
FET
SET
60 mph Cruise
Catalyst cars (49-State
Standard) ^
50 mph Cruise
Mean
Median
Range
Catalyst cars (Calif.
Standard)4
FTP
Dual catalyst
FTP
FET
SET
4
Lean Burn Engine
FTP
FET
SET
4
Stratified charge
FTP
FET
SET
4
Rotary
FTP
Diesel (#2 Natl. Avg. Fuel)
FTP
FET
SET
Pb
(rag/mi)
40.5
19.8
20.0
0.028
0.015
0-0.325
0.03
0.18
0.03
0.07
6.69
3.76
7.00
N.D.
N.D.
0.12
0.40
6
2.55
2.50
2.00
Mn
(mg/mi)
N.A.a
N.A.
N.A.
N.A.
N.A.
N.A.
N.D.
N.A.
N.A.
N.A.
0.05
0.12
0.07
N.D.
N.D.
0.01
0.09
1.46
1.45
1.19
Cu
(mg/mi)
N.A.
N.A.
N.A.
0.016
0
0-0.293
0.18
0.12
0.01
0.02
0.06
0.08
0.08
N.D.
N.D.
N.D.
0.04
1.56
2.03
1.54
Fe
(mg/mi)
N.A.
N.A.
N.A.
0.029
0.007
0-0.341
2.28
0.56
0.03
0.07
1.17
0.07
0.15
0.13
0.23
0.05
0.70
1.56
0.12
0.08
Ni
(mg/mi)
N.D.b
N.D.
N.D.
N.A.
N.A.
N.A.
0.01
4.12
0.27
0.50
N.D.
N.D.
N.D.
0.01
0.01
N.D.
0.06
N.A.
N.A.
N.A.
? N.A.: not available
*"^ IkT "IN » u ... x. J^A^^-i-j-— .13
1 Unpublished data by R.L. Bradow cited in Lee and Duffield, 1977b.
2 Dow, 1970
3 DEC, 1976
4 Gabele ejt^ al., 1977
5 EPA, 1977
6 Braddock and Bradow, 1975
28
-------
TABLE 3.7. SULFATE EMISSION RATES (mg/mile) FROM DIESEL
VERSUS GASOLINE PASSENGER CARS
Vehicle type
Diesel vehicles:
V.W. Rabbit1
Peugeot 204D2
Peugeot 504 3
2
Mercedes 220D (comprex)
Mercedes 240D2
Mercedes 300D2
Oldsmobile 3501
Gasoline vehicles:
Non-catalyst, no air pump
Chevy, 1972*
Non-catalyst, air pump
Granada, 19754
Dodge, 19 754
Catalyst, pelletized, no air
Average of 1975-76 cars4
Catalyst, pelletized, air
/
Average of 1975-76 cars
Catalyst, Monolith, no air
L
Average of 1975-76 cars
Catalyst, Monolith, air
L
Average of 1975-76 cars
Advanced non-catalyst
A
Stratified charge
/
Rotary, THM reactor
4
Lean burn
4
Lean burn, THM reactor
Advanced catalyst
3-way, F-l4 (fuel injected)
1 3-way4
Start cat.
Duel Cat.4
/
Lean bum oxd. cat.
Sulfate trap
Eaee metal cat.
Emission rate
(mg/mlle)
1.7
8.2
6.5
9.2
14.2
16.6
21.0
0.2
0.7
2.7
10.1 "
12.8 *
7.7
33.5
1.9
1.5
1.0
6.3
1.5
0.6"
26.3
36.1
88.9
4.7
6.3
a. One high value rejected In averaging the result.
1. Ref. Springer and Balnes, 1977
2. Ref. Springer and Stahman, 1977
3. Ref. Braddock and Gabele, 1977
4. Ref. goners et al., 1977
29
-------
an air injected oxidation catalyst downstream of a three-way catalyst, or
addition of an oxidation catalyst to a lean burn non-catalyst system, will
result in higher sulfate emission rates typical of air pump equipped oxidation
catalyst systems.
The sulfate emission rates for both classes of cars (diesel and
gasoline) are dependent on four variables. These are: (1) vehicle design and
engine displacement, (2) sulfur content of the fuel, (3) vehicle operation
mode, and (4) engine mileage. The emission rates tend to increase proportion-
ately to vehicle size, which is reasonable because this is the order of in-
creasing fuel consumption. The effect of vehicle size, and mode of operation on
sulfate emission rates has been demonstrated by Springer and Baines (1977).
The national average for diesel and gasoline fuel supply sulfur
content are 0.23% and 0.03%, respectively (Springer and Baines, 1977). Of the
total fuel sulfur, 1-3% is converted to sulfate (Braddock and Gabele, 1977) in
diesel engine operation. The conversion rate is similar in gasoline engines
equipped with a catalytic .converter (Springer and Baines, 1977). However,
Wilson et al. (1977) have shown that the conversion rate can be as high as
13 + 3% in gasoline cars fitted with a catalyst and air pump. Since diesel
fuels contain about 8 times more sulfur than gasoline fuels, the increase of
sulfate emission with increase in fuel sulfur content is pronounced in diesel
vehicles. This is demonstrated in Table 3.8.
The effect of engine mileage on sulfate emission as noted by Lee
and Duffield (1977) is evident in gasoline vehicles with catalytic converters.
This is due to sulfate storage in the catalyst as shown by the reaction:
A£0 + 3 S0 —' A£
30
-------
TABLE 3.8. SULFATE EMISSION RATES WITH THE VARIATION
OF FUEL SULFUR CONTENT3
Fuel
Jet A
No. 1 No. 2 No. 2D° High-Sulfur
No. 2D
Fuel % sulfur
0.04
0.13
0.23
0.29
0.49
Sulfate emission
rates in SET mode,
mg/mile 2.6
5.4
6.5
7.0
11.7
a. Ref. Braddock & Gabele, 1977
b. National average for No. 2 fuel oil.
c. With the exception of 26.5% aromatics content versus the minimum 27%
specified, this fuel conforms with the No. 2D fuel specifications given
in Federal Register for certification of light duty diesel engines.
31
-------
On a fresh catalyst, much of the SC>3 that is formed is stored as
AH (SO.) . However, as the catalyst nears saturation, the reverse reaction
becomes prominent. The equilibrium storage/release level of a catalyst is a
function of catalyst temperature, catalyst space velocities, feed gas composi-
tion and time (Somers et^ ai. , 1977).
Although the irate of sulfate emission is shown in Table 3.7,
nothing has been said about the form of sulfate emitted from automobiles, that
is, how much of it is emitted in the free acid and salt state. This could be
important in relation to the study of any possible effects on health. It is
not known what types of sulfate are emitted by the diesel. In the case of
gasoline cars, essentially all the sulfate is in sulfuric acid form (Lee and
Duffield, 1977). A striking increase in the acidity of the particulate has
been observed by Stara et al. (1974) when gasoline engine exhausts are treated
with a catalytic converter. Comparison of the exhaust particulate acidity has
shown that the acidity of catalyst-fitted engines is 65-260 times greater than
engines with no catalytic converter (Stara et ajl., 1974).
The sulfuric acid mist emitted from vehicles is partly neutral-
ized by ammonia to form (NH,)«SO, and NH.HSO,. A part of the acid reacts with
other metallic elements or compounds in the atmosphere to form metal sulfates.
However, it has been estimated by Wilson et al. (1977) that more than 70% of
the sulfate emitted by the vehicles remains in the form of H9SO, at 20 meters
downwind from the point of emission.
The emission rates of hydrocyanic acid (HCN) as studied by
Braddock and Gabele (1977) is discussed in Section 3.2.
Certainly, anions other than sulfate, such as nitrate and
carbonate are emitted from automobile exhaust. Campbell and Dartnell (1972)
32
-------
have estimated that the nitrate ion from non-catalyst gasoline cars comprises
7.3% by weight of the total particulate. Stara e_t ad. (1974) have reported
nitrate emission rates from gasoline cars with and without a catalytic con-
verter as 0.01 mg/mile and 0.04 mg/mile, respectively, at 15 MPH speed.
3.1.7 Elemental Carbon and Unburned and Partially Burned Fuel
and Lubricant
The combustion of hydrocarbon fuels under "non-ideal" conditions
involves complex processes. Some fuel passes the combustion zone unaltered,
while some cracking processes in the ignition zone produce lower molecular
weight compounds. Cracking processes out of contact with Cv may produce ele-
mental carbon. Some of the fuel is chemically rearranged by pyrolysis to
produce fragments. Consequently, some new products can be formed by the inter-
action of various fragments of the fuel molecules. A part of the fuel under-
goes oxidation producing partially oxidized products. Besides the fuel,
lubricating oil is partially responsible for hydrocarbon emissions from auto-
mobiles.
Both absolute and relative concentrations of combustion products
are influenced by numerous factors. Some of the most prominant factors are:
(1) fuel-to-air ratio, (2) ignition timing, (3) inlet mixture density, (4)
combustion chamber geometry, and (5) the variable parameters, such as speed,
load and engine temperature. The fuel-to-air ratio influences the principal
combustion products more than any other factors.
Although the literature has abundant data on the percent of
carbon in the particulate matter from automobile emissions, the quantitative
distribution of carbon into elemental and organically bound forms is not well
established. One of the better methods for determining elemental carbon has
33
-------
been developed by Schreck £t al_. (1978). According to this method, the exhaust
particulate matter is subjected to thermogravimetric analysis in a NZ atmos-
phere from room temperature to 700°C. The particulate matter will start losing
weight until all of the volatile organic matter has decomposed and/or volatil-
ized and pure carbon particulates are left as residue. At 700°C, the atmos-
phere is changed to air. Oxidation of the remaining carbon particles will
occur. When this process is complete, the. remaining residue from the sample
may be assumed to be due only to nonvolatile metallic compounds, chiefly as
oxides. The residue from soxhlet extraction of particulate matter with benzene;
ethanol (4:1) will also indicate the amount of unburned carbon particles and
metallic compounds.
These methods were applied by Schreck et_ al. (1978) for the
determination of elemental carbon in diesel exhaust. With a Peugeot 504D
engine, these authors found that 40% of the particulate matter consisted of
unburned elemental carbon and metallic compounds. Of the 40%, 31% was elemen-
tal carbon and 9% was metallic residue.
The value for percent elemental carbon from unleaded gasoline
cars with emission controls was determined by Springer and Baines (1977) as
0.08%. It is evident from these results that the percent of elemental carbon
from emission-controlled gasoline cars is very small compared to diesel auto-
mobiles. This is important to keep in mind since the PNA's usually have a
tendency to remain adsorbed on the particulate phase, including the elemental
carbon. Because of the absence of abundant particulate elemental carbon, the
determination of PNA's in gasoline exhaust (with emission control) becomes
especially difficult.
34
-------
Determination of total hydrocarbons (THC) emissions from gaso-
line vehicles is usually performed by collecting the integrated cyclic emissions
in a bag. Measurement of THC emissions from diesel motor vehicles is more
difficult than with gasoline vehicles. The .high boiling range fuels used in
diesel preclude integration of the cyclic emissions in a bag since a major
portion of the. high molecular weight hydrocarbons are lost to the walls of the
bag and other cool surfaces contacted. Thus, the cyclic emissions in diesel
must be "real-time" integrated electronically.
Filtration of particulate matter from diesel emissions for
hydrocarbon determinations poses several problems. First, the particulate
adsorbed form must be distinguished from the free unadsorbed form. If the
particulates are filtered at ambient temperature, the potential exists for
adsorption of gaseous hydrocarbons on the carbon particles during their inti-
mate contact on filtration. On the other hand, heating the filters could
result in desorption of hydrocarbons which would remain particulate-bound under
ambient conditions. This sensitivity of the gas phase-particulate phase parti-
tioning of the organics to a variety of controllable sampling parameters was
examined by Black and High (1978). The findings of their investigations are
summarized below.
1. Instead of sample collection on hot (375°F) or cold (85°F)
filters, a turbulent flow tunnel mixing system described
in this work appears to adequately simulate the important
parameters of short-term ambient dilution. The small
variation of filter temperature under this sampling con-
dition did not affect the particulate collection efficiency.
2. The probability of organic adsorption on the retained parti-
culates on the filters was extremely low under this collec-
tion technique.
3. Isokinetic sampling is not important with diesel particu-
lates under most conditions due to their very small aero-
dynamic size.
35
-------
4. The distribution of organics between the gas and particulate
phases is not sensitive to dilution ratio (dilution factor
8.7 to 17.9). There was indication of a small shift towards
somewhat small particulate size at higher dilution rates.
5. The filtration media is also important as it was found that
glass fiber media acted as sorption media for some gas phase
organics. Best results were obtained with Teflon-coated
glass fiber filter media (Pallflex Type T60 A20).
Determination of organic emissions by the above authors from
diesel passenger cars driven under the cyclic patterns of the FTP showed that
the hydrocarbons range from CL to about C,n. The hydrocarbons in the CL to
C range result from the combustion process, that is, cracking from higher
molecular weight organics. The CL to C organics are dominated by C. , CL, and
C hydrocarbons. The C_n to C, organics are dominated by uncombusted fuel and
lubricant and by partial combustion and rearrangement of these compounds. Some
of the organics in this range (C to C, ) are particle bound and some are gas
phase. Typically 15 to 20% of the THC from the vehicles are associated with
the particles, but this can range to 40% with vehicles emitting significant
levels of lubricant.
NMR determination of the solvent extracted portion of the parti-
culate matter from a Peugeot 504D car showed that 45 wt% of the extract con-
sisted of aromatic compounds with two or more rings (Schreck et^ al., 1978).
Only 8% of the 45% consisted of three or more rings. Liquid exclusion chroma-
tography (LEG) with styragel columns showed that the particle bound organics
contained compounds with carbon numbers more than CL,,. The origin of the
higher carbon number compounds was probably lubricating oil (Schreck et al., 1978)
The values for the hydrocarbons from automobile emissions are
normally reported as total hydrocarbon (THC) and volatile hydrocarbons. These
36
-------
values are given in later sections. Quantitative concentration values for the
individual compounds in the particulate adsorbed hydrocarbons are not available.
Various authors have attempted to identify these compounds. Boyer and Laitenen
(1975) have detected hundreds of compounds in the molecular weight range 300 to
500 in gasoline automobile exhaust particulates by extraction and fractionation
of the extract. The first fraction consisted of straight chain aliphatic
hydrocarbons from Cn,H0/ to C_.H,0. The second fraction contained branched
lo J4 Jj Oo
chain aliphatics, unsaturated aliphatics and small ring compounds. The indiv-
idual components in this fraction have not been identified. The third fraction
consisted of PNA's. Because of the carcinogenic properties of some of these
compounds the PNA's will be discussed individually (see Section 3.1.9). The
fourth fraction contained mostly oxygenates. The individual components detected
in this fraction by GC-MS are: 1) diethyl phthalate, 2) di-isobutyl phthalate,
3) di-n-butyl phthalate, 4) triphenyl phosphate, 5) di-n-octyl phthalate, 6)
methyl triphenyl phosphate, 7) trimethyl triphenyl phosphate, 8) dimethyl
triphenyl phosphate, 9) benzanthrone, 10) suspected 8-capryophyllene, benzo[cj-
cinnoline, benzoic acid, 2,6-di-tert-butyl hydroquinone or nonylphenol, and a
number of other unidentifed compounds including oxygenated PNA's. The relative
amounts of the classes of compounds has been estimated to be 50% saturated
aliphatics, 5% PNA's and 30% oxygenated hydrocarbons (Boyer and Laitinen, 1975).
Organic compounds adsorbed on diesel exhaust particulates have
been studied by Mentser and Sharkey (1977). The list of compounds detected by
these authors by high resolution MS with diesel fuel oil No. 1 and No. 2 is as
follows: 1) crotonaldehyde, 2) g-propriolactone, 3) unresolved 2-butanone,
tetrahydrofuran, 4) pentane, 5) unresolved ethylformate, 2,3-epoxy-l-propanol,
37
-------
methyl acetate, 6) carbon disulfide, 7) benzene, 8) pyridine, 9) cyclohexane,
10) cyclohexene, 11) methylacrylate, 12) 2-pentanone, 13) hexane, 14) unresolved
dioxane, ethyl acetate, 15) toluene, 16) aniline, 17) phenol, 18) furfural, 19)
furfuryl alcohol, 20) unresolved mesityl oxide, cyclohexanone, 21) methyl
cyclohexane, 22) unresolved cyclohexanol, 2-hexanone, 23) unresolved ethyl
acrylate, methyl methacrylate, 24) heptane, 25) styrene, 26) unresolved ethyl
benzene, xylene, 27) unresolved monomethylaniline, 0-toluidine, 28) cresol, 29)
hydroquinone, 30) methylcyclohexanone, 31) allylglycidyl ether, 32) octane, 33)
unresolved vinyl toluene, a-methyl styrene, 34) cumene, 35) isophorone, 36) p-
tert-butyl toluene, 37) phenylglycidyl ether, 38) camphor, 39) phenyl ether,
and 40) dinitro-o-cresol. The lower molecular weight compounds in the list
have not been reported by other workers. Besides the above listed compounds,
sulfur containing compounds, namely benzothiophenes and dibenzothiophenes have
been detected in diesel exhaust (NIOSH, 1978).
The distribution of organic compounds in. particulates changes in
a consistent and characteristic manner as the speed and loading of the engines
are increased from the idle to rated speed and full load. The composition
profiles are not largely affected by the type of diesel engine or the type of
fuel used (Mentser and Sharkey, 1977).
The above hydrocarbons in the presence of NO and light can be
A
responsible for the formation of photochemical smog. Among the oxygenates,
aldehydes are extremely reactive. Low molecular weight saturated ketones,
alcohols, esters, and ethers are unreactive (Seizinger and Dimitriades, 1972).
No reactivity data for heavier or unsaturated ketones, alcohols, ethers, and
nitroalkanes have been reported. Seizinger and Dimitriades (1972) have sugges-
ted that the unsaturated oxygenates might possess significant reactivity.
38
-------
3.1.8 Polycyclic Aromatic Compounds
This class of compounds is found in the automotive exhaust
mostly in the particulate adsorbed phase. These compounds are discussed
separately because of the demonstrated animal carcinogenic effects of some of
these compounds.
One group of polycyclic aromatic compounds, the polycyclic
aromatic hydrocarbons (PNA's), have been detected in both diesel and gasoline
exhaust. These compounds originate from three sources: (1) PNA's present in
original fuel, (2) synthesis from lower molecular weight hydrocarbons during
fuel ignition, and (3) pyrolysis of lubricating oil. The mechanism of PNA
formation in automotive engines has been demonstrated by Laity e_t^ al. (1973).
PNA's apparently can exist in the quench zone at the surfaces of the combustion
chamber. Some of these PNA's are vaporized from the walls or deposits during
engine operation. Anything that increases the heat input to the combustion
chamber walls, for example, advanced ignition timing, knock, use of hydrocarbon
fuels, or high speed operation, leads to enhanced PNA emissions (Laity et ajl.,
1973). From the examination of the soot particles obtained from a gasoline-
powered passenger car and a diesel-powered omnibus, Lyons qualitatively detec-
ted a series of PNA compounds shown in Table 3.9.
Quantitative comparison of the PNA emission levels from diesel
and gasoline exhaust requires that the vehicles at least be of similar duty
category and operated under typical driving and fuel conditions. Although
several publications have reported the levels of various PNA in the exhaust
from gasoline- and diesel-powered vehicles, the results in most cases cannot be
used for comparative purposes because either the engine, fuel, or driving
39
-------
TABLE 3. 9. PNA1 s DETECTED IN VARIOUS ATMOSPHERIC POLLUTANT SAMPLES3
Compound Gasoline Diesel soot Atmospheric
SOOt 800t
Naphthalene +
Acenaphthylene +• + "*"
Anthracene + + +
Phenanthrene 4-
Anthracene derivatives + + +
Pyrene + + +
Fluoranthene + + • +
Alkyl pyrene +
Benz (a) anthracene + + H"
Chrysene + - +
Benzo(e)pyrene + •)• +
Perylene 4- + +
Benzo(a)pyrene + + +
Benzo(ghi)perylene + + +
Benzo(b)f luoranthene 4- + +
Anthranthrene H- + +
Tetracene +
Coronene 4-4- 4-
Dibenz(a,h)anthracene + -
Dibenzo(a,l)pyrene +• +
Benzo(k)f luoranthene + 4- +
Pentaphene 4- +
Dibenzo(a,l)naphthacene +
Dibenzo (a,h)pyrene • 4-
Dibenzo(a,e)pyrene +
Dibenzo (b.pqr)perylene 4- -
Dibenzofluorene 7 - + +
Tribenzo(h,rst)pentaphene 4-
Indeno-1,2,3-f luoranthene ? 4-
a. Re if. Lyons, 1962
b. Detected in sample
c> Not detected in sample
40
-------
mode were not comparable. The average PNA emissions from typical diesel
vehicles run under the 13-mode federal cycle, and typical gasoline automobiles
run under European and American city driving schedules are presented in Table
3.10. The data from older cars are included since current data on detailed PNA
analysis are not available.
The fuels used in the tests given in Table 3.10 are as follows:
Diesel-2D diesel fuel with 26% aromatics; American cars - typical regular grade
gasoline fuels with 25% aromatics; European cars - a blend containing 47.7%
aromatics and 52.3% paraffins. It should be pointed out that the diesel
vehicles in Table 3.10 are the heavy duty variety. Due to unavailability of
data from light duty vehicles, the heavy duty vehicle has been used for the
purpose of comparison.
A large uncertainty in PNA levels from automobile exhaust can be
expected. The discrepency between reported PNA values exists primarily because
of difficulties in sample collection and analytical procedures, and the depen-
dence of PNA emissions on engine operating conditions. This is reflected in
Table 3.11.
Compounds other than those listed in Table 3.9 and Table 3.10
have also been detected in exhausts from automobiles operated on both types of
fuel. For example, Grimmer (1977) has reported six PNA's, two of which are
unknown compounds of molecular weight 300 and the rest are cyclopento[cd]-
pyrene, methylenebenzo[a]pyrene, methylenebenzo[e]pyrene, and methylenebenzo-
[ghijperylene. Grimmer (1977) felt that this group of compounds accounted for
the predominant part of the carcinogenic effect observed with gasoline engine
exhaust extracts in mouse skin-painting studies.
41
-------
TABLE 3.10. COMPARISON OF PNA EMISSION RATES FROM HEAVY DUTY
DIESEL- AND GASOLINE-POWERED VEHICLES
Compounds
Anthracene
Phenanthrene
Phenanthrene derivatives
Fluor anthene
Pyrene
Benz (a) anthracene
Chrysene
Benzo ( j+k) f luoranthene
Benzo(a)pyrene
Benzo (e)pyrene
Indeno (1,2,3 , -cd) pyrene
Benzo (ghi)perylene
Anthranthrene
Coronene
Perylene
Emission rates, pg/gal fuel
Diesel
vehicles3
N.D.d
6410
8280
253
349
35f
5
N.R.
22
4
N.R.
7
N.R.
N.R.
N.R.
Typical 6 cyclinder
1956-1962 American
cars (gasoline)
41
176
N.R.
872
1145
N.R.
N.R.
N.R.
147
205
N.R.
649
27
256
12
burned
1970 European
car (gasoline)
1486
N.R.6
N.R.
891
2159
123
246
33
63
147
97
423
N.R.
197
N.R.
a. Ref. Spindt, 1974
b. Ref. Hangebrauck, 1967
c. Ref. Candeli et al., 1974
d. N.D.: not detected
e. N.R.: not reported
f. Ref. Spindt, 1977
42
-------
TABLE 3.11. FREQUENCY OF OCCURRENCE OF PNA's IN
DIESEL EXHAUST PARTICIPATES a
1 •" ' •" ..~^-~- * — . ' ' ^ ' '
Formula
C18H12
C20H12
C20H14
C20H16
C21H14
C20H13N
C22H12
C22H14
Compound Carcino- Mol. wt. Frequency of
genicityk occurrence in
30 samples
Chrysene + 228.0936
Benzo ( c) phenanthrene +++
Benz( a) anthracene +
Benzo(a)pyrene +++ 252.0936
Benzo (b) f luoranthene -H-
Benzo(j)fluoranthene 4+
Benz(j)aceanthrylene 4+ 254.1092
7, 12-Dimethylbenz( a) anthracene +44+ 256.1248
Dibenzo(a,g)fluorenfe 4 266.1092
Dibenzo(c,g)carbazole 444- 267.1045
Indeno(l,2,3-cd)pyrene 4 276.0936
Dibenz (a, h) anthracene 4+4 278.1092
Dibenz (a, j) anthracene +
Dibenz ( a, c) anthracene +
28
16
2
1
2
1
3
2
a. Ref. Menster & Sharkey, 1977
b. Carcinogenicity: +, uncertain; +, carcinogenic; ++, H-44-, Mil, strongly
carcinogenic, as per NAS notation.
43
-------
In a recent publication Wang et_ al. (1978) have speculated on
the presence of 6-nitrobenzo[a]pyrene in gasoline automobile exhaust. The Ames
Salmonella tjrphimurium assay of this compound has shown that this compound is
a direct-acting mutagen with activity comparable to benzo[a]pyrene (Wang
ejt aJL. , 1978). In fact, the formation of these direct-acting mutagens (nitro-
BaP) upon exposure of PNA to gaseous pollutants in smog has been demonstrated
by Pitts £t al. (1978). In simulated atmospheres containing 1 ppm NO and
traces of HNO., direct-acting mutagens are readily formed from both BaP and
perylene, a non-mutagen in the Ames reversion assay (Pitts e£ a!L , 1978). The
nitration reaction produces 6-nitro, 1-nitro and 3-nitro-isomers of BaP and 3-
nitro-isomers of perylene. These authors also suggest that the nitro-deriva-
tives of PNA may eventually photooxidize to polycyclic quinone.
Primarily because of its carcinogenicity and frequency of occur-
rence, BaP has typically been measured as an indicator of PNA emission from
automobile exhausts. Consequently, the bulk of available data is in terms of
BaP, although the use of BaP data as an indicator of other PNA's is highly
questionable.
Polynuclear aza heterocyclics is another class of compounds in
automobile exhaust which can contribute to carcinogenic activity. Sawicki
et al. (1965) determined the amounts of poly aza arenes in gasoline automobile
exhaust which are summarized in Table 3.12. However, these data were generated
with cars not equipped with a catalyst and may be subject to change.
So far, the emission levels of PNA have been discussed without
any specific reference to the dependency of emissions on other parameters. In
fact, the PNA emissions, like all other exhaust emissions, are dependent on a
44
-------
TABLE 3.12. CONCENTRATION OF POLY AZA ARENES IN AUTOMOTIVE EXHAUST3
Compound Cone, in ug per g
exhaust particulate
Benz(h)quinoline 0.2
Benz(c)acridine 0.4
Indenoquinolines 0.9
Dibenz(a,j)acridine < 0.2
Dibenz(a,h)acridine • < 0.2
Alkylbenz(c)acridines < 0.2
a. Ref. Sawicki et al., 1965
45
-------
number of parameters. These are: (1) vehicle characteristics and engine
design, (2) engine operation mode, (3) engine maintenance, (4) fuel composi-
tion, and (5) exhaust emission control system. The effect of each individual
parameter is discussed in the following sections.
3.1.8.1 Dependency of PNA Emission on Vehicle Characteristics
Since the objective of this report is to consider
emissions from light duty vehicles only, emissions from heavy duty vehicles
will not be discussed. Even in light duty automobiles, PNA emissions may be
dependent on the engine displacement capacity of the automobile. Comparing a
number of 1956-1964 V-8 and V-6 engines, Hangebrauck et_ ail. (1967) have failed
tb detect any statistically significant difference in PNA emission rates
between the two engines operated with gasoline. A'Similar conclusion has been
reached by Springer and Baines (1977) from the comparison of two catalytically-
equipped cars, one with a V-8 and the other with an 1-4 engine, and both
powered with gasoline.
In the case of diesel engines, BaP emission rates and
their dependency on engine type are shown in Table 3.13. It can be seen from
Table 3.13 that BaP emission rates may not only depend on engine displacement
but also on the engine type. The Peugeot 504D engine with larger engine dis-
placement showed a lower BaP emission rate per mile. However, this may be due
to the fact that the Peugeot 504D results may have been subjected to sampling
error.
3.1.8.2 Dependency of PNA Emission on Engine Operation
Mode
The dependency of PNA emission on engine speed and
load for a gasoline- and diesel-powered vehicle is shown in Table 3.14 and
Table 3.15.
46
-------
TABLE 3.13. DEPENDENCY OF BaP EMISSION RATES WITH ENGINE TYPE
Vehicle type Engine displace- BaP emission rate,
ment, CID yg/mile
Oldsmobile, V-8 350 7.3a
VW Rabbit, 1-4 90 4.3a
Peugeot 504D 129 1.6b
a. Ref. Springer & Baines, 1977
b. Ref. Braddock & Gabele, 1977
47
-------
The marked contrast between gasoline and diesel
engine exhausts can be noted from Table 3.14 and Table 3.15. Increasing the
engine load resulted in a rapid decrease in PNA emission in the former, and a
marked increase in the latter. As the speed of the engine increased, the
quantity of PNA emitted decreased for the gasoline-powered automobile. No
uniform variation of PNA production with speed was noted for diesel emissions.
The PNA emission rates given in Table 3.14 and Table
3.15 are for older model gasoline cars and the characteristics of the diesel
engine are not identified. With new model gasoline cars equipped with a cataly-
tic converter, the PNA emission rates can be expected to be substantially lower
(see Section 3.1.8.6). The BaP emission rates for diesel-powered passenger
cars under various recently-developed cyclic modes of operations are shown in
Table 3.16. Corresponding results for gasoline cars are not available. That
the BaP emission rates in the FTP mode are higher than in SET and FET modes is
obvious from Table 3.16.
3.1.8.3 Variation of PNA Emission with Engine Maintenance
Both deposits in the combustion chamber and fouling of
the fuel injection system (improper fuel-to-air ratios) can dramatically
increase PNA emissions in vehicular exhaust. With a gasoline-powered auto-
mobile having combustion chamber deposits, Gross (1972) has shown that the
amount of BaP emission could be as much as 6.5 times greater than for clean
engines. The dramatic effect of fuel-to-air ratio, which controls the effi-
ciency of diesel engine operation, on the .variation of PNA emission is shown in
Table 3.17. It is evident from Table 3.17 that diesel engines with improper
48
-------
TABLE 3.14. VARIATION OF PNA EMISSION RATES WITH INCREASING LOAD
AND CONSTANT SPEED OF 1000 r.p.m.3
Engine load
Emission rate, yg/min. for PNA compound
Pyrene
t
0
1/4
1/2
3/4
full
Gasoline
439
59
26
17
21
Diesel"
137
267
536
1800
2500
Benzo (a) py rene
Gasoline
61
0
1
0
0
Diesel
146
465
772
1320
876
Benzo (ghi)perylene
Gasoline
177
45
5
3
2
Diesel
22
42
124
640
1265
Anthranthrene
Gasoline
102
17
0.3
0.3
0.3
Diesel
0
43
223
472
469
a. Ref. Kotin et al., 1955 & Kotin et al., 1954
b. All the diesel results were run under inefficient fuel injection systems.
-------
TABLE 3.15. VARIATION OF PNA EMISSION RATES WITH INCREASING SPEED
AND NO LOAD3
Ui
o
r.p.m.
Emission rate, ug/min. for PNA compound
Pyrene
500
1000
1200
1400
1500
2000
2500
3000
Gasoline
225
439
N.R.
N.R.
507
374
346
121
Dieselb
N.R.°
137
208
188
N.R.
N.R.
N.R.
N.R.
Benzo(a)pyrene
Gasoline
120
61
N.R.
N.R.
33
40
25
13
Diesel
N.R.
146
9
80
N.R.
N.R.
N.R.
N.R.
Benzo (ghi ) p erylene
Gasoline Diesel
235
177
N.R.
N.R.
60
73
70
85
N.R.
22
79
0
N.R.
N.R.
N.R.
N.R.
Anthranthrene
Gasoline
153
102
N.R.
N.R.
36
27
31
14
Diesel
N.R.
0
4.3
20
N.R.
N.R.
N.R.
N.R.
a. Ref. Kotin et al., 1955 and Kotin et al., 1954
b. All the diesel results were run under inefficient fuel injection system.
c. N.R.: not reported.
-------
TABLE 3.16. BENZO(a)PYRENE EMISSION RATES UNDER VARIOUS
MODES OF ENGINE OPERATION
BaP emission rate, yg/mile
Cycle
Oldsmobile V-8a V.W. Rabbit I-4a
FTP • 7.3 4.3
FET , 3.1 2.4
SET 4.0 2.5
a. Springer and Baines, 1977
51
-------
TABLE 3.17. DEPENDENCY OF BaP EMISSION UNDER DIFFERENT DIESEL
ENGINE MAINTENANCE CONDITIONS3
Load
0
1/4
1/2
3/4
full
BaP Emission
Efficient condition*3
0
0
0
0
0
rate, yg/min.
Inefficient condition0
9
47
437
432
1706
a. Ref. Kotin et al., 1955
b. Clean fuel injection system and obtaining samples from completely
warmed-up engine.
c. Fouling of the fuel injection system and/or engine deterioration.
52
-------
maintenance can be a significantly greater source of PNA pollution. However,
well maintained diesel engines may preclude PNA emission into the atmosphere to
a remarkable degree. Optimization of fuel-to-air ratio, except during the
warm-up period, can make a diesel engine almost completely free from PNA
emissions into the atmosphere.
3.1.8.4 Effect of Fuel Composition
Increased fuel aromaticity generally causes an in-
crease in PNA emissions from gasoline engines (Griffing e£ al^., 1971; Candeli
e£ al., 1974). Gross (1972) has shown that an increase in fuel aromaticity
from 11% to 46% causes an increase of 134% in PNA emissions from uncontrolled
gasoline cars. Begeman and Colucci (1970) have demonstrated that the emission
of BaP and benz[a]anthracene increased by 5 and 3.5 times, respectively, by
increasing the BaP content of fuel from 1.1 ppm to 4 ppm. Analogous results
have been obtained by Rinehart e_t al. (1970). According to Gross (1972) fuel
rich in BaP enhances the emission of the same compounds, only if there are
deposits in the combustion chamber of the engine. Stichting Concawe (1974),
however, has contradicted this result and has shown that PNA emission is on the
average 40% lower with fuels without PNA than with fuels containing 1.3 ppm of
BaP at about the same level of aromaticity. Candeli et_ al. (1975) have attempted
to resolve the problem but have been unable to ascertain whether the observed
increase in BaP emission is due to an increase in fuel aromaticity or to an
increase in fuel PNA content.
Tests with two gasoline cars by Gross (1972) have
shown that fuels containing a high-boiling naphtha displayed increased PNA
emissions compared to fuels without the naphtha but with the same fuel aromatics
53
-------
and PNA levels. In a third vehicle, the naphtha effect has been shown to be
reversed.
The immediate effect of the presence of lead in
gasoline on PNA emission has been examined in several laboratories. Begeman
and Colucci (1970) have shown both small increases and small decreases in PNA
emission for the presence of lead in Indolene fuel. Griff ing es£ al_. (1971),
employing two different 1967 vehicles, have not found any effect of lead on BaP
emissions. A similar conclusion has been reached by Gross (1972) from examina-
tion of later model gasoline cars.
3.1.8.5 Effect of Engine Mileage on PNA Emission
The effect of engine mileage on PNA emissions from
gasoline cars is evident from Table 3.18.
Hoffman et al. (1965) have reported BaP emission rates
at two levels of oil consumption for the same V-8 engine. BaP emission rates
equivalent to 19 and 250 ng/mile have been determined for oil consumption of
1 quart per 1600 miles and 1 quart per 200 miles, respectively. The 13 times
greater emission for the high-oil-consumption test suggests that the source of
BaP might have been from burning of oil.
3.1.8.6 Effect of Exhaust Emission Control
Table 3.19 shows the effectiveness of engine modifi-
cation and emission control devices on PNA emission rates.
54
-------
TABLE 3.18. EFFECT OF GASOLINE ENGINE MILEAGE ON PNA EMISSION3
Car Mileage
19000
26000
49000
58000
BaP
5.6
4.2
3.9
21.5
Pyrene
81
70
27
119
PNA emissions,
BeP
9.5
8.1
8.6
23.5
Perylene
0.28
0.78
0.57
1.38
yg/mile
B(ghi)P
26.0
35.0
14.3
77.0
Anthranthrene
2.3
0.64
0.3
3.17
Coronene
9.6
10.7
4.1
32.2
a. Ref. Hagenbrauck et al., 1967
55
-------
TABLE 3,19. EFFECT OF EXHAUST EMISSION CONTROL ON PNA EMISSION
Type of control BaP emission rate, % Reduction
pg/gal fuel consumed
Gasoline:
Uncontrolled (1956-64) 170a
Uncontrolled, 1966 70b 0
Engine modification, 1968 19-25 ^ 70
Air-injected RAM thermal ,
reactor, 1968 1.6 ^98
Catalyst equipped, 1970 1.1 ^ 98
Diesel:
Catalyst treated 80-90°
Water scrubber , 30C
Catalyst + water scrubber 80-90°
a. Hangebrauck et al., 1967
b. Gross, 1972
c. NIOSH, 1978
56
-------
3.2 Volatile Emissions
Table 3.20 summarizes the regulated gaseous emissions data from
diesel-powered passenger cars. The dependence of these emissions on engine
class and operating modes is obvious from this table.
The corresponding values from gasoline vehicles is shown in Table 3.21.
To make comparison easy, the federal light-duty emission standards
are presented in Table 3.22.
,From the emission rates given in these tables it can be concluded
that diesels (without emission controls) can be a higher source of hydrocarbon
pollution than catalyst-equipped gasoline cars. The CO emission rates for both
diesel and gasoline cars are about equal with gasoline cars emitting more CO in
the FTP and less in the FET and SET modes than diesel cars'. The NO, emission
X
rates for the gasoline cars, on the other hand, are higher than for diesel cars
in all modes of cyclic operation.
The individual hydrocarbon emission rates for diesel-powered passen-
ger cars are given in Table 3.23. The corresponding values for gasoline cars
are presented in Table 3.24. With the exception of methane and toluene, in-
dividual hydrocarbon emission levels are higher for diesel than for gasoline
cars.
The emission rates of earbonyl compounds for diesel cars are shown in
Table 3.25. Tha emission rates for carbonyl compounds for gasoline cars are
shown in Table 3.26. It is evident from these tables that with the exception
of crotonaldehyde, diesel cars emit more carbonyl compounds than catalyst-
equipped gasoline cars. That the emission rates for aromatic aldehydes increase
with fuel aromaticity is shown in Table 3.27 for cars without emission controls.
37
-------
TABLE 3.20, GASEOUS EMISSIONS DATA FROM A VARIETY OF DIESEL CARS
UNDER DIFFERENT ENGINE MODES
00
Yehicle
Mercedes 220Da
Mercedes 240Da
Mercedes 300Da
Peogeot 204Da
Peugeot 504D
Nissan 220C°
Oldsmobile
V.W. Rabbitd
Engine displac(
ment, lit
2.
2.
3.
1.
2.
-N.
5.
1.
20
4
0
36
11
A.e
74
47
Emission rates, g/km
Hydrocarbons
FTP
0.11
0.18
0.10
0.69
0.29
0.25
0.47
0.23
FET
0.08
0.06
0.06
0.48
0.07
N.A.
0.21
0.08
SET
0.06
0.06
0.08
0.54
0.12
N.A.
0.27
0.09
FTP
0.81
0.60
0.53
1.06
0.88
1.10
1.24
0.49
CO
FET
0.48
0.38
0.36
0.57
0.37
N.A.
0.63
0.31
SET
0.55
0.45
0.39
0.71
0.51
N.A.
0.79
0.34
PXP
0.65
0.79
1.07
0.42
1.63
1.37
0.70
0.54
NOX
FET
0.56
0.80
0.99
0.34
1.20
N.A.
0.59
0.52
SET
0.57
0.78
0.98
0.33
1.33
N.A.
0.59
0.50
a. Ref. Springer and Stahman, 1977
b. Ref. Braddock and Gabele, 1977
c. Ref. EPA result cited in b
d. Ref. Springer and Baines, 1977
e. N.A.: Not available
-------
Ut
VO
TABLE 3.21. GASEOUS EMISSIONS DATA FROM A VARIETY OF GASOLINE CARS
WITH AND WITHOUT CATALYST
Vehicle Emission rates,
Hydrocarbons CO NOX
FTP FET SET FTP FET SET FTP FET SET
1977 Catalyst equipped Olds. Cutlassa 0.24 0.06 0.08 1.34 0.12 0.53 0.85 0.88 0.86
1977 Catalyst equipped V.W. Rabbit3 0.14 0.03 -0.03 2.30 0.03 0.19 0.63 1.22 1.01
1970 Mercedes (no catalyst)b 1.66 N.A.d N.A. 20.05 N.A. N.A. 2.19 N.A. N.A.
1968 Air-injected RAM thermal
reactor vehicle^ 0.04 N.A; N.A. 2.60 N.A. N.A. 1.18 N.A. N.A.
1970 Catalyst equipped carc 0.25 N.A. N.A, 4.96 N.A. N.A. 0.43 N.A. N.A.
a. Ref. Springer and Baines, 1977. This is a prototype automobile.
b. Ref. Springer, 1971
c. Ref. Gross, 1972
d. N.A. : not available
-------
TABLE 3.22. FEDERAL LIGHT-DUTY EMISSION STANDARDS3
Emission Standards, g/mile
ItJcir
1977-79
1980
1981
Hydrocarbons
1.5
90% reduction from
1970 value
90% reduction from
1970 value
CO
15
7
90% reduction from
1970 value
NO
X
2
2
1
a Ref. Public Law 95-95 issued Aug. 7, 1977.
60
-------
TABLE 3.23. DETAILED HYDROCARBONS EMISSION RATES (mg/km) FOR
DIESEL CARS DURING TRANSIENT CYCLES
Emission
Methane
Ethylene
Acetylene
Propylene
Ethane
Propane
Benzene
Toluene
Cycle
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
Mercedes
220D a
19.65
13.82
12.58
21.63
15.26
14.56
8.65
6.51
6.07
N.D.C
N.D.
N.D.
N.A/
N.A.
N.A.
N.A.
N.A.
N.A.
6.07
3.26
4.99
N.A.
N.A.
N.A.
Mercedes
24 OD a
5.57
2.75
3.66
17.74
12.07
12.14
1.31
5.02
8.94
N.D.
N.D.
N.D.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.D.
N.D.
N.D.
N.A.
N.A.
N.A.
Mercedes
300D a
3.94
3.34
4.25
14.49
9.39
8.49
2.81
trace
3.50
N.D.
N.D.
N.D.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
2.51
N.D'.
3.10
N.A.
N.A.
N.A.
Peugeot
204D a
9.30
4.44
4.04
38.13
27.76
24.20
7.57
5.08
4.77
19.98
10.93
9.37
N.A.
N.A.
N.A.
N.A.
N.A.
.N.A.
N.D.
N.D.
N.D.
N.A.
N.A.
N.A.
Cutlass
12.7
5.1
3.4
49.2
28.5
21.8
5.3
2.6
1.9
17.1
8.9
6.5
4.2
1.7
0.3
0.1
N.D.
N.D.
11.6
6.4
4.9
2.6
N.D.
0.9
v.w.b
Rabbit
6.7
3.3
4.7
28.1
15.3
15.1
1.5
1.2
1.7
9.6
4.5
4.5
0.9
0.4
0.6
N.D.
N.D.
N.D.
5.1
2.9
3.2
0.6
1.6
N.D.
b. Ref. Springer and Baines, 1977
c. N.D. not detected
d. N.A. not available
61
-------
TABLE 3.24. DETAILED HYDOCARBON EMISSION RATES (mg/km) DURING
TRANSIENT CYCLES OF GASOLINE CARS3
Emission
Methane
Ethylene
Acetylene
Propylene
Ethane
Propane
Benzene
Toluene
Cycle
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
Cutlass
29.5
24.2
18.4
18.2
6.3
2.9
1.1
N.D.b
N.D.
8.2
N.D.
N.D.
14.8
9.2
6.7
N.D.
N.D.
N.D.
5.6
2.7
0.8
13.6
2.4
1.4
V.W. Rabbit
33.0
17.4
14.6
15.2
1.1
2.0
2.6
N.D.
N.D.
4.0
N.D.
N.D.
6.4
2.4
N.D.
N.D.
N.D.
N.D.
9.1
'N.D.
0.4
12.3
1.2
0.9
a. Ref. Springer and Baines, 1977
b. N.D.; not detected.
62
-------
TABLE 3.25. EMISSION RATES (mg/km) FOR CARBONYL COMPOUNDS
FROM DIESEL CARS DURING TRANSIENT CYCLES
Emission
Formaldehyde
Acetaldehyde
Acetone
Iso-butalde-
hyde
Crotonalde-
hyde
Hexanalde-
hyde
Benzaldehyde
Cycle
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
Mercedes
220Da
2.52
1.50
1.55
1.00
N.D.C
N.D.
8.37
10.15
4.31
11.08
15.96
7.90
2.37
2.82
1.75
0.47
1.81
1.55
N.D.
N.D.
N.D.
Mercedes
240Da
3.96
3.08
3.57
1.13
0.55
1.19
1.47
2.99
2.46
2.19
3.58
4.26
0.67
2.27
1.51
N.D.
0.13
0.67
N.D.
2.29
2.00
Mercedes
300D3
3.80
5.81
3.95
1.11
N.D.
N.D.
9.41
6.53
1.45
N.D.
N.D.
1.50
1.17
1.17
N.D.
N.D.
N.D.
0.45
N.D.
N.D.
1.22
Peugeot
204Da
11.25
8.50
7.76
4.28
3.75
4.05
3.01
1.58
4.61
8.75
6.54
7.92
4.10
2.75
3.05
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
Cutlass
15.8
12.3
8.2
6.5
6.3
3.0
35.7
5.1
3.3
18.5
10.4
8.9
4.2
2.4
N.A.
N.A.
N.A.
N.A.
1.8
1.1
N.A.
V.W.b
Rabbit
16.0
6.0
4.3
5.0
1.5
1.1
2.6
1.0
2.7
16.0
1.9
3.3
N.A.d
N.A.
N.A.
N.A.
N.A.
N.A.
2.7
N.A.
N.A.
a. Ref. Springer and Baines, 1977
b. Ref. Springer and Baines, 1977
c. N.D.: not detected.
d. N.A.: not available.
63
-------
TABLE 3.26. DETAILED CARBONYL EMISSION RATES (mg/km) FOR
GASOLINE CARS DURING TRANSIENT CYCLES3
Emission
Formaldehyde
Acetaldehyde
Acetone
Iso-butanaldehyde
Crotonaldehyde
Hexanaldehyde
Benzaldehyde
Cycle
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
FTP
SET
FET
Cutlass
2.6
1.3
1.6
0.4
N.A.
N.A.
N.A.
N.A.
0.5
3.8
1.8
6.5
7.2
0.6
1.2
N.A.
N.A.
N.A.
N.A.
5.3
0.8
V.W. Rabbit
0.4
0.3
0.5
N.A.b
N.A.
N.A.
N.A.
N.A.
N.A.
2.6
2.1
6.4
32.1
4.3
3.7
N.A.
N.A.
N.A.
2.7
2.2
1.0
a. Ref. Springer and Baines, 1977
b. N.A.: not available
64
-------
TABLE 3,27. INCREASE IN AROMATIC ALDEHYDE EMISSION RATES FOR
GASOLINE CARS WITH INCREASE IN FUEL AROMATICITY3
Fuel Aromatics,
mole %
Unleaded Premium 46 . 6
Leaded Premium 30.8
Leaded Regular 27.3
Total aldehydes,
ppm
65
72
69
Aromatic
aldehyde ,
ppm
13.6
6.1
5.8
a. Ref. Hinkamp et al., 1971
65
-------
TABLE 3.28. PHENOL IN EXHAUST GAS
Phenol emission range, mg/gal
Test vehicle
11%
Aromatic fuels
Aromatic fuels
46%
Aromatic fuels
1966, no emission control
1968, engine modification
1970, engine modication
with spark retard
1968, air injected RAM
thermal reactor
1970, catalyst equipped
110-179
78-165
74-123
279-435
314-406
222-287
653-776
535-691
370-485
Ref. Gross (1972).
66
-------
TABLE 3.29. SO- EMISSION RATE FROM VARIOUS CARS UNDER
CYCLIC OPERATIONS3
SO,, Emission Rate mg/km
Cycle Mercedes 220D Mercedes 240D Mercedes 300D Peugeot 204D
FTP 350 320 . 310 260
FET 270 270 320 210
SET 250 250 260 200
o
Ref. Springer and Stahman, 1977.
67
-------
TABLE 3.30. COMPARISON OF HCN AND COS HUSSIONS (mg/mlle) FROM DIESEL AND GASOLINE CARS*
00
Emission Cycle
HCN FTP
FET
SET
e
COS FTP
FET
SET
Peugeot 504D
1.32 + 0.31°
0.63 + 0.39
0.43 + 0.05
0.55 + 0.15
0.19 + 0.22
0.24 + 0.14
Honda CVCC Lean-burn Chrysler
11.5 + 2.0
7.6 + 0.5
NR
0.06 + 0.06
0.51 + 0.67
NR
4.44 + 2.67
NRd
NR
0.40 + 0.35
0.26 + 0.16
NR
Dual Catalyst Hornet
10.7 + 1.6
7.2 + 0.3
5.2 + 0.8
NR
NR
NR
a Ref. Braddocfc & Gabele (1977).
The tiiesel tests were run with national average diesel, and gasoline tests were run with
unleaded gasoline.
£»
Means standard deviation from mean.
The diesel tests were run with 0.46 weight% fuel sulfur, and gasoline tests were run with
0.03 weight% fuel sulfur.
-------
Phenols have also been detected in exhaust gases from gasoline cars.
Table 3.28 shows the phenol emission rates as the gasoline cars became more and
more modified. It is evident from Table 3.28 that phenol emission increases
with increase in fuel aromaticity. However, with the present catalyst equipped
cars the effect may not be pronounced since the phenol emission rate is too
low.
Sulfur dioxide emission rates from diesel cars are dependent on fuel
sulfur content. As the fuel sulfur increases, SO- emission also increases
(Braddock and Gabele, 1977). The SCL emission rate from a number of cars
operating with national average fuel sulfur (0.23%) is shown in Table 3.29.
The emission rates of HCN and COS from both diesel and gasoline
passenger cars are listed in Table 3.30.
69
-------
3.3 Fuel Economy
For comparison purposes, Springer and Baines (1977) have used one
large (Oldsmobile V-8) and one small car (V.W. Rabbit 1-4) in each diesel and
gasoline category. Their results are summarized as follows: fuel consumption
(&/100 Km) of the diesel Cutlass is consistently 26 to 29% lower than the
gasoline car regardless of the driving cycle. In terms of fuel economy (mpg),
the percent increase in miles per gallon for the diesel is 35 to 40% greater
than the gasoline car. In the case of the Rabbit, the fuel consumption rates
for the diesel are 42%, 39% and 33% lower for FTP, SET and FET tests, respec-
tively, compared to the gasoline Rabbit. In terms of fuel economy, the corres-
ponding percent increase amounts to 74%, 65% and 49%, respectively.
70
-------
3.4 Smoke Results
The results of a diesel smoke test on the larger Oldsmobile Cutlass
and smaller V.W. Rabbit car are presented in Table 3.31.
It should be noted that 3 to 4% opacity by the EPA smokemeter is at
the limit of smoke visibility. Most of the time, both cars operated in this
area with brief excursions during rapid throttle movement.
71
-------
TABLE 3.31. PERCENT EXHAUST SMOKE OPACITY FOR TWO DIESEL CARS
Cutlass
Condition
Start
Idle
First acceleration peak
Idle at 1255 sec.
Second acceleration peak
Cold start
cycle
16v.3
4.4
21.4
5.2
19.4
Hot start
cycle
7.8
4.1
7.5
4.3
16.6
Rabbit
Cold start
cycle
72.9
4.5
7.4
0.5
39.4
Hot start
cycle
27.4
0.4
3.0
0.3
37.7
Ref. Springer and Baines, 1977
72
-------
3.5 Odor Rating
The odor rating by the Turk Kit method includes an overall "D" odor
which is comprised of burnt-smoky "B", oily "0", aromatic "A", and pungent "P"
qualities as determined by an odor panel. On an odor intensity series of one
through four, the last is considered the strongest odor. The odor intensity
determined by this method (odor panel) for a number of diesel cars under
various engine modes of operation is given in detail by Springer and Stahman
(1977) and Springer and Baines (1977). Since the objective of this report is
to identify the chemical components, the odor rating based on this scale will
not be discussed. The reader is referred to the previously mentioned investi-
gations. Another system of odor rating called the Diesel Odorant Analytical
System (DOAS) which expresses odorant as Total Intensity of Aroma (TIA) has
been used by Springer and Baines (1977) for diesel cars.
At IIT Research Institute, Dravnieks and coworkers (Dravnieks et al.,
1971; O'Donnell _et_ al., 1970) have employed two high-resolution chromatographic
columns for separation of diesel exhaust components in order to identify diesel
odorants. Table 3.32 lists the odorants they determined quantitatively.
Another group which has been conducting odor-related research on
diesel exhaust for a number of years is Arthur D. Little Co. Based on the
results of their odor studies (Spicer et^ al., 1975), ADL investigators have
identified a large number of aromatic compounds and their isomers which are
listed in Table 3.33.
73
-------
TABLE 3.32. EXHAUST CONCENTRATIONS OF SOME ODORANTS AS DETERMINED BY
IITRI WITH HIGH-RESOLUTION CHROMATOGRAPHY3
Exhaust component Concentration, ppm
Acetaldehyde 0.00003
n-Butanol 0.00017
Decane 0.00344
Methyl benzene 0.000009
Cg Substituted benzene 0.000032
Allyl toluene 0.000037
Methylindan 0.000074
Benzaldehyde 0.000345
Naphthalene 0.00038
Methyl naphthalene 0.00034
Ref. Dravnieks et^ al_. (1971), O'Donnell e£ al. (1970). It is not clear
from the original reference whether these estimates refer to actual
exhaust concentrations or to exhaust which has been diluted 11 to 1.
74
-------
TABLE 3.33. DIESEL EXHAUST ODORANTS IDENTIFIED BY ADL
Compound
Methylindan
Tetralin
Dimethylindan
Methyltetralin
Dimethyltetralin
Trimettiylindan
Alkyltetralin
Trimethyltetralin
Alkyltetralin
Alkylindene
Alkylindene
Monomethyl napthalene
Composition
C- _H _
10 12
C10H12
C11H14
C11H14
C12H16
C12tt16
C12H16
C13H18
C13H18
C12H14
C13H16
C11H16
Odor note
Irritation
Rubbery sulfide
Kerosene
Naphthenate
Kerosene
Kerosene, irritation
Kerosene
Irritation
Kerosene, pungent, acid
Heavy oil
Heavy oil
Mothball, irritation
Ref. Spicer e£al. (1975)
75
-------
3.6 Noise
A summary of sound levels from diesel and gasoline cars under differ-
ent driving conditions is shown in Table 3.34.
The driveby exterior rating for a diesel Cutlass has been found to be
5 dBA higher than for a gasoline Cutlass, while the Rabbit has shown the same
dBA level under this driving condition. Interior noise levels are slightly
higher with diesels of both makes during acceleration. The exterior driveby at
a constant 48.3 Km/hr speed has shown slightly higher interior and exterior
noise for the Cutlass, while the opposite is true for the gasoline Rabbit.
Idle noise levels are noticeably higher with the diesel Rabbit.
76
-------
TABLE 3.34. SUMMARY OF SOUND LEVEL MEASUREMENTS
Noise at
Exterior
Interior
Interior
Exterior
Interior
Interior
Exterior
Interior
Interior
Driving mode
Accel, driveby
Blower on
Blower off
48.3 km/hr, driveby
Blower on
Blower off
Idle
Blower on
Blower off
Oldsmobile
Gasoline
68.8
73.2
68.8
58.8
71.5
60.5
64.5
71.5
48.5
Noise on
Cutlass
Diesel
73.8
74.2
70,5
61.2
72.2
64.0
70.0
71.0
51.5
dBA scale
V.W. Rabbit
Gasoline
71.0
78.2
76.5
60.5
73.5
70.5
65.0
69.5
58.0
Diesel
71.5
80.0
79.5
58.5
71.8
68.0
67.0
69.5
62.5
Ref. Springer & Baines (1977).
77
-------
3.7 Engine Modification and Antipollution Devices for Diesel Cars
Diesel powered cars discussed so far do not include automobiles
equipped with antipollution devices. However, for diesel cars, the variation
of combustion systems may result in different pollution characteristics and
fuel economy. Table 3.35 shows qualitatively the characteristics of the two
combustion systems without any other emission control system.
With the introduction of further emission control device(s), the
pollution level can be further decreased in diesel cars. This is indicated in
Table 3.36. Although the emission rates given in this table are for diesel
vehicles run during mining operations, the results can be qualitatively applied
towards diesel passenger cars. The effects of catalytic reactors, certain
types of traps, and a combination of these, on diesel exhaust composition have
been studied in detail by Marshall e_t al. (1978) and Seizinger (1978).
78
-------
TABLE 3.35. FUEL ECONOMY AND EMISSION CHARACTERISTICS OF TWO
DIESEL CARS WITH DIFFERENT COMBUSTION SYSTEMS3
Characteristics Direct Injection Indirect Injection
Fuel economy Favorable Less favorable
CO Less favorable Favorable
Hydrocarbons Less favorable Favorable
NO Less favorable Favorable
Aldehydes Less favorable Favorable
SO- Approximately same Approximately same
a. Ref. NIOSH, 1978
79
-------
TABLE 3.36. DIESEL EMISSION FACTORS WITH AND WITHOUT
EMISSION CONTROL3
Pollutant
CO
Hydrocarbons
NO
N02
Carbon
Phenols
Aldehydes
so2
H2S°4
Trace metals
PNA
Odor
Irritancy
co2
Noise, dBA
Emission level, grams/brake horsepower-hour
Untreated
engine
0.6-2.7
0.03-0.17
1.25-4.1
0.3-0.7
0.17-0.67
Trace
0.02-0. ,2
0.5
Trace
510-600
96-104
Catalyst
treated
0.6-0.3
0.003-0.017
1.25-3.5
0.15-1.1
0.17-0.67
80-90%
reduction
0.005
0.25
0.37
0.025 max
80-90%
reduction
substantial
reduction
—
no reduction
no reduction
Water
scrubber
0.6-2.7
0.02-0.12
1.25-4.1
0.3-0.7
0.12-0.47
30% reduc-
tion
0.01
0.096
30% reduc-
tion
—
some reduc-
tion
no reduction
no reduction
Catalyst and
water scrubber
0.6-0.3
0.003-0.017
1.25-3.5
0.15-1.1
0.08-0.33
80-90% reduction
0.005
<0.09
<0.24
80-90% reduction
—
—
no reduction
no reduction
a. Ref. NIOSH, 1978
80
-------
3.8 Effect of Irradiation of Automobile Exhaust
It has been long known that nitrogen oxides and hydrocarbons present
in automobile exhaust can react photochemically in the presence of sunlight to
produce 'photochemical smog.1 The primary source of Los Angeles smog has been
attributed to automobile emissions arising from evaporative fuel losses and
exhaust discharges. Substantial research efforts have been devoted on this
subject. No attempt has been made in this report to describe all these inves-
tigations. Instead, only a few investigations which demonstrate the altera-
tions of major components as a result of photoreaction of exhaust emissions are
presented below.
3.8.1 Photoreactivity of Gasoline Emissions
Light irradiation of gasoline emissions usually results in oxi-
dation of products which are present in the original emissions. The reactivity
criteria expressed in terms of rate of formation of 0 , N09, peroxyacetyl
nitrate (PAN), peroxypropionyl nitrate, peroxybenzoyl nitrate (PBzN), formalde-
hyde and increase in eye irritation. In the case of gasoline cars, the effects
of increased fuel aromaticity on photoreactivity of exhaust emissions have been
studied by Heuss et_ aJL. (1974). This study is particularly important since
unleaded gasoline used for catalytically equipped cars contains a higher per-
centage of aromatics in order to maintain, the equivalent octane rating. The
study by Heuss et al. (1974) has shown that the presence or absence of tetra-
ethyl lead (TEL) in gasoline does not affect the photochemical reactivity of
the exhaust hydrocarbons produced from the gasoline. Certain aromatics, when
added to a low-aromatic gasoline, greatly increase the eye irritation and PBzN
yield of the exhaust, although they do not increase other reactivity criteria.
81
-------
The six aromatics tested by Heuss et al. (1974) have been ranked in the follow-
ing order for their effect on eye irritation: isopropylbenzene > (o-xylene, n_-
propylbenzene, and ethylbenzene) > toluene > benzene. These authors concluded
that the specific aromatics in gasoline, not the total, is the important factor
affecting eye irritation.
Similar results on photochemical reactivity of gasoline exhaust
in relation to increased fuel aromaticity have been estimated by Altshuller
(1972). His results are summarized in Table 3.37.
3.8.2 Photoreactivity of Diesel Emissions
The photoreactivity of diesel emissions has been studied by EPA
(1978). Their results show small but positive differences in the measured
values of component concentrations between irradiated and non-irradiated
diesel emissions. These differences exist between NO , S09, hydrocarbon and
X £•
particulate matter in both atmospheres. Low molecular weight aliphatic hydro-
carbons and solvent extractables do not show any significant differences. The
EPA (1978) studies are summarized in Table 3.38.
82
-------
TABLE 3.37. EFFECT OF PHOTOCHEMICAL REACTIVITY RESULTING
FROM 10% INCREASE IN FUEL AROMATICITY3
Photochemical Reactivity
Probable Results
Ozone or oxldant
Peroxyacyl nitrate
Formaldehyde and other aldehydes
Eye irritation
Aerosol formation
Plant damage
No increase overall and decrease for
evaporative loss contribution
No increase overall and decrease for
evaporative loss contribution
Decrease with increase in fuel
arpmaticity
2 to 5% increase
10% increase
No change
a. Ref. Altschuller, 1972
83
-------
TABLE 3.38. EFFECT OF IRRADIATION OF DIESEL EMISSIONS'
Component
Concentration in exposure chamber
Non-irradiated
Irradiated
co2%
CO, ppm
Total hydrocarbons, ppm C
75°F
350°F
NO, ppm
N02, ppm
S02, ppm
03, ppm
o
Total particulate, mg/m
3
Sulfate, mg/m
0.252
15.7
15.6
31.2
5.85
2.19
2.13
6.32
0.57
0.255
15.4
15.0
26.0
4.94
2.73
1.91
<0.01
6.83
0.57
a. Ref. EPA, 1978
84
-------
3.9 Research Gaps and Recommendations
The present review of the state-of-the-art knowledge on light duty
vehicular emissions has detected the following areas of research gaps with
regard to physical and chemical characterization.
3.9.1 Definition of Particulate Matter
There is no general agreement about the specific fraction of
exhaust emissions which constitutes the particulate matter. Certainly, the
nature and quantity of the particulate matter will depend on both the temper-
ature of .sample collection and the nature of the filtering medium. Therefore,
a universally acceptable definition of particulate matter requires a standard-
ization of these parameters.
3.9.2 Inadequate Particulate Sampling Procedure
This is perhaps one of the major reasons for both inter- and
intralaboratory inconsistencies between the reported particulate matter con-
centrations from automobile exhausts. The turbulent flow tunnel mixing system
described by Black and High (1978) seems promising. Although the filter paper
and the cooling system for hot exhaust is optimized to accomplish consistent
quantitative collection of particulates, hard data showing the actual collec-
tion efficiency are not available. Data by Spindt (1977) have shown that the
collection efficiency for BaP can be less than 10%. In their experiments with
gasoline exhausts, Springer and Baines (1977) point out the difficulty of
collecting BaP present in particulate matter. Use of radioactive, tracer to
establish the recovery of particulate matter by the collection method used may
be helpful.
The possibility of interactions among pollutants during collec-
tion (namely oxidants with PNA's) and during storage should be further
85
-------
investigated. This interaction may be responsible for the production of arti-
facts during these processes.
3.9.3 Better Storage Method
In cases where real time analyses are not possible, a better
storage method for HLSO, and other reactive components is needed.
3.9.4 Improvement of Analytical Methodology
For certain components where quantifications are hindered by
interferences, namely, aldehydes, PNA's and phenols, better analytical method-
ologies need to be developed. There is a need for an instrument which will
specifically measure NO concentration.
Better analytical procedures are also needed for the reasonable
recovery of adsorbed components in the particulate matter during the solvent
extraction procedure (perhaps ultrasonic extraction). Recovery during eva-
porative concentration of the extract can be improved by such well-established
procedures as Kuderna-Danish evaporation.
3.9.5 Identification and Quantification of New Components
More research is required to identify and quantify suspected new
carcinogenic components in the exhaust, namely, methylene-PNA's, (Grimmer,
1977) and nitro-PNA's, (Pitts e£ al., 1978), and hitherto undetected nitro-
soamines. Fractionation of the extractables in the particulate matter into
acid, base, and neutral fractions and the subsequent separation of each indivi-
dual fraction may be helpful for this purpose. EPA has ongoing programs on the
latter subject.
3.9.6 Analysis of Sulfates
A better method for the analysis of H-SO, and other neutral
sulfates from automobile exhaust is needed. Ion chromatography may be helpful
86
-------
for the detection of these and other ionic components. The time scale for
neutralization of H-SO, in ambient air has to be established.
3.9.7 Effect of NO and 0_ on Pollutant Formation
The effect of NO and 0_ on PNA and other compounds formed
through free radical mechanisms should be investigated. Since NO and 0, are
well-known free radical quenchers, a relationship establishing NO and 0-
concentration and rate of PNA formation in automobile exhaust should be estab-
lished.
3.9.8 Quantification of Different PNA Levels
More research effort should be directed towards establishing
different PNA levels and measuring the effects of variables on PNA's emitted
from light duty vehicles.
3.9.9 Uniformity in Data Reporting
EPA should establish a uniform method for data reporting. Com-
parison between various results sometimes becomes impossible because of the
various units used to express the data.
3.9.10 Diesel Odor Characterization
Diesel odor characteristics as measured by DOAS and total "D"
odor ratings by a human odor panel needs further evaluation because of their
inherent inadequacies. Identification of odor components by chemical methods
and rating odor on the basis of analytical quantification of a few representa-
tive odorants may be a solution to this problem.
3.9.11 Necessity for Using Additives
Use of a smoke suppresant for diesel exhaust is thought to be
neither widespread nor essential. The use of MMT (C&EN News, 1978) or Ce(thd)
87
-------
(Sievers and Sadlowski, 1978) to increase the anti-knock value for gasoline
fuels needs careful examination. The best ways to control emission of pollu-
tants from automobiles may be: (1) control of fuel parameters, (2) introduc-
tion of anti-pollution devices, and (3) engine modification.
3.9.12 Regulation of Pollutants
Finally, decisions have to be made regarding the necessity of
regulation of some of the components in automobile emissions of possible
interest in relation to health effects. The presently regulated gaseous
emissions may be toxic and photochemically reactive, but are not principally
mutagenic/carcinogenic. The regulation of vaporous components has uninten-
tionally reduced the emission of some carcinogenic compounds (PNA's) in gaso-
line exhaust. Whether the presence of residual trace amounts of carcinogenic
compounds poses any long term health hazards has to be assessed both by in
vitro and long term in vivo biological studies.
88
-------
References for Section 3.0
Altshuller, A.P. (1972), "Effects of Reduced Use of Lead in Gasoline on Vehicle
Emissions and Photochemical Reactivity," EPA National Environmental Research
Center, Research Triangle Park, N.C., February, 1972.
Apostolescu, N.D., R.D. Matthew, and R.F. Sawyer (1977), "Effects of a Barium-
Based Fuel Additive on Particulate Emissions from Diesel Engines," SAE
Paper No. 770828.
Begeman, C.R. and J.M. Colucci (1968), "Polynuclear Aromatic Hydrocarbon Emissions
from Automotive Engines," SAE Transactions, Vol. 79, Paper 700469, p. 1682-
1698.
Black, F. and L. High (1978), "Diesel Hydrocarbon Emissions, Particulate and
Gas Phase," Symposium on Diesel Particulate Emissions Measurement Char-
acterization, Ann Arbor, Michigan, May 17-19.
Boyer, K.W. and H.A. Laitinen (1975), "Automobile Exhaust Particulates: Proper-
ties of Environmental Significance," Environ. Sci. Technol., £:457-469.
Braddock, J. and R. Bradow (1975), "Emissions Patterns of Diesel-Powered
Passenger Cars, Part I," SAE Paper No. 750682.
Braddock, J.N. and P.A. Gabele (1977), "Emission Patterns of Diesel Powered
Passenger Cars. Part II," SAE (Tech. Pap.) 770168, 11 pp.
Broome, D. and I.M. Khan, (1971), "Mechanisms of Soot Release from Combustion of
Hydrocarbon Fuels with Particular Reference to the Diesel Engine," Insti-
tutipn of Mechanical Engineers Symposium on Air Pollution in Transport
Engines, Paper C140/71, Nov. 1971.
Campbell, K. and P.L. Dartnell (1972), "Vehicle Particulate Emissions," Air-
Pollut. Control Transp. Engines, Symp., pp. 14-20.
Candeli, A., G. Morozzi, A; Paolacci, and L. Zoccolillo (1975), "Analysis
Using Thin Layer and Gas-Liquid Chromatography of Polycyclic Aromatic
Hydrocarbons in the Exhaust Products from a European Car Runnning on
Fuels Containing a Range of Concentrations of These Hydrocarbons,"
Atmos. Environ., j):843-849.
Candeli, A., V. Mastrandrea, G. Morozzi, and S. Toccaccli (1974), "Carcinogenic
Air Pollutants in the Exhaust from a European Car Operating on Various
Fuels," Atmos. Environ., JJ:693-705.
Chemical and Engineering News, (1978), .56(28), 24 pp.
DEC, New York State (1976), "Sulfate and Particulate Emissions from In-Use
Catalyst Vehicles," EPA Grant No. R803520.
89
-------
Dolan, D.F., D.K. Kittelson, and K.T. Whitby (1975), "Measurement of Diesel
Exhaust Particle Size Distributions," ASME Publication 75-WA/APC-5.
Dow Chemical Co. (1970), "Effects of Fuel Additives on the Chemical and Physical
Characteristics of Particulate Emissions in Automotive Exhaust," EPA
Contract No. EPA-22-69-145.
Dravnieks, A., A. O'Donnell, R. Scholz, and J.D. Stockham (1971), "Gas
Chromatographic Study of Diesel Exhaust Using a Two-Column System,"
Presented at A.C.S. Meeting, Los Angeles, California.
EPA (1977), "Characterization of Exhaust Emissions from a Dual Catalyst-Equipped
Vehicle," Report No. 600/2-77-068.
EPA (1978), Toxicological Assessment of Diesel Emissions, U.S. Environmental
Protection Agency, HERL, Cincinnati, Ohio, April, 1978.
Energy and Environmental Analysis, Inc. (March 6, 1978), Draft Report to EPA,
Atmospheric POM: Source and Population Exposure Estimate.
Frey, J.W. and M. Corn (1967), "Diesel Exhaust Particulates," Nature, 216(5115);
615-616.
Gabele, P., J. Braddock, and R. Bradow (1977), "Characterization of Exhaust
Emissions from Lean Burn, Rotary, and Stratified Charge Engines," SAE
Paper No. 770301.
Ganley, J.T. (1973), "Particulate Formation in Spark Ignition Engine Exhaust Gas,"
Dissertation Work, University of Michigan, available from Univ. Microfilms,
Ann Arbor, Mich., Order No. 73-24,565, 148 pages.
Greeves, G., I.M. Khan, and G. Onion (1977), "Effect of Water Introduction on
Diesel Engine Combustion and Emissions," Symposium (Int.) Combust. (Proc.),
Vol. 16, p. 321-336.
Griffing, M.E., A.R. Maler, J.E. Borland, and R.R. Decker (1971), "Applying a
New Method for Measuring Benzo[a]pyrene in Vehicle Exhaust to the Study
of Fuel Factors," presented at Div. Petrol. Chem., Inc., ACS, Los Angeles
Meeting, March 28 - April 2, 1971.
Grimmer, G. (1977), "Analysis of Automobile Exhaust Condensates," in Air
Pollution and Cancer in Man, IARC Sci. Publ. No. 16.
Gross, G.P. (1972), "The Effect of Fuel and Vehicle Variables on Polynuclear
Aromatic Hydrocarbons and Phenol Emissions," Paper No. 720210. Presented
at the SAE Automotive Engineering Congress and Exposition. Detroit, Michigan.
Habibi, K. (1970), "Characterization of Particulate Lead in Vehicle Exhaust -
Experimental Techniques," Environ. Sci. Technol., 4_:239-248.
90
-------
Habibi, K., E.S. Jacobs, W.G. Kunz, Jr., and D.L. Pastell (1970), Characteriza-
tion and Control of Gaseous and Particulate Emissions from Vehicles,
Presented at the Air Pollution Control Association, Fifth Technical Meet-
ing, San Francisco, October, 1970, available from E.I. DuPont de Nemours
& Co., Wilmington, Delaware.
Hangebrauck, R.P., D.J. von Lehmden, and J.E. Meeker (1967), "Sources of Poly-
nuclear Hydrocarbons in the Atmosphere," Environmental Health Series,
U.S. Dept. of HEW, Cincinnati, Ohio.
Henein, N.A. (1973), "Diesel Engines Combustion and Emission," in Engine
Emissions. Chapter 6, G.S. Springer and D.J. Patterson (eds.), Plenum
Press, New York.
Heuss, J.M., G.J. Nebel, and B.A. D'Alleva (1974) "Effects of Gasoline Aromatic
and Lead Content on Exhaust Hydrocarbon Reactivity," Environ Sci. Technol.,
£: 641-647.
Hinkamp, J.B., M.E. Griffing, and D.W. Zutut," Aromatic Aldehydes and Phenols
in the Exhaust from Leaded and Unleaded Fuels," Presented at Div. Pet.
Chem., Inc., ACS, Los Angeles Meeting, March 28 - April 2, 1971.
Hoffman, D., E. Theisz, and E.L. Wynder (1965), "Studies on the Carcinogenicity
of Gasoline Exhaust," J. Air Poll. Control Assoc., 15;162-165.
Kotin, P., H.L. Falk, and M. Thomas (1955), "Aromatic Hydrocarbons III. Presence
in the Particulate Phase of Diesel-Engine Exhausts and the Carcinogenicity
of Exhaust Extracts," AMA Arch. Ind. Health, 11:113-120.
Kotin, P., H.L. Falk, and M. Thomas (1954), "Aromatic Hydrocarbons II. Presence
in the Particulate Phase of Gasoline-Engine Exhausts and the Carcinogenicity
of Exhaust Extracts," AMA Arch. Ind. Hygiene and Occup. Med., J3:164-177.
Laity, J.L., M.D. Malbin, W.W. Haskell, and W.I. Doty (1973), "Mechanisms of
Polynuclear Aromatic Hydrocarbon Emissions from Automotive Engines,"
SAE Paper No. 730835.
Laresgoiti, A, A.C. Loos, and G.S. Springer (1977), "Particulate and Smoke
Emission from a Light Duty Diesel Engine," Environ. Sci. Technol., 11:973-
978.
Lee, R.E., Jr. and F.V. Duffield (1977), "EPA's Catalyst Research Program:
Environmental Impact of Sulfuric Acid Emissions," J. Air Poll. Control
Assoc., 2^:631-635.
Lee, R.E. and F.V. Duffield (1977b), "Sources of Environmentally Important
Metals in the Atmosphere," Presented at ACS National Meeting, August 29-
September 2, Chicago, Illinois.
91
-------
Lyons, M.J. (1962), "Comparison of Aromatic Polycyclic Hydrocarbons from Gaso-
line Engine and Diesel Engine Exhausts, General Atmospheric Dust, and
Cigarette-Smoke Condensate," NCI Monogr. No. 9, p. 193-199.
Marshall, W.F., D.E. Seizinger, and R.W. Freedman (1978), "Effects of Catalytic
Reactors on Diesel Exhaust Composition," Bureau of Mines Technical Progress
Report No. 105, U.S. Dept. of the Interior.
Menster, M. and A.G. Sharkey, Jr. (1977), Chemical Characterization of Diesel
Exhaust Particulates, NTIS, PERC/RI-77/5.
Millington, B.W. and C.C.J. French (1966), "Diesel Exhaust - A European View-
point," SAE Paper No. 660549.
Moran, J.B. and O.J. Manary (1970), Effect of Fuel Additive on the Chemical
and Physical Characteristics of Particulate Emissions in Automotive
Exhaust, Interim Technical Report to the National Air Pollution Control
Administration, submitted by the Dow Chemical Co., Midland, Michigan,
July, 1970.
Mueller, P.K. (1970), "Characterization of Particulate Lead in Vehicle Exhaust -
Experimental Techniques," Environ. Sci. Technol., 4.:248-251.
Mueller, P.K., H.L. Helwig, A.E. Alcocer, W.K. Gong, and E.E. Jones (1962),
"Concentration of Fine Particles and Lead in Car Exhaust," in Symposium
on Air Pollution Measurement Methods, ASTM Special Technical Publication
352, Philadelphia, A.S.T.M., p. 60-77.
National Academy of Sciences (1976), "Medical and Biologic Effects of Environ-
mental Pollutantss Vapor-Phase Organic Pollutants," NRC, Washington, D.C.
NIOSH (1978), "The Use of Diesel Equipment in Underground Coal Mines," Work
Group Reports from a NIOSH Workshop, Morgantown, W.Va., Sept. 19-23,
NIOSH Publication Feb., 1978.
Ninomiya, J.S., W. Bergman, and B.H. Simpson (1970), "Automotive Particulate
Emissions," Presented at the Second International Clean Air Congress,
International Union of Air Pollution Prevention Association, Washington,
D.C., Dec., 1970, available from Automotive Emissions Office, Ford Motor
Co., Dearborn, Michigan.
O'Donnell, A. and A. Dravnieks (1970), "Chemical Species in Engine Exhaust and
Their Contributions to Exhaust Odors," Report No. IITRI C6183-5, for
NAPCA and CRC, Nov., 1970.
PEDCO Environmental, Inc. (1978), Report on Air Quality Assessment of Particulate
Emissions from Diesel-Powered Vehicles, prepared for Pollutant Strategies
Branch, U.S. Environmental Protection Agency, Research Triangle Park N C
March, 1978. '
92
-------
Pitts, J.N., Jr., K.A. Van Cauwenberghe, D. Grosjean, J.P. Schmid, D.R. Fitz,
W.L. Belser, Jr., G.B. Knudson, and P.M. Hynds (1978), "Atmospheric Reac-
tions of Polycyclic Aromatic Hydrocarbons!: Facite Formation of Mutagenic
Nitro-Derivatives," Science (in press).
Reinhart, W.E., S.A. Gendermalik, and L.F. Gilbert (1970), "Fuel Factors in
Automobile Tailpipe Emissions," presented at American Industrial Hygiene
Conference, Detroit, Michigan, Paper No. 127, 15 pp.
Sampson, R.E. and G.S. Springer (1973), "Effects of Exhaust Gas Temperature and
Fuel Composition on Particulate Emission from Spark Ignition Engines,"
Environ. Sci. Technol., j^:55-60.
Sawicki, E., J.E. Meeker, and M. Morgan (1965), "Polynuclear Aza Compounds in
Automotive Exhaust," Arch. Environ. Health, 11:773-775.
Schreck, R.M. (1978), "Health Effects of Diesel Exhaust," Biomedical Sciences
Dept., General Motors Research Laboratories, Warren, Michigan.
Schreck, R.M., J.J. McGrath, S.J. Swarin, W.E. Bering, P.J. Groblicki, and
J.S. MacDonald (1978), "Characterization of Diesel Exhaust Particulate
for Mutagenic Testing," G.M. Research Report No. 78-33.5, General Motors
Research Labs., Warren, Michigan.
Seizinger, D.E. (1978), "Analysis of Carbonaceous Diesel Emissions," Presented
at the Conference on Carbonaceous Particles in'the Atmosphere, March 20-
22, Berkeley, California.
Seizinger, D.E. and B. Dimitriades (1972), Oxygenates in Automotive Exhaust
Gas; Estimation of Levels of Carbonyls and Noncarbonyls in Exhaust
from Gasoline Fuels, NTIS, PB-212 600, Springfield, Virginia.
Sievers, R.E. and J.E. Sadlowski (1978), "Volatile Metal Complexes: Certain
Chelates Are Useful as Fuel Additives, as Metal Vapor Sources, and in
Trace Metal Analysis," Science, 201:217-223.
Somers, J.H., R.C. Garbe, R.D. Lawrence, and T.M. Baines (1977), "Automotive
Sulfate Emissions - A Baseline Study," SAE Paper No.770166, 20 pp.
Spicer, C.W. and A. Levy (1975), The Photochemical Reactivity of Diesel Exhaust
Organics. Battelle Columbus Laboratories, NTIS, PB-244 166, Springfield,
Virginia.
Spindt, R.S. (1977), Polynuclear Aromatic Content of Heavy Duty Diesel Engine
Exhaust Gases, prepared for Coordinating Research Council, Inc., CRC
APRAC Contract CAPE-24-72, NTIS, PB-267 774.
Spindt, R.S. (1974), First Annual Report on Polynuclear Aromatic Content of
Heavy Duty Diesel Engine Exhaust Gases, Gulf Research and Development Co.,
prepared for Coordinating Research Council, Inc., EPA Contract No. 68-01-'
2116.
93
-------
Springer, K.J. and T.M. Baines (1977), "Emissions from Diesel Versions of
Production Passenger Cars," SAE Paper No. 770818.
Springer, K.J. and R.C. Stahman (1977), "Diesel Car Emissions - Emphasis on
Particulate and Sulfate," SAE Tech. Paper, Vol. 770254, 29 pp.
Springer, G.S. (1973), "Particulate Emission from Spark-Ignition Engine,"
in Engine Emissions, Chapter 6, G.S. Springer and D.J. Patterson (eds.),
Plenum Press, New York.
Springer, K.J. (1971), Emissions From Gasoline- and Diesel-Powered Mercedes
220 Passenger Cars, report to EPA, Contract No. CPA-70-44, June, 1971.
Stara, J.F., W. Moore, and A.W. Breidenbach (1974), "Toxicology of Atmospheric
Pollutants Resulting from Fuel Additives and Emissions Associated with
the Use of Automobile Catalytic Converters," Recent Advances in the
Assessment of the Health Effects of Environmental Pollutants, Vol. II.,
pp. 751-772. Proceedings of an International Symposium, Paris, June 24-
28, 1974.
Stichting Concawe (1974), Effect of Gasoline Aromatic Content on Polynuclear
Aromatic Exhaust Emissions, Report 6-74, 60 Von Hogenhouckleaan, The
Hague, Sept., 1974.
Ter Haar, G.L., D.L. Lenane, J.N. Hu, and M. Brandt (1972), "Composition,
Size, and Control of Automotive Exhaust Particulates," J. Air Poll.
Control Assoc., 22:39-46.
Vuk, C.T. and J.H. Johnson (1975), "Measurement and Analysis of Particles
Emitted from a Diesel Combustion Process." Paper presented at the
Combustion Institute, Central States - Western States, 1975 Spring
Technical Meeting, San Antonio, Texas, April 20-21, 1975.
Vuk, C.T., M.A. Jones, and J.H. Johnson (1975), "The Measurement and Analysis
of the Physical Characteristics of Diesel Particulate Emissions," SAE
Paper No. 760131.
Wang, Y.Y., S.M. Rappaport, R.F. Sawyer, R.E. Talcott, and E.T. Wei (1978),
"Direct-Acting Mutagens in Automobile Exhaust," Cancer Letters (in press).
Wilson, W.E., L.L. Spiller, T.G. Ellestad, P.J. Lamothe, T.G. Dzubay, R.K.
Stevens, E.S. Macias, R.A. Fletcher, J.D. Husar, R.B. Husar, K.T. Whitby,
D.B. Kittelson, and B.K. Cantrell (1977), "General Motors Sulfate Disper-
sion Experiments: Summary of EPA Measurements," J. Air Poll. Control
Assoc., 27;46-51.
94
-------
4.0 Biological Effects
It has been known for many years that the exhaust emissions from both
gasoline- and diesel-powered vehicles contain a variety of potentially toxic
materials. Most prominent among these are several gaseous emissions: carbon
monoxide, sulfur dioxide, oxides of nitrogen, aldehydes, and hydrocarbons. In
*
addition, the presence of sulfates, metals, particulates, and polycyclic
organic matter (POM) can be detected in varying amounts depending upon the type
of fuel, engine load, and efficiency of operation (see Section 3.0).
The major public health concern regarding the use of diesel engines
presently involves the particulate fraction of diesel exhaust. Recent analy-
tical studies have shown that particulate emissions in diesel exhaust can be up
to 82 times as much as in gasoline exhaust using paired vehicles (Springer and
Baines, 1977). Emissions of carbon monoxide and volatile hydrocarbons are
lower from diesel than from gasoline engines, although the use of an oxidation
catalyst can substantially reduce most of these emissions from gasoline engines
(Stara e£ _al_., 1974; Lee et_ _al., 1976). There are several important reasons
why increased exposure to particulates derived from diesel engines may con-
stitute a potential health hazard (Schreck, 1978):
1) Carbonaceous particles from diesel exhaust are reportedly
composed in part of high molecular weight polycyclic
aromatic hydrocarbons.
2) These particles have high surface areas, theoretically
enabling them to adsorb large quantities of gaseous
exhaust products, most importantly the carcinogenic
POM's such as benzo[a]pyrene.
3) The particles themselves may be degraded by atmospheric
oxidation to yield lower molecular weight POM's which
are potentially carcinogenic.
95
-------
4) Diesel particulates are primarily in a size range
(0.2-0.3 ym mean diameter) which would allow for
deposition in the deep lung compartments, and
possible retention in the lung.
The discussions presented in the following sections of this report summar-
ize the major studies conducted thus far which indicate potential toxic reac-
tions to diesel exhaust mixtures, particulate extracts, and fuel additives.
Significant related studies using gasoline exhaust are included for comparison
and clarification of toxicologic hazards resulting from combustion processes.
Extensive health effects reviews have recently been published for many of the
individual components of diesel exhaust such as carbon monoxide (National
Academy of Sciences, 1977a), oxides of nitrogen (National Academy of Sciences,
1977b), particulates (National Academy of Sciences, 1977c), and polycyclic
organic matter (Santodonato jst: a^., 1978). Selected individual toxicants will
be considered which make a particularly significant contribution to the overall
toxic potential of the diesel exhaust mixture.
96
-------
4.1 In Vitro Studies
4.1.1 Mutagenicity in Bacterial Systems
It has been shown recently that organic extracts of airborne
particulate matter are mutagenic to histidine-requiring strains of Salmonella
typhimurium (Teranishi eit al. , 1978; Dehnen £t al., 1977). Because of the
strong formal relationship between molecular events involved in mutagenesis and
carcinogenesis (Miller, 1978), the demonstration of mutagenic activity for a
substance is generally taken as strong presumptive evidence for the existence
of carcinogenic activity as well. Therefore, it is believed that an investi-
gation of the mutagenicity of foreign substances: (1) may be predictive of
carcinogenic potential, (2) may be used to identify the most biologically
active fractions of complex organic pollutants (e.g., diesel exhaust), and (3)
may serve as an early warning of a possible threat to human health in cases
where positive results are obtained.
The Ames Salmonella mutagenicity assay incorporating a mammalian
microsomal preparation for activation of promutagens has received widespread
use in environmental research. Studies sponsored by the U.S. Environmental
Protection Agency have applied this assay to guide the fractionization of
heavy-duty diesel exhaust by identifying biologically active components of the
particulate fraction (Huisingh ejt al., 1978). Five histidine-requiring tester
strains of Salmonella typhimurium were employed: TA 1535, TA 1537, TA 1538, TA
98, and TA 100. Strains TA 1537 and TA 1538 are reverted to histidine-inde-
pendence by frameshift mutagens, while TA 1535 is reverted by mutagens causing
base-pair substitutions. Strains TA 98 and TA 100, which contain a plasmid to
increase sensitivity, respond to mutagens acting either by frameshift mutation
or base-pair substitution.
97
-------
Studies were carried out on fractions of a dichloromethane (DCM)
extract of diesel exhaust particulate collected from two different engines on
glass fiber filters. The DCM extract was divided into ether insoluble (INT),
acidic (ACD), basic (BAS), and neutral (NUT) fractions; the NUT fraction being
by far the largest and was further subdivided into paraffins (PRF), aromatics
(ARM), a transitional (TEN) fraction, and a polar oxygenated (OXY) fraction
(Figures 4.1 and 4.2).
Some mutagenic activity was demonstrated in the insoluble,
basic, and acidic fractions extracted from both particulate samples. However,
the neutral fraction showed most of the mutagenic activity. Within the neutral
fraction, the paraffins subfraction was not mutagenic, whereas the transitional
and oxygenated subfractions were highly active. Both direct-acting components
and components requiring metabolic activation were apparent, although most of
the mutagens were of the direct-acting frameshift type. These results are
summarized in Figure 4.3. The observation of direct-acting mutagens is signi-
ficant in that it excludes unsubstituted polycyclic hydrocarbons (e.g., benzo-
[a]pyrene) as causative agents, since they require metabolic activation for
expression of mutagenic effects. Analysis of the specific components of the
TRN and OXY subfractions was difficult, however, it was suggested that polar
neutral compounds such as substituted polynuclear aromatics, phenols, ethers,
and ketones were the major components.
Caution must be exercised before implicating polar neutral
compounds in diesel particulate fractions as major health hazards to man. The
human body, when confronted with airborne particulate pollutants, has no
physiologic means to chemically fractionate these complex organic mixtures.
98
-------
FILTERS
SOXHLET
EXTRACTION
CH2CI2 (DCM)
DCM EXTRACT
1. EVAPORATES WEIGH RESIDUE
I 2. REDISSOLVE IN ETHER
ETHER SOLUTION WITH
SOME INSOLUBLES
EXTRACT WITH BASE
vo
VO
BASIC FRACTION
BAS
FILTERS
1 SOXHLET
EXTRACTION CH3CN (ACN|
ACN EXTRACT
* 1
AQUEOUS PHASE ETHE
1. ACIDITY
2. EXTRACT ETHER
1
ACID FRACTION AQUEOUS PHASE
(ACD) (DISCARD)
R SOLUTION
EXTRACT H3P04
.1-2.08-5.15% AQUEOUS PHASE
11. ADD BASE
2. EXTRACT ETHER
f
ETHER INSOLUBLES
(INT)
I -0.12-0.008
II - 1.18-0.700
~1
ETHER SOLUTIONS
NEUTRALS (NUT)
1 -53.38-51.81
II . Id •>-? . 17 Tfl
AQUEOUS PHASE
(DISCARD)
I - 0.03 -
II -0.09
0.03
0.05
Figure 4.1.
Isolation and fractionation organics from diesel exhaust particulates
(Huisingh e£ al., 1978)
-------
NEUTRALS
SILICA
GEL
CHROMATOGRAPHY
HEXANE ELUTION
NO FLUORESCENCE
1% ETHER/HEXANE
INCIPIENT FLUORESCENCE
CONTINUED
1% ETHER HEXANE
STRONG FLUORESCENCE
50/50
ACETONE/METHANOL
ELUTION
MODERATE FLUORESCENCE
N-PRF
I - 36.26
II - 8.89
34.35%
8.94%
7.51%
1.61%
2.96%
1.22%
I - 7.40
II - 7.43
6.31%
5.64%
Figure 4.2. Silica gel chromatography fractionation of the neutral
organics from diesel exhaust particulate (Huisingh e_t^ al., 1978)
100
-------
800
120
240
CONCENTRATION OF COMPOUND ADDED TO PLATE IN MICROGRAMS
DOSES ARE 10, 33. 100. 333. 1000
1000
(+) = with metabolic activation; (-) = without metabolic activation
Figure 4.3.
Comparison of the mutagenic response of various organic fractions from the
4-stroke cycle diesel truck exhaust particulate in Salmonella typhimurium
strain TA 1538.
-------
Target organs are thus simultaneously exposed to large numbers of environmental
chemicals. It is well-established that various chemical components of polluted
air, automobile exhaust, and tobacco smoke may interact with each other to
either increase or decrease the carcinogenic response (Falk &t_ al_., 1964;
Pfeiffer £t al., 1973; 1977; Van Duuren jet al., 1976). Therefore, the biologi-
cal activity of chemical mixtures cannot be reliably predicted based on the
specific actions of individual components. Furthermore, factors governing ease
of absorption and biotransformation are important determinants of carcinogenic
potency. These factors cannot be accounted for in bacterial systems. The
apparent lack of correlation between potency in the Ames assay under certain
conditions with carcinogenic potency in animals (Ashby and Styles, 1978a,b)
further emphasizes the need for restraint in extrapolation of results.
The formation of direct-acting mutagenic compounds by combustion
processes was confirmed in studies involving non-catalyst treated automobile
exhaust (Wang e£ al., 1978). Acetone extracts of particulates collected from
six different gasoline engines showed direct-acting mutagenic activity to
Salmonella typhimurium strains TA 98, TA 100, and TA 1537. In contrast,
unused motor oil and various fuels (leaded, unleaded, diesel) were not muta-
genic. The postulated formation of nitro-substituted polycyclic aromatic
hydrocarbons during combustion led to the synthesis and examination of 6-
nitrobenzo[a]pyrene as a potential mutagen. This compound was found to be a
direct-acting mutagen in strains TA 98, TA 100, and TA 1537. Mutagenic activity
of 6-nitrobenzo[a]pyrene was comparable to that obtained with benzo[a]pyrene in
the presence of a liver enzyme activating system. Since the mutagenic activity
of particulate fractions of city air was correlated to the lead content of air
102
-------
it was suggested that automobile emissions may be a primary source of direct-
acting mutagens in the ambient atmosphere. However, nitro-substituted poly-
cyclic compounds per se have not been monitored in the urban atmosphere.
Taken together, the results of bacterial mutagenicity assays on
diesel and gasoline engine exhaust indicate that direct-acting mutagens are
formed during combustion. The chemical identity of these substances is un-
known, although substituted polycyclic aromatic hydrocarbons seem to be likely
candidates in both cases. Presently there is no way to compare the mutagenic
potency of gasoline versus diesel particulates since different collection and
chemical fractionation schemes have been employed. Even if results were avail-
able from parallel Ames bioassays, the extrapolation of these data to support a
health risk assessment would be limited to a qualitative judgement concerning
cancer risk.
103
-------
4.2 In Vivo Studies
4.2.1 Absorption, Metabolism, and Excretion
Exposure to diesel and gasoline engine exhaust occurs primarily
by inhalation of gaseous and particulate emissions. Whereas highly water-
soluble vapor phase organic emissions are generally absorbed across the moist
surfaces of the upper respiratory tract, particulate material (depending upon
size), water-insoluble compounds, and gases adsorbed to particulates may pene-
trate to the deeper lung compartments.
Studies on the deposition and retention of diesel particulate
have considerable significance in light of the preponderance of this emission
in comparison to that produced by the gasoline engine. Moreover, the likeli-
hood that diesel particulate will contain adsorbed oxidants and POM presents a
further dimension to the problem of potential toxic interaction with, or
absorption across, the respiratory epithelium. In the absence of adsorbed
substances with toxic potential, pure carbon particles as are formed during the
diesel combustion process may not present a significant health threat.
Preliminary results are available concerning the physical
characterization and clearance of diesel particulate from lungs of rats exposed
to a 1:13 dilution of automotive diesel exhaust (Moore e_t. aJL., 1978). The
nature of the diesel particulate when collected on nucleopore membranes was
examined by scanning electron microscopy. Particles smaller than 0.01 ym
resembled spherical cotton balls while the larger respirable particulate had a
flaky appearance, presumably due to the presence of adsorbed and condensed
organic matter. Daily eight-hour exposures of rats to diesel exhaust lasting
from one to 54 days produced a grey to black pigmentation of the lungs which
104
-------
varied in intensity with the duration of exposure. Black granular particulate
material was observed histologically in the cytoplasm of the alveolar macro-
phages from all exposed animals. The diesel particulate could no longer be
found in alveolar macrophages examined 28 days after a single eight-hour exposure,
Phagocytosis has been found to occur following the inhalation of
diesel particulates. Examination of macrophages containing the phagocytosed
particles has recently been conducted by transmission electron microscopy
(Orthoefer £t al., 1978). At 5000 times magnification, diesel particulate
appears in the macrophage as an aggregation of small particles. Macrophages
containing the particulate aggregates were also distinguished by the lack of
primary vacuoles.
The quantitative aspects of particle clearance by macrophage
ingestion are not well understood (NAS, 1977). It is generally believed that
phagocytosed particles are transported by macrophages to the pharynx where they
are subsequently swallowed. Thus, exposure to adsorbed chemicals may also
occur via the gastrointestinal tract. In addition, it is suggested that ;
clearance of macrophages from the lung may also lead to localization in various
organs and tissues (Moore £t al_., 1978; Lauweryns and Baert, 1977). This
process may have important implications for toxic effects in non-respiratory
tissues. Insoluble particles, and presumably diesel particulate as well, can
actually be partially digested by lysosomal hydrolases in the macrophage or
remain trapped for the life of the cell. It is likely that the chemical nature
of the diesel particulate will be the critical factor in determining its fate
within the macrophage. Examination of the physical and chemical characteristics
of diesel particulate, however, are complicated by the fact that particles and
105
-------
their agglomerates may be altered during the collection process prior to
analysis.
The fate of POM adsorbed to diesel particulate can be inferred
from studies involving BaP-coated carbon particles intratracheally instilled in
mice (Creasia et_ al., 1976). When radiolabelled BaP was adsorbed to large
carbon particles (15-30 ym) and instilled in the lungs, 50 percent of both the
BaP and the carrier particles were cleared from the lungs in four to five days.
Little carcinogen was released from the carbon particles in this case, and
therefore, contact with the respiratory epithelium (and carcinogenicity) was
low. With smaller carbon particles (0.5-1.0 urn),. however, 50 percent particle
clearance was not achieved until seven days after instillation. In this case,
15 percent of the adsorbed BaP was eluted from the particles and left free to
react with the respiratory tissues. No measurements were made in this study of
the phagocytic uptake of the particles by alveolar macrophages. In the complete
absence of carrier particles, however, BaP was cleared from the lungs at 20 times
the rate of adsorbed BaP.
In addition to mucociliary clearance and phagocytosis by alveo-
lar macrophages, processes occurring in the lung which also determine the fate
of adsorbed POM include metabolism by the respiratory tissues, and systemic
absorption across the respiratory epithelium. It is known that BaP when
administered intratracheally to rats appears in the body tissues with the same
pattern of distribution as when given parenterally (Kotin e_t al., 1959).
Similarly, Vainio and coworkers (1976) reported that unchanged BaP quickly
appears in the perfusion fluid of isolated perfused rat lungs following intra-
tracheal administration of a 200 nmole dose. The presence of particulate
106
-------
matter, however, can profoundly affect the rate and pathways of BaP metabolism
in the isolated perfused lung (Warshawsky, 1978). When BaP and crude air
particulate or ferric oxide were administered together, the rate of BaP meta-
bolism was inhibited (Table 4.2). Pretreatment with particulate, on the other
hand, caused a significant increase in the subsequent rate of BaP metabolism
(Table 4.3). The enhancement of BaP metabolism by pretreatment with parti-
culate apparently resulted from increased enzyme activity. Particulate-induced
inhibition of the metabolism of co-administered BaP may have been due to the
sequestering of adsorbed BaP in macrophages.
4.2.2 Acute Toxicity
4.2.2.1 Inhalation Exposure
The first comprehensive examination of the acute inhalation
toxicity of diesel exhaust was conducted by Pattle and coworkers (1957). Their
objective was to determine the principle toxic constituents of diesel exhaust
generated under four conditions of engine operation: light load; moderate
load; moderate load with "worn" fuel injector; light load with high fuel-to-air
ratio. Mice, rabbits, and guinea pigs were severely exposed for five hours to
the undiluted diesel exhaust. Under a light load, a highly acrid exhaust was
produced which caused no mortality and minimal damage to the lungs. Exposures
of greater duration (7 to 14 hours) under light load conditions produced nearly
complete mortality in all species, accompanied by mild pathologic alterations
in the trachea and lungs. Aldehydes (16 ppm) and oxides of nitrogen (46 ppm)
were presumed to be the primary toxic agents in this case. Under moderate load
conditions, a less irritating but more lethal exhaust was produced. Only
slight alterations were seen in the trachea, but severe lung damage occurred,
107
-------
g
Table 4.1 Influence of Particulates Administered to Isolated Perfused Lung
on BaP Metabolism* (Warshawsky et al., 1978)
Pretreatment :
IPL:
No. of animals:
Total rate of appearance
of metabolites in blood
(ng/hr/g lung + SE)
Metabolic pattern in
blood (%+SE)b
7,8-Dihydrodiol
9,10-Dihydrodiol
4 , 5-Dihydrodiol
Monohydroxylated
Diones
Nonextractable
All three columns compared
1 mg/kg.
BaP
9
256+38
6.6+0.9
15.4+4.0
3.3+0.6
9.7+1.1
10.6+1.8
54.4+5.4
to each other. All metabolites
BaP + Fe 0 a
5 *
165+51
14.0+2.9°
20.4+1.9
6.0+3.1
5.4+1.66
5.6+1.5-
48.0+4.5
separated by TLC.
BaP + CAP3
5
156+42
19.1+4.4d
28.3+7.9
3.0+1.3
5.1+1.4e
5.2+2.6
39.3+13.8
Metabolite pattern values expressed as percent of total rate of appearance of metabolite in blood +SE.
° p = 0.05.
° p = 0.01.
^ p = 0.1 (by Student-Newman-Keuls test).
abbreviations: IPL, isolated perfused lung; CAP, crude air particulate* BaP benzo[a]pyrene
-------
Table 4.2. Influence of Particulate Pretreatment on BaP Metabolism"
(Modified from Warshawsky et_ al., 1978)
o
VO
Pretreatment :
IPL:
No. of animals
Total rate of appearance
of metabolites in blood
(ng/hr/g lung + SE) :
Metabolite pattern in
blood (Z + SE)d
7,8-Dihydrodiol
9,10-Dihydrodiol
4,5-Dihydrodiol
Monohydr oxy lated
Diones
Nonextractable
BaP
9
256+37
6.640.9
15.4+4.0
3.3+0.6
9.7+1.1
10.6+1.8
54.4+5.4
Fe2°3ITa
BaP
5
637+203
13.3+2.3
26.3+5.6
2.9+2.0
4.9+0.7c
14.3+4.4
37.7+5.8
CAPITa
BaP
5
830+100°
I.
18.2+5.6°
32.6+4.3
0.9+0.5
3.4+0.6c
5.5+1.8
39.4+8.0
CAPIT*
BaP+CAP
5
143+29C
23.6+7.6
17.0+7.2
1.9+0.7
6.5+2.3
10.4+2.3
40.6+6.6
All three columns compared to each other. All metabolites separated by TLC.
? 10 mg/kg, once/week x 5.
p = 0.05 (by Student-Newman-Keuls test).
": p = 0.01.
a Metabolite pattern values expressed as percent of total rate of appearance of metabolite in blood +SE.
abbreviations: IPL, isolated perfused lung; CAP, crud air particulate; BaP, benzo[a]pyrene
-------
probably resulting from the high levels of nitrogen oxides (174-209 ppm) present
in the exhaust. A rich combustion mixture produced the most lethal exhaust,
killing all animals within five hours. The exhaust was high in aldehydes
(154 ppm) and carbon monoxide (0.17%) and produced extreme irritation, only
mild lung damage, but severe tracheal damage in rabbits and guinea pigs. Death
was most likely due to carbon monoxide poisoning. Hydrocarbon levels were not
determined in these experiments, and thus it is not known to what extent they
may have contributed to the effects observed.
The acute effects of irradiated and nonirradiated gasoline
engine exhaust in rats and hamsters were qualitatively similar to those pro-
duced by diesel exhaust as described above (Stara et al., 1974). After seven
days of continuous inhalation exposure to gasoline exhaust (diluted at a ratio
of 10:1), mortality in groups of infant rats reached 100% for irradiated exhaust
and 77% for nonirradiated exhaust. The specific toxicant(s) responsible for
the lethal effect was not determined, although death apparently did not result
from carbon monoxide poisoning. In adult rats and hamsters exposed continuously
for five days, various changes in lung morphology were noted, as well as vacuolar
changes in hepatic parenchymal cells (irradiated and nonirradiated exhaust) and
renal tubular cells (irradiated exhaust) of hamsters. In marked contrast to
these results, gasoline exhaust from engines equipped with an oxidation catalyst
produced virtually no mortality in infant rats or pathologic alterations in the
tissues of adult rats and hamsters.
The importance of the particulate fraction in contributing
to the acute effects of inhaled diesel exhaust was suggested in studies by
110
-------
Battigelli and coworkers (1966). The effect of diluted diesel exhaust on
tracheal clearance by mucociliary action was examined in rats exposed for
cumulative periods of 4 to 100 hours. It has been previously established that
noxious irritants and such gases as NO and SO can inhibit ciliary clearance
and thereby render an organism more susceptible to respiratory infection and
the actions of inhaled carcinogens (e.g., benzo[a]pyrene). As might be expected,
inhalation of diluted diesel exhaust produced varying degrees of mucociliary
inhibition which appeared to correlate with levels of N0? (1.9 •* 15.0 ppm) and
S0~ (0.1 - 3.0 ppm) in the exhaust. Even more noteworthy, however, was the
observation that the inhibition of clearance was markedly less in animals
inhaling particle-free (filtered) exhaust. The particulate material was a
complex mixture composed primarily of inorganic carbon (55-60%) and normal
paraffins with chain lengths greater than four carbon atoms (35-38%). The
authors concluded that the particulates in diesel exhaust may contribute
directly to an adverse effect on host defenses, even in the absence of other
gaseous emissions. The authors demonstrated that the inhibitory effect of a
single exposure to diesel exhaust on mucociliary clearance was completely
reversible within a few days.
Recent tests have now shown that female mice (CD-I, Charles
River) inhaling diesel exhaust displayed enhanced mortality from respiratory
infection by Streptococcus pyogenes (Campbell et^ al., 1978). Mice inhaled
either irradiated or nonirradiated diesel exhaust, diluted with clean air at a
ratio of 1:13, for a six hour period. Following the diesel exhaust exposure
(within 1-2 hours), animals were briefly exposed to an aerosal of a broth
culture of the test pathogen. Enhanced susceptibility to lethal infection was
111
-------
observed in all exhaust-treated groups, with those animals exposed to irradi-
ated exhaust apparently being more severely affected. Enhanced susceptibility
was not displayed when animals were challenged 22 hours after acute exposure.
The contribution of N0_ exposure to the observed increase in mortality is not
known.
Both catalyst- and noncatalyst-treated gasoline engine
exhaust can also enhance infective susceptibility in mice. Coffin and Blommer
(1967) reported that mice exposed to diluted irradiated gasoline exhaust (no
oxidation catalyst) for four hours experienced increased mortality from sub-
sequent immediate exposure to streptococci. Exhaust diluted to yield carbon
monoxide levels as low as 25 ppm and oxidant levels as low as .15 ppm was
effective in increasing mortality to infectious pneumonia. In related studies
using catalyst-treated gasoline engine exhaust (average dilution 1:14.1),
irradiated exhaust caused a consistent and significantly greater susceptibility
to infection than non-irradiated exhaust (Campbell e£ al., 1978). It was
concluded, however, that relative to mortalities produced in clean air-treated
controls, diesel exhaust was somewhat more effective than catalyst-treated
gasoline engine exhaust in producing increased infection mortality.
Several studies have been concerned with the health effects
of potential diesel fuel additives. Gutwein and coworkers (1972, 1974) con-
ducted distribution and retention studies in rats acutely exposed to exhaust
generated from diesel fuel containing a radiolabelled barium-based antismoke
additive. Barium contained in the diesel exhaust (average concentration
3
1.39 mg/m ) was transferred to the lungs, gastrointestinal tract, and bone
during a 10 hour exposure. Whole body levels of barium reached 0.1 yg/g of
112
-------
tissue; clearance of barium from the lungs and gastrointestinal tract was
rapid, whereas accumulation of the compound occurred in the bone. No observa-
tions were made for toxic symptoms resulting from absorption of barium or other
components of the diesel exhaust.
4.2.3 Subacute Toxicity
4.2.3.1 Inhalation Exposure
A series of extensive studies has been initiated by
the U.S. Environmental Protection Agency (EPA) regarding the effects of repeat-
ed inhalation of automotive diesel exhaust in animals. Preliminary results are
available from several of these investigations where biological effects on
selected parameters were measured at various intervals following initiation of
exposure. Diesel exhaust used for these studies was generated with a Nissan
CN6-33 engine coupled to a Chrysler Torque-flite automatic transmission. The
engine was operated in a modified "California Cycle" using number 2 diesel
fuel. A summary of exhaust component concentrations measured in the animal
exposure chambers is presented in Table 4.3.
Moore and coworkers at EPA (1978a) exposed infant rats
for 54 days (8 hours per day) to irradiated and non-irradiated diesel exhaust
diluted at a ratio of 1:13. Clinical laboratory determinations were made
during the study and selected animals were sacrificed for histologic examina-
tion of tissues at the termination of exposure. During the exposure period
there was no mortality or adverse effects on body weight gain and general
appearance of the animals. No significant differences in hematologic para-
meters or plasma electrolyte values could be shown between treated and control
groups. However, reduced levels of alkaline phosphatase, serum glutamic-
oxaloacetic transaminase (SCOT), and lactate dehydrogenase were evident in rats
113
-------
TABLE 4.3 . EXHAUST CONSTITUENTS AND CONDITIONS IN
EXPOSURE CHAMBERS3 (LEE et al., 1978)
Atmosphere
Constituent or
Condition
Carbon monoxide
(CO), ppmb
Total hydrocarbons
(THC)C, ppm (as
carbon)
Nitric oxide
(NO), ppm
Nitrogen dioxide,
(N02), ppm
Sulfur dioxide
(S02), ppm
Total suspended
particulates
(TSP), mg/m3
Sulfate ,
SO- mg/ni
Ozone
ppm
Carbon dioxide,
(C02), mol %
Temperature, °C
Relative humidity,
per cent
Exposure Chamber Atmosphere
Pur.ified Air Nonirradiated Irradiated
(Control Diesel Exhaust Diesel Exhaust
atmosphere)
2.0 15.7 15.4
2.0 15.6 15.0
0.11 5.85 4.94
0.07 2.19 2.73
NAd 2.13 1.91
NA 6.32 6.83
0.0 0.57 0.57
<0.01
0.040 0.252 0.255
24.0 23.7 24.1
51.8 51.3 48.2
, Averages of weekly means
ppm values are v/v
By ambient-temperature probe flame ionization detector. Values
using heated (350°F) probe are higher and may be approximated by
. multiplying by 1.9
Data not yet available; values should be much lower than in exhaust
chambers.
114
-------
exposed to irradiated and non-irradiated diesel exhaust. Tissue damage gener-
ally results in increased serum levels of these enzymes, thus the toxicologic
relevance of these observations is not known. Upon necropsy, minor histo-
pathologic lesions of the respiratory tract were found. These included an
accumulation of black pigmented alveolar macrophages throughout the lung.
Black pigment found in the bronchial lymph nodes of one animal suggested clear-
ance of particulate-laden macrophages. In the absence of any observed func-
tional impairment, it is not possible to make a definitive statement regarding
the severity of these pathologic changes. On the other hand, the presence of
diesel particulate in .the alveolar region and their clearance via the lymphatics
indicates that adsorbed carcinogens are also delivered to these sites.
In further studies at EPA conducted with rats, the
effect of a 28-day exposure to diesel exhaust on pulmonary function and arterial
blood gases was evaluated (Pepelko et al., 1978). Groups of rats were exposed
to diluted, irradiated and non-irradiated, diesel exhaust 20 hours per day,
seven days per week. The slope of the static lung compliance curve and resi-
dual lung volume, both indicators of emphysematous change, were not affected
by exposure to either irradiated or non-irradiated exhaust. Vital capacity
and total lung capacity, which are non-specific indicators of change in pul-
monary function, were both significantly increased in the group exposed to
non-irradiated exhaust. This observation is consistent with previous results
from studies with catalyst-treated automotive gasoline engine exhaust. It was
not considered a serious effect in light of the duration of exposure. However,
since these studies did not allow adequate time for the development of chronic
lung disease, definitive conclusions cannot yet be drawn. Likewise, a lack of
115
-------
significant treatment effects on arterial blood gases may have reflected func-
tional integrity only under conditions of short-term exposure. In addition,
most clinical measures of pulmonary function are not ideally suited for the
detection of very early lung damage.
More extensive studies on pulmonary function were
conducted at EPA with cats exposed for 28 days to a 1:13 dilution of diesel
exhaust (Pepelko e£ a_l. , 1978a). Following completion of exposure, measure-
ments were made of expiratory flow-volume curves, dynamic compliance, resis-
tance, and pulmonary diffusing capacity. In addition, hematologic parameters
were recorded and pathologic tissue evaluations conducted. No exposure-related
physiologic effects were found other than a decrease in maximum expiratory flow
rate at 10% of vital capacity; probably resulting from a slight increase in
small airway resistance. This change can result from airway constriction under
conditions such as smoking, chronic exposure to coal dust, or subclinical
emphysema. However, pathologic examination of the respiratory tissues did not
reveal emphysematous changes. The most prominent finding was the presence of
focal alveolitis characterized by an accumulation of black pigmented clusters
of one to 50 alveolar macrophages. These results support the conclusion that
diesel exhaust particulate may penetrate to deep lung compartments and produce
histopathologic changes, but do not allow for the prediction of possible
chronic effects.
More serious physiologic and pathologic effects were
found in infant guinea pigs exposed continuously (20 hours/day) to the diluted
diesel exhaust for 28 or 56 days (Wiester e£ al., 1978). After four weeks of
exposure to irradiated diesel exhaust, pulmonary flow resistance was substantially
116
-------
increased while dynamic compliance, minute volume, breathing rate, and tidal
volume did not differ between exposed and control groups. The reason for the
paradoxical increase only in pulmonary flow resistance was not evident. Expo-
sure to either irradiated or nonirradiated exhaust caused increased lung weight
to body weight ratios. In addition, guinea pigs exposed to irradiated exhaust
displayed a slight but significant sinus bradycardia on electrocardiogram
tracings. Other parameters of cardiac function and histopathology were normal.
Thus the significance of the observed sinus bradycardia may have been more
statistical than clinical. Animals sacrificed after 56 days of exposure re-
vealed a characteristic focal alveolitis accompanied by pigmented macrophage
accumulation, as was observed in cats and rats similarly exposed to diesel
exhaust. The presence of black pigment in draining bronchial and carinal lymph
nodes indicated a similar clearance mechanism for inhaled diesel particulate as
was seen in the rat. Tissue response to diesel exhaust irritation was mani-
fested by hypertrophy of goblet cells in the tracheobronchial tree; possible
tissue damage was suggested by the presence of focal hyperplasia of alveolar
lining cells, presumably Type II granular pneumocytes. There was no evidence
of other changes such as squamous metaplasia, emphysema, peribronchitis, or
peribronchiolitis.
It is apparent that the irritant effects produced by
inhalation of diesel exhaust cannot be attributed solely to the action of
carbonaceous particles, even though it is apparent that diesel particulate
becomes widely distributed throughout the lung. At least some of the tissue
damage produced by inhalation of diesel exhaust may be attributable to oxides
of nitrogen, and particularly N02- On the other hand, water-soluble gases such
as S02, as well as sulfate aerosols may contribute to functional disturbances
117
-------
(i.e., increased pulmonary flow resistance) but probably do not act in the
alveolar region (NAS, 1977). The intervention of carrier particles having the
proper size, however, can deliver irritant substances to the deepest lung
compartments. Thus, diesel exhaust contains particulate which can adsorb
irritant gases which are responsible for producing histopathologic damage that
may lead to the development of emphysema. Moreover, the relative abundance of
NO and particulates produced by diesel engines in comparison to gasoline
X
engines suggests that the risk for chronic respiratory disease may be increased
due to the greater potential for carrying adsorbed oxidants into the parenchymal
region. The participation of adsorbed POM in eliciting irritation of the
alveolar tissue is probably minimal (Santodonato et^ a.^., 1978).
Lee and coworkers at EPA (1978) have examined several
biochemical parameters relevant to pulmonary fibrosis/emphysema and carcino-
genesis that are influenced by exposure to diesel exhaust. In the rat, sub-
chronic exposure to diesel exhaust produced a doubling in aryl hydrocarbon
hydroxylase (AHH) activity, a measure of mixed-function oxidation, in the
prostate and lung, and a 30% increase in the liver. Epoxide hydrase activity
in these tissues was not increased by the diesel exposure. These microsomal
enzyme systems are normally involved with detoxification of xenobiotics in
conjunction with various P-450 type cytochromes. However, this system is also
directly involved with the metabolism of carcinogenic polycyclic hydrocarbons
to their active species (epoxides and diol-epoxides). Epoxide hydrase, also a
microsomal enzyme, converts epoxides into vicinal glycols. Since some glycols
are further metabolized by the mixed-function oxidases to form ultimate carcino-
genic forms (i.e., diol-epoxides). this enzyme would likely affect both carcino-
genesis and detoxification. Figure 4.4 presents a schematic representation of
118
-------
3
l
S
CM
CO
(ENDOPLASMIC
RETICULUM)
GLUTATHIONE
CYTOCHROME P-450
MIXED-FUNCTION OXIDASE (MFO)
MFC
BaP-
-SG
(DETOXIFICATION TRANSFERASE
PRODUCTS) (CYTOSOL)
BaP OXIDES
EPOXIDE
HYDRASE
(ENDOPLASMIC
RETICULUM)
BaP PHENOLS
MFO
BaP QUINONES
MFO
BaP DIOL EPOXIDES
(PROPOSED ULTIMATE
CARCINOGENS)
BaP DIHYDRODIOLS (PROPOSED PROXIMATE CARCINOGENS)
UDP-GLUCURONOSYL TRANSFERASE
(ENDOPLASMIC RETICULUM)
H2O-SOLUBLE CONJUGATES
(DETOXIFICATION PRODUCTS)
Figure 4.4. Enzymatic pathways involved in the
activation and detoxification of BaP.
119
-------
the various enzymes involved in activation and detoxification pathways for BaP,
that is also representative of the known mechanisms of POM metabolism in
general.
Further biochemical aspects of the subacute toxicity
of diesel exhaust concerned early changes produced in the lung tissue (Lee
£t al_. , 1978). Inhalation of a 1:13 dilution of diesel exhaust by rats caused
increases in the rate of collagen and protein synthesis, and enhanced prolyl-
hydroxylase activity in the lungs. These alterations were indicative of
fibrogenic changes, consistent with a large increase in connective tissue
proliferation and continuous scar formation in response to injury. Such dis-
turbances in the integrity of lung structure, although not necessarily linked,
are important indicators of potential emphysema development. Accompanying
these biochemical alterations were noticeable changes in the appearance of the
diesel-exposed lungs. These lungs were rubbery to the touch, charcoal grey in
color, and much more difficult to homogenize than control lungs.
Subacute exposure studies with gasoline engine exhaust
have clearly shown the difference in toxicity between catalyst- and noncatalyst-
treated exhaust (EPA, 1978). Infant guinea pigs exposed continuously for 35
days to diluted (1:10) catalyst-treated exhaust displayed reduced growth rate
(0-20%), increased airway resistance (3-47%), and no change in lung compliance.
Removal of the catalyst, however, resulted in growth rate reductions of 34-36%,
increased airway resistance by 63-68%, and a significant decrease in lung
compliance (35-39%). A severe increase in bronchial constriction was indicated
in the non-catalyst group, which might represent a simple defense mechanism.
Pathologic changes in the lungs of clean air controls and catalyst-treated
exhaust groups included inflammatory lesions, focal thickening of the alveolar
120
-------
walls, and a focal pneumocytic hyperplasia. Similar, but more severe, changes
were seen in the lungs of guinea pigs inhaling noncatalyst-treated exhaust.
The changes observed in the control animals complicates the analysis of these
results. Lactating female rats and their newborn offspring exposed to catalyst-
treated gasoline exhaust for four and 12 weeks, respectively, displayed no
treatment-related effects on mortality, body weight, hematology, or histo-
pathology (EPA, 1978). The ability of an oxidation catalyst to reduce carbon
monoxide levels in gasoline exhaust is credited with preventing the cardiac
hypertrophy, and polycythemia which results from subchronic (4 week) exposure
of rats to noncatalyst-treated exhaust. In addition, damage to the lung and/or
kidney as shown by increased serum lactate dehydrogenase levels is also pre-
vented by the catalyst.
Overall it is apparent that health-related benefits
are derived from the use of an oxidation catalyst with gasoline engines. The
prevention of significant subchronic effects by a catalytic converter can
almost certainly be attributed to the substantial reductions which are realized
in the emission of gaseous exhaust components. Subchronic exposure to diesel
exhaust, on the other hand, produces damage which is apparently more severe
than that produced by catalyst-treated gasoline exhaust, but somewhat less than
that resulting from exposure with the catalyst removed.
Fuel additives have the potential to alter the chemi-
cal composition of engine emissions and possibly modify the biological effects
produced by inhalation of exhaust. Moore and coworkers (1975) studied the
biological effects of automotive emissions containing Mn particulate introduced
by the fuel additive, methylcyclopentadienyl manganese tricarbonyl (MMT). MMT
is used in unleaded gasoline as an antiknock additive, and is marketed as a
121
-------
smoke suppressant for diesel engines and stationary jet fuel power sources.
Rats and hamsters were exposed eight hours per day for 56 consecutive days to
gasoline engine exhaust (1:25 dilution) derived from fuel containing MMT at
0.25 g (as Mn) per gallon. Although increased tissue concentrations of Mh were
produced by the exposure, no gross changes or histopathologic lesions could be
attributed to the presence of MMT in the fuel. The primary lesion produced was
a thickening of the cuboidal epithelium in the terminal bronchioles of the
lung; an effect which was not considered particularly severe. Lesions did not
become more severe with length of exposure, but occurred in 21% of the animals
exposed to irradiated exhaust, 14% exposed to non-irradiated exhaust, and 60%
of the clean air controls. It is noteworthy that the incidence of lesions in
control animals costs doubts on the entire experiment.
4.2.3.2 Dermal Exposure
An early study has demonstrated that an organic
extract of diesel exhaust particulates can produce severe systemic toxicity
when applied to the skin of mice (Kotin et_ akL., 1955). Extracts of exhaust
from a grossly inefficiently operating diesel engine when applied to the inter-
scapular area of C57 black mice produced immediate tremors, followed by a
reversible lethargy and loss of neuromuscular responses. Deaths began to
result after about ten weeks of treatment (3 applications per week). Post-
mortem examination revealed a combined hepatotoxic and nephrotoxic effect.
This was characterized by liver cord cell degeneration, and tubular degenera-
tion in the lower nephron of the kidneys. The immediate cause of death in
exhaust-treated mice was pneumonia, which was taken as evidence of decreased
host resistance, although a mechanism was not postulated. When mice of the
same strain were treated with extracts from gasoline engine exhaust (no oxidation
122
-------
catalyst) in the same manner as in the diesel studies, severe toxicity was not
encountered (Kotin e_t al., 1954). The effect of the solvent employed in these
experiments must be carefully considered, however.
4.2.3.3 Behavioral Effects
Behavioral alterations in rats have been produced by
exposure to catalyst- and noncatalyst-treated gasoline engine exhaust (Cooper
et ajL. , 1977) as well as diluted diesel exhaust (Laurie e£ al., 1978). Con-
centrations of the various exhaust components are summarized in Table 4.4.
Exposure of adult rats to diesel exhaust for 20 hours per day for six weeks
caused a significant reduction in spontaneous locomotor activity measured at
the end of the treatment period (Figure 4.5). In addition, decrements in
forced activity performance resulted from the diesel exhaust exposure. During
a four week recovery period, however, the difference in spontaneous locomotor
activity between exhaust-treated and control rats was reduced.
In related studies, neonatal rats were exposed to
diesel exhaust from day one after birth, 20 hours per day, for 17 days (Laurie
je_t al.., 1978). Measurements of surface righting and ear detachment taken after
one or two days of exposure showed no differences from control values. Like-
wise., air righting, measured on days 14, 15, and 16, was not affected by the
treatment. However, significant reductions were observed for both pivoting
(measured on days 6 and 7) and eye opening (measured on days 14, 15, and 16)
behavior when compared to controls. If data collected on female rats were
considered alone, eye opening behavior was not significantly delayed.
It was concluded from these studies that spontaneous
locomotor activity in adult rats was depressed further by exposure to diluted
123
-------
CD
(M
o
30^
26-
~ 22-\
o
if «H
£
W
| 141
tt 10-
• Control
o Exposed
I
1
I
I
2
I I
3 4
I I I I I I
561 2 3 t
Weeks
Figure 4.5. Effect of diesel exhaust on spontaneous locomotor
activity in rats (Laurie et_ al., 1978)
124
-------
diesel exhaust than by exposure to catalyst- and noncatalyst-treated gasoline
engine exhaust (1:11 dilution). The causative agent(s) which affects rat
behavior cannot be specified. It was postulated, however, that hydrocarbon
components of diesel exhaust most likely accounted for the behavioral altera-
tions observed.
125
-------
4.2.4 Chronic Toxicity
The preliminary results of only two studies have been published
thus far concerning the pulmonary damage resulting from chronic inhalation of
diesel exhaust. These have employed high concentrations of emission products.
Stuart and coworkers (1978) have exposed male rats (48 per group) to the fumes
3
of an inefficiently operating diesel engine (50 ppm carbon monoxide, 10 mg/m
3
soot) alone and in combination with bituminous coal mine dust (6 mg/m ) for six
hours daily, five days per week for periods up to 20 months. Serial sacrifice
and histopathologic examination of the lungs in rats inhaling diesel exhaust
revealed particulate accumulations, vesicular emphysema, and beginning inter*-
stitial fibrosis. Inhalation of diesel exhaust together with coal dust pro-
duced similar alterations in the lungs as well as bronchiolar epithelial pro-
liferation and inflammatory reaction. Carboxyhemoglobin levels were elevated
in both treatment groups.
Parallel studies have also been conducted by Stuart and co-
workers with Syrian golden hamsters exposed to diesel exhaust, and their results
were summarized at a recent workshop (NIOSH, 1978). Exposures to diesel exhaust
3
containing 4-6 ppm NO^ and respirable particulates at 6-10 mg/m were conducted
five hours daily, five days per week, for up to 20 months. Histopathologic
changes which resulted in the lungs included marked particulate aggregation in
alveoli or macrophages, pulmonary consolidation, vesicular emphysema4 inter-
stitial fibrosis, and cuboidal metaplasia.
Very few studies have been conducted which involve chronic
inhalation exposure to gasoline exhaust, usually because the carbon monoxide
exposures involved are too great. Thus, direct comparison of histopathologic
damage with that produced by diesel exhaust cannot be validly performed. In
126
-------
early reports of studies where mice were chronically exposed (>2 years) to
diluted gasoline exhaust, most toxicologic observations were negative (Campbell,
1936). No adverse effects were noted on death rate, body weight, or rate of
growth. When death occurred, it apparently involved heart failure accompanied
by lung congestion. Pneumonia and pathologic alterations in the liver (con-
gestion, atrophy with fibrosis or necrosis) were more common in exhaust-treated
mice than in controls. More recent studies involving life-long exposure of
rats to diluted (1:100) gasoline exhaust (containing 58 ppm CO and 23 ppm of
total nitrogen oxides) have yielded several significant results (Stupfel et
al., 1973). The most important finding was the presence of bilateral renal
sclerosis in more than half of the animals autopsied. In addition, a greater
number of emphysematous lesions and spontaneous tumors of various organs were
observed in exhaust-treated rats than in controls. However, no tumors of the
respiratory tract were found. It is noteworthy that the levels of nitrogen
oxides were very high in this study.
Most recently, a chronic study has been completed which involved
the exposure of beagle dogs to raw or photochemically reacted gasoline engine
exhaust 16 hours daily for 68 months (Hyde e_t^ aJ^., 1978). In all exposure
groups, reversible lung damage was encountered. Dogs inhaling raw exhaust dis-
played hyperplasia of nonciliated bronchiolar cells, which apparently persisted
long after cessation of exposure. With irradiated exhaust, incipient emphysema
was produced, which was thought to result from exposure to nitrogen oxides
(1.77 + 0.68 mg/m3 N02; 0.23 + 0.36 mg/m3 NO) and ozone (0.39 + 0.18 mg/m3).
4.2.5 Bioassays for Carcinogenicity
Pioneering work by Kotin and coworkers (1955) established that
the particulate fraction of exhaust from an inefficiently operating diesel
127
-------
engine contains carcinogenic POM which are capable of producing tumors in
experimental animals. Although the production of polycyclic hydrocarbons in an
efficiently running diesel engine was extremely low, the exhaust from an in-
efficient diesel engine contained significant amounts of pyrene, benzo[a]-
pyrene, benzo[e]pyrene, benzo[ghi]perylene, anthanthrene, coronene, and an
unidentified "compound X." Acetone solutions of benzene extracts from the
particulate exhaust fraction from an inefficiently operating diesel engine were
repeatedly applied (3 times weekly for more than 60 weeks) to the skin of mice
(C57 black and A strain). A high incidence of skin cancers resulted in A
strain mice when they were pa,inted with particulate extracts obtained during
full-load engine operation. These results corresponded with chemical analyses
showing that polycyclic hydrocarbon emissions are greatest under conditions of
full load and inefficient engine operation.
Kotin and coworkers noted that the diesel engine can be a
greater source of polycyclic hydrocarbons than the gasoline engine (without
oxidation catalyst) depending on engine operating conditions. However, pre-
vious studies by these same investigators demonstrated that benzene extracts of
gasoline exhaust particulates are also capable of producing large numbers of
skin cancers using C57 black mice (Kotin eit^ a3L , 1954).
Subsequent studies conducted on the potential carcinogenicity of
diesel exhaust fractions have reportedly produced negative results. Clemo and
Miller (1955) briefly mentioned that two fractions from diesel bus smoke yielded
no carcinomas in mice, but experimental details were not reported. Mittler and
Nicholson (1957) collected diesel and gasoline engine exhaust condensates
(without using filters) and applied benzene extracts of this material twice
weekly for eleven months to mice. They obtained a 76% incidence of skin tumors
128
-------
(presumably papillomas) in mice receiving a 4.0% gasoline exhaust extract, and
no tumors in mice receiving a 2.26% diesel exhaust extract. These results are
difficult to interpret, however, for several reasons: a) conditions of engine
operation were not reported, b) the collection efficiency for the exhaust
particulate fraction by the method employed is not known, and c) no chemical
analyses were conducted on the engine exhaust condensates.
Thus far, attempts to produce tumors of the respiratory tract by
the inhalation of either diesel exhaust (Stuart ejt al., 1978) or gasoline
exhaust (Campbell, 1936; Stupfel et_ al., 1973) have not been successful.
Nevertheless, the carcinogenicity of gasoline exhaust fractions by dermal
application or subcutaneous injection in mice has been repeatedly confirmed
(Wynder and Hoffmann, 1962; Brune, 1977; Pott et^ al_., 1977).
Most investigators agree that the demonstrated carcinogenicity
of particulate diesel and gasoline exhaust fractions, as well as particulate
air pollutants from fossil fuel combustion, is largely (but not entirely) due
to the presence of POM. Consequently, an intensive research effort has been
mounted over the past several decades to thoroughly characterize the biological
activity of chemicals in this class. A recent review of the published litera-
ture on POM indicated that benzo[a]pyrene is the most well-studied of all these
compounds (Santodonato ^t al., 1978). Attention has focused on benzo[a]pyrene
primarily because: a) it is an ubiquitous contaminant in atmospheric emissions
from the combustion of fossil fuels, b) it is easily detected, and c) it is a
potent animal carcinogen. It is evident that POM's in general and benzo[a]-
pyrene in particular, when present in air, are nearly always found as adsorbed
material on particulate matter. Thus, the inhalation of particulate matter
129
-------
from combustion processes can deliver a number of carcinogenic POM's, including
benzo[a]pyrene, into direct contact with the respiratory tissues. Furthermore,
numerous studies have indicated that the ability of suspensions of benzo[a]-
pyrene to induce experimental lung tumors can be considerably enhanced by
concomitant exposure to particulate matter. It is suggested that particulates
increase the carcinogenic response to benzo[a]pyrene by providing increased
retention in the lung. In addition, simultaneous exposure to ciliastatic gases
(e.g., SO , N0«) can further enhance the respiratory tumor reponse to benzo[a]-
pyrene administration, possibly by inhibiting normal lung clearance mechanisms.
Consequently, both gasoline and diesel exhaust are rightfully suspect as
carcinogenic mixtures, and the amount of particulate material and POM's con-
tained in them might be regarded as critical determinants of carcinogenic
potential.
Nevertheless, in the absence of positive bioassay data involving
chronic inhalation exposures, a direct comparison cannot be made between the
carcinogenic potential of diesel versus gasoline exhaust. Based on the limited
number of animal studies which have thus far been conducted, there are no data
which suggest an increased carcinogenic threat from the substitution of diesel
for gasoline exhaust.
130
-------
4.3 Human Studies
4.3.1 Controlled Exposures
Little is known concerning the specific effects of diesel ex-
haust on various physiologic parameters in humans. Battigelli (1965) exposed
volunteers to several dilutions of diesel exhaust containing 0.2-7.0 ppm NO^,
0.2-2.8 ppm SO , 20-80 ppm CO, 900-15,000 ppm CO , 19.5-20% 0 , total aldehydes
£* £* £•
less than 1-2 ppm, and total hydrocarbons less than 5-6 ppm. Inhalation of the
diluted exhaust for periods up to one hour had no effect on pulmonary resistance
and produced no complaints of respiratory distress from the experimental sub-
jects. Although the diesel exhaust produced no complaints when inhaled, the
same subjects found that eye exposure to the diesel exhaust dilutions produced
conjunctival irritation which often became intolerable.
4.3.2 Epidemiologic Studies
Attempts to establish an association between exposure to speci-
fic pollutants and adverse non-occupational health effects, especially cancer,
are seldom successful. However, in occupational situations the effects of
long-term exposure to high levels of a toxicant are more easily quantified;
thus allowing for extrapolation back to the effect of small doses present in
the ambient atmosphere. Unfortunately, only a few occupational studies in-
volving diesel exhaust exposure have been published, and in none of these could
the exposure be considered particularly- intense in terms of concentration
and/or duration.
4.3.2.1 Occupational Studies
Raffle (1957) authored a general review paper which made
use of a number of examples to argue for the mutual benefits which can accrue
131
-------
to industry, medicine and the worker through the study of occupational and
health records of employees. Absences from work because of sickness were shown
to vary in frequency, duration and type depending upon the age, sex and occupa-
tional category of the worker. While acknowledging the diagnostic inaccuracy
of individual absence records, Raffle emphasized the validity of comparisons
between large groups.
Using records of the London Transport staff, he recalled
the classical studies of coronary heart disease in conductors and drivers, in
which uniform sizes at the times of initial employment provided critical infor-
mation for the interpretation of self-selection. Raffle used health records of
the conductors and drivers to show differences in their rates of absence attri-
buted to "bronchitis" (undefined). Lastly, he reported on the incidence of
cancer of the lung in relation to occupational exposure to the exhausts of
diesel engine buses. When the London Transport staff, aged 45-64, were grouped
according to expected low to high exposure there was no discernible gradient in
the rates of death, retirement or transfer to alternative work during the period
1950 to 1954 due to lung cancer. There was an observed tendency for lung
cancer death rates to follow the workers' residential patterns with respect to
urban density and air pollution carried by prevailing winds. This is in agree-
ment with the concept that "the amount of carcinogen in town air depends on the
density of the population (possibly the number of coal fires) and that it is
also driven by the prevailing wind." Thus, diesel exhaust could not be speci-
fically implicated as a serious contributing factor. This clearly does not
rule out possible health hazards of diesel engine exhausts, especially if
diesel-powered urban vehicles were to predominate at some time in the future.
132
-------
It could be argued, in connection with the London Transport
staff work, that the period 1950 to 1954 was too early for any possible effects
of diesel emissions to have become evident; diesel buses having been introduced
gradually over a period from about 1935 to 1952. Therefore, the collection of
lung cancer data has continued for this same group of workers, and by now a 25
year series is available (Raffle and Waller, to be published). The findings
broadly support those for the first five years, and overall the lung cancer
incidence rates are slightly lower than expected in the general population of
London. This feature is common to other studies among occupational groups
(Kaplan, 1959), and to some extent is may reflect the selection of relatively
fit people for employment.
At about the same time as the paper by Raffle, air sampling
data for two London Transport garages were reported by Commins and coworkers
(1957). The Merton garage housed about 200 diesel buses and Dalston housed
120. Each garage was monitored from 6:00 P.M. to 7:00 A.M. on two nights:
Merton in April and again in June, 1956; Dalston in October, 1956, and in June,
1957. Each 13 hour session was divided into 4 periods typified by fueling,
down-time, departures and returns. Smoke samples for analysis of polycyclic
hydrocarbons were taken at one site in each garage, and one outside (on the
roof) as a control. A long-term smoke record (one week at Dalton and two weeks
at Merton) was also taken at each garage.
On each occasion in every period, the average concentration
of smoke was higher inside than outside each garage, though only slightly so
during the down periods. Typically, the concentrations of hydrocarbons (pyrene,
fluoranthene, 1:2- and 3:4- and l:12-benzpyrene) were greater inside than out,
but by a lower value than would be indicated by the inside to outside smoke
133
-------
concentration ratio. No marked excess of inside over outside sulfur dioxide
concentration was observed. Variations in nitrogen dioxide concentrations
followed the same pattern as smoke concentrations. Thus all pollutants were at
their lowest levels in the down periods.
In their introductory remarks and at the conclusion of the
report, the authors underscore the limited usefulness of their data in attempt-
ing to generalize: "The results are not to be applied without qualification to
air pollution under the different conditions obtained in streets in the open
air."
Kaplan (1959) analyzed the records of 154 lung cancer
deaths among employees of the relief department of the Baltimore and Ohio
Railroad from January 1, 1953 to December 31, 1958. Three groups, in order
from greatest to least putative exposure, were compared: (1) those with direct
occupational exposure to diesel or steam engine exhaust; (2) service workers or
laborers in shops or roundhouses; (3) clerks and others rarely occupationally
exposed. 96% of the employees were males and the few females belonged mainly
to group (3) with low exposure. Group 3 was found to have a slightly greater
age-adjusted rate of lung cancer deaths than group 1, while the rate for group
2 was considerably lower than for the others. Thus, the pattern did not con-
form well with the concept of exposure to diesel fuel or coal engine exhausts.
Overall, the age-adjusted lung cancer death rate fell somewhat below that
estimated for the United States male population.
Kaplan pointed out that these results were in strong con-
flict with those of a similar study by Heuper in 1955, i.e., that, while only
25% were operating employees (group 1), they accounted for 75% of the lung
cancer cases. One of the reasons advanced for the relatively high lung cancer
134
-------
rate in group 3 was its greater proportion of urban dwellers. This factor was
as least partly offset, however, by its higher proportion of females.
Not mentioned by the author, but clearly of concern is the
limited character of the data. Selection in comparing local railway employees
with the United States as a whole may operate through variations in the coding
of causes of death due to lung cancer. In addition, the extent to which selec-
tion operated to retire ill employees was not evaluable. Lastly, it would have
been desirable to consider causes of death which compete with lung cancer as
well as to analyze inedical records of the occurrences of chronic bronchitis and
other illnesses, both of long and short-term character.
In a later review article Battigelli (1963) reported that
none of the several measured components of air samples taken over several
months in various studies of confined areas polluted by diesel engine exhausts
(e.g., roundhouses, railway tunnels, bus garages) exceeded threshold limits
established by the American Conference of Governmental Industrial Hygienists.
He recognized, however, that in every major episode of air pollution with
adverse effects to humans, none of the measured contaminants had exceeded
accepted maximum concentrations.
Diesel exhaust was characterized as being distinct from the
two major types of health-threatening polluted atmospheres typified by London
and Los Angeles. Air polluted by diesel engine exhausts contains relatively
low levels of carbon monoxide and of carcinogenic polycyclic hydrocarbons, as
compared with gasoline engine polluted air. Diesel engines produce higher
levels of objectionable odorants, conjunctival irritants and smoke, but Battigelli
expressed the opinion that diesel exhausts were no more harmful than alternative
forms of pollution.
135
-------
The author pointed out that diesel engines discharge more
nitrogen oxides and aldehydes per hour than comparable gasoline engines. This
disadvantage was tempered by noting that the concentration of both of these
gases in diesel exhaust is lower because of a greater air-to-fuel ratio in
diesel motors. However, the air-to-fuel dilution factor is valuable only with
respect to localized sources of pollution in confined spaces. When considering
the environmental effects of a predominance of diesel powered vehicles, hourly
discharges of nitrogen oxides and aldehydes would seem to be the more relevant
issue.
In 1964, Battigelli teamed with two colleagues to report a
cross-sectional study involving physical examinations and medical histories of
210 workers occupationally exposed for an average of 10 years to diesel exhausts
in three Pittsburgh railroad engine houses. These workers were compared to
154 yard workers comparably distributed with respect to age, cigarette smoking
histories, and extrapulmonary medical problems.
The internal air was sampled systematically at locomotive
roof levels and at head-high "floor" levels at locations near to and distant
from the locomotives. While NCL, SCL, total aldehydes, acrolein and hydro-
carbons from C- to Cfi varied somewhat according to type of location, the gen-
eral pattern indicated an extensive dilution of exhaust products. For example,
the average concentration of total hydrocarbons was nearly identical for all
three types of sample locations (engine roof, near floor and distant floor
levels). Winter concentrations consistently exceeded those in summer.
The medical data were based on a physical examination,
chest X-ray, electrocardiogram, spirometr.y, a standardized medical history with
special focus on chronic respiratory diseases, and other standardized pulmonary
136
-------
function tests. The investigators reported that no significant differences
were discernible between the 210 exposed workers and the 154 controls. In
order to allay suspicions of insensitivity to existing major differences, they
presented parallel comparative analyses of the workers who smoked versus those
who did not smoke cigarettes within the previous 10 years. These analyses
revealed consistently higher relative frequencies of dyspnea, cough and measures
of bronchitis together with lower pulmonary function levels in cigarette smokers
than in non-smokers.
Nonetheless, this study has shortcomings which resulted
from an inability to pursue the original study plan. A low number of examinees
was obtained, and participation was strictly on a volumtary basis. While the
exposed workers were processed with a participation of better than 90% of the
specific employed population, the non-exposed group showed a much smaller
participation frequency. Thus, any attempt to carry out a formal analysis of
statistical significance on a group of this composition is useless. However,
based on their collected observations, particularly considering a mean exposure
of 10 years, the fact that no major adverse effects were found should not be
considered as trivial.
Several cohort mortality studies have been conducted with
workers from underground mines in which diesel equipment was routinely used
(Waxweiler e_t al_., 1973; NIOSH, 1978). Although deaths attributed to malignant
and non-malignant respiratory disease were elevated in certain instances, it
was not possible to determine the potential contribution of diesel exhaust to
this result. It was concluded that, "no excess mortality was attributable to
the presence of diesel engines in some mines; however, there has probably been
too little elapsed time for this observation to be definitive" (Waxweiler et al.
137
-------
1973). Further epidemiologic studies involving metal and non-metal miners
exposed to diesel exhaust are in progress (NIOSH, 1978) and may provide further
clarification of the risk for lung cancer and respiratory disease.
A preliminary analysis of ventilatory function among 60
coal miners exposed to diesel emissions has recently been performed (Reger,
1978). Decrements in lung function over the period of a work shift could be
demonstrated among miners exposed to coal dust either with or without concomi-
tant exposure to diesel emissions. The decrements shown were no greater in the
presence of diesel emissions than for exposure to ordinary mine atmospheres in
the absence of diesel emissions.
The combustion of fossil fuels, resulting in exposure to
POM, has long been associated with an increased cancer risk. More recently,
however, inhalation of vehicular emissions has received attention as a specific
factor in the etiology of malignant disease. In this regard it was reported
that moderately elevated relative risks for cancers of the nose, pancreas, and
prostate were observed in mechanics and repairmen (Viadana et^ al^., 1976). In
addition, bus, taxi, and truck drivers showed increases in cancer of the pan-
creas; locomotive engineers were at increased risk for lymphomas and cancers of
the buccal cavity and pharynx. Other investigators observed an elevated lung
cancer rate in Los Angeles County for workers in the auto repair industry
(Menck and Henderson, 1976).
Occupational exposure to gasoline engine emissions have
also been of interest to epidemiologists for the development of non-malignant
disease. One recent report concerns a retrospective cohort study of mortality
among approximately 1600 motor vehicle examiners employed in New Jersey during
the period 1938 through 1973 (Stern and Lemen, 1978). Because previous studies
138
-------
had shown that chronic low-level exposures to carbon monoxide may exacerbate
coronary heart disease, measurements were made of carbon monoxide levels
(average concentration 22 ppm) and heart disease mortality was specifically
pursued. The study, however, failed to show a significant excess of heart
disease deaths, but instead revealed a significant excess of cancer deaths (13
observed vs. 6.69 expected) for individuals with greater than 30 years since
onset of employment. The excess of cancers was not associated with a particu-
lar organ site. Since cause-specific mortality rates for the general United
States white male population were used for comparisons in this study, the
consistently higher New Jersey cancer death rate was not taken into considera-
tion. Thus, it is not possible to conclusively attribute the excess of cancer
deaths in this study to occupational exposure to automotive exhaust.
A study of 386 children who died of malignant disease in
the province of Quebec during the years 1965 through 1970 has suggested a
correlation with specific types of occupation of the father (Fabia and Thuy,
1974). Prior to undertaking the study it had been postulated that the risk of
malignant disease may be greater among children whose fathers are occupation-
ally exposed to petroleum products. The study revealed that excess cases of
cancer existed among the children of motor vehicle mechanics, machinists,
miners, and painters. These striking results do not provide an explanation of
the mechanism by which such a phenomenon might occur. Nevertheless, in other
studies involving occupational exposure to known carcinogens (e.g., vinyl
chloride) it has been reported that fetal deaths and congenital defects may be
increased in the offspring of exposed fathers (Infante eit al., 1976; U.S.
Public Health Service, 1976).
139
-------
4.3.2.2 Community Studies
A suitable community population has not yet been studied
which might reveal the impact of diesel emissions in the ambient atmosphere on
human health. However., the effect of diesel emissions on the general population
may be too small to be accurately detected and separated from other variables.
Similarly, the question of whether gasoline engine emissions represent a signi-
ficant carcinogenic threat to man has not yet been resolved (Lawther and Waller,
1976). This is primarily due to the diversity of pollutants in the environment,
and the inability to identify a suitable population having significant exposure
to a specific pollutant type. Nevertheless, a report implicating emissions
from automobiles (presumably gasoline-powered) as an important cause of cancer
has been published in Switzerland (Blumer et_ al_., 1972). In this study of a
very small population, deaths due to cancer of various sites were nine times
more frequent (11% vs. 1.2%) among residents of a Swiss mountain town living
near a highway as compared to residents living in an area remote from traffic.
The authors concluded that differences in age, occupation, exposure to non-
automotive combustion products, sex, and smoking habits could not entirely
account for the increased cancer mortality rate. However, the population
studied was too small to allow for determination of age/sex/site specific cancer
rates. Therefore, the reported observations may have'little overall significance
In a follow-up study to clarify these observations, it was
found that soil content of polycyclic aromatic hydrocarbons in this region
showed a correlation with proximity to a highway (Blumer ejt a!L., 1977). The
compounds identified were present as a complex mixture of unsubstituted three-
to eight-membered rings and heavily alkyl-substituted derivatives, which the
140
-------
author concluded had originated primarily from automobile exhaust. Thus it was
believed that the observed mortality from cancer in this area might indeed be
associated with exposure to automotive exhaust. Little can be said, however,
concerning the levels of exposure to automobile-derived carcinogens among those
residents living near the highway. Since the Swiss town was situated in a deep
valley with frequent thermal inversions, it is likely that these persons had
unusually high exposures to exhaust emissions. On the other hand, deaths due
to non-malignant disease (heart and circulatory) did not seem to vary greatly
between roadside and non-roadside residents.
141
-------
References for Section 4.0
Ashby, J. and J.A. Styles (1978), "Does Carcinogenic Potency Correlate with
Mutagenic Potency in the Ames Assay?" Nature, 271(5644);452-455.
Ashby, J. and J.A. Styles (1978a), "Factors Influencing Mutagenic Potency
In Vitro," Nature, 274;20-22.
Battigelli, M.C. (1963), "Air Pollution from Diesel Exhaust," J. Occup. Med.,
5:54.
Battigelli, M.C. (1965), "Effects of Diesel Exhaust," Arch. Environ. Health,
10:165-167.
Battigelli, M.C., F. Hengstenberg, R.J. Mannella, and A.P. Thomas (1966),
"Mucociliary Activity," Arch. Environ. Health, JL2_:460-466.
Battigelli, M.C., R.J. Mannella, and T.F. Hatch (1964), "Environmental and
Clinical Investigation of Workmen Exposed to Diesel Exhaust in Railroad
Engine Houses," Ind. Med. Surg., 33:121-124.
Blumer, M., W. Blumer, and T. Reich (1977), "Polycyclic Aromatic Hydrocarbons
in Soils of a Mountain Valley: Correlation with Highway Traffic and
Cancer Incidence," Environ. Sci. Technol., 11:1082-1084.
Blumer, W., R. Jauman, and T. Reich (1972), "Automobile Exhausts - The Most
Important Cause of Cancer?" Schweiz. Rundsch. Med. Prax., 61:514-518.
Brune, H.F.K. (1977), "Experimental Results with Percutaneous Applications of
Automobile Exhaust Condensates in Mice," in Air Pollution and Cancer in
Man, IARC Sci. Publ. No. 16:41-47.
Campbell, K., E. George, I. Washington, and Y. Yang (1978), Enhanced Suscepti-
bility to Respiratory Infection in Mice Exposed to Automotive Diesel
Emissions, U.S. Environmental Protection Agency, unpublished report.
Campbell, J.A. (1936), "The Effects of Exhaust Gases from Internal Combustion
Engines and of Tobacco Smoke Upon Mice, with Special Reference to Inci-
dence of Tumours of the Lung," Brit. J. Exper. Pathol., 17:146-158.
Candeli, A., G. Morozzi, A. Paolacci, and L. Zoccolillo (1975), "Analysis Using
Thin Layer and GLC of Polycyclic Aromatic Hydrocarbons in the Exhaust
Products from a European Car Running on Fuels Containing a Range of Con-
centrations of These Hydrocarbons," Atmos. Environ., 9^:843-849.
Clemo, G.R. and E.W. Miller (1955), "The Carcinogenic Action of City Smoke,"
Chem. Ind. (London), January 8, 1955, p. 38.
Coffin, D.L. and E.J. Blommer (1967), "Acute Toxicity of Irradiated Auto
Exhaust," Arch. Environ. Health, 15:36-38.
142
-------
Commins, B.T., R.E. Waller, and P.J. Lawther (1957), "Air Pollution in Diesel
Bus Garages," Brit. J. Ind. Med., 14_: 232-239.
Cooper, G.P., J.P. Lewkowski, L. Hastings, and M. Malanchuk (1977), "Cata-
lytically and Noncatalytically Treated Automobile Exhaust: Biological
Effects in Rats," J. Toxicol. Environ. Health, 2-923-934.
Creasia, D.A., J.K. Poggenburg, Jr., and P. Nettesheim (1976), "Elution of
Benzo[a]pyrene from Carbon Particles in the Respiratory Tract of Mice,"
J. Toxicol. Environ. Health, .1(6) :967-975.
Dehnen, W., N. Pitz, and R. Tomingas (1977), "The Mutagenicity of Airborne
Particulate Pollutants," Cancer Letters, 4;.:5-12.
Dolan, D.F., D.B. Kittelson, and K.T. Whitby (1975), Measurement of Diesel
Exhaust Particle Size Distributions, ASME Publication 75-WA/APC-5.
Energy and Environmental Analysis, Inc. (March 6, 1978) Draft Report to EPA,
Atmospheric POM: Source and Population Exposure Estimate.
EPA (1978), Biological Effects of Auto Exhaust Emissions With and Without
Catalysts, U.S. Environmental Protection Agency, unpublished report.
Fabia, J. and T.D. Thuy (1974), "Occupation of Father at Time of Birth of
Children Dying of Malignant Diseases," Brit. J. Prev. Soc. Med., 28;98-
100.
Falk, H.L., P. Kotin, and S. Thompson (1964), "Inhibition of Carcinogenesis.
The Effect of Polycyclic Hydrocarbons and Related Compounds," Arch.
Environ. Health, £: 169-179.
Frey, J.W. and M. Corn (1967), "Diesel Exhaust Particulates," Nature, 216(5115);
615-616.
Grimmer, G. (1977), "Analysis of Automobile Exhaust Condensates," in Air
Pollution and Cancer in Man," IARC Sci. Publ. No. 16.
Gross, G.P. (1972), "The Effect of Fuel and Vehicle Variables on Polynuclear
Aromatic Hydrocarbons and Phenols Emissions," Paper No. 720210. Presented
at the SAE Automotive Engineering Congress and Exposition. Detroit, Mich.
Gutwein, E.E., R.R. Landolt, and D.L. Brenchly (1974), "Barium Retention in
Rats Exposed to Combustion Products from Diesel Fuel Containing a Barium-
Based Antismoke Additive," J. Air Pollut. Control Assoc., 24:4043.
Gutwein, E.E., R.R. Landolt, and D.L. Brenchly (1972), "The Retention and
Distribution in the Rat of Barium Emitted from the Combustion of Diesel
Fuel Containing a Barium-Based Antismoke Additive," Presentation at the
65th Annual Meeting of the Air Pollution Control Association, June 18-22.
143
-------
Hangebrauck, R.P., D.J. von Lehmden, and J.E. Meeker (1967), "Sources of Poly-
nuclear Hydrocarbons in the Atmosphere," Environmental Health Series,
U.S. Dept. of HEW, Cincinnati, Ohio.
Hoffmann, D., E. Theisz, and E.L. Wynder (1965), "Studies on the Carcinogenicity
of Gasoline Exhaust," J. Air Pollut. Control Assoc., 15;162-165.
Hueper, W.C. (1955), "A Quest Into the Environmental Causes of Cancer of the
Lung," Public Health Monograph No. 36, U.S. Dept. of Health, Education,
and Welfare, Public Health Service.
Huisingh, J., R. Bradow, R. Jungers, L. Claxton, R. Zweidinger, S. Tejada,
J. Bumgarner, F. Duffield, and M. Waters (1978), "Application of
Bioassay to Characterization of Diesel Particle Emissions," Symposium
on Application of Short-term Bioassays in the Fractionation and Analysis
of Complex Environmental Mixtures, Feb. 21-23, Williamsburg, VA.
Hyde, D., J. Orthoefer, D. Dungworth, W. Tyler, R. Carter, and H. Lum (1978),
"Morphometric and Morphologic Evaluation of Pulmonary Lesions in Beagle
Dogs Chronically Exposed to High Ambient Levels of Air Pollutants,"
Laboratory Investigation, 38(4);455-469.
Infante, P.F., J.K. Wagoner, A.J. McMichael, R.J. Waxweiler, and H. Falk, (1976),
"Genetic Risks of Vinyl Chloride," Lancet, !_(7972):1289-1290.
Kaplan, I. (1959), "Relationship of Noxious Gases to Carcinoma of the Lung in
Railroad Workers," J. Am. Med. Assoc., 171.:2039-2043.
Kotin, P., H.L. Falk, and R. Busser (1959), "Distribution, Retention, and
Elimination of C14-3,4-Benzpyrene After Administration to Mice and Rats,"
J. Natl. Cancer Inst., _23;541-555.
Kotin, P., H.L. Falk, and M. Thomas (1955), "Aromatic Hydrocarbons III.
Presence in the Particulate Phase of Diesel-Engine Exhausts and the
Carcinogenicity of Exhaust Extracts," AMA Arch. Ind. Health, 11;113-120.
Kotin, P-, H.L. Falk, and M. Thomas (1954), "Aromatic Hydrocarbons II. Presence
in the Particulate Phase of Gasoline-Engine Exhausts and the Carcinogeni-
city of Exhaust Extracts," AMA Arch. Ind. Health, 9^:164-177.
Laresgoiti, A., A.C. Loos, and G.S. Springer (1977), "Particulate and Smoke
Emission from a Light Duty Diesel Engine," Environ. Sci. Technol., 11:973.
Laurie, R.D., J.P. Lewkowski, G.P. Cooper, and L. Hastings (1978), Effects
of Diesel Exhaust on Behavior of the Rat, U.S. Environmental Protection
Agency, unpublished report.
Lauweryns, J.M. and J.H. Baert (1977), "Alveolar Clearance and the Role of
Pulmonary Lymphatics," Am. Rev. Resp. Dir., 115:625-683.
Lawther, P.J. and R. Waller (1976), "Coal Fires, Industrial Emissions and Motor
Vehicles as Sources of Environmental Carcinogens," in Environmental
Pollution and Carcinogenic Risks, IARC Sci. Publ. No. 13, pp. 27-40.
144
-------
Lee, S.D., K.I. Campbell, D. Laurie, R.G. Hinners, M. Malanchuk, W. Moore,
R.J. Bhatnagar, and I. Lee (1978), "lexicological Assessment of Diesel
Emissions," Presentation at the 71st Annual Meeting of the Air Pollution
Control Association, Houston, Texas, June 25-30, 1978.
Lee, S.D., M. Malanchuk, and V.N. Finelli (1976), "Biologic Effects of Auto
Emissions. I. Exhaust from Engine With and Without Catalytic Converter,"
J. Toxicol. Environ. Health, l.:705-712.
Menck, H.R. and B.E. Henderson (1976), "Occupational Differences in Rates of
Lung Cancer," J. Occup. Med., 18(12):797-801.
Miller, E.G. (1978), "Some Current Perspectives on Chemical Carcinogenesis in
Humans and Experimental Animals: Presidential Address," Cancer Res.,
3_8:1479-1496.
Mittler, S. and S. Nicholson (1957), "Carcinogenicity of Atmospheric Pollutants,
Ind. Med. Surg., 26^:135-138.
Moore, W., J. Orthoefer, J. Burkart, and M. Malanchuk (1978), "Preliminary
Findings on the Deposition and Retention of Automotive Diesel Particulate
in Rat Lungs," Presentation at the 71st Annual Meeting of the Air Pollu-
tion Control Association, Houston, Texas, June 25-30, 1978.
Moore, W., J. Orthoefer, and P. Berez (1978a), Subchronic Exposure of. Infant
Rats to Irradiated and Non-Irradiated Diesel Exhaust, U.S. Environmental
Protection Agency, unpublished report.
Moore, W., D. Hysell, R. Miller, M. Malanchuk, R. Hinners, Y. Yang, and
J.F. Stara (1975), "Exposure of Laboratory Animals to Atmospheric Mangan-
ese from Automotive Emissions," Environ. Res., £:274-284.
Mustafa, M.G., A.D. Hacker, J.J. Ospital, M.Z. Hussain, and S.D. Lee (1977),
"Biochemical Effects of Environmental Oxidant Pollutants in Animal Lungs,"
in Biochemical Effects of Environmental Pollutants, Chapter 6, S.D. Lee
(ed.), Ann Arbor Science Pub., Ann Arbor, Michigan.
NAS (1972), "Biological Effects of Atmospheric Pollutant: Particulate Poly-
cyclic Organic Matter," National Academy of Sciences, Washington, D.C.
NAS (1976), "Medical and Biologic Effects of Environmental Pollutants. Vapor-
Phase Organic Pollutants," National Academy of Sciences, Washington, D.C.
NAS (1977a), "Medical and Biologic Effects of Environmental Pollutants. Carbon
Monoxide," National Academy of Sciences, Washington, D.C.
NAS (1977b) "Medical and Biologic Effects of Environmental Pollutants. Nitrogen
Oxides," National Academy of Sciences, Washington, D.C.
NAS (1977c), "Committee on Medical and Biologic Effects of Environmental Pollu-
tants. Airborne Particles," National Academy of Sciences, Washington, D.C.
145
-------
NIOSH (1978), "The Use of Diesel Equipment in Underground Coal Mines." Work
Group Reports from a NIOSH Workshop, Morgantown, W.Va., Sept. 19-23,
NIOSH Publication Feb. 1978.
Orthoefer, J., W. Mo'ore, and H. Ball (1978), Electron Microscopic Observations
of Macrophages Containing Diesel Emission Particulate, U.S. Environmental
Protection Agency, unpublished report.
Pattle, R.E., H. Stretch, F. Burgess, K. Sinclair, and J.A.G. Edginton (1957),
"The Toxicity of Fumes from a Diesel Engine Under Four Different Running
Conditions," Brit. J. Ind. Med., 14_:47-55.
Pepelko, W., J. Mattox, Y.Y. Yang, and W. Moore (1978), Effects of 28 Days
Exposure to Diesel Exhaust Emissions Upon Arterial Blood Gases and Pul-
monary Function in Rats, U.S. Environmental Protection Agency, unpub-
lished report.
Pepelko, W.E., J.K. Mattox, Y.Y. Yang, and W. Moore, Jr. (1978a), Pulmonary
Function and Pathology in Cats Exposed 28 Days to Diesel Exhaust, U.S.
Environmental Protection Agency, unpublished report.
Pfeiffer, E.H. (1977), "Oncogenic Interaction of Carcinogenic and Non-Carcino-
genic Polycyclic Aromatic Hydrocarbons in Mice," in Air Pollution and
Cancer in Man, V. Mohr, D. SchmMhl, L. Tomatis (eds.), International
Agency for Research on Cancer, Scientific Publ. No. 16, p. 69-77.
Pfeiffer, E.H. (1973), "Investigations on the Carcinogenic Burden by Air Pollu-
tion in Man. VII. Studies on the Oncogenic Interaction of Polycyclic Aro-
matic Hydrocarbons," Zbl. Bakt. Hyg., I. Abt. Orig. B., 158;69-83.
Pott, F., R. Tomingas, and J. Misfeld (1977), "Tumours in Mice After Subcutan-
eous Injection of Automobile Exhaust Condensates," in Air Pollution and
Cancer in Man, IARC Sci. Publ. No. 16:79-87.
Raffle, P.A.B. (1957), "The Health of the Worker," Brit. J. Ind. Med., 14_:
73-80.
Reger, R. (1978), "Ventilatory Function Changes Over a Work Shift for Coal
Miners Exposed to Diesel Emissions," Presentation at the First Annual
NIOSH Scientific Symposium, Cincinnati, Ohio, April 17-19, 1978.
Rockette, H. (1977), "Mortality Among Coal Miners Covered by UMWA Health and
Retirement Funds," NIOSH Research Report, March 1977.
Santodonato, J., P. Howard, D. Basu, S. Lande, J.K. Selkirk, and P. Sheehe
(1978), "Health Assessment Document for Polycyclic Organic Matter,"
Draft report submitted to the U.S. Environmental Protection Agency,
Office of Research and Development, Washington, D.C.
Sawicki, E., J.E. Meeker, and M. Morgan (1965), "Polynuclear Aza Compounds in
Automotive Exhaust," Arch. Environ. Health, 31:773-775.
146
-------
Schreck, R.M. (1978), "Health Effects of Diesel Exhaust," Biomedical Sciences
Dept., General Motors Research Laboratories, Warren, Mich. 48090.
Springer, K.J. and T.M. Baines (1977), "Emissions from Diesel Versions of
Production Passenger Cars," Society of Automotive Engineers, Passenger
Car Meeting, Detroit, Mich., Spet. 26-30, 1977.
Stara, J.F., W. Moore, and A.W. Breidenbach (1974), "Toxicology of Atmospheric
Pollutants Resulting from Fuel Additives and Emissions Associated with the
Use of Automobile Catalytic Converters," Recent Advances in the Assessment
of the Health Effects of Environmental Pollutants, Vol. II. pp. 751-772.
Proceedings of an International Symposium, Paris, June 24-28, 1974.
Stern, F.B. and R.A. Lemen (1978), Exploratory Mortality Study Among Workers
Exposed to Carbon Monoxide from Automotive Exhaust, U.S. Dept. of Health,
Education, and Welfare, Public Health Service.
Stuart, B.O., R.F. Palmer, R.E. Filipy, K. Mapstead, and D. Teats (1978),
"Biological Effects of Chronic Inhalation of Coal Mine Dust and/or
Diesel Engine Exhaust in Rodents," Battelle Pacific Northwest Laboratory
Annual Report for 1977 to the DOE Assistant Secretary for Environment.
Part 1. Biomedical Sciences.
Stupfel, M., M. Magnier, F. Romary, M-H. Tran, and J-P. Moutet (1973), "Life-
long Exposure of SPF Rats to Automotive Exhaust Gas," Arch. Environ.
Health, ^6:264-269.
Teranishi, K., K. Hamada, and H. Watanabe (1978), "Mutagenicity in Salmonella
typhimurium Mutants of the Benzene-Soluble Organic Matter Derived from
Air-Borne Particulate Matter and Its Five Fractions," Mutat. Res., 56;273-
280.
U.S. Public Health Service, Center for Disease Control, unpublished report,
July 26, 1976.
Vainio, H., P. Uotila, J. Hartiala, and 0. Pelkonen (1976), "The Fate of
Intratracheally Installed Benzo[a]pyrene in the Isolated Perfused Rat
Lung of Both Control and 20-Methylcholanthrene Pretreated Rats," Res.
Commun. Chem. Path. Pharmacol., 13_(2): 259-271.
Van Duuren, B.L., C. Katz, and B.M. Goldschmidt (1973), "Brief Communications:
Cocarcinogenic Agents in Tobacco Carcinogenesis," J. Natl. Cancer Inst.,
51(2):703-705.
Viadana, E., I.D. Bross, and L. Houten (1976), "Cancer Experience of Men
Exposed to Inhalation of Chemicals or to Combustion Products," J. Occup.
Med., 18(12):787-792.
147
-------
Vuk, C.T. and J.H. Johnson (1975), "Measurement and Analysis of Particles
Emitted from a Diesel Combustion Process." Paper presented at the
Combustion Institute, Central States - Western States, 1975 Spring
Technical Meeting, San Antonio, Texas, April 20-21, 1975.
Wang, Y.Y., S.M. Rappaport, R.F. Sawyer, R.E. Talcott, and E.T. Wei (1978),
"Direct-Acting Mutagens in Automobile Exhaust," Cancer Letters, in press.
Warshawsky, D., R.W. Niemeier, and E. Bingham (1978), "Influence of Particu-
lates on Metabolism of Benzo[a]pyrene in the Isolated Perfused Lung,"
in Carcinogenesis, Vol. 3: Polynuclear Aromatic Hydrocarbons, P.W. Jones
and R.I. Freudenthal (eds.), Raven Press, New York.
Waxweiler, R.J., J.K. Wagoner, and W.C. Archer (1973), "Mortality of Potash
Workers," J. Occup. Med., 15;406-409.
Wiester, M.J., R. Iltis, J.F. Swan, and W. Moore (1978), Altered Function
and Histology in Guinea Pigs After Inhalation of Diesel Exhaust, U.S.
Environmental Protection Agency, unpublished report.
Wynder, E.L. and D. Hoffmann (1962), "A Study of Air Pollution Carcinogenesis.
III. Carcinogenic Activity of Gasoline Engine Exhaust Condensate," Cancer,
15:103-108.
148
-------
5.0 Identification of Knowledge Gaps
5.1 Biological Effects
There are no areas concerning the biological activity of diesel
emissions in which complete information is available. These gaps in the health
effects data base are a reflection of the limited number of studies which have
been conducted, rather than an indication that the data are not obtainable.
Nevertheless, this lack of information is sufficient to prevent the formulation
of any definitive health risk assessment at this time. Recognizing that data
in certain areas are more valuable than in others for the purpose of risk
assessment, selected gaps in our body of knowledge are discussed (but not nec-
essarily in order of priority) below.
A population study has not yet been conducted in which a diverse
human group has been studied with respect to chronic low-level exposure to
diesel emissions. Therefore, nothing can be said regarding the existence of
susceptible groups, effects on mental and physical development, interference
with reproductive success, exacerbation of pre-existing disease, or interaction
with other common environmental pollutants.
Data derived from human exposure studies are of paramount importance
in establishing a chemical threat to health. Unfortunately, the available epi-
demiologic evidence is insufficient to define the effect of diesel emissions on
human populations. The historical epidemiologic literature regarding morbidity
and mortality has not adequately addressed the factors of exposure intensity
and duration, contial populations, and possible bias in reporting of results.
The comparative analysis of diesel emissions health effects is further com-
plicated by uncertainty over the consequences of human exposure to gasoline
149
-------
engine exhaust. Moreover, the impact of catalyst-treated gasoline engine
exhaust on human health is virtually unknown.
Thus far only one investigator has examined the clinical symptoms and
metabolic alterations produced by controlled human exposures to diesel exhaust.
Numerous questions still remain to be answered including: lung deposition,
retention and clearance of diesel particulate; alterations in enzyme activity;
effects on hematologic parameters; and correlation of physiologic effects with
concentration of specific components in the diesel exhaust mixture.
Animal bioassays conducted with diesel exhaust have provided important
data regarding the production of histopathologic lesions and biochemical altera-
i
tions in the lung, behavioral disturbances, susceptibility to infection, effects
on the heart, and decrements in pulmonary function. However, little is known
concerning the reversibility of this damage or the correlation between extent
of tissue damage and degree of functional impairment. In addition, dose-
response studies have not been conducted, nor is it known whether the toxic
effects produced by subchronic exposures are indeed life-shortening.
Recognizing that the lung is the primary organ which contacts air-
borne diesel emissions, several parameters of its interaction with diesel
exhaust components should be classified. Furthermore, it is difficult to re-
late morphologic alterations to biochemical events produced in response to
toxic insult. Since the respiratory epithelium is a major site for the sys-
temic absorption of airborne chemicals, we should also know the extent to
which diesel particulate and its adsorbed materials (e.g., carcinogenic POM)
reach the systemic circulation via the lung. This is especially relevant in
light of studies implicating automobile exhaust as a contributing factor in
cancers of various internal organs.
150
-------
Thus far it has not been possible to attribute most of the toxic
effects of diesel exhaust to the action of specific components in the mixture.
Likewise, a similar problem exists with most studies involving exposure to
gasoline engine exhaust, and thus comparisons between the two systems are
difficult.
The behavior of environmental pollutants in isolated organs and
individual cells often provides critical data concerning mechanisms of toxic
action. Numerous _in vitro studies with extracts.of airborne particulate pol-
lutants and their individual components (e.g., POM) have revealed the basis by
which damage is produced. With diesel emissions, however, nothing is known
regarding cytotoxicity, interaction with critical cellular macromolecules, cell
transformation, or cytogenetic damage.
15.1
-------
6.0 Recommended Research
6.1 Biological Effects
Animal studies conducted with diesel exhaust have revealed a diver-
sity of toxic effects in several species. These results are sufficient to
warrant further investigations in several areas. A list of suggested bio-
logical research projects is presented below which would fill many of the
information gaps identified in Section 5.1. This, however, is not a list of
research priorities.
1. Conduct of occupational and community-based epidemiologic
studies to provide morbidity and mortality data regarding
diesel exhaust exposure - Cancer as a biologic endpoint
should be of primary interest, but chronic respiratory
disease such as emphysema must also be carefully evalua-
ted. Whenever possible, information should be obtained
regarding current and past employment, respiratory symp-
toms, smoking histories, and other health information
such as genetic factors. For morbidity studies, a battery
of tests which includes pulmonary function measurements
and sputum cytology should be conducted. In addition,
comparison of the health status of persons exposed to
gasoline engine emissions with those exposed to diesel
emissions is highly desirable. Furthermore, a means
to quantitate observed exposures to diesel exhaust will
be necessary for the formulation of valid health risk
exposure criteria. It must be recognized, however, that
there are few opportunities for separating diesel and
gasoline exposure.
2. Cytogenetic testing of workers having high occupational
exposures to diesel emissions - Analysis, should be made for
chromosomal aberrations in cultured lymphocytes taken from
exposed workers. These determinations are felt by many
scientists to provide an indication of increased cancer
risk and potential for transmission of birth defects and
mutations. The use of somatic cells to predict a muta-
genic effect that may occur in germinal cells is not
entirely valid. Nevertheless, numerous examples of the
positive correlation between a chemical's ability to pro-
duce chromosome aberrations in somatic cells and its car-
cinogenic/mutagenic activity indicate that cytogenetic
data should be carefully evaluated. Moreover, the widely
held view that cancer arises as a result of somatic muta-
tion emphasizes the need for cytogenetic testing.
152
-------
3. Conduct of inhalation exposure studies in animals using
various dose levels - These studies should be designed
to detect dose-response parameters and ascertain the
reversibility of treatment-induced damage. Included in
the experimental protocols should be means to detect
neurotoxicity, reproductive effects, cardiovascular
function, effects on host defenses and threshold dose
levels for toxic response.
4. Evaluation of diesel exhaust mutagenicity in vivo and in
vitro - For the detection of gene mutations the use of
mammalian somatic cells in culture (with and without
metabolic activation) should be employed. Chromosomal
aberrations should be measured by in vivo cytogenetic
tests in animals, dominant lethal effects in rodents,
and heritable translocation tests in rodents. Primary
DNA damage should be detected using tests for unsched-
uled DNA repair synthesis and sister chromatid exchange
in mammalian cells (with and without metabolic activa-
tion), DNA repair in bacteria, and mitotic recombina-
tion and/or gene conversion in yeast.
5. Evaluation of in vitro carcinogenesis by diesel emissions -
Previous studies have shown that organic extracts of certain
samples of airborne particulate matter can transform
mammalian cells in culture. To establish the presence
of carcinogenic materials in diesel exhaust particulate,
various systems could be employed such as early passage
hamster embryo cells, baby hamster kidney cells, C3H10T 1/2
mouse fibroblasts, and several organ culture systems
using respiratory tract epithelium. Results from such
tests would provide important data to support the
observed mutagenic effect of diesel particulate extracts
in the Ames assay.
6. Evaluation of in vivo carcinogenesis by diesel emissions -
Studies using various fractions of diesel exhaust and
employing various routes of administration should be
pursued. In addition, an evaluation of the potential
involvement of cocarcinogens should be conducted.
7. Evaluation of the pulmonary deposition, clearance, and
transport of diesel particulate - Animal models should
be employed to ascertain the fate of inhaled diesel
particulate matter.
8. Evaluation of the toxicity of the vapor phase components
of diesel exhaust emissions - Standard inhalation experi-
ments in animals using reconstituted mixtures of gaseous
components at ratios found in raw exhaust should be con-
ducted.
153
-------
9« Evaluation of the effect of variation in fuels and engine
operating parameters on the resultant toxicity of diesel
exhaust emissions - These parameters can be incorporated
into nearly all of the suggested research studies listed
above.
10. Conduct of parallel toxicity studies with diesel and
gasoline engine exhaust - Using identical test conditions,
a comparison should be made between the relative hazards
to health of diesel versus gasoline engine exhaust. These
studies would provide important data concerning the ulti-
mate environmental impact and public health implications
of a major changeover to the use of diesel vehicles.
154
------- |