PB92-153055
Emissions and Fuel Economy of DOE Flex-Fuel Vehicles
(U.S.) Environmental Protection Agency, Research Triangle Park, NC
1992
V
I
U.S. Deoartment of Cmmktcc
wf WlfVI mnnfl «r»	w^r

-------
TECHNICAL REPORT DATA
1. REPORT NO.
EPA/600/A-92/042
2.
3.'
PB92-153055
4. TITLE AND SUBTITLE
EMISSIONS AND FUEL ECONOMY OF DOE FLEX-FUEL VEHICLES
5.REPORT DATE
6.PERFORMING ORGANIZATION C00E
7. AUTHOR(S)
F.M. Black, USEPA
T. Kleindienst, ManTech
8.PERFORMING ORGANIZATION REPORT NO.
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Atmospheric Research and Exposure Assessment Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
10.PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Atmospheric Research and Exposure Assessment Lab - RTP
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
13.TYPE OF REPORT AND PERIOD COVERED
•
Cmnf. ri i—
U. SPONSORING AGENCY COOE J
EPA/600/09
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The U.S. Department of Energy and the U.S. Environmental Protection Agency have
established, through a Memorandum of Understanding, a coordinated framework for
collaborative research examining the impact of alternative motor vehicle fuels on air
quality and risk to public health and welfare. A cooperative effort to examine the
emissions and fuel economy of DOE flex-fuel vehicles, capable of operating on a variety
of liquid fuels, and the atmospheric chemistry of the emissions, will begin in January,
1992. During the first year, emissions will be characterized for 6 vehicles, 2
conventional fuel baseline vehicles and 4 flex-fuel vehicles, using up to 9 fuels.
Additionally, a dual-chamber irradiation facility will be constructed to support future
study of the atmospheric chemistry of the emissions. These studies will examine the
formation of ozone and toxic compounds. A detailed description of the experimental
procedures to be used is provided.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b. IDENTIFIERS/ OPEN ENDED TERMS
c.COSAT I
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC

19. SECURITY CLASS (This Report)
UNCLASSIFIED
21.NO. OF PAGES
15

20. SECURITY CLASS (This Paqe)
UNCLASSIFIED
22. PRICE

-------
EPA/600/A-92/042
Emissions and Fuel Economy of DOE Flex-Fuel Vehicles
Frank Black	Mobile Source Emissions Research Branch
U.S. Environmental Protection Agency
Research Triangle Park, N.C.
Taduesz Kleindienst
ManTech Environmental Technology, Inc.
Research Triangle Park, N.C.

-------
ABSTRACT
The U.S. Department of Energy and the U.S.
Environmental Protection Agency have established,
through a Memorandum of Understanding, a coordi-
nated framework for collaborative research examining
the impact of alternative motor vehicle fuels on air
quality and risk to public health and welfare. A coop-
erative effort to examine the emissions and fuel econ-
omy of DOE flex-fuel vehicles using a variety of
potential fuels, and the atmospheric chemistry of the
emissions, will begin in January, 1992. During the first
year, emissions will be characterized for 6 vehicles, 2
conventional fuel baseline vehicles and 4 flex-fuel vehi-
cles, using up to 9 fuels. Additionally, a dual-chamber
irradiation facility will be constructed to support future
study of the atmospheric chemistry of the emissions.
These studies will examine the formation of ozone and
toxic compounds. A detailed description of the experi-
mental procedures to be used is provided.
THE ALTERNATIVE MOTOR FUELS ACT OF 1 988
requires the Secretary of the Department of Energy
(DOE) to ensure that Federal Government motor vehicle
fleets include the maximum number practical of vehicles
compatible with alternative fuels such as methanol,
ethanol, and compressed natural gas. The Act further
requires the Secretary, in cooperation with the Environ-
mental Protection Agency (EPA) and the National High-
way Traffic Safety Administration (NHTSA), to conduct
a study examining the safety, fuel economy, and emis-
sions of such vehicles. DOE and EPA have established
a Memorandum of Understanding agreeing to a frame-
work for collaborative research examining the charac-
teristics of emissions from motor vehicles using
alternative fuels, and the atmospheric chemistry of the
emissions. This paper describes experimental protocols
for planned 1992 activities.
Initial program efforts will include characterization
of tailpipe and evaporative emissions from 6 motor
vehicles with up to 9 fuels. Laboratory simulations of
roadway driving conditions will be used to produce
samples repressntatjve of automotive evaporative and
tailpipe emissions. Emissions characterization will in-
clude measurerr»ents 0f total hydrocarbon (THC), car-
bon monoxide (CO), carbon dioxide (CO2), nitrogen
oxides (NOx). methanol (MeOH), ethanol (EtOH),
methyltertiarybijty| ether (MTBE), aldehydes (RCHO),
and over 200 in«^jvj,jUal hydrocarbon compounds. Ve-
hicle fuel econoi>,y wj|| also be characterized. Addition-
ally, irradiation chamber facilities suitable for studying
the atmospheric chemistry of the emissions will be
designed, constructed, and characterized.
technical approach - emissions
CHARACTERISATION
Emission ^ests will be conducted at the EPA
Mobile Source EEmjcSjons Research Branch (MSERB)
laboratory located at Research Triangle Park, N.C. In
addition to all Of the equipment required for measuring
regulated emissions (THC, CO, NOx, CO2) from auto-
mobiles, this laboratory is equipped with a temperature
controlled chassis dynamometer enclosure permitting
variation of driving simulation ambient temperature, and
all of the equipment and instrumentation required for
measuring aldehyde, alcohol, ether, and detailed HC
emission rates. Although both exhaust and evaporative
tests will be conducted according to the Federal Test
Procedure (FTP) ^ the fuels and ambient test tempera-
tures may be varied outside the parameters prescribed
for emissions certification (1). *
TEST VEKjicles AND FUELS-Six 1991 motor
vehicles will be studied including a conventional gaso-
line and 2 alcohol f|ex-fuel Ford Taurus automobiles,
and a conventional fUel and 2 alcohol flex-fuel Chevrolet
Lumina automobiles. Nine fuels with specifications
* Numbers in Parentheses indicate references at end
of the paper.
Black
P.ttc 1

-------
provided in Table 1 will be used during the 1992
program. The 4 reformulated gasolines will be provided
from a set of fuels examined in the Auto/Oil Air Quality
Improvement Research Program (2). A total of 40
emissions tests will be completed as indicated in Table
2. Table 3 provides an overview ot the activities of a
typical test week.
Table 1. Program Test Fuels.
1.	Indolene (EPA emissions certification fuel)
2.	CAAA Summer Baseline Unleaded Gasoline
8.7 psi RVP, 87.3 (R + MW2 octane,
339 ppm sulfur, 1.53% benzene,
32.0% aromatic, 9.2% olefin, 58.8%
paraffin
3.	M85 (85% methanol, 15% unleaded gasoline)
4.	E10 (10% ethanol, 90% unleaded gasoline)
5.	ESS (85% ethanol, 15% unleaded gasoline)
6.	Reformulated gasoline 1 (Auto/Oil code C)
8.7	psi RVP, 288 F T90, 15.4% MTBE,
43.8% aromatic, 3.3% olefin, 37.5% paraffin
7.	Reformulated Gasoline 2 (Auto/Oil code J)
8.6	psi RVP, 356 F T90, 14.9% MTBE,
21.4% aromatic, 4.0% olefin, 59.7% paraffin
8.	Reformulated Gasoline 3 (Auto/Oil code N)
8.8	psi RVP, 292 F T90, 13.9% MTBE,
21.4% aromatic, 5.7% olefin, 59.0% paraffin
9.	Reformulated Gasoline 4 (Auto/Oil code M)
8.7	psi RVP, 356 F T90, 14.5% MTBE,
18.0% aromatic, 21.8% olefin, 45.7% paraffin
VEHICLE PREPARATION-^ major goal of this
program is assessment of the impact of varied fuel
formulations on motor vehicle emissions. Each vehicle
will be tested with multiple fuels. To assure that fuel
memory effects are minimized, each vehicle will be
preconditioned with each test fuel prior to emissions
evaluation using the following sequence;
1. Remove the evaporative canister from the vehicle
Table 2. Year 1 Test Matnx.
Vehicles
Fuels
No. of


Tests
Conv. Baseline -1,2
*ndolene. Reform, Gas. 3
4
FFV • 3,4,5,6
Indolene, Summer Bise.,

Reform. Gas, 1,2,3,4
36

M85. E10, £85

Table 3. Typical Test Week.
Day 1 - quality assurance
Day 2 - vehicle conditioning with fuel 1
Day 3 - emission tests with fuel 1
Day 4 • vehicle conditioning with fuel 2
Day 5 - emission tests with fuel 2
Week 2 would involve a simibr sequence
with fuels 3 and 4, and so on.
and purge with 300°F nitrogen at 20 l/min until the
incremental weight loss is less than 1 g in 30 min
(typically takes 3-4 hrs and removes 100 to 120 g of
adsorbed gasoline vapors).
2.	Drain the vehicle fuel tank of the previous test fuel,
add 5 gal of the following test fuel, and complete an
Urban Dynamometer Driving Schedule (UDDS) (initial
1372 sec of the FTP driving schedule to be described
later) (1). Drain and refuel to 40% of capacity with the
test fuel. Return the "purged" canister to the vehicle.
Heat the vehicle fuel tank from 72°F to 120°F using a
2-hr linear temperature ramp. Repeat as necessary
(with refueling between each heat build) until the can-
ister reaches a "break-through" load. "Break-through"
is defined by monitoring the evaporative emission rate
as a function of time, and noting when the slope of
emissions versus time changes abruptly. Figure 1 pro-
vides a typical "break-through" trace.
3.	Drain the vehicle fuel tank and refuel to 40% of
capacity. Complete a UDDS driving sequence followed
Black
Page 2

-------
o
E
a
a
ti
c
o
U
u
c
cs
0
UJ
1
(/)
72 to 120 F 2 hr ramp
canister break-through
7K SO *0 ftC 9Q fO « »C 100 HO 'JO
Time, min.
Figure 1. Evaporative Canister "Break-through" Trace.
by overnight soak in preparation for the FTP emissions
tests described in the following discussion.
EMISSIONS CHARACTERIZATION-Tailpipe and
evaporative emissions will be examined using proce-
dures defined for Federal light-duty motor vehicle emis-
sions certification {1). Figure 2 provides a flow diagram
of the the test sequence. After an overnight soak at
the test temperature, a diurnal (Di) evaporative emis-
sions test is completed, followed by a urban transient
driving tailpipe emissions test, followed by a hot soak
(HS) evaporative emissions test. Figure 3 provides a
schematic of the chassis dynamometer test cell, and a
speed versus time trace for the FTP transient driving
schedule used to simulate urban driving conditions. The
FTP driving schedule includes a cold engine start (after
an overnight soak, see Fig. 2), 21.3 mi/h average speed,
2.4 stops/mi, 19% idle operation, 11.1 mi traveled, and
31.3 min duration (plus 10 min engine off soak period).
The first 505 sec of the FTP driving schedule is com-
monly refered to as test phase 1, the next 867 sec as
test phase 2, and the final 505 sec as test phase 3.
Evaporative Emissions Determination-Motor ve-
hicle evaporative emissions are measured using a
Sealed Housing for Evaporative Determination (SHED).
The vehicle is sealed within the SHED enclosure and the
Di arid HS emissions determined in accordance with
the FTP (1). At the conclusion of each evaporative test,
samples are taken from the SHED into a 60L Tedlar bag
for gas chromatographic (GC) analysis of methanol,
(^START^
FUEL DRAIN AND RU
UN AND RLU	1
[PYNQ PRECONDITIONING}"
1 hour maximum

~1-
COLD SOAK PARKING-
'S minutes maximum
FUELING
~ DRAIN
• 40 % RU
t

DIURNAL HEAT BUILD
*	HEAT FUEL 1 HOUR
*	60 TO B4 DEG F

EVAPORATIVE TEST
NOT REQUIRED
DIURNAL
ENCLOSURE TEST
"12-36 hours
_r\
ICOID START EXHAUST TEST
•0-1 hour
EVAPORATIVE TEST
HYDROCARBON
NOT PERFORMED
RUNNING LOSSES

AS REQUIRED
* *
HOT start exhaust test
HOT SOAK
ENCLOSURE TEST
10 minutes
"7 minutes
Figure 2. Emissions Test Sequence.
ethanol, MTBE, and detailed HCs, as appropriate. Sam-
ples for THC analysis are taken directly from the SHED
to a heated (235±15°F) FID (HFID). Evaporative THC
emissions are reported as non-oxygenated THC by
correcting the THC value for HFID response to methanol
ior other oxygenates). The organic emission rates may
also be reported as Organic Material Hydrocarbon Equiv-
alent (OMHCE) according to equation 1 for methanol
fuels (1):
				14.3594	. . 14.228*
OMHCEevap = HCoimass + 32 042 OHoi mass + HChS mass + ^2 042
14.2284
CHiOHhs mass (1)
Black
Page 3

-------
EVAPORATIVE
SHED
BAG FOflHC
ROH ANALYSIS
HLATfD
HC no
TEMPERATURE
CONTROLLED
TESTCaL
Winn DYNAMOMETER
SAGS • fOH
HC. «0«. ROM
ANALVS6
ANALYZERS
ALDCHYDC
CARTROGES
KLATlD F®
HLATfD TRANSH.R IHC
HEATED TRANSFER UNES
DILUTION
TUNNEL
tcriow ine
ANALYZER
Phase 1
Phase 2
JZ
Q.
E
"O 30
0)
a
a
cr>

BOO
Time, sec
rrp
10 min Soak
Phase 3
	ft!
Figure 3. Test Cell Configuration and Driving Cycle.
where:
HC Di mass and HC HS mass = Di and HS emissions
hydrocarbon mass in grams, respectively, and
CH3OH Di mass and CH3OH HS mass = Di and HS
emissions methanol mass in grams, respectively;
which assume that the evaporative Di emissions hydro-
gen to carbon ratio is 2.33 and HS emissions hydrogen
to carbon ratio is 2.2 (conforming with conventional
gasoline standards). Similar calculations can be com-
pleted for ethanol fuels using appropriate coefficients
for ethanol.
Di and HS evaporative emissions tests are con-
ducted in conjunction with the FTP, as shown in Figure
2. Following an overnight soak in the Temperature
Controlled Test Chamber (TCTC) at the test tempera-
ture, the vehicle is pushed into the SHED for the Di test.
During the Di test, tank fuel temperature is elevated
using a 24°F/hr ramp, e.g., 40 to 64°F for a 40°F test,
60 to 84°F for a 70°F test, and 72 to 96°F for a 90°F
test. The initial 1 992 program matrix (see Table 2) will
examine Di evaporative emissions from 60 to 84°F,
with tailpipe emissions examined at 70°F (as in Federal
emissions certification). Following the Di test, the vehi-
cle is pushed back into the TCTC and allowed to
equilibrate at the test temperature. After temperature
equilibiium is reached, the UDDS is run, followed im-
mediately by the HS evaporative test. The SHED is not
Black
Page 4

-------
equipped to be operated at reduced ambient tempera-
tures; therefore, a subambient (e.g. 40°F) evaporative
emission test would be conducted at laboratory temper-
atures near 70°F. This should have little effect on Di
tests which are conducted with engine and fuel cold.
HS tests, which begin with engine and fuel warm,
should be somewhat affected in that the engine and
fuel will not cool as rapidly as they would in a cooler
environment. For a high temperature evaporative emis-
sions test (e.g. 90°F), an appropriate Di temperature
ramp (e.g. 72°F to 96°F) would be used and the SHED
temperature for HS maintained at the elevated test
temperature {e.g. 90°F). The Di and HS evaporative
emission rates are combined according to equation 2 to
permit comparisons with tailpipe exhaust rates.
non-oxygenated HC by using a procedure to correct the
THC value for FID response to alcohols (5). Tailpipe
organic emissions rates may also be reported as
OMHCE wherein total organic carbon mass is calculated
according to equation 3 for methanol fuels which as-
sumes that the hydrogen to carbon ratio of all tailpipe
organic emissions is 1.85 (permits conformity with
conventional gasoline emission standards) (1). Similar
calculations can be completed for ethanol fuels using
appropriate coefficients for ethanol and acetaldehyde.
The FTP driving schedule includes three test
phases: a cold start transient phase (505 sec.), a
stabilized phase (867 sec.), and a hot start transient
phase (505 sec.). There is a 10 minute engine-off soak
period between phases two and three. Emissions from
Evap. Emissions, a r
(3.05'"P^dayX-hot soak emissions, a/trip) + diurnal emissions, s^ay
31.1 ™¥day
(2)
OMHCEt3;ip;pe =» HCmass +	(CHzOHmass) +	(HCHOmass) (3)
Exhaust Emissions Determinations -- Vehicie ex-
haust emission tests will be conducted using an electric
chassis dynamometer {Horiba Instruments, Inc.) to sim-
ulate vehicle road load. The dynamometer rolls are
enclosed within a TCTC permitting vehicle soak and
operation at temperatures from 20°F to 1 1 0°F as
illustrated in Figure 3.
Exhaust gases are sampled using a constant
volume sampling (CVS) technique commonly used for
vehicle emissions certification tests (1). A heated
transfer tube is used within the TCTC to direct vehicle
raw exhaust to a "dilution tunnel" where the exhaust is
cooled and diluted prior to sampling for analysis. Ex-
haust gas is thoroughly mixed in the dilution tunnel with
70°F dilution air. Total flow through the dilution tunnel
is held constant (e.g. 750 CFM) . Aliquotes of the
diluted exhaust are collected directly from the dilution
tunnel at a constant flow rate over the duration of the
test, permitting determination of pollutant mass emis-
sion rates (g/mi) from sample concentration, total di-
luted exhaust volume, and distance traveled. The
sampling system has previously been qualified for quan-
titative transfer of methanol, formaldehyde, and other
compounds of interest at varied ambient temperatures
from the motor vehicle tailpipe to the analytical instru-
mentation (3,4).
Regulated emissions (THC, CO, NOx, CO2) are
sampled and analyzed using standard Federal certifica-
tion procedures (1). THC emissions are reported as
each phase are analyzed separately and then combined
to calculate a "weighted" emission rate according to
equation 4:
,/	r,	+	Yht+Ys , AS
Ywm = 0.43 -^— + 0.57 n (4)
Dq(+Ds
Dfjt+Ds
where:
Ywm = weighted mass emissions of each pollutant, i.e.
HC, CO, NOx, CO2, MeOH, etc. in grams per vehicle
mile,
Vet = mass emissions calculated from the transient
phase of the cold start test, in grams per test phase,
Ys = mass emissions calculated from the stabilized
phase of the cold start test, in grams per test phase,
Yht = mass emissions calculated from the transient
phase of the hot start test, in grams per test phase,
Dct = the measured driving distance during the tran-
sient phase of the cold start test, in miles,
Ds = the measured driving distance during the stabi-
lized phase of the cold start tust, in miles, and
Black
Page 5

-------
Dht = the measured driving distance during the tran-
sient phase of the hot start test, in miles.
Fuel Economy-Fuel economy is evaluated both in
terms of miles traveled per gallon of fuel consumed
(mi/gat) and in terms of BTUs of energy consumed per
mile traveled (BTU/nni). Mi/gal fuel economy is deter-
mined using classical carbon balance equations with
appropriate carbon weight fraction for the varied fuels.
Equations are presented in Figure 4. Energy based fuel
economy is ralculated using appropriate energy densi-
ties for the van.?d fuels. Equations are presented in
Figure 5.
Because of the lower energy densities of the
alcoho' fuels, reduced mi/gal fuel economies are ex-
pected, depending on the fraction of the fuel that is
alcohol. However, improved energy effiencies are pos-
sible with alcohol fuels, off-setting somewhat, the
reduced fuel energy densities. Table 4 provides exam-
ples of fuel economies observed with FFVs using gas-
oline and M85 {05% methanol, 15% gasoline) fuels.
Analytical Chemistrv--As previously discussed,
THC, CO, NOx, and CO2 are sampled and analyzed
using standard Federal emissions certification proce-
dures (1). Samples for exhaust detailed HC measure-
ments are collected by pumping a constant aliquot of
the diluted exhaust from the CVS into Tedlar bags for
subsequent analysis by gas chromatography. Gas chro-
matographs equipped with flame ionization detectors
(FIDs) are used for the detailed HC analysis (6). Each
instrument uses three analytical columns -two packed
columns that resolve C1 and C2 HCs, and a capillary
column that resolves C3-C 12 HCs. The method pro-
vides quantitation of over 200 HC compounds. Figure
6 provides a shematic of the chromatographic system
for detailed HC analysis.
Aldehyde compounds, sampled from the dilution
tunnel through a heated (235±15°F) sample line (at 1
LPMl, are collected on dinitrophenyl- hydrazine (DNPH)-
coated silica gel cartridges. Individual aldehydes,
which are collected on the cartridge as DNPH aldehyde
derivatives, are subsequently analyzed by high perfor-
mance liquid chromatography. This sampling technique
and analytical method permits quantitative determina-
tion of 1 5 individual aldehydes (7). Figure 7 provides a
schematic of the chromatographic system for aldehyde
analysis.
Alcohols are sampled using water impingers and
analyzed using a previously described GC method (8).
Ethers are sampled into Tedlar bags, similar to the
detailed HC practice, for subsequent GC analysis using
previously described GC procedures (9,10). Figures 8

g Carbon / gal fuel

n.tYgal = 	

0 Carbon in exhaust / mi

K1 (fuel g 1 flail
mi/gal =
K1 (a OM / mi! + K2 (g CO / mi) + K3 Ig C02 t ml!
K1
= Fuel carbon weiQht fraction
= 0.866 (gasoline), 0.375 (MeOH). 0.449 IM05)
K2
= CO carbon weight fraction = 0.429
K3
= C02 carbon weight fraction = 0.273
Figure 4. Carbon-Balance Fuel Economy.
8.34 * heating value. BTU/lb * fuel density. g/ml
BTU/mi = 				
fuel economy, mi/gal
where:

Heating


Value.
Density,

BTU/lb
g/ml
MeOH
8.600
0.79
Gasoline
18.700
0.74
M85
10,115
0.78
Figure 5. Energy-Based Fuel Economy.
Table 4. Example FFV Fuel Economies.
; Fuel
mi/gal
BTU/mi J
1
j MO
i
21.8
5,305.0 j
1
! MS5
I
13.3
i
4,996.8
j
and 9 provide schematics of the chromatographic sys-
tems used for alcohol and ether analyses, respectively.
Black
Page- 5

-------
T <5>
Gauge
Ak
Solenoid 1
Solenoid 2

Solenoid 3
¦ Vacuum
Helium
Carrier
/Q_)
\5cm3y
Aio^/T
Pcropak
Catumn
je
Sample
In
Sample
) Pump
FO
Air
Hydrogen
Column
Helium
Gamer
Silica
Column
Air
Hydrogen
Figure 6. Ci to C12 Hydrocarbon Chromatography.
Quadrex 007 Silicone
Fused Silica Column
FID
Air
Hydrogen
Back Pressure
	, / Regulator
Split
Injector
Split
Vent
| Flow
Helium	Controller
Carrier
Figure 8. Alcohol Chromatography.
Injection
Valve
Waste
¦ Sample Loop
— Guard Column
rumps
Water Acctonitrile
Reservoir Reservoir
OO,
Autosamplor
Zorbax 00S
Analytical Column
360 nm
UV
Detector
Waste
Liquid
Nitrogen
Cooling
Hydrogen
Nutech
Temperature
Controller
Toggle
Sample
Valve
Toggle
Valve
Hydrogen
Flush
Vacuum
Gauge ¦>.
Vacuum -*	O-
Pump Toggle
Valve
Vacuum
Reservoir
Togo1®
Valve
~'10 1\ 002
9* ® 2 Cryogenic
	 g £ 3 p Focusing
Hydrogen [ /
I 7 ( (p Jjl FlD
Flow
Air
Hydrogen
i / (	MU
15 % TCLP \
Firebrick *
Column	P'E 25 m
Methyl Silicone
Cofumn
Figure 7. Aldehyde Chromatography.
QUALITY ASSURANCE--The Quality
Assurance/Quality Control (OA/QC) Plan and Proce-
dures include Organization and Responsibility, assigning
QC responsibilities to program staff. Objectives for
Figure 9. Ether Chromatography.
Measurements and Performance establishing accuracy
and precision goals for all program measurement sys-
tems, Outputs providing both the outputs which are
necessary to assure that equipment is properly main-
B.ack
Pace 7

-------
tained, and outputs needed to assess and monitor QC,
Statistical Methods describing how outputs, such as
accuracy and precision, are to be calculated. Reports
identifing all QC reporting requirements and milestones,
and Audits describing system audit procedures,
cal program at Research Triangle Park, N.C. is sup-
ported by an onsite contract with ManTech
Environmental Technology, Inc. QC/QA organization
for this program provides overall project QA responsi-
bility to the EPA Project Officer. All QC reports or
outputs related to measurements performed by Man-
Tech personnel are the responsibility of the ManTech
Technical Supervisor.
mance-The quality assurance objectives for accuracy
and precision are presented in Table 5. If at any time
it is noted that deviations in measured values exceed
the objectives, testing is stopped, equipment is exam-
ined, and testing is resumed after the prob" 3m has been
corrected.
Table 5. Quality Assurance Objectives.
PARAMETER
ACCURACY,
%
PRECISION, %
THC Analysis
10
2
CO Analysis
10
2
N0X Analysis
10
2
CO2 Analysis
10
2
Alcohol Analy-
sis
10
5
Ether Analysis
10
5
Aldehyde Anal-
ysis
10
5
Detailed HC
Analysis
10
5
Dyne Speed
5
5
Dyno Torque
5
R
Reid Vapor
Pressure
10
5
PDP Counter
10
5
SHED Volume
2
2
SHED Leak
Rate
10
5
Gravimetric
Balance
5
1
Gravimetric
Weights
1
1
SHED Temper-
ature
5
5
Dyno Cell Tem-
perature
5
5
Veh. Coolant
Temperature
5
5
Fuel Tempera-
ture
5
5
Catalyst Tem-
perature
10
10
CVS Tempera-
ture
5
5
Dyno Cell Pres-
sure
5
2
CVS Pressure
5
2
QC Procedures and Outputs-The QC outputs
required for this program are given in Tables 6 and 7,
All outputs should be completed within the specified
periods for the duration of the program. Outputs given
in Table 6 are "nondeliverable" which means that the
QC work, when completed, is signed off in the QC
Notebook with no other report required. The Project
Officer reviews all QC Notebooks monthly insuring that
all equipment is being properly maintained and quality
controlled. Outputs given in Table 7 are reported
directly to the Project Officer since these indicate the
status of compliance with the data specifications stated
in the previous section.
Table 6. QC Nondeliverable Outputs.
OUTPUTS 1
TIME PERIOD
Calibrate THC An-
alyzer (86.121-82,90)
Daily
Adjust THC FID for op-
timum HC response
(86.121-82!
Annually
THC Analyzer linearity
checks (86.121-82,90)
Monthly
THC Analyzer MeOH
response (86.1 21 -90)
Monthly
Black
Page 8

-------
Calibrate CO Analyzer
(86.122-78)
Daily
CO Analyzer H2O inter-
ference check (86.122-
78)
Annually
CO Analyzer linearity
check (86.122-78)
Monthly
Calibrate NOx An-
alyzer (86.123-78)
Daily
NOx Analyzer con-
verter efficiency check
(86.123-78)
Weekly
NOx Analyzer linearity
check (86.123-78)
Monthly
Calibrate CO2 An-
alyzer (86.124-78)
Daily
CO2 Analyzer linearity
check
Monthly
Calibrate CVS (86,119-
78,90)
Semi-annually
Calibrate temperature
transducers (ASTM
E220-80)
Monthly
Calibrate pressure
transducers (CVS Pro-
tocol)
Monthly
Calibrate dry test me-
ters (86.120-82)
Monthly
Verify currency of
NBS cylinder certifi-
cates
Monthly
Calibrate dyno speed
signal (EPA 650/4-75-
024d, TP 202)
Monthly
Calibrate dyno load
cell (86.118-78 &
manuf. recommenda-
tions)
Monthly
Calibrate weights
I ASTM E617-81)
Quarterly
Calibrate RVP (80.Ap-
pendix D, ASTM D323-
89)
Monthly
Calibrate GCs (GC
RPM)
Daily
Calibrate HPLC (Alde-
hyde RPM)
Daily
Calibrate SHED
(86.117-78,90)
Annually
Characterize SHED
leak rate (86.117-
78,90)
Monthly
Perform dyno preventa-
tive maintenance
As Scheduled
Perform SHED preven-
tative maintenance
As Scheduled
Oxygenate methods
cross-checks
Monthly
Detailed HC methods
cross-checks
Monthly
Calibrate Ether an-
alyzer (80-Appendix F)
Daily
1 numbers in parenthesis are Federal Register refer-
ences unless otherwise indicated
Table 7. QC Deliverable Outputs.
OUTPUTS 1
TIME PERIODS
THC Analysis
Monthly
CO Analysis
Monthly
NOx Analysis
Monthly
CO2 Analysis
Monthly
Methanol Analysis
Monthly
Aldehyde Analysis
Monthly
Detailed HC Analysis
Monthly
Ether Analysis
Monthly
Dyno Speed
Monthly
Dyno Torque
Monthly
SHED Temperature
Monthly
Cell Temperature
Monthly
Coolant Temperature
Monthly
Fuel Temperature
Monthly
Catalyst Temperature
Monthly
CVS Temperature
Monthly
SHED Pressure
Monthly
Cell Pressure
Monthly
Black
Page 9

-------
CVS Pressure
Monthly
SHED Volume (and in-
tegrity)
Monthly
Balance
Monthly
Reid Vapor Pressure
Monthly
PDP Counter
Monthly
HC Blind Audit Results
Quarterly
CO Blind Audit Results
Quarterly
NOx Blind Audit Re-
sults
Quarterly
CO2 Blind Audit Re-
sults
Quarterly
Gravimetric Weights
Annually
QC Notebook
Monthly
1 measures of precision and accuracy unless otherwise
indicated
Accuracy and precision for most of the parame-
ters listed in Table 5 are determined using "blind"
samples which have been referenced to the National
Institute of Standards and Technology (NIST). Because
NIST standards are not available for detailed HC and
aldehyde measurements, research protocol methods
(RPMs) are used to assure the accuracy and precision
of these measurements.
Any parameter which fails to achieve its specified
accuracy or precision goal shall be corrected before
testing can proceed. Data whose integrity has been
compromised due to the malfunctioning of any instru-
ment or system during its collection shall be discarded
and the tests rerun. For this reason, the Project Officer
reviews all data being generated daily and ceases
testing when trends reveal the likelihood of some com-
ponent malfunction or other system irregularity.
QC Statistical Methods-All accuracy determina-
tions (except for SHED Volume) are made by comparing
the mean measured value from three separate measure-
ments of the reference material with the actual value
of the reference material according to equation 5:
determining accuracy and precision. All reference ma-
terials are NIST, directly referenced to NIST, or prepared
in accordance with an accepted Standard Operating
Procedure (SOP) when NIST standards are not available.
Measures of accuracy and precision involve pollutant
concentrations, temperatures, and pressures near the
median values experienced in the empirical program.
Accuracy of the SHED volume is assured in accordance
with standard practice (1).
Precision is calculated by taking the standard
deviation (SD) of ten measurements for emissions an-
alyzers and three for all other devices, and dividing it
by the mean value (MV) according to equation 6:
SD
% Precision = —— x 100 (6)
MV
Reports-Routine "deliverable" outputs (Table 7)
are reported to the Project Officer in accordance with
the QC procedures and outputs section; and other
outputs entered :n a QC Notebook which is presented
monthly to the Project Officer as a "deliverable" output.
The Project Officer makes a QC Evaluation Report
quart >y and within the Final Report. The Evaluation
Report is a brief summary of QA/OC within the project
and is meant to highlight problem areas, their resolution
or nonresolution, and recommended action to be taken
in the event of unresolved issues.
Audits—An annual systems audit of the project is
conducted by the MSERB QA Officer and/or the Branch
Chief. The systems audit focuses on the project's
adherence to required procedures. For example, instru-
ment logs or notebooks are checked to see if the
equipment is being properly maintained and calibrated,
procedures for determining accuracy and precision are
discussed with personnei who actually perform these
measures, and QA equipment, such as calibration gases
and meters, are examined for documented certification.
Instrument operators are questioned about the daily or
routine procedures they follow when running a test.
Particular attention is directed to insuring that SOPs and
RPMs are being followed. As deficiencies are noted,
the person responsible is instructed to insure that
immediate corrective action is taken.
MV-RV
% Accuracy = ——— x 100 (5)
n v
where MV - RV = absolute magnitude of mean value
minus reference value.
The reference material is presented to the analyzer
operator as an "unknown"; span gases are not used for
TECHNICAL APPROACH - ATMOSPHERIC
CHEMISTRY
Comprehensive evaluation of the impact of alter-
native fuels on risk to public health and welfare must
include examination of the atmospheric chemistry of
the emissions. The formation of photochemical oxi-
Black
Page 1 0

-------
dants in urban environments including ozone, results
from chain reactions involving hydrocarbons and oxides
of nitrogen (NOx) in the presence of sunlight. These
reactions produce many organic compounds, including
aldehydes and ketones, peroxyacetyl nitrate (PAN),
organic nitrates and peroxides, and others. Some of
the products have different genotoxic (i.e., mutagenic
or possibly carcinogenic) properties than the reactant
compounds, that is, the original emissions.
ManTech Environmental Technology has devel-
oped experimental protocols permitting the direct study
of the atmospheric chemistry of motor vehicle emis-
sions, including the formation of ozone, and toxic and
mutagenic compounds (11). The protocols involve
irradiation of mixtures of motor vehicle tailpipe exhaust
(generated by prescribed driving cycles) and surrogates
of evaporative emissions or background urban air or-
ganic compound mixtures. The surrogate addition is
necessary to achieve HC/NOx ratios typical of urban air
mixtures. Automobile emissions typically have HC/NOx
ratios of about I to 3, whereas urban atmospheres have
ratios in the 7 to 1 2 range. The formation of ozone and
other toxic chemicals is sensitive to this ratio. At the
lower ratios (i.e., 1 - 3), atmospheric chemistry pro-
ceeds extremely slowly.
In these studies, continuous measurements are
made of the major inorganic chemical species present
in the chamber, which include NO, NOx, O3, and CO.
The total hydrocarbon signal is also measured continu-
ously. Hydrocarbons in the mixture are speciated by
gas chromatography using two columns (DB1 and Car-
bowax) in series. This GC is also capable of measuring
numerous reaction products following irradiation includ-
ing organic nitrates and nitro compounds. PAN and
other peroxyacyl nitrates are measured using a dedi-
cated GC having a packed carbowax column and elec-
tron capture detection. Carbonyl compounds are
sampled by impinger collection through DNPH derivatiz-
ing agent and quantified by HPLC. Nitric acid can also
be formed through photochemical reactions and is
measured by collection on nylon filters and analysis by
ion chromatography. These measurements are made
before irradiation and periodically during the progress
of the photochemical reaction.
During 1992, ManTech will construct and char-
acterize an irradiation chamber facility interfaced with
a motor vehicle similar to that illustrated in Figure 10.
In the experiments for this study, the chamber design
will be portable permitting both indoor irradiations with
UV-A and UV-B blacklights, and outdoor irradiations
with actual sunlight. The design uses two identical
8,000 L chambers permitting contrasts between the
reaction products of the motor vehicle exhaust mixtures
and reference mixtures. The initially conceived design
allows the chamber to be operated with or without
dilution, depending on the objectives of individual ex-
periments. The chamber will be characterized by irra-
diating mixtures of single hydrocarbons and NOx ,
which have well-studied profiles of reactant disappear-
ance and product formation.
Following construction and characterization of
the chamber facility, an extensive three-phase testing
program will be initiated examining the atmospheric
chemistry of emissions from the previously described
vehicles and fuels. In most experiments, direct com-
parisons will be made of the oxidant and/or toxic
compound(s) formation, by comparison of the irradiated
reference and the test mixtures. The major two phases
include: oxidant formation studies and detailed chemi-
cal characterization of the photooxidation products. In
an optional third phase, characterization of the mutage-
nicity of the reactants and products can be made using
Ames bioassays by procedures already developed.
REFERENCES
1.	Code of Federal Regulations, Title 40, Part 86,
Control of Air Pollution from New Motor Vehicle and
New Motor Vehicle Engines: Certification and Test
Procedures, Office of the Federal Register National
Archives and Records Administration, Washington,
D.C., July, 1990.
2.	Burns, V.R., Bensen, J.D., "Description of
Auto/Oil Air Quality Improvement Research Program,"
SAE 912320, International Fuels and Lubs Meeting,
Toronto, Canada, October, 1991.
3.	Baugh, J., Ray, W., Black, F., Snow, R., "Motor
Vehicle Emissions Under Reduced Temperature Idle
Conditions", Atmos. Env., Vol. 21 (10), pp 2077-2082
(1987).
4.	Sigoby, J.E., McArver, A., Snow, R., "Evalua-
tion of a FTIR Mobile Source Measurement System,"
EPA/600/S3-89/036, U.S. Environmental Protection
Agency, Research Triangle Park, NC, August, 1989.
5.	Gabele, P.A., Baugh, J.O., Black, F.M., Snow,
R., "Characterization of Emissions Using Methanol-Gas-
oline Blended Fuels," JAPCA, Vol. 35, pp. 1168-1175
(1985).
6.	Sigsby, J.E., Duncan, J., Crews, W., Burton,
C., "Research Protocol Method for Analysis of Detailed
Hydrocarbons Emitted from Automobiles by Gas Chro-
Black
Page 11

-------
Ambient air
Exhaust
Compressor Air
3.2 cm Teflon Tubing
Dilution Tunnel
20 m /min
Turbine
Dynamometer
Clean Air
Generator
Flow mete:
Nitrogen
Panicle Filter
Inlet
Manifold
150 L
Dewar
273 K
Evaporative
Surrogate
60-65
L/min
Filter I i I
Effluent
Exposure
Chamber
Clean Air
Exposure
Chamber
1 40 L
Reactants
Exposure
Chamber
140 L
Reaction Chamber
Carryover
Exposure
Chamber
140 L
140 L
8,000 L
Samples
~ 14 Umin
y 14 L/min
Teflon
14 L/min
Effluent
Exposure
Chamber
Reaction Chamber
8,000 L
Carryover
Exposure
Chamber
140 L
Samples
140 L
14 L/min Y
Figure 10. Schematic of Test Apparatus for Examining the Photochemical Oxidant and Toxic Compound
Products from Irradiated Automobile Exhaust Mixtures.
Frank Black
Page 12

-------
matography," U.S. Environmental Protection Agency,
Research Triangle Park, NC, June 1989.
7.	Tejada, S.B., "Evaluation of Silica Gel Car-
tridges Coated In Situ with Acidified 2,4-
Dinitrophenylhydrazine for Sampling Aldehydes and
Ketones in Air", Intern. J. Environmental Anal. Chem.,
Vol. 26, pp 167-185 (1986).
8.	U.S.EPA, "Calculation of Emissions and Fuel
Economy When Using Alternative Fuels - Appendix A-5:
Measurement of Methanol," EPA 460/3-83-009, Office
of Mobi'e Sources, Ann Arbor, Ml, March, 1983
9.	Federal Register, "Test Method for Determina-
tion of C1 to C4 Alcohols and MTBE in Gasoline by Gas
Chromatography," Vol. 54(54), pp 11904-11 91 U,
March 22, 1989.
10.	Duncan, J.W., Burton, C.D., Crews, W.S., "A
Method for Measurement of Methanol, Ethanol, and
Methyltertiarybutyl Ether Emissions from Motor Vehi-
cles," Northrop Services, Inc., Research Triangle Park,
NC, 1988.
11.	Kleindienst, T.E., Smith, D.F., Hudgens, E.E.,
Snow, R.F., Perry, E., Bufalini, J.J., Claxton, L.D.,
Black, F.M., Cupitt, L.T., "The photooxidation of au-
tomotbile emissions: measurements of the transforma-
tion products and their mutagenic activity," In Review,
Atmos. Environ., (1991).
The information presented in this document has in part
been funded by the United States Environmental Protection
Agency. It has been subjected to the Agency's peer and
adminisrtative review, and has been approved for publication.
Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.

-------