I nt A h f) m;-; T-T-10, il QW I O;d 1 d Q_ 1 k
Tnt 111
A r bleoi ' VI
N Fn
Up flenlth flO If t 1
Review articles
Combustion of diesel fuel from a toxicological perspective
I Origin of incomplete combustion products
P T J Scheepers and R P Bos
Toxicology Department, University of Nijmegen, P O Box 9101, NL-6500 HB Nijmegen, The Netherlands
Received November 7, 1991 / Accepted April 29, 1992
Summary Since the use of diesel engines is still increas-
ing, the contribution of their incomplete combustion
products to air pollution is becoming ever more impor-
tant The presence of irritating and genotoxic substances
in both the gas phase and the particulate phase consti-
tuents is considered to have significant health implica-
tions The quantity of soot particles and the particle-as-
sociated organics emitted from the tail pipe of a diesel-
powered vehicle depend primarily on the engine type
and combustion conditions but also on fuel properties.
The quantity of soot particles in the emissions is deter-
mined by the balance between the rate of formation and
subsequent oxidation Organics are adsorbed onto car-
bon cores in the cylinder, in the exhaust system, in the
atmosphere and even on the filter during sample collec-
tion Diesel fuel contains polycyclic aromatic hydrocar-
bons (PAHs) and some alkyl derivatives Both groups of
compounds may survive the combustion process PAHs
are formed by the combustion of crankcase oil or may be
resuspended from engine and/or exhaust deposits The
conversion of parent PA Hs to oxygenated and nitrated
PA Hs in the combustion chamber or in the exhaust sys-
tem is related to the vast amount of excess combustion
air that is supplied to the engine and the high combus-
tion temperature Whether the occurrence of these de-
rivatives is characteristic for the composition of diesel
engine exhaust remains to be ascertained After the emis-
sion of the particles, their properties may change be-
cause of atmospheric processes such as aging and resus-
pension The particle-associated organics may also be
subject to (photo)chemical conversions or the compo-
nents may change during sampling and analysis Mea-
surement of emissions of incomplete combustion pro-
ducts as determined on a chassis dynamometer provides
knowledge of the chemical composition of the particle-
associated organics This knowledge is useful as a basis
for a toxicological evaluation of the health hazards of
diesel engine emissions.
Correspondenceto: P T J Scheepers
' Internatonal Archvesof
OeC Upationaland
Enironmentad
Health
© Springer-Verlag 1992
Key words: Diesel exhaust Combustion Soot parti-
cles Particle-associated organics Polycyclic aromatic
hydrocarbons
Introduction
For centuries horsepower was the most important aid in
heavy duty work At the turn of the eighteenth century
the steam engine was introduced, triggering the industrial
revolution From the end of the nineteenth century fossil
fuel-powered engines started to replace coal-powered
steam engines Of these new generation engines, the
diesel engine has been the predominant power source
for heavy equipment machinery, most high-powered
farm tractors, and a large proportion of trucks and buses
(Encyclopedia Brittanica 1984) Diesel-powered engines
are also penetrating the private car market Favorable
price developments in several countries, together with
the higher efficiency and long life of diesel engines as
compared to spark-ignition engines, have favored the
sale of diesel-powered cars for private use Profession-
ally and privately used diesel-powered vehicles contri-
bute up to 6 % and 15 %-20% of the total number of
vehicles in car parks in the United States and European
countries, respectively (Hare 1986).
The diesel engine has recently attracted more atten-
tion, especially since the introduction of a catalytic ex-
haust gas purification system in gasoline-powered engines
and "clean" fuels such as liquified petroleum gas (LPG).
At the moment particulate traps and exhaust gas purifi-
cation systems are not widely used in diesel-powered
vehicles to clean exhaust emissions Because of this,
diesel engines have become a relatively important con-
tributor to air pollution.
In order to develop vulnerability and risk estimates
for exposure to diesel exhaust, knowledge of the forma-
tion and composition of diesel exhaust is essential In this
review we intend to combine both current knowledge
150
from the technological field (fuel quality and motor
conditions as determinants of exhaust composition) and
the toxicological field (the relevance of individual ex-
haust components to the health hazard) This should be
seen as an effort to identify (groups of) chemicals that
can be used as markers for environmental and biological
monitoring of (occupational) exposure to diesel exhaust.
Animal studies have shown that the carcinogenicity of
diesel exhaust is primarily attributable to the particulate
phase Because of this, emphasis is given to the emission
of particles and the organic compounds associated with
them.
This part of the review will be limited to the origin of
incomplete combustion products Part II will deal with
the toxicological characteristics of diesel exhaust emis-
sions.
Fuel, motor oil, and combustion
Fuel quality
During production, diesel fuel is distilled at higher tem-
peratures ( 160 °-390°C) than gasoline (80 °C-120 °C).
Therefore it contains predominantly aliphatic hydrocar-
bon chains of 9-28 carbon atoms (IARC 1989) Diesel
fuel may also contain a considerable amount of polycyclic
aromatic hydrocarbons (PA Hs) Another characteristic
of diesel fuel is its high sulfur content (0 1-0 5 wt%)
when compared to that of gasoline (about 0 02 wt%).
Comparison of commercial quality diesel fuels collected
at five different fuel stations all over the United King-
dom showed that the concentrations of individual PA Hs
relative to each other remain fairly constant (Williams et
al 1986 a) However, major components such as alkyl
derivatives of naphthalene, fluorene, and phenanthrene
show wide variation (by a factor of 2-3) between the
fuels This can be explained by their origin from differ-
ent refineries and different crude oil feedstocks (Wil-
liams et al 1986 a) In agreement with the analysis of the
United Kingdom fuels, Nelson ( 1989) reported that fluo-
renes and phenanthrenes and their alkyl derivatives were
the major components of a commerical fuel from a Syd-
ney refinery Alkyl-substituted naphthalenes were not at
all important constituents in this fuel In addition to the
major contribution of 2-3 rings, 4-5 rings such as py-
rene, fluoranthene, and benzolalpyrenes are detected in
diesel fuel at far lower levels (Henderson et al 1982) It
must be emphasized that fuel quality differences exist on
a worldwide basis In the United States diesel fuel vola-
tility is higher, the cetane number (which reflects the
ignition rate on a scale where methylnaphthalene corre-
sponds to the combustion rate of 0 and cetane to that of
100) is lower, and both density and aromatic content are
greater than in fuels currently traded in Europe and Ja-
pan.
The quality of diesel has changed over the years since
its introduction as a fuel Wade and Jones (1984) noted
that between 1960 and 1984 in the United States the
cetane number of # 2 fuel fell five numbers (from 50 to
45) and the 90 % boiling point rose 20 °C (from 560 to
580 °C) Both quality changes led to higher particulate,
N Ox and PAH emissions For the near future (until the
year 2000) a slight decrease in cetane number (two num-
bers combined with a small rise in density (0 006) as a re-
sult of a lower H/C ratio is expected, due to the increased
use of cracked components Because of this, the future
increase in emissions from current technology engines is
estimated to be no more than 2 % for N Ox, 7 % for HC,
12% for CO, and 6 % for particles (CONCAWE 1987).
Fuel combustion
Attempts to use multiple regression analyses to separate
the effects of various fuel properties on the particulate
emission composition have led to the following results
(Van Beckhoven 1991): for heavy duty direct injection
(DI) engines the 50 % and 90 % boiling points and viscos-
ity at 40 °C have been identified as important fuel factors
determining particulate emissions Aromatics and cetane
number were found to have no significant influence on
particles In contrast, regression analysis on indirect in-
jection (IDI) engines used in passenger cars shows that
particulate emissions correlate well with the cetane num-
ber or the aromatic content of the fuel From calcula-
tions with a multiple regression model it could be con-
cluded that the aromatic content and the back-end vola-
tility (as measured by the fraction of the fuel boiling
above 356 °C) affected particulate emissions in two dif-
ferent IDI engines (Hare and Baines 1979) Table 1 sum-
marizes the influence of diesel fuel composition on ex-
haust composition Testing of diesel exhaust revealed a
wide variation in mutagenicity Huisingh and co-workers
( 1978) combusted five types of commerical fuels ranging
in cetane number from 41 8 (low quality # 2 fuel) to 53 0
(high quality # 2 fuel) in two types of light duty vehicles
(LD Vs) The low quality fuel with high aromatic content
and low cetane number contained most mutagens In
this type of fuel the benzolalpyrene content of the par-
ticulate extract was shown to be more than twice as high
(20 ng/mg particle) as that of the fuels with higher cetane
numbers (< 10ng/mg particle) In another study both
Table 1 Influence of fuel characteristics on exhaust composition:
, much lower; -, lower; 0, neutral; +, more; ++, much more
(adopted from Weidmann et al 1988)
Influence on emission of:
HC CO N Ox Par PA Hs Alde-
ticles hydes
Rise in:
Cetane number +/0
Aromatic content ++ + O + + +
Density ++ ++ + ++ ± +
Initial boiling point O 0/ + ++b ++ +/-
90% boiling point O + +c + 0/-
Final boiling point O + +d + O
a 0 5 %-1 9 % per 1 0% aromatics (Hare 1986)
b Nearly neglible (Hare 1986)
c 2% per 10 °C (Hare 1986)
d 2% for each volume % of fuel boiling over 356 °C (Hare 1986)

aromatic content and initial boiling point of the fuel
were recognized as determinants of mutagenicity of the
particulate extracts (Clark et al 1982).
At a fuel sulfur content of 0 1%-0 2 %, sulfur is re-
covered in the exhaust, mostly as sulfur dioxide A small
portion ( 1 %-2%) of the sulfur is converted to sulfate
(Truex et al 1980) In heavy duty vehicles (HDVs) a
lower fuel sulfur content tends to decrease the sulfate
and water content of the particles A reduction of the
fuel sulfur content by 0 1 % would lead to a reduction in
particulate emissions of about 10 % in DI engines (Hare
1986) However, the magnitude of this effect varies with
the load conditions used in the test procedure (Van Beck-
hoven 1991) In several tests using low sulfur fuels in IDI
engines, no significant effect on particulate emissions
was observed (Van Beckhoven 1991).
The effect of decreased sulfur content tends to be
greater at the lower particulate emission levels In addi-
tion, the diesel sulfate emissions have been found not to
be proportional to the fuel sulfur level: the percentage of
sulfur converted to sulfate increases as the sulfur level
decreases At a fuel sulfur content of 0 009 % this con-
version amounts to 8 6 % (Truex et al 1980).
Fuel additives
The performance of diesel fuel is improved by various
additives such as amine detergents, polymeric dispersants,
metal deactivators, demulsifiers, cetane improvers or ig-
nition accelerators, flow improvers, antistatics, smoke
suppressants, antioxidants, and biocides in the 1-1000
ppm range The latter two are added to ensure that fuel
quality is maintained during transport and storage Col-
lectively these chemicals are usually referred to as fuel
additives Many chemical structures with toxicological
significance are found among the additives: esters, phe-
nols, amines, organic nitrates, both inorganic and organ-
ic metal compounds, etc The influence of these addi-
tives on the exhaust composition has scarcely been stud-
ied Certain (unspecified) detergent type fuel additives
are known to clean up the injector deposits, thereby low-
ering particulate emissions (Yoshida et al 1986) The
use of cetane or ignition improvers was found to result in
higher particulate emissions (Ryan et al 1981 ; Hare 1986)
and higher benzolalpyrene emissions (Hare et al 1976).
A barium and manganese-based fuel additive, sold for
smoke suppression, resulted in statistically significant in-
creases in both exhaust mutagens (Salmonella mutage-
nicity test) and exhaust PAH concentrations when added
to the diesel fuel of a heavy duty IDI engine (Draper et
al 1988 ; Hare et al 1976) When an ignition improver
was added to two different fuel types combusted in a
heavy duty DI engine no significant effect was observed
(Van Beckhoven 1991) Addition of a smoke suppres-
sant, containing calcium and small quantities of barium,
to # 2 diesel fuel resulted in markedly lower visible
smoke This was the reverse of the finding in respect of
the emission of particles, higher fuel specific particulate
emissions being reported in virtually all conditions tested
in two types of diesel engines (Hare et al 1976).
151
Lubricatingoil
The contribution of lubricating oil to the magnitude of
particulate emissions has been studied by adding a radio-
active tracer to the oil of a motor running on a chassis
dynamometer at steady state conditions (Mayer et al.
1980) Depending on speed and load, between 1 5 and
25 wt% of the particulate matter consisted of material
from the engine oil Williams et al (1987) reported that
lubricating oil contributed up to 70% of the total exhaust
particles at low load and high speed The contribution to
the extractable organic portion of the particulate matter
ranged between 16 % and 80% Zweidinger (1982) re-
ported an increase in the emission of the soluble organic
fraction of 82%, due to the combustion of motor oil.
Cuthbertson and co-workers ( 1987) analyzed the par-
ticle-associated hydrocarbons from two HD Vs operating
at standard injection timing and fueled with # 2 type fuel.
They estimated the unburned oil to contribute 40 % of
the emission of a DI engine and 28 % of the emission of
IDI type engines The contribution of heavy molecular
combustion products was neglected in this estimation.
The oil compartment can be considered as a sink for
heavy molecular incomplete combustion products such
as PA Hs As will be further explained in the section
"Polycyclic aromatic hydrocarbons and their alkylated
derivatives," these compounds are scrubbed from the
exhaust gas and accumulate in the oil rapidly (Williams et
al 1987 ; Abbass et al 1987) Later, when small amounts
of oil leak into the combustion chamber, because of a
poor fit of the piston in the cylinder, PA Hs and analogues
may contribute to the exhaust emissions.
Engine design and motor conditions
A basic differentiation when classifying engine types is
that between LDVs and HD Vs In the first category
emission data are numerous, showing a wide variation in
exhaust composition Samples from different makes of
LDV contain similar chemical components but the concen-
trations of the constitutents are highly variable (Schuetzle
1983) However, from analysis of the PAH composition
of exhaust emissions from 20 different car models (LDVs)
Grimmer and Hildebrandt ( 1975) concluded that it was
impossible to distinguish different car models by their
PAH profile LD Vs of the same make and model yielded
extracts of similar cytotoxicty in the Chinese hamster
ovary cell assay Extracts collected from cars of different
manufacturers, however, yielded extracts with a three-
fold difference in cytotoxicity (Li et al 1982).
That engine design is an important determinant of
emission performance of HD Vs as compared to fuel qual-
ity differences is illustrated in Fig 1 This shows the mag-
nitude of the variation in particle and NOx emissions due
to engine design In the same figure the relative influ-
ence of diesel fuel quality changes is depicted.
Another factor influencing the emission performance
is the method of injection, either direct or indirect The
first is mainly used for HD Vs, while the latter is used for
LDVs such as passenger cars In each of the designs, fuel
volatility, aromaticity, and cetane number determine soot
152
Particulates (g/k Wh) Predicted impact
1.1 of year 2000
el
1.0-
0.9-
0.8-
0.7-
0.6'
0.5'
0.4-
0.3-
0.2-
0.1-
I I I I I I
2.0 40 60 80 10 0 12 0 14 0
NOx (g/k Wh)
Fig 1 Particulate NO trade-off for HDV DI engines tested over
the US transient cycle (CONCAWE 1987)
formation in a different way because of the different char-
acteristics of the combustion process: in DI engines fuel
volatility is an important determinant of soot formation.
High volatility leads to a greater amount of fuel burning
in lean premixed conditions avoiding soot formation (Bar-
bella et al 1988) Fuel injected into the prechamber of
an IDI engine vaporizes and burns in low air/fuel condi-
tions and high temperatures favoring pyrolysis an thus
soot formation In this type of engine the enhancing ef-
fects of aromatic content and volatility on soot formation
are remarkable, whereas the fuel cetane number seems
to have no effect (Barbella et al 1988) Yamaki et al.
( 1986) explain the higher emissions of 1-nitropyrene from
an IDI as compared to a DI engine as a direct result of
differences in the combustion process.
In addition to the general design of the engine, several
other factors such as piston cavity diameter, swirl ratio,
number of injection nozzle openings, and injection direc-
tion determine the soluble organic fraction and particu-
late matter in the exhaust emissions (Murayama et al.
1986).
The combustion system of HDVs has been improved
by the introduction of turbo charging (see Fig 1) Re-
duction of particulate emissions is achieved by (a) in-
creasing the supply of oxygen to improve the air/fuel
mixing, (b) enhancing particle oxidation during the ex-
pansion stroke, and (c) using higher intake charge tem-
peratures which have the potential to improve fuel va-
porization (Wade 1980).
Injection timing Poor timing of the injection of fuel into
the combustion chamber may lead to emissions with
higher mutagenicity (Zweidinger 1982) A low oxidation
rate of particles formed in the combustion chamber of an
IDI engine at advanced timing may contribute to the in-
creased particulate emission (Du et al 1986) Changes in
timing, however, were reported to have little effect on
PAH emissions in IDI engines (Schuetzle and Frazier
1986).
When comparing retarded, standard, and advanced
timing, in succession, 1-nitropyrene emissions and direct
mutagenicity increase several times This increase can be
correlated with the increase in emitted N Ox Saito and
co-workers (1982) found an increase in 1-nitropyrene
and benzolalpyrene emission with the occurrence of sec-
ondary injection induced by injector clogging in an IDI
engine Injection timing was found to have no influence
on the emission of alkyl-PA Hs (Jensen and Hites 1983).
Load and temperature PA Hs emitted during idling are
enriched with respect to 3 and 4-ring PA Hs (Rosenkranz
and Mermelstein 1985) At high engine loads PA Hs and
total particulate emissions tend to increase gradually At
mid load conditions PAH emissions are lowest At high
load, particles are oxidized during the expansion stroke.
At very high temperatures resulting from extremely high
load conditions, the oxidation rate of the particles pro-
duced in the combustion process could be higher, de-
creasing particulate emissions (Pipho et al 1986).
Temperature, together with the air/fuel ratio, seems
to be the most important determinant of emitted PAHs
(Grimmer 1979) The temperature of a running engine
and its exhaust system is more or less proportional to its
load Therefore observations made for assessing the in-
fluence of load are applicable to temperature conditions.
Fuellairratio Increasing the fuel/air ratio led to higher
particulate emissions in both DI and IDI engines tested
with four different fuels The unburnt liquid hydrocarbon
emission, however, increased with the fuel/air ratio in
IDI engines whereas a decrease was observed in the DI
engine (Barbella et al 1988) In DI engines, especially
PAH emissions increase rapidly with decreasing fuel/air
ratio In IDI engines the relation between the fuel/air
ratio and PAH emissions is weak Increasing the amount
of pure oxygen gas in the combustion air of a DI engine,
from 21 % to 29%, was found to cause a decrease in par-
ticulate emission, especially at high load conditions How-
ever, N Ox increased with the intake of oxygen (Iida et
al 1986).
General condition of the engine For both LD Vs and
HDVs, elevated emissions of PA Hs, hydrocarbons, and
particles have been reported in more intensively used or
older engines as compared to extensively used or new
engines (Steenlage and Rijkeboer 1985 ; Stenberg et al.
1983 ; Braddock and Perry 1986 ; Zweidinger 1982) Some
of these investigators have suggested that the combus-
tion of oil explains the enhanced emissions in these en-
gines when compared to new ones (Stenberg et al 1983 ;
Zweidinger et al 1982).
Emissions determined on a chassis dynamometer
In Table 2, average emission data are presented from
several European diesel LD Vs tested on a chassis dyna-
mometer Despite differences in test cycle, diesel fuel,
and engine fabrication, emission factors are of the same
order of magnitude For this kind of test the choice of
simulation conditions, such as driving cycle and sampling
 153

Table 2 Emission data of some LDVs tested on a chassis dynamometer

Egback 1983 Lies 1989 Steenlage and
Rijkeboer 1985
Vehicle + odometer Volve 244 D 14087 VW Rabbit/VW Dasher/VW Jetta/ Opel Rekord 2 3D < 50000
count (km) Mercedes 125554 VW Jetta TD/Audi 5000 D (3 x) Opel Rekord 2 1 D > 100000
Fuel:
Cetane number (-) 51 51 8
Density (g/cm 3 ) 0 837 0 834
Sulfur content (wt%) 0 31 0 23
Viscosity (c ST/mm3 /s) 13 3 12
Test cycle U578 a U 575/SET/HD T a EC Ea
Test condition: Volvo Mercedes U 575 SET HDT < 50000 > 100000
n = 2-3 N = 2-3 N = 3-6 N = 3-6 N = 3-6 n=1 N =1
Compound (glkm):
Hydrocarbons 0 19 0 73 0 18 0 20 0 14 30 29
CO 0 93 1 87 0 87 0 85 0 65 17 14
N Ox 0 95 0 92 0 74 0 62 0 55 09 09
Compound (mglkm):
Particles 238 70 221 91 166 42
SO2 260 92 242 61 196 78
Sulfate 5 41 5 76 4 94
Cyanides 0 68 0 62 0 50
Ammonia 1 55 1 04 0 92
Methane 5 10 6 30 5 63 24 1
Ethane 0 22 0 31 0 16 1 -
Ethylene 25 6 90 7 15 68 16 88 13 54 31 19
Propylene 8
1/i-Butane 4 4
n-Decane 2 2
n-Undecane 4 3
n-Dodecane 5 3
n-Tri/heptadecane 14 12
Methanol 16 16
Ethanol 22 2 < 12
Formaldehyde 79 20 7 8 13
Acetaldehyde < 22 36 3 14
Acrolein < 1 18
Propionaldehyde O99
Methyl ethyl ketone O78
+ isobutylaldehyde
Crotanaldehyde 1 22
Benzaldehyde O55
Hexanaldehyde O 80
Methyl nitrite 0 217 0 302
Ethyl nitrite 0 136 0 125
Benzene 95 18 1 3 77 4 33 2 58 3 6
Toluene <1 0 <1 0 1 92 1 26 1 08 2 <1
Naphthalene 2 6

Compound (glkm):
Anthracene 17 63
b
Phenanthrene 140 295 524
Fluoranthene 172 b 79 69 60 38 65 60 58 200
Pyrene 186 b 73 58 66 59 87 13 <0 9 22
Chrysene 40 b 13 60 10 74 22 25 14 67
Benzolb, klfluoranthene 40 b 10 79 10 26 15 21 26 47
154

Table 2 (continued)
Test condition: Volvo Mercedes U575 SET HD Vs < 50000 > 100000
n = 2-3 n = 2-3 N = 3-6 N = 3-6 N = 3-6 n=1 n=1
b
Benzolghilfluoranthene 40
b
Cyclopentalcdlpyrene 2 0.81c
Benzolelpyrene 16.87 13.80 14.84 3 38
Benzolalpyrene 10 b 5.32 4.01 4.68 <1 19
Benzolalanthracene 25 b 3.93 3.51 2.74 8 43
Indeno l 1,2,3-cdlpyrene 15 b 4.35 3.54 3.78
Benzolghilperylene 20 b 10.94 11.67 8.73 <1 18
Perylene 0.62 0.49 0.55 <1 2
Coronene 5.43C
Triphenyl 9 <1
3,6-Dimethylphenanthrene 10 19
2-Nitrofluorene 0.99c
2,7-Dinitrofluorene 0.13 c
9-Nitroanthracene 1.00 C
3-Nitrofluoranthene 0.06c
1-Nitropyrene 3.43C
1,3-, 1,6 and 0.12c
1,8-Dinitropyrene
6-Nitrobenzolalpyrene 0 12c
Mutagenicity of particulateextract in Salmonella mutagenicity assay (revl/g)
TA 98 -59 398 8b
TA 98 +59 360 O"
TA100 -59 501 6b
TA100 +59 439 Ob
-, Not analyzed; <, below detection limit b Average of two vehicles (n = 5)
a U578 = Federal Test Procedure; U575 = Federal Test Proce- c U572 cycle with hot start, four cylinder engine
dure with cold start; SET = US Sulfate Emission Test; HDT = US
Highway Driving Test; ECE = European city driving cycle

system, is not critical for the sampling of organic vapors,
particle-associated organics, or mutagenic activity (Bra-
dow 1982) Among PAHs, differences are somewhat
greater, especially when vehicles with high and low odom-
eter counts are compared It is not known to what extent
the differences found in this study were caused by the
contribution of PA Hs from the higher oil consumption
rate of the more intensively used vehicles The profile of
the PA Hs from the three studies appears to be similar,
showing the slightly volatile phenanthrene, fluoranthene,
and pyrene (with vapor pressures of 10-6 to 10- 4 mm Hg
and Mw = 178-202) at an emission level one order of
magnitude greater compared to the PA Hs with a relatively
lower vapor pressure (below 10- 7mm Hg, Mw = 228-300).
The emission level of 1-nitropyrene is of the same or-
der of magnitude as the low volatile PA Hs Artificial
formation of this compound is not expected under the
sampling conditions employed in the German study (Lies
1989).
Table 3 presents the amount of hydrocarbons emitted
in relation to the amount consumed as fuel every kilome-
ter of distance in LDVs, driving on a chassis dynamome-
ter The fuel profiles are typical of # 2 quality diesel fuel:
high levels of C8-C 18 and an abundance of the 3 and 4-ring
PA Hs From the tail pipe low molecular combustion
products are emitted in the mg/km range, while PA Hs
are emitted in the jlg/km range The conversion of high
molecular fuel constituents to low molecular exhaust
components is consistent with the observations of Nelson
( 1989).
The amounts of soluble organic compounds emitted
by an HDV, tested on a chassis dynamometer, are pre-
sented in Table 4 The influence of both fuel type and par-
ticle trap was tested when simulating a bus cycle The in-
crease in PAH emissions from the combustion of a fuel
with a higher cetane number is not consistent with previ-
ous observations (see Table 1) However, fuel B also
contains more high boiling point components (final boil-
ing point 355 °C compared to 259 0 C for fuel A), suggest-
ing a greater contribution of higher molecular constituent
in fuel B.
Mutagenicity testing of the particulate extract with
the bacterial Salmonella mutagenicity assay (Salmonella
typhimurium strains TA100 and TA100 NR in the ab-
sence of an activating enzyme system) suggests that the
nitrated aromatics contribute 35 %-65% of the total di-
rect mutagenicity of the particulate extract The profile
of the PA Hs shows the same bimodal shape as in the
LDV 1-Nitropyrene is present in the same range as the
5 and 6-ring PAHs Both the particle traps tested re-
 155

Table 3 Comparison of fuel composition (mg/km) on the basis of fuel consumption and the composition of exhaust (mg/km) Data from
two Opel Rekord passenger cars tested on a chassis dynamometer (Steenlage and Rijkeboer 1985)
Compound EC E a ,c ECEb, c 90 km/h a' 90 km/hb c 120 km/h a c 120 km/hb c
Fd Ed Fd Ed Fd Ed Fd Ed Fd Ed Fd Ed
Methane <1 24 1 <1 <1 <1 <1 <1 <1 <1 <1
Ethane <1 1 11 < <1 <1 <1 < <1 < 1 1<1
Ethylene <1 31 19 <1 6 <1 13 <1 18 < 1 <1
Propane <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1
n/i-Butane <1 118 8 <1 53 <1 1 <1 57 <1 6
1/i-Butane 6 <1 <1 2 <1 2 <1 2 <1 2 <1
n/1-Pentane 48 <1 <1 17 <1 18 <1 25 <1 24 < 1
2,2,4-Trimethyl 140 <1 <1 49 <1 54 <1 72 <1 71 < 1
pentane
2/3-Methylhexane 46 <1 <1 16 <1 18 <1 24 <1 24 <1
Cyclohexane 16 <1 <1 6 <1 6 <1 8 <1 8 <1
n-Heptane 46 <1 <1 16 <1 17 <1 24 <1 23 <1
n-Octane 134 <1 <1 48 <1 52 <1 69 <1 68 <1
n-Nonane 482 <1 <1 171 <1 186 <1 247 <1 244 <1
n-Decane 1396 2 2 494 <1 538 <1 715 <1 706 <1
n-Undecane 2068 4 3 732 <1 797 <1 1060 <1 1045 <1
n-Do/heptadecane 19140 5487 5136 7946 7839 -
Benzene 6 3 6 2 <1 3 3 4 2 4 5
Toluene 56 2 <1 20 <1 22 <1 28 <1 28 2
p,m,o-Xylene 400 <1 <1 142 <1 154 <1 206 <1 203 <1
Ethylbenzene 72 <1 <1 26 <1 28 <1 37 <1 37 <1
Phenanthrene 33000 0 295 05 23000 0 115 25000 0 115 34000 0 061 33000 0 280
Anthracene 1034 0 017 0 062 731 0 007 795 0 022 1061 0 015 1043 0 029
Benzolal 7500 0 008 0 043 5304 0 006 5762 0 005 7699 0 007 7565 0 008
anthracene
Fluoranthene 21000 0 058 0 200 15000 0 049 16000 0 071 21000 0 019 21000 0 058
Benzolkl 129 < 0 001 0 013 91 < 0 001 99 0 002 133 0 002 130 0 002
fluoranthene
-, Not analyzed; <, below detection limit 120 km/h = simulated driving test, 15 min at 120 km/h constant
a less than 50000 km on odometer speed
b more than 100000 km on odometer d F = fuel composition calculated on the basis of consumed amount
c ECE = simulated driving test, European city driving cycle; 90 of fuel (mg/km); E = exhaust composition (mg/km)
km/h = simulated driving test, 15 min at 90 km/h constant speed;

duce PAH emissions by a factor of 10 and mutagenicity
in the Salmonella mutagenicity assay by more than 50%
(Westerholm et al 1986).
Particles as vehicles for transport
of combustion products
Particleformation and oxidation
Three stages have been proposed during the process of
soot particle formation in the diesel engine and exhaust
system (Lies 1989 ; Amann et al 1980) In the first
phase, graphite-like particles are formed from simulta-
neous cracking, pyrolysis, dehydrogenation, and pyro-
synthesis at temperatures ranging from 2000° to 2300°C
(Kamimoto et al 1980 ; Aoyagi et al 1980) These parti-
cles reach a critical size by surface growth with diameters
ranging from about 0 001 to 0 01 gm and are called car-
bonaceous nuclei In a second phase, basically spherical
primary soot particles of 0 01-0 05 pm diameter are
formed by agglomeration The primary soot particles
form sponge-like clusters and chains in a third phase,
that of coagulation, reaching diameters of 0 1-0 2 gm.
These particles have a very large surface area (50-200
m 2/g soot) similar to charcoal (Lies 1989) This mecha-
nism of particulate formation is supported by observa-
tions made by Dolan et al (1980) They noted a bimodal
size distribution with particles of 0 02 gm diameter origi-
nating from agglomeration (second phase) and particles
of 0 14 gm diameter representing the products of coagu-
lation (third phase) The particles do not vary greatly in
size with engine design, speed, or load (Groblicky and
Begeman 1979).
The quantity of exhaust soot depends upon how rap-
idly the soot formed during the combustion process is
destroyed by oxidation (Uyehara et al 1980 ; Amann et
al 1980 ; Pipho et al 1986) This process involves sub-
micron particles and is temperature directed rather than
diffusion directed Oxidant species in this process are
156

Table 4 Emission data of an HDV tested on a chassis dynamometer molecular 02 and OH radicals and less reactive mole-
(Westerholm et al 1986) cules such as CO 2 and H 2 0 (Serageldin 1981) Measure-
Fuel: A B ments in the cylinder of an IDI engine show that particu-
Cetane number (-) 42 51 late formation begins 1-5 crank angle degrees after the
Density (g/cm 3 ) 0 801 0 841 start of combustion A peak in particle concentration is
Sulfur content 0 03 0 26 reached at about 10-20 crank angle degrees after the
(% wt) start of combustion During the next 20-30 crank angle
degrees the particles are oxidized to a level of 5-15 times
Test condition: No trap No trap Particu Particu- lower concentrations in the exhaust system (Pipho et al.
late trapa late trapb
1986) In accordance with this, Hayano and co-workers
Fuel A B A B
(1985) reported the PAH levels in the exhaust gas flow
No of 5c 3c 3c 3C
observations
to be 2-3 orders of magnitude lower than in the combus-
tion chamber Du et al ( 1984) noted a tenfold reduction
Compound (glkm): in the concentrations of pyrene, fluoranthene, benzolkl
Fuel consumption 397 397 393 409 fluoranthene, and benzolalpyrene between the cylinder
Hydrocarbons 3.6 3 2 19 22 and the exhaust manifold.
CO 5.4 4 9 49 16
CO 2 1244 1282 1200 1310 Aging and resuspension The particles emitted from both
NO 2 16.1 12 5 13 6 15 1 LDV and HDV engines running at high speed on a chas-
Particulates 0.71 0 97 01 0 20 sis dynamometer were found to be divided by diameter
into two groups (Israel et al 1982): submicron particles
Compounds (mglkm):
with diameters ranging from 0 02 to O 5 ~tm and super-
Formaldehyde 140 150 170 110 micron particles with diameters greater than 2 tm Be-
Acetaldehyde 50 40 80 30 tween 50 % and 80 % of the total emission is discharged
Acrolein 4 4 4 <1 in the submicron range PA Hs are almost entirely bound
Acetone/pro- 11 17 8 70 onto these submicron particles (Israel et al 1982) The
pionaldehyde process of coagulation is usually referred to as aging It
Benzaldehyde 10 14 12 25 is not dependent on engine or exhaust conditions but
Tolualdehyde 14 17 14 2 rather on atmospheric factors such as photochemical
Compound (g/glkm): processes and humidity.
Anthracene 1.2 3O O4 O1 Lach and Winckler ( 1988) observed resuspension of
Phenanthrene 4.8 25 1O 13 diesel exhaust particles from a dilution tunnel after mea-
Fluoranthene 13 38 49 57 surements had been carried out on several diesel en-
Pyrene 20 104 62 71 gines The same process might also occur in indoor work
Benzolghil- 4.0 12 20 14 environments As an explanation for the presence of
fluoranthene large particles it has been suggested that frequent move-
Cyclopentalcdl- 3.6 39 0 07 O2 ment of vehicles causes resuspension of particles depo-
pyrene sited on the floor (Lehmann et al 1989).
Benzolal- 4.0 70 04 04 In a dilution tunnel the median diameters of particles
anthracene leaving the tail pipe were found to be between 0 13 and
Chrysene 11 25 20 15 O 19 jm for both HDV and LDV engines (Israel et al.
Benzolb, klfluorene 3.9 61 09 12 1982) The results of size distribution studies on particles
Benzolelpyrene 2.5 41 O3 O4 collected in a dilution tunnel do not appear different
Benzolalpyrene 1.6 22 02 02 from on-the-road measurements of particle size distribu-
Indenol1,2,3-cdl- 1.5 23 05 03 tions (Dolan et al 1980 ; Wade 1980).
pyrene
Benzolghil- 1.9 32 O5 O4 Photochemical reactions When emitted into the open
perylene air, diesel soot particles and the associated organic com-
Coronene 0.3 20 7 01 03 pounds are exposed to ambient factors, such as sunlight,
1-Nitropyrene 0.9 83 0 35 0 25 which may change their chemical composition Photo-
chemical transformations of PA Hs exposed to sunlight
Mutagenicityofparticular
'es in Salmonella mutagenicityassay (rev/#g) have been studied in smog chambers.
TA 98 59 189 384 74 67 In a rotary photoreactor, photolysis of PA Hs adsorbed
TA 98 + 59 2
218 448 79 61 onto the surface of carbon black showed half-lives for
TA 100 -59 445 764 128 221 eight individual PA Hs to exceed 1000 h (Behymer and
TA 100 + 59 4435 947 197 201 Hites 1985) Under the same conditions PA Hs adsorbed
TA 100NR -59 219 326 107 129 on silicagel, alumina, and fly ash were shown to be 10-
a Degussa particulate trap 100 times more vulnerable to photodegradation This in-
b Johnson/Matthey partticulate trap dicates that the carbon black surface stabilizes some
c Arithmetic means as calculated from the data reported by Wes PA Hs, facilitating their transport from combustion sources
terholm et al. into the atmosphere.

Hisamatsu et al ( 1986) have identified 1-nitropyrene,
1,3-dinitropyrene, 1,6-dinitropyrene, and 1,8-dinitropy-
rene after exposure of pyrene to UV light for 3 h in a
reaction vessel containing 10 ppm NO 2 diluted with air.
These results indicate that photochemical transforma-
tion is probably one of the processes contributing to the
conversion of particle-associated PA Hs to nitro-PA Hs.
On the other hand observations have been made of the
decomposition of 1-nitropyrene under irradiation, lead-
ing to a reduction in direct mutagenicity with increasing
irradiation energy (Stark et al 1985).
The origin of the soluble organic fraction of the particles
Aliphatic compounds are found in the vapor and the par-
ticulate phase of diesel exhaust Among the n-alkanes,
Cll-C16 products are the most abundant, having emis-
sion rates of 10 mg/km in HDVs The partitioning be-
tween vapor and particulate phase was evaluated for
C 14 -C 17 alkanes by Hampton et al ( 1983) The vapor/
particle ratio for these alkanes was found to be propor-
tional to the compounds' vapor pressure.
It is not clear whether the hydrocarbons associated to
soot particles are acquired within the cylinder, in the
exhaust system, in the atmosphere, or even on the soot
collection filter MacDonald et al (1980) and Plee and
MacDonald ( 1980) found that adsorption onto soot par-
ticles trapped on the filter was the dominant mechanism
responsible for the soluble hydrocarbon fraction appear-
ing on the filter Condensation of high boiling point com-
pounds directly onto the filter could also contribute to
the hydrocarbon fraction under conditions of low filter
temperature and high hydrocarbon concentrations (Amann
et al 1980).
The hydrocarbons adsorbed on soot particles could
originate from the same tail pipe as the soot particles,
but since soot particles travel through the atmosphere
acting as charcoal adsorbents, they may also adsorb atmo-
spheric contaminants The contribution of atmospheric
contaminants can be ruled out in the dilution tunnel by
supplying filtered air or pure N 2/O 2 as a dilution medi-
um In the atmosphere the contribution of ambient con-
taminants may become substantial, especially at realistic
outdoor dilution ratios of several thousands (Dolan et al.
1980).
Another question which must be answered is to what
extent the particle-associated hydrocarbon fraction is
modified by N Ox, SO 2, or 02 These reactants are con-
tinuously drawn past soot particles trapped on the filter.
Despite many environmental factors, the emission rates
of some n-alkanes, aromatics, and two methyl-PA Hs de-
termined for HD Vs and LD Vs in a highway mountain tun-
nel were generally consistent with tail pipe emission rates
determined from chassis dynamometer studies (Hampton
et al 1983) However other investigators have demon-
strated that the gas chromatographic profiles of extracts
of roadside samples were quite different from those
produced by dilution tunnel samples (Williams et al.
1988).
157
Polycyclic aromatic hydrocarbons
and their alkylated derivatives
In principle, four possible explanations exist for the pres-
ence of PA Hs in diesel exhaust:
1 Parent PA Hs in the fuel survive the combustion pro-
cess to appear in the exhaust From comparisons of fuel
and exhaust composition it has been suggested that some
fuel components such as fluoranthene, phenanthrene,
pyrene, and benzolalpyrene can pass unchanged into
the exhaust (Henderson et al 1982, 1984 ; Nelson 1989).
However, in none of these studies could the contribution
of PAHs contained within lubricating oil be excluded.
Methylnaphthalene or acenaphthene added to a pure
hexadecane fuel were not recovered in the exhaust after
combustion in a single cylinder DI engine (Henderson et
al 1984) Methylphenanthrenes and methylfluorenes
are suggested to originate from unburnt fuel components
(Williams et al 1986 a).
2 PA Hs are formed during the combustion process by
pyrolysis followed by pyrosynthesis reactions of fuel in
the combustion chamber Meyer and Grimmer ( 1974)
found no differences between PAH profiles of emissions
from a fuel containing PA Hs and otherwise identical fuel
free from PAHs in a spark ignition engine Apparently
PA Hs detected in the exhaust were produced during the
combustion process However, in this case, also, the
contribution of PA Hs from lubricating oil could not be
excluded When taking samples directly from the com-
bustion chamber, PA Hs were found at 200-fold higher
concentrations as compared to the PAH concentrations
in the exhaust mixture (Hayano et al 1985) This sug-
gests the formation of PA Hs in the combustion chamber
followed by decomposition in the exhaust system In a
recent study Abbass and co-workers (1988) estimated
the contribution of pyrosynthesis to be 5 % at most.
3 Crankcase oil constituents are known to be possible
suppliers of PA Hs Williams et al (1987) and Abbass et
al (1987) investigated lubricating oil contamination with
PA Hs in two types of DI engines used for mining Un-
used oil contained negligible concentrations of these
contaminants However, once the oil was used, the con-
tamination accumulated rapidly, reaching substantial
concentrations after only 10h use A steady state is reached
after about 100 h use A second older technology engine
showed a much higher accumulation rate and a very
much higher contamination level than did the current
technology engine that was tested.
Since all of the PA Hs are accumulated in the oil dur-
ing the combustion process, their profile is expected to
be similar to the PAH profile in the fuel or the profile of
PA Hs produced during the combustion process This is
why variable amounts of engine oil in the combustion
chambers may change the level of PA Hs emitted, but
are not expected to affect the exhaust PAH profile.
4 PA Hs are also thought to originate from engine and/
or exhaust system deposits (Abbass et al 1988) In a
study using pure hexadecane as a fuel and lubricating oil
free from PA Hs, it has been suggested that the evapora-
158
tion of high boiling point compounds deposited in the
engine or exhaust system might be a source of PA Hs in
addition to pyrolysis and pyrosynthesis.
Oxygenated polycyclic aromatichydrocarbons
Hydrocarbons are oxidized to a greater extent during
their combustion in a diesel engine than a comparative
combustion of gasoline or liquid petroleum gas because
of the vast amount of excess air supplied to the combus-
tion chamber This could be the major explanation of the
observation by Jensen and Hites ( 1983) that oxy-PA Hs
outnumber the abundance of parent PA Hs by a factor of
2 Oxygen-containing derivatives may include hydroxy,
dihydroxy, aldehyde, carboxyaldehyde, anhydride, qui-
none, ketone, and acetoxy functional groups (Choud-
hury et al 1982) A problem hindering the determina-
tion of the contribution of oxy-PA Hs to diesel engine
exhaust is their relative instability On the one hand oxy-
PA Hs may be formed artificially on the filter of the sam-
pling device On the other hand these compounds may
be converted during sampling, storage, and chemical
analysis (Schuetzle et al 1981).
9-Fluorenone and 9,10-anthracenequinone are two of
the most abundant oxygenated PAH derivatives in diesel
exhaust (Schuetzle et al 1981 ; Schuetzle and Frazier
1986 ; Lies 1989 ; Choudhury 1982) 9-Fluorenone, how-
ever, is not characteristic for diesel exhaust as it is more
abundant in other combustion products such as coal fire
exhaust (Moriske and Ruiden 1988).
Nitro-substitutedpolycyclic aromatichydrocarbons
Nitro-substituted PA Hs are formed from the reactions
of PA Hs with nitrating species that are provided by the
conversion of N2 and 02 at high temperatures in the
combustion chamber Two possible mechanisms are pos-
tulated for this (Schuetzle 1983 ; Olah et al 1981): (a) the
nitration of PA Hs with HNO 3 and (b) the addition of
NO or NO 2 to PAH free radicals generated during the
combustion process.
The role of nitrogen in the nitro-substitution of PA Hs
was investigated in a single cylinder diesel engine run-
ning on nitrogen-free air The amount of 1-nitropyrene
formed in the exhaust was substantially lower as compared
with the emission from an engine using ambient air (Herr
et al 1982) A frequently used method to investigate the
influence of fuel composition is the use of a synthetic fuel
consisting of one compound When fueling a diesel en-
gine with n-hexadecane, a relative increase in nitropyrenes
and nitrofluoranthenes was observed when pyrene was
added (Henderson et al 1984) In the case of a similar
treatment of the hexadecane fuel with phenanthrene an
increase in the emission of nitrophenanthrenes and dini-
trophenanthrenes was found However, a simultaneous
increase in direct mutagenicity in the Salmonella muta-
genicity assay was not observed.
The mutagenicity in the Salmonella mutagenicity as-
say with and without metabolic activation of the particles
collected from the exhaust system was observed to be
one order of magnitude lower as compared to the parti-
cles captured from the combustion chamber Since it is
known that nitro-PA Hs are directly mutagenic and un-
substituted PA Hs need to be metabolized into mutagens,
the relatively high direct mutagenicity in the combustion
chamber suggests a possible contribution of nitro-PA Hs,
already formed in the combustion chamber, in addition to
PA Hs Du et al ( 1984) described the quite different be-
havior of a specific nitro-PAH ( 1-nitropyrene): a four-
fold increase in concentration between the cylinder and
the exhaust manifold This suggests that 1-nitropyrene is
formed during the expansion stroke and/or in the ex-
haust manifold Nitro-PA Hs could also be formed after
leaving the tail pipe or even on the filter of the sampling
device It has been established that the latter contribute
less than 20 % to the 1-nitropyrene recovered on a filter
during short sampling times (Lies 1989).
Crankcase oils of diesel engines have been found to
exert very high direct mutagenic activity in the Salmonel-
la mutagenicity assay When NO 2 was bubbled slowly
through the crankcase oil the mutagenicity increased sub-
stantially (Abdelnasser et al 1986) This increase was
not found when the oil was treated with air, ammonia,
nitrous oxide, or SO 2 This suggests that the products of
the reaction of PA Hs with NO2 (nitro-PA Hs?) contri-
bute to the increase in mutagenicity.
1-Nitropyrene and 1,6-dinitropyrene are identified in
oil samples (Manabe et al 1984) They are found to con-
tribute 0 9% and 12 % to the mutagenicity, respectively.
In an LDV running 9000 km, the motor oil was found to
contain 0 5 ppm 1-nitropyrene, while this was not detect-
ed in new oil (Jensen et al 1986) In the exhaust of this
vehicle running on new and used oils the 1-nitropyrene
emission was 10 and 16 mg/km, respectively These levels
could not be explained by the contribution of 1-nitro-
pyrene contained within the oil, since no oil consump-
tion was observed over a driving distance of 9000 km.
From these observations Jensen and co-workers concluded
that it was unlikely that 1-nitropyrene in oil contributed
substantially to the exhaust levels of 1-nitropyrene.
In a comparative characterization of the nitro-PAH
content from diesel particles, coke oven mains, roofing
tar vapour, and cigarette smoke condensate, Williams
and co-workers (1986 b) reported significant levels of 18
different species of nitro-PAH in extracts of diesel parti-
cles of LDVs that are not detected in the other combus-
tion products at a detection level of < 50 pg In this com-
parison nitro-PAHs appear to be a group of PAH deriva-
tives typical in diesel exhaust emissions This could very
well be the consequence of the vast amount of excess air
supplied to the diesel fuel combustion.
Concluding remarks
The carbon cores of soot particles are an important ve-
hiculum for high molecular incomplete combustion pro-
ducts They are formed during the combustion process
by radical mediated pyrosynthesis reactions A fraction
of these particles is subsequently destroyed by oxidation
during the expansion stroke of the piston The yield of

soot particles eventually leaving the tail pipe depends on
the balance between soot formation and destruction.
Particle emissions increase with advanced injection
timing and fuel/air ratio The net yield of particles also
increases with engine load At very high load the par-
ticulate emission might go down again, since high tem-
peratures enhance the oxidation rate of carbon Fuel
characteristics associated with high particulate emissions
include high density, high initial boiling point and high
aromatic content Fuel characteristics that are more
weakly associated with particle emission are the 90 %
and final boiling point of the fuel.
Engine conditions causing high PAH and PAH de-
rivative emissions are: advanced timing of the injection,
extremely low and high load conditions, cold start, and
decreasing fuel/air ratio (specifically in DI engines).
Incomplete combustion products such as PAHs are
formed during the combustion process or originate from
the fuel, surviving the combustion proces Fuel charac-
teristics associated with high PAH emissions are the
same as those associated with high particulate emissions.
However, PAH emissions are also known to vary with
the amount of oil taking part in the combustion This
points to the lubricating oil as an important secondary
source of PAHs Since PA Hs are not present in unused
lubricating oil, they are obviously formed by pyrosyn-
thesis in the cylinder and accumulate in the oil rapidly.
Another secondary source of PA Hs are the engine and ex-
haust deposits: fuel oil and/or exhaust residues that con-
tain PA Hs or scrub PAHs from the gas phase Eventu-
ally these deposits are resuspended, causing PA Hs to ap-
pear in the emissions.
The vast amount of excess air as well as very high
temperatures (up to 23000 C) in the combustion chamber
leads to the appearance of reactive oxygen and nitrogen
oxide species In this reactive mixture, PA Hs are con-
verted to oxy and nitro-PA Hs Some studies suggest
that this conversion also takes place in the exhaust sys-
tem Since the excess combustion air supply and high
combustion temperature are characteristic for the com-
bustion of diesel fuel, the appearance of nitrated and
oxygenated PA Hs could be characteristic for the com-
bustion of diesel fuel This is confirmed by comparison
of the emissions from diesel engines with those from
other sources.
Knowledge of the origin and conditions determining
the emission of incomplete combustion products is use-
ful in the evaluation of health hazards that may derive
from the use of diesel engines as power sources in indus-
try and transport.
Acknowledgements Financial support was obtained from the Gen-
eral Directorate of Labor, Dutch Ministry of Social Affairs and
Employment The authors wish to acknowledge the critical review
by and discussions with R C Rijkeboer and J Kruithof.
References
Abbass MK, Williams PT, Andrews GE, Bartle KD (1987) The
aging of lubricating oil, the influence of unburnt fuel and par-
159
ticulate SOF contamination Society of Automotive Engineers
paper no 872085
Abbass MK, Andrews GE, Williams PT, Bartle KD, Davies IL,
Tanui LK (1988) Diesel particulate emissions: pyrosynthesis of
PAH from hexadecane Society of Automotive Engineers pa-
per no 880345
Abdelnasser M, Hyland M, Jespersen ND (1986) Synthesis of
mutagenic compounds in crankcase oils Environ Sci Technol
20: 145-149
Amann CA, Stivender DL, Plee SL, MacDonald JS (1980) Some
rudiments of diesel particulate emissions Society of Automo-
tive Engineers paper no 800251
Aoyagi Y, Kamimoto T, Matsui Y, Matsuoka S (1980) A gas sam-
pling study on the formation processes of soot and NO in a DI
diesel engine Society of Automotive Engineers paper no.
800254
Barbella R, Bertoli C, Ciajolo A, D'Anna A (1988) Soot and un-
burnt liquid hydrocarbon emissions from diesel engines Com-
bust Sci Technol 59: 183-198
Behymer TD, Hites RA (1985) Photolysis of polycyclic aromatic
hydrocarbons adsorbed on simulated atmospheric particulates.
Environ Sci Technol 19:1004-1006
Braddock JN, Perry Jr NK (1986) Gaseous and particulate emis-
sions from gasoline and diesel-powered heavy duty trucks.
Society of Automotive Engineers paper no 860617
Bradow RL (1982) Diesel particle and organic emissions: engine
simulation, sampling, and artifacts In: Lewtas J (ed) Carcino-
genic and mutagenic effects of diesel engine exhaust Elsevier
Science, Amsterdam, pp 33-50
Choudhury DR (1982) Characterization of polycyclic ketones and
quinones in diesel emission particulates by gas chromatography/
mass spectrometry Environ Sci Technol 16:102-106
Clark CR, Henderson TR, Rover RE, Brooks AL, McClellan
RO, Marschall WF, Naman TM (1982) Mutagenicity of diesel
exhaust particle extracts: influence of fuel composition in two
diesel engines Fundam Appl Toxicol 2:38-43
CONCAWE (1987) Diesel fuel quality and its relationship with
emissions from diesel engines CONCAWE, The Hague
Cuthbertson RD, Shore PR, Sundstr 6 m L, Heden PO (1987) Di-
rect analysis of diesel particulate-bound hydrocarbons by gas
chromatography with solid sample injection Society of Auto-
motive Engineers paper no 870626
Dolan DF, Kittelson DB, Pui DYH (1980) Diesel exhaust particle
size distribution measurement techniques Society of Automo-
tive Engineers paper no 800187
Draper WM, Phillips J, Zeller HW (1988) Impact of a barium fuel
additive on the mutagenicity and polycyclic aromatic hydrocar-
bon content of diesel exhaust particulate emissions Society of
Automotive Engineers paper no 881651
Du C-J, Kittelson DB, Zweidinger RB (1984) Measurement of
polycyclic aromatic compounds in the cylinder of an operating
diesel engine Society of Automotive Engineers paper no
840364
Egebtck KE, Bertilsson BM ( 1983) Chemical and biological char-
acterization of exhaust emissions from vehicles fueled with
gasoline, alcohol, LPG and diesel National Swedish Protec-
tion Board, report no snv pm 1635
Encyclopedia Brittanica (1984) The new Encyclopadia Brittanica,
15th edn Encyclopedia Brittanica, Chicago
Grimmer G (1979) Sources and occurrence of polycyclic aromatic
hydrocarbons In: Egan H (ed) Environmental carcinogens.
Selected methods of analysis, vol 3 Analysis of polycyclic
aromatic hydrocarbons in environmental samples IARC Publ
no 29 IARC, Lyon, pp 31-125
Grimmer G, Hildebrandt A (1975) Investigations on the carcino-
genic burden by air pollution in man XIII Assessment of
the contribution of passenger cars to air pollution by carcino-
genic polycyclic hydrocarbons Zentralbl Bakt Hyg 161:104-
124
Groblicky PJ, Begeman CR (1979) Particle size variation in diesel
exhaust Society of Automotive Engineers paper no 790421
160
Hampton CH, Pierson WR, Schuetzle D, Harvey TM (1983)
Hydrocarbon gases emitted from vehicles on the road 2 De-
termination of emission rates from diesel and spark-ignition
vehicles Environ Sci Technol 17:699-708
Hare CT (1986) The effects of diesel fuel properties on particulate
emissions In: Lee SD, Schneider T, Grant LD, Verkerk PJ
(eds) Aerosols Lewis, Chelsea, pp 501:514
Hare CT, Baines TM (1979) Characteristics of particulate and
gaseous emissions from two diesel automobiles as functions of
fuel and driving cycle Society of Automotive Engineers paper
no 790424
Hare CT, Springer KJ, Bradow RL (1976) Fuel and additive ef-
fects on diesel particulate development and demonstration of
methodology Society of Automotive Engineers paper no.
760130
Hayano S, Jang-Ho L, Furuya K et al (1985) Formation of hazardous
substances and mutagenicity of PAH produced during the com-
bustion process in a diesel engine Atmos Environ 19:1009-
1015
Henderson TR, Royer RE, Clark CR, Harvey TM, Hunt DF
(1982) MS/MS analysis of diesel emissions and fuels treated
with NO 2 J Appl Toxicol 2:231-237
Henderson TR, Sun JD, Li AP et al (1984) GC/MS and MS/MS
studies of diesel exhaust mutagenicity and emissions from
chemically defined fuels Environ Sci Technol 18:428-434
Herr JD, Dukovich M, Lestz SS, Yergey JA, Risby TH, Tejada
SB ( 1982) The role of nitrogen in the observed direct microbial
mutagenic activity for diesel engine combustion in a single-cy-
linder DI engine Society of Automotive Engineers paper no
820467
Hisamatsu Y, Nishimura T, Tanabe K, Matsushita H (1986) Muta-
genicity of the photochemical reaction products of pyrene with
nitrogen dioxide Mutat Res 172:19-27
Huisingh J, Bradow R, Jungers R et al (1987) Application of bio-
assay to the characterization of diesel particle emissions In:
Waters MD, Nesnow S, Huisingh JL, Sandhu SS, Claxton L
(eds) Application of short-term bioassays in the fractionation
and analysis of complex environmental mixtures Plenum, New
York, pp 382-418
IARC ( 1989) IARC monographs on the evaluation of carcinogenic
risk to humans Diesel and gasoline engine exhausts and some
nitroarenes IARC, Lyon
Iida N, Suzuki Y, Sato GT, Sawada T (1986) Effects of intake oxy-
gen concentration on the characteristics of particulate emis-
sions from a D I diesel engine Society of Automotive En-
gineers paper no 861233
Israel GW, Zierock K-H, Mollerhauer K (1982) Gro13 enverteilung
und chemische Zusammensetzung der Partikelemissionen ver-
schiedener Dieselmotoren VDI-Berichte 429:279-286
Jensen TE, Hites RA (1983) Aromatic diesel emissions as a func-
tion of engine conditions Anal Chem 55: 594-599
Jensen TE, Richert JFO, Cleary AC, La Couse DL, Gorse Jr RA
(1986) 1-Nitropyrene in used diesel engine oil J Am Pollut
Control Assoc 26: 1255-1256
Kamimoto T, Aoyagi Y, Matsui Y, Mastuoka S (1980) The effects
of some engine variables on measured rates of air entrainment
and heat release in a DI engine Society of Automotive En-
gineers paper no 800253
Lach G, Winckler J (1988) Specific problems of sampling and
measuring diesel exhaust emissions Society of Automotive En-
gineer paper no 881763
Lehmann E, Rentel K-H, Allescher W, Homann R (1989) Gefihr-
liche Arbeitsstoffe GA 33: Messung der beruflichen Exposition
gegenfiber Dieselabgas Bundesanstalt ftir Arbeitsschutz, Dort-
mund
Li AP, Royer RE, Brooks AL, McClellan RO ( 1982) Cytotoxicity
of diesel exhaust particle extracts a comparison among five
diesel passenger cars of different manufacturers Toxicology
24:1-8
Lies K-H (1989) Unregulated motor vehicle exhaust gas compo-
nents Volkswagen AG, Wolfsburg
MacDonald JS, Plee SL, D'Arcy JB, Schreck RM (1980) Experi-
mental measurements of the independent effects of dilution
ratio and filter temperature on diesel exhaust particulate sam-
ples Society of Automotive Engineers paper no 800185
Manabe Y, Konouchi T, Wakisaka T, Tahra I, Ohnishi Y (1984)
Mutagenic 1-nitropyrene in wastewater from oil-water separat-
ing tanks of gasoline stations and in used crankcase oil Envi-
ron Mutagen 6:669-681
Mayer WJ, Lechman DC, Hilden DL (1980) The contribution of
engine oil to diesel exhaust particulate emissions, Society of
Automotive Engineers paper no 800256
Meyer JP, Grimmer G (1974) Einflisse PAH-haltiger and PAH-
freier Kraftstoffe auf die Emission von polycyclischen aroma-
tischen Kohlenwasserstoffen eines Kraftfahrzeugs mit Otto-
motor im Europa-test DGMK-Forschungsbericht 4547, II,
Hamburg
Moriske HJ, Riden H (1988) Polare neutrale organische Verbin-
dungen (POCN) im Stadtaerosol Mitteilung: Vergleichende
Untersuchungen von Emissions und Immissionsstauben in
Berlin (West) Zentralbl Bakt Hyg 185:452-468
Murayama T, Miyamoto N, Chikahisa T, Yamane K (1986) Ef-
fects of combustion and injection systems on unburnt HC and
particulate emissions from a DI diesel engine Society of Auto-
motive Engineers paper no 861232
Nelson PF (1989) Combustion-generated polycyclic aromatic hy-
drocarbons in diesel exhaust emissions Fuel 68:283-286
Olah GA, Narang SC, Olah JA (1981) Nitration of naphthalene
and remarks on the mechanism of electrophilic aromatic nitra-
tion Proc Natl Acad Sci USA 78: 3298-3300
Pipho MJ, Ambs JL, Kittelson DB (1986) In-cylinder measure-
ments of particulate formation in an indirect injection diesel
engine Society of Automotive Engineers paper no 860024
Plee SL, MacDonald JS (1980) Some mechanisms affecting the
mass of diesel exhaust particulate collected following a dilu-
tion process Society of Automotive Engineers paper no.
800186
Rosenkranz HS, Mermelstein R (1985) The genotoxicity, metabo-
lism and carcinogenicity of nitrated polycyclic aromatic hydro-
carbons J Environ Sci Health 3:221-272
Ryan TW, Storment JO, Wright BR, Waytulonis R (1981) The ef-
fects of fuel properties and composition on diesel engine ex-
haust emissions a review Society of Automotive Engineers
paper no 810953
Saito T, Tolureo N, Katoh T (1982) Analysis of factors affecting
the formation of major mutagenic substances in diesel particu-
late extracts Society of Automotive Engineers paper no.
821244
Schuetzle D (1983) Sampling of vehicle emissions for chemical
analysis and biological testing Environ Health Perspect 47:
65-80
Schuetzle D, Frazier JA ( 1986) Factors influencing the emission of
vapor and particulate phase components from diesel engines.
In: Ishinishi N, Koizumi A, McClellan RO, St6 ber W (eds)
Carcinogenic and mutagenic effects of diesel engine exhaust.
Elsevier Science, Amsterdam, pp 41-63
Schuetzle D, Skewes LM, Fisher GE, Levine SP, Gorse Jr RA
( 1981) Determination of sulfates in diesel particulates Anal
Chem 53:837-840
Serageldin MA (1981) Soot formation in small flames Society of
Automotive Engineers paper no 811197
Strk G, Stauff J, Miltenburger HG, Stumm-Fischer I (1985)
Photodecomposition of 1-nitropyrene and other direct-acting
mutagens extracted from diesel exhaust particulates Mutat
Res 155:27-33
Steenlage R, Rijkeboer RC (1985) Orinterende metingen van de
emissie van schadelijke en hinderlijke organische verbindingen
afkomstig van gemotoriseerd wegverkeer TNO, Delft, Report
no G 1115
Stenberg U, Alsberg T, Westerholm W (1983) Emission of car-
cinogenic components with automobile exhausts Environ Health
Perspect 47:53-56

Truex TJ, Pierson WR, McKee DE (1980) Sulfate in diesel ex-
haust J Environ Sci Technol 14:1118-1121
Uyehara OA (1980) Effect of burning zone A/F, fuel H/C on soot
formation and thermal efficiency Society of Automotive En-
gineers paper no 800093
Van Beckhoven LC (1991) Effects of fuel properties on diesel en-
gine emissions a review of information available to EEC-
MVEG group Society of Automotive Engineers paper no.
910608
Wade WR ( 1980) Light-duty diesel NO-HC-particulate trade-off
studies Society of Automotive Engineers paper no 800335
Wade WR, Jones CM (1984) Current and future light duty diesel
engines and their fuels Society of Automotive Engineers paper
no 840105
Weidmann K, Menrad H, Reders K, Hutcheson RC (1988) Diesel
fuel quality effects on exhaust emissions Society of Automo-
tive Engineers paper no 881649
Westerholm R, Alsber T, Strandell M et al (1986) Chemical anal-
ysis and biological testing of emissions from a heavy duty diesel
truck with and without two different particulate traps Society
of Automotive Engineers paper no 860014
Williams PT, Bartle KD, Andrews GE (1986 a) The relation be-
tween polycyclic aromatic compounds in diesel fuel and ex-
haust particulates Fuel 65:1150-1158
161
Williams R, Sparacino C, Petersen B, Bumgarner J, Jungers RH,
Lewtas J (1986 b) Comparative characterization of organic
emissions from diesel particles, coke oven mains, roofing tar
vapors and cigarette smoke condensate Int J Environ Anal
Chem 26:27-49
Williams PT, Andrews GE, Bartle KD (1987) The role of lubricat-
ing oil in diesel particulate and particulate PAH emissions.
Society of Automotive Engineers paper no 872084
Williams PT, Abbass MK, Tam LP, Andrews GE, Ng KL, Bartle
KD (1988) A comparison of exhaust pipe, dilution tunnel and
road side diesel particulate SOF and gaseous hydrocarbon
emissions, Society of Automotive Engineers paper no 880351
Yamaki N, Kohno T, Ishiwata S (1986) The state of the art on the
chemical characterization of diesel particulates in Japan In:
Ishinishi N, Koizumi A, McClellan RO, St 6ber W (eds) Car-
cinogenic and mutagenic effects of diesel engine exhaust Else-
vier Science, Amsterdam, pp 17-40
Yoshida E, Nomura H, Sekimoto N (1986) Fuel and engine effects
on diesel exhaust emissions Society of Automotive Engineers
paper no 860619
Zweidinger RB (1982) Emission factors from diesel and gasoline
powered vehicles: correlation with the Ames test In: Lewtas J
(ed) Toxicological effects of emissions from diesel engines.
Elsevier Science, Amsterdam, pp 83-96