States
Environmental Protection
Agency
V **f-
Environmental Sciences Research -^ j x />, C
Laboratory /• j|- ^
Research Triangle Park NC 27711 ''
Research and Development
EPA-600/S3-83-087 Jan. 1984
Project Summary
Characterization of
Emissions and Fuel Economy of
In-Use Diesel Automobiles
Richard E. Gibbs, James D. Hyde, Robert A. Whitby, and Delip R. Choudhury
Exhaust emissions from twenty
1977-1980 in-use light-duty diesel ve-
hicles were measured to determine the
effects of driving cycle, mileage ac-
cumulation and test conditions. Hydro-
carbons, carbon monoxide, carbon
dioxide, nitrogen oxides and particu-
lates were measured from the Federal
Test Procedure (FTP), Highway Fuel
Economy Test (HFET), Congested Free-
way Driving Schedule (CFDS), and
New York City Cycle (NYCC), 50 mph
cruise (50C), and idle. Individual partic-
ulate samples were Soxhlet extracted
with dichloromethane to partition the
particulate into extract (soluble) and
residue (insoluble). The extracts were
tested for mutagenicity by the Ames
Salmonella typhimurium/microsome
method. Detailed chemical analysis and
subsequent bioassay was performed on
selected composite particulate samples.
Emissions (g/mi) and fuel consump-
tion by driving cycle generally increased
in the order 50C< HFET < CFDS < FTP
< NYCC. Vehicles in the General
Motors group generally had higher
emissions than the Mercedes-Benz and
Volkswagen groups and were more
sensitive to driving cycle. The extract
showed very little cycle dependence but
the residue was very cycle dependent.
NOx emissions decreased with mileage
accumulation, while other emissions
increased or were unaffected.
Fuel economy was determined by the
carbon balance method, by fuel meters,
and by fueling records. Over-the-road
fuel economy was always lower than
carbon-balance fuel economy and was
best approximated by the FTP.
A new method for real-time particulate
measurements is described using a
Tapered Element Oscillating Microbal-
ance (TEOM). The TEOM mass agreed
to within 10% of the gravimetric mass
on average with a response time of 8-
15s.
Short studies were performed on the
effects of driving cycle sequence, the
dilution tunnel, sub-FTP temperature
and mutagenic artifact formation.
Bulk extract samples were fraction-
ated and analyzed by GC, GC/MS and
HPLC/UV. The highest specific activity
was in the acidic fraction, but most total
activity was in the neutral fraction
which contained fluorenones and oxy-
PAH's.
This Project Summary was developed
by EPA's Environmental Sciences
Research Laboratory. Research Triangle
Park, NC. to announce key findings of
the research project that is fully docu-
mented in a separate report of the same
title (see Project Report ordering
information at back).
Introduction
The diesel passenger car has only
recently become a significant contributor
to automobile pollution. From 1975 to
1980 diesels increased from 0.05% to
0.64% of the light-duty fleet in New York
State and penetration has been predicted
to be as high as 25% by the year 2000.
This study grew out of a need for
comprehensive emissions data from in-
use diesel automobiles as opposed to
certification data. Particulate emissions
have been of special concern because
they are a visible pollutant with a high
-------
potential for adverse health effects. Few
data were available to assess the effects
of non-FTP driving, vehicle aging and
real-world use/abuse of light-duty diesel
automobiles. The compounds responsible
for the mutagenicity in diesel particulate
extract have not been adequately charac-
terized and the applicability of current
emissions test methods to such areas as
mutagenicity testing required further
study.
Several short experiments were con-
ducted to investigate special data inter-
pretation situations. The topics investi-
gated were the effects of driving cycle
sequence on emissions of the dilution
tunnel on emissions and mutagenicity, of
filtered exhaust gas on particulate and
extract mutagenicity, and of sub-FTP tem-
perature soaking on particulate emissions.
These experiments were conducted with
a limited number of vehicles and,
therefore, the results may not be generally
applicable.
Experimental Approach
Table 1 lists the test vehicles by their
car number designations which are used
in subsequent data presentation. Cars 1
and 5 were purchased new to be used as
loan cars for people whose private
vehicles were being tested. Cars 2 and 3
were purchased new for use by the New
York State Thruway Authority to serve as
high mileage accumulation vehicles. Car
10 was also operated by the Thruway
Authority. Except for Car 21, the other
vehicles were privately owned and
maintained. Vehicles were divided into
groups according to manufacturer as
follows:
GM: Cars 2, 3, 4, 5, 7, and 16
VW: Cars 1, 6, 8, 9, and 11
MB: Cars 12, 13, 14, and 19
Other: Cars 10,15,17, 18, and 21.
(Car 18 was excluded from the GM
group and Car 20 was excluded
from all groups.)
The test protocol employed two or
three replicate driving cycle sequences to
test different fuel/lubricating oil test
conditions. Each sequence of driving
cycle was called a "Phase." The three
phases were:
Phase 1 -vehicle tested as received;
Phase 2 - project control fuel, as re-
ceived oil;
Phase 3 - project control fuel, fresh
oil of manufacturer spec-
ifications.
Table 2 shows the phases and driving
cycle sequences used.
Hydrocarbons, carbon monoxide, carbon
dioxide, nitrogen oxides, andparticulates
were measured from the Federal Test
Procedure (FTP), Highway Fuel Economy
Test (HFET), Congested Freeway Driving
Schedule (CFDS), New York City Cycle
(NYCC), 50 mph cruise (50C), and idle.
Fuel economy was measured by the
carbon balance method for dynamometer
tests and by fuel meters and fueling
records for over-the-road tests.
Particulate samples were collected on
50- x 50-cm Pallflex T60A20 Teflon-
coated fiberglass filters. The particulate
was Soxhlet extracted with dichlorome-
thane for 24 hours. The resultant extract
was dried, weighed and an aliquot tested
in dimethylsulfoxide for mutagenicity by
Table 2. Vehicle Test Driving Cycle
Sequences
B
Table 1. Vehicle Specifications and Dynamometer Test Conditions
Car it
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Year
79
79
79
79
79
80
79
78
79
78
77
77
78
79
79
79
79
80
79
78
78
Make
VW
Olds
Olds
Olds
Olds
VW
Cadillac
VW
VW
Dodge
VW
M-B
M-B
M-B
Audi
Olds
Peugeot
Olds
M-B
Olds
Dodge
Model
Rabbit
Cutlass Cruiser
Cutlass Cruiser
98 Regency
Cutlass Cruiser
Rabbit
Eldorado
Rabbit
Rabbit
D-JO Mitsubishi
Rabbit
240-D
300-CD
240-D
5000
Delta 88
504
Cutlass Cruiser
3OO-SD (Turbo)
Delta 88
Tradesman 200
Engine
Displacement
1-4
V-8
V-8
V-8
V-8
1-4
V-8
1-4
1-4
1-6
1-4
1-4
1-5
1-4
1-5
V-8
1-4
V-8
1-5
V-8
1-6
1.5L
5.7 L
5.7 L
5.7 L
5.7 L
1.5L
5.7 L
1.5 L
1.5L
4.0 L
J.5L
2.4 L
3.0 L
2.4 L
2.0 L
5.7 L
2.3 L
5.7 L
3.0 L
5.7 L
3.3 L
Trans-
mission
M4
A3
A3
A3
A3
MS
A3
M4
M4
A3
M4
M4
A4
M4
MS
A3
M4
A3
A4
A3
A3
Dynamometer
H.P.
7.3
12.5
12.5
12.8
12.5
6.8
10.6
7.3
7.3
14.4
7.3
12.3
13.2
12.6
11.8
13.3
10.7
12.6
13.0
13.3
12.0
I.W.
2250
4000
4000
4500
4000
2250
4500
2250
2250
5500 (+)
2250
3500
4000
3500
3000
4500
3500
4000
4000
4500
4000
Vehicle Tests
1-34
Vehicle Tests
35-80
50C, 30 min*
50C, 30 min.
CFDS
HFET X 3
SOAK, overnight
FTP
CFDS
HFET X 3
IDLE, 30 min
Repeated for
each of 3 fuel/
oil combinations=
Phase 1, 2. 3
50C. 15 min*
HFET X 3
SOAK, overnight
FTP
CFDS
HFET
NYCC
50C. 15 min
IDLE, 15 min
Repeated for
each of 2 fuel/
oil comt>inations=
Phase 1, 3
"Pre-Test conditioning, no data taken.
the Ames Salmonella typhimurium/micro-
some method. Most work was performed
with tester strain TA98 without metabolic
activation. Duplicate plates were run at
extract doses of 0, 10, 20, 30,40, 50, 75,
100, and 200 /ug. Linear correlation
coefficients and their significance levels
were computed for bioactivity and
extract/residue parameters.
Real-time diesel particulate mass
measurements were made for the first
time using a Tapered Element Oscillating
Microbalance (TEOM). The TEOM is a
hollow glass rod, fixed at a wide base,
with a removable filter element attached
to the narrow top, and oscillating in an
electric field (Figure 1). The TEOM has
been shown to behave as a harmonic
oscillator with a frequency dependent
upon the mass collected by the filter
element.
Physical, chemical and bioactive charac-
teristics of the particulate were measured
to assess the effect of the dilution tunnel
on the particulate. The effects of sub-FTP
temperature soaks on FTP emission were
investigated by cold-soaking a vehicle
outdoors during the winter.
Large samples of particulate were
collected to provide large amounts of
extract (1 -27 g) for chemical fractionation
and identification by GC, GC/MS and
HPLC/UV. The Ames assay was used as a
biological monitor to identify important
chemical fractions. The extracts were
first fractionated into acidic, basic, and
neutral compounds by liquid-liquid
partitioning. The neutral fraction (about
90-95% by weight) was further divided
into seven subfractions by silica gel
adsorption chromatography. All subfrac-
tions were bioassayed with tester strains
-------
Side View
Top View
Filter Element
Field Plates
LED
Tapered
Element
Conductive
Path to Fiber
I/////////////////
Sample Flow
TEOM Operation
1. Electric field is set up between Held plates.
2. Image of tapered element is projected on phototransistor.
3. Oscillation of element initiated electrically or mechanically produces an AC voltage output
from phototransistor.
4. AC voltage is amplified and applied to conductive path on element which maintains the
oscillation due to interaction with field set up in Step 1.
5. Frequency of oscillation and hence mass on filter element is determined by frequency counter.
Figure 1. Schematic representation of TEOM instrumentation.
TA98 and TA100 with and without S9
activation.
Results and Discussion
Gaseous and Paniculate
Emissions
Table 3 summarizes the mean values
of paniculate and gaseous emissions and
related parameters for the three groups
of vehicles for Phase 3. Paniculate refers
to the material collected by the EPA
procedure while residue and extract are
respectively the insoluble and soluble
fractions of the particulate after extrac-
tion.
Figure 2 illustrates the effect of driving
cycles on particulate emissions. In
general the other emissions (g/mi)
followed the same general trends of 50C
< HFET < CFDS < FTP < NYCC and VW <
MB < GM. The most prominent aspects
for all emissions were the very large
increase for the NYCC (about double the
FTP) and the large decrease for the IDLE
relative to the driven cycles. The General
Motors group's emissions were greater
and much more sensitive to driving cycle
than those of the Volkswagen and
Mercedes-Benz groups. The cycle varia-
tion of the particulate emission rates
were principally due to variations in the
residue emission rate for all vehicle
groups. Residue emission rates were
very cycle dependent for the General
Motors group but only slightly cycle
Table 3. Fuel Economy and Particulate and Gaseous Emissions Summary All Cycles - Phase 3
-------
o Car 02
* Car 03
+ Car 04
x Car 05
» Car 07
* Car 16
h
a ~J
.s> o
o
o"
90
Figure 3.
15 30 45 60 75
Test Mileage. 1000's
Mileage accumulation effects for
paniculate from the General
Motors group.
Fuel Economy
Pue\ economy was measured by the
carbon balance method and by under-
hood flow totalizing meters. Figure 4
shows the variations of carbon-balance
fuel economy with driving cycle. For each
vehicle group, the fuel economy increases
as the average speed of the cycle
increases. Fuel economy for a given cycle
was always highest for the Volkswagen
group, followed by the Mercedes-Benz
group. Mileage accumulation generally
had no effect on the FTP fuel economy.
Over-the-road fuel economy was most
closely approximated by the FTP economy
even though over-the-road useage was
at a higher average speed. At comparable
average speeds, over-the-road fuel
economy was about 15-20% less than
carbon-balance fuel economy. In general,
the fuel economy for the CFDS very
closely approximated the maximum over-
the-road fuel economy.
Bioassay Characterization
The Ames activity of the extract itself
generally increased in the order GM<
MB < VW for all test cycles and all
methods of expressing activity. Figure 5
shows the Ames activity in terms of
revertants//ug extract as a function of
cycle. The activity generally increased as
the average speed of the cycle decreased
except that the NYCC generally showed
less activity than the other driven cycles.
4
8-
f
o.
• General Motors
A Volkswagen
+ Mercedes-Benz
NYCC FTP CFDS HFET 50 C
Figure 4. Cycle variations of fuel economy,
miles/gallon, by vehicle group.
The FTP or the CFDS usually had the
highest activity. Some large differences
in activity were noted with fuel/lubricat-
ing oil changes and with mileage accum-
ulation, but no definite trends could be
established.
Linear correlations of activity parame-
ters and exhaust parameters were
generally very weak (r values typically
ranging from 0.2 to 0.4), although often
statistically significant. Some parameter
pairs showed a consistency in the sign of
the correlation coefficient which was
independent of vehicle type and driving
cycle. The bioactivity parameter which
most frequently yielded a statistically
significant correlation coefficient was
revertants per /ug extract and it was
usually correlated negatively with extract
and correlated positively with residue.
Real-Time Paniculate
Measurements
The TEOM (Figure 1), as tested, could
respond to dilution tunnel concentrations
as low as 1 to 2 mg/m3 with a response
time on the order of 8 to 15 seconds. Cars
1 and 5 were driven over the FTP Bag 3
and NYCC using both standard 47-mm
filter collection and the TEOM. On
average (29 tests) the TEOM mass was
within 10% of the gravimetric determina-
tions. Figure 6 shows a trace of paniculate
mass emissions rate vs. time and the
corresponding acceleration vs. time
trace for the FTP Bag 3. Periods of high
particulate emissions corresponded to
periods of rapid acceleration while the
particulate emissions decreased very
rapidly during deceleration. Portions of
the curve which seemingly indicate
negative particulate emission rates are
Revertants/UG SOF
o General Motors
* Volkswagen
+ Mercedes-Benz
* Other Cars
o.
s
"
I
NYCC FTP CFDS HFET 50 C IDLE
Figure 5. Cycle variations of specific Ames
activity.
thought to be caused by the desorption of
water held on the particulate and filter.
Chemical Characterization of
Extracts
The acidic fraction had the highest
specific activity, but was a small fraction
of the total mass, and therefore, was not
the main contributor to the mutagenicity
of the extract. The neutral fraction (about
90 to 95% by weight) had a lower specific
activity; but due to its mass was the main
contributor to the mutagenicity of the
extract. The subfraction containing
PAH's showed very little activity. Twenty-
one PAH's were identified by gas chroma-
tography. The highest specific mutagenic
activity was found in the fourth subfrac-
tion which contained about 2-4% of the
mass of the neutral fraction, but con-
tained 42-52% of the direct-acting
mutagenicity of the neutral fraction.
GC/MS and HPLC/UV showed this frac-
tion to contain alkylfluorenones with
benzo(a)fluorenone the major constituent.
The fifth subfraction comprised 4 to 5%
(predominantly oxy-PAH's of the mass
and 13-20% of the activity of the neutral
fraction.
For experiments conducted only with
Car 5 the cycle driven prior to the FTP cold
soak had no effect on gaseous emissions
and little effect on particulate emissions
except when that cycle was an idle. Idling
prior to a driven cycle always increased
particulate, but did not affect gaseous
emissions. Previously driven cycles had
no effect on gaseous emissions. When a
cycle was repeated consecutively during
-------
8
05
I §
1 N
-S o
8 3
o *-
18
<0
Mass Rate tig/sec
Acceleration mph/sec
0. 33. 67. 100. 133. 167. 200. 233. 267. 300. 333. 367. 400. 433. 467. 500.
<
Time (seconds)
Figure 6. TEOM mass rate vs. acceleration for Car 5 and Phase 3 of the FTP.
a day, the fuel economy gradually
increased.
The day-to-day variability of particulate
and gaseous emission measurements
over a period of six or seven test-days
usually did not exceed 0.04 g/mi (standard
deviation). Multiple repetition of a cycle in
a single day had ranges in the order of
0.01 to 0.04 g/mi. The ranges for fuel
economy over a six test-day period were
0.5 to 0.8 mpg for Car 5.
The dilution tunnel collected particulate
of varying physical, chemical, and biologi-
cal character. Proceeding from the
warmest to the coolest section, particulate
matter decreased in bulk density, increased
in soluble organic content and decreased
in Ames specific activity. The specific
activity of the particulate extract from the
tunnel walls was higher than that of most
of the particulate matter collected with
filters.
No effect on specific activity or total
mutagenic activity of the extract was
observed when particulate or its extract
alone were re-exposed to filtered dilute
exhaust. The effect of cold ambient
temperature on FTP emissions was
investigated by cold soaking a vehicle
outdoors during the winter. Comparison
of FTP data for soaks at 0°C as opposed to
20°C showed an increase in particulate
of 18-75%, a decrease in percent extract-
able of 13-36%, little change in the mass
of the extract, and an increase in specific
bioactivity of 230-400%.
Richard E. Gibbs. James D. Hyde, and Robert A. Whitby are with the New York State
Department of Environmental Conservation, Albany, NY 12233; Delip R.
Choudhury is with the New York State Department of Health, Albany, NY 12201.
Peter Gabele is the EPA Project Officer (see below).
The complete report, entitled "Characterization of Emissions and Fuel Economy of
In-Use Diesel Automobiles," (Order No. PB 83-262-071; Cost: $ 17.50, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Sciences Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
-------
United States
Environmental Protection
Agency
Official Business
Penalty for Private Use $300
Center for Environmental Research
Information
Cincinnati OH 45268
111!
HH
111!
ill!
U&OFFjgAyjAIL
m <3'4 ] ^Mll±-li.
^-QHJ>-
PS 0000329
U S ENVJR PROTECTION AGENCY
REGION 5 U8RA*Y
Z30 S DEARBORN STREET
CHICAGO IL 60604
U.S. GOVERNMENT PRINTING OFFICE: 1984-759-102/830
------- |