-------
Table of Contents
Page
Number
I. Summary 1
II. Introduction 2
III. Description of Catalytic Converter Technology 3
IV. Description of Test Vehicles 4
V. Test Facilities and Analytical Methods 6
VI. Test Procedures 6
VII. Discussion of Test Results 7
A. M100 Volkswagen Rabbit Vehicle 7
B. M100 Toyota Corolla Vehicle 16
VIII. Evaluation Highlights 24
IX. Future Efforts 25
X. Acknowledgments . 25
XI. References 26
APPENDIX A - EMITEC EHC System Specifications A-l
-------
I.
A fresh electrically heated catalytic converter (EHC) was
furnished by EMITEC to the U.S. Environmental Protection Agency
(EPA) for evaluation on two methanol-fueled vehicles. The EHC
consisted of a compact resistively heated quick light-off catalyst
followed by a larger conventional three-way main converter. The
main catalyst was similar in volume to catalysts found on most
late-model U.S. automobiles.
The EMITEC EHC was evaluated on two neat methanol-fueled
vehicles, a 1981 Volkswagen Rabbit and a 1988 Toyota Corolla.
Emission testing was conducted over the Federal Test Procedure
(FTP) CVS-75 test cycle. The emissions of primary interest were
cold start methanol (unburned fuel), CO and formaldehyde.
The EHC was evaluated in several modes. First, the EHC was
emission tested without resistive heating or air assist. Different
heating sequences with/without heat applied prior to starting the
engine were then evaluated. Catalyst air assist with/without
resistive heating was evaluated. Initially, resistive heating and
air assist were limited to the cold start segment (Bag 1) of the
FTP. However, as emission levels of unburned fuel over the Bag 1
portion with the Rabbit vehicle were reduced below Bag 3 levels,
resistive heating and air assist were also used during the initial
portion of Bag 3 on some tests.
Resistively heating the EHC without air assist provided mixed
results on the VW vehicle. Bag l unburned fuel emissions were
reduced 55 percent from levels obtained with the unassisted
catalyst when a 15/40-second heat sequence (resistive heating 15
seconds prior to and 40 seconds following start) was used. Bag 1 CO
emissions, however, increased above unassisted catalyst levels.
When secondary air injection was utilized with 15/40 heating,
Bag l unburned fuel emissions were reduced to 0.20 grams, a 93
percent reduction from heated catalyst only levels. CO emissions
over Bag 1 were 1.2 grams with this configuration, a 90 percent
reduction from unassisted catalyst levels.
Formaldehyde emissions were also significantly reduced when
the catalyst was assisted in Bag 1 by resistive heating and air
addition. Bag 1 formaldehyde levels with the unassisted catalyst
were approximately 96 milligrams. With catalyst resistive heating
and air assist, Bag 1 formaldehyde levels were reduced to 17
milligrams, an 83 percent reduction.
Bag 1 unburned fuel emissions were lower than those from Bag
3 when catalyst heat ing/ air addition were used. Tests were then
conducted with catalyst heating/air assist provided during Bag 3 as
well as during Bag 1. This additional catalyst assist reduced Bag
3 methanol, CO, and formaldehyde emissions and lowered composite
FTP emission levels of unburned fuel and formaldehyde to 0.03
grams/mile and 1 milligram/mile respectively.
-------
-2-
The EJMITEC EHC was then installed underfloor on a lean-burn
M100 Toyota Corolla vehicle and the test sequence was repeated.
With catalyst resistive heating only, Bag 1 unburned fuel emissions
were reduced 70 percent from unassisted catalyst levels. Resistive
heating also significantly lowered Bag 1 formaldehyde levels. (With
the 15/40 heat sequence, Bag 1 formaldehyde was measured at 69
milligrams, approximately 12 percent less than the 245 milligrams
obtained with the unassisted catalyst.) Resistive heating without
catalyst air assist did not significantly lower Bag 1 CO levels.
Catalyst efficiency improved when secondary air injection was
used together with 15/40 heating. Unburned fuel emissions were
reduced to 0.71 grams over Bag 1, almost 89 percent below
unassisted catalyst levels and 96 percent below engine-out levels.
Bag 1 CO was also reduced to 5.3 grams, approximately 65 percent
lower than the 14.9 grams measured with the unassisted catalyst.
Formaldehyde emissions over Bag 1 were reduced to 28 milligrams
with this catalyst configuration, 97 percent below baseline levels.
Some testing was also conducted on the Corolla with heat/air
assist supplied to the catalyst during the Bag 3 portion of the FTP-
as well as Bag 1. FTP composite emissions of unburned fuel here
were measured at 0.05 grams/mile, CO at 0.3 grams/mile, and
formaldehyde at 3 milligrams/mile. FTP NOx emissions remained
relatively constant at 0.4 grams/mile.
II. Introduction
Cold start accounts for the most significant portion of
unburned fuel, carbon monoxide, and formaldehyde emissions over the
Federal Test Procedure (FTP) from catalyst-equipped, methanol-
fueled vehicles.[1,2] Recent enactment of Federal and California
clean air legislation has focused attention on reducing these cold
start emissions.[3,4]
One strategy to reduce cold start emissions uses an
electrically heated catalyst (EHC) to shorten the time to catalyst
light-off (the time in which the converter becomes catalytically
active). Excess emissions of unburned fuel and CO at cold start
occur because the engine and catalytic converter have not warmed to
relatively steady state conditions, and a period of fuel enrichment
is necessary to ensure good starting and driveability. EPA has
evaluated several electrically heated metal monolith catalytic
converters to reduce Bag 1 emissions of unburned fuel, CO, and
formaldehyde from methanol-fueled vehicles.[1,2,5,6,7,8] These
evaluations involved the use of low mileage catalyst substrates
with various volumes and active catalyst loadings.
-------
-3-
EPA conducts and publishes the results from emission control
technology" evaluations to enhance interest in new technologies by
automakers and industry hardware suppliers. Interest in
resistively heated catalyst technologies for mobile sources has
continued to grow, and several industry sources have provided EPA
with samples of their catalysts for evaluation. One of these
catalyst suppliers, EMITEC, recently furnished EPA with an
electrically heated catalyst for evaluation on two methanol-fueled
vehicles. A preliminary evaluation of this converter was
conducted, and results of this evaluation are presented in this
report.
III. Description of Catalytic Converter Technology
Figure 1 below is a cut-away diagram of the EMITEC heated
catalyst system. The compact electrically heated substrate has a
cell density of 200 cpsi, with a catalyst loading of 60 g/ft3 of
5:1 platinum:rhodium. Downstream of the EHC is the main converter,
which consists of two separate metal foil bricks. The first brick
(Catalyst No. 1) is an oxidation catalyst, and the second brick
(Catalyst No. 2) functions as a 3-way catalyst. A more detailed
description of the EMITEC EHC system is provided in Appendix A.
Figure 1
EMITEC Electrically Heated Catalyst System
Citalytt No. 1
Catalyti No. 2
Electrically Hiatrt Catalyst
Figure 2 below is a photograph of the EMITEC electrically
heated catalyst system. The small light-off converter on the left
is resistively heated; the larger main converter is on the right.
The overall length of the entire unit (EHC plus main catalyst) was
565 millimeters. This system is described in more detail by the
manufacturer in an earlier paper. [9]
-------
Figure 2
Photograph of EMITEC EHC System
^e>!
^ ; • "; " 'tiifcis, ' " "" 1=' /- *£
EPi+P1 S
.^Vra
,
1 ?\' 'T .li'hv r i'*'!"*'-"""- lilTn o-h
M ,, ,,,:.,,,*',, A;> fc^p. M ,^ „
*
A 12-volt DC automotive battery was used to supply the energy
for resistively heating the EHC. Resistance across the can
electrical connection posts was measured at 1.2 ohms. The
application of 12 volts from the fully charged battery caused a
current of approximately 300 amps at start in the circuit which
decreased to about 260 amps after 50 seconds of resistive heating.
The other circuit components consisted of #06 gauge copper cables
six feet in length and #00 gauge copper cable 9 feet in length.
IV. Description of Test Vehicles
The first test vehicle (Figure 3) was a 1981 Volkswagen Rabbit
sedan, equipped with automatic transmission, air conditioning, and
radial tires. The 1.6-liter engine had a rated maximum power
output of 88 horsepower at 5,600 rpm on neat methane1 fuel. The
vehicle was tested at an equivalent test weight of 2,500 Ibs. and
an actual dynamometer horsepower of 7.7. This vehicle was loaned
to the U.S. EPA by Volkswagen of America; a detailed description of
the vehicle and its modifications were provided in an earlier
report.[5]
The second test vehicle was a 1988 4-door Toyota Corolla
equipped with Toyota's second-generation methanol lean-burn system.
The 1.6-liter engine is equipped with 4-valves/cylinder, compact
combustion chamber technology, swirl control valve system, lean-
mixture sensor, sequential fuel injection, and exhaust gas
recirculation for NOx control. A detailed description was provided
in previous papers.[2,10,11] The Corolla was tested at an
equivalent test weight of 2,750 Ibs. and an actual dynamometer
horsepower of 8.9. This vehicle was loaned to the EPA by Toyota
Motor Corporation; a picture of the car is given in Figure 4.
-------
-5-
Figure 3
M100 Volkswagen Rabbit Test Vehicle
Figure 4
M100 Toyota Corolla Test Vehicle
-------
-6-
V. Test Facilities and Analytical Techniques
Emissions testing at EPA was conducted on a Clayton model ECE-
50 double-roll chassis dynamometer, using a direct-drive variable
inertia flywheel unit and road load power control unit. A Philco
Ford constant volume sampler with a nominal capacity of 600 cfm was
used. Exhaust hydrocarbon emissions were measured with a Beckman
Model 400 flame ionization detector (FID). CO was measured using
a Bendix Model 8501-5CA infrared CO analyzer. NOx emissions were
determined by a Beckman Model 951A chemiluminescent NOx analyzer.
Exhaust formaldehyde was measured using a dinitrophenol-
hydrazine (DNPH) technique.[12,13] Exhaust carbonyls including
formaldehyde are reacted with DNPH solution forming hydrazone
derivatives. These derivatives are separated from the DNPH
solution by high performance liquid chromatography (HPLC), and
quantization is accomplished by spectrophotometric analysis of the
LC effluent stream.
The procedure developed for methanol sampling and presently in
use employs water-filled impingers through which are pumped a
sample of the dilute or evaporative emissions. The methanol in the
sample gas dissolves in water. After the sampling period is
complete, the solution in the impingers is analyzed using gas
chromatographic (GC) analysis.[14]
VI. Test Procedures
The goal of this test program was an initial evaluation of an
EMITEC resistively heated catalyst system on two methanol-fueled
vehicles. The test procedures used here were similar to those used
in two recent evaluations of resistively heated catalysts conducted
by EPA.[7,8] These procedures were used because the EMITEC EHC has
a total volume similar to the Kemira Oy and Camet EHCs evaluated
previously, and all three EHC's had platinum:rhodium three-way
catalyst formulations.
The testing was conducted in several phases, each succeeding
phase using an additional catalyst assist in an attempt to further
lower emission levels. Cold start emissions of particular interest
were unburned fuel (methanol), CO, and formaldehyde.
Several baseline emission tests were first conducted.
Baseline for the Rabbit refers to emission tests over the FTP
driving cycle with a straight pipe inserted in place of the
underfloor catalyst. A dummy substrate was inserted in place of
the underfloor emissions catalyst with the Corolla vehicle. After
these tests, the straight pipe/dummy catalyst was removed and the
EMITEC catalyst system installed in its place. Several tests were
then conducted over the FTP without catalyst heating or secondary
air injection. This unassisted testing provided a reference for
determining the improvement in catalyst conversion efficiency
provided by catalyst heating and/or air assist.
-------
-7-
The next phase of testing involved catalyst resistive heating
without secondary air injection. Two heat sequences, 0/40 and
15/40 were evaluated on the Volkswagen vehicle. 15/40 refers to a
15-second period of heating prior to starting the engine followed
by 40 seconds of resistive heating after cold start. The heated
catalyst testing on the Corolla was limited to the 15/40 sequence.
Secondary air assist was then added in front of the EHC to
assist the oxidation reactions. The air was added from a shop air
line immediately following engine start in Bag 1. The air assist
was limited to 100 seconds following cold start; it was assumed
that limiting air addition would minimize any undesired effect on
the ability of the three-way catalyst to convert NOx emissions.
A gas rotameter was inserted in the shop air line to the
vehicle to measure excess air flowrate to the catalyst. This meter
also provided an indication of the effect of exhaust backpressure
on air flowrate. A bypass valve in the air line controlled
secondary air flow to an average 5.0 ft3/minute over the period of
air addition.
The EHC was tested on each vehicle with a 15/40 heating
sequence and catalyst air addition for 100 seconds following Bag 1
cold start. Initially, resistive heating and air assist were
limited to the Bag 1 portion of the FTP. Unburned fuel emission
levels over Bag 1 with catalyst heat/air assist were lower than Bag
3 levels in some tests with the Rabbit vehicle. Therefore, the
last phase of testing utilized the same Bag 1 resistive heating
(15/40) and air assist (100 seconds after start) sequences as well
as 5/30 heating and 30 seconds of air assist at the start of Bag 3.
This configuration was expected to result in the lowest FTP
composite emission levels of unburned fuel, CO, and formaldehyde
for both vehicles.
VII. Discussion of Test Results
A. M100 Volkswagen Rabbit Vehicle
This section provides the results of emission testing
utilizing the methanol-fueled Volkswagen Rabbit vehicle. The
following section describes similar emissions testing on the lean-
burn methanol-fueled Toyota Corolla vehicle.
The emissions of primary interest here are unburned fuel, CO
and formaldehyde emissions related to the cold start and warm-up
period. Unburned fuel emissions are a result of the large
quantities of fuel inducted during the cold enrichment period, poor
vaporization and mixing, and cold cylinder walls and intake
manifold runners. CO and formaldehyde emissions are related to
incomplete combustion, and their formation is also enhanced by the
fuel enrichment period.
-------
-8-
All test results presented below were obtained over the FTP
driving cycle. Bag l emissions are presented in grams over the
test segment, except for formaldehyde, which is presented in
milligrams. Composite FTP emission levels are presented in
grams/mile, except for formaldehyde, which are presented in
milligrams/mile.
Figure 5 below presents formaldehyde emissions obtained over
the cold start portion of the FTP with the M100 Volkswagen vehicle.
Resistive heating and air assist were limited to Bag 1. Following
baseline testing, the vehicle was emissions tested with the
catalyst in the unheated mode and without air assist. The vehicle
was then tested without resistive heating but with 100 seconds of
catalyst air assist provided immediately following cold start in
Bag 1. The EHC system was then tested using two different
resistive heating sequences. No air assist was provided to the
catalyst during this testing. The final catalyst configuration
evaluated combined Bag l resistive heating with air assist.
Figure 5
EMITEC EHC, M100 VW Vehicle
Formaldehyde Emissions, Bag 1
EHC Configuration
No Heat/No Air
Air Assist Only
0/40 Heat Only
15/40 Heat Only
15/40 Heat/100 Air
50 100 150
Formaldehyde (milligrams)
200
The two resistive heating sequences evaluated here involved
catalyst heating during the cold start portion of the FTP. Heating
was limited to Bag 1 because of the importance of cold start
emissions to weighted FTP levels. The numerator in the heat scheme
fraction represents the number of seconds of catalyst preheating
prior to cold start; the denominator represents the number of
seconds of catalyst heating following start. The last catalyst
configuration combined the 15/40 resistive heating sequence with
100 seconds of air assist after start in Bag 1.
-------
-9-
The unassisted catalyst (no heat/no air mode) reduced Bag 1
formaldehyde levels to 96 milligrams, a reduction from baseline of
over 95 percent. Bag 1 formaldehyde levels increased above
unassisted catalyst levels when 100 seconds of secondary air
addition was used. Preheating the catalyst for 15 seconds had
little effect on Bag 1 formaldehyde when compared to the unassisted
catalyst. However, when 40 seconds of catalyst resistive heating
were used after starting the engine, Bag l formaldehyde levels
decreased to 64 milligrams. This represents a 33 percent reduction
from levels obtained with the unassisted catalyst.
A significant reduction in formaldehyde was noted when both
resistive heating and air assist were provided to the catalyst.
Bag 1 formaldehyde was measured at 17 milligrams when the 15/40
resistive heating sequence and 100 seconds of secondary air assist
were used. This represented an 83 percent reduction from
unassisted catalyst levels and approached the Bag 3 emissions
level.
Figure 6 presents the results from exhaust methanol sampling
over the same test sequence in Figure 5. The fresh unassisted.
catalyst was very effective in converting unburned fuel. The
methanol level measured without catalyst assist, 6.73 grams,
represents a 58 percent reduction from the baseline level of 16.09
grams (not shown).
Figure 6
EMITEC EHC, M100 VW Vehicle
Methanol Emissions, Bag 1
EHC Configuration
No Heat/No Air
Air Assist Only
0/40 Heat Only
15/40 Heat Only
15/40 Heat/100 Air! 0.2
6.73
5.2:
3.66
3.04:
246
Methanol (grams)
8
-------
-10-
Secondary air injection during the initial portion of Bag i
also reduced unburned fuel emissions below unassisted catalyst
levels. Bag 1 methanol emissions were reduced to 5.20 grams when
air was added before the catalyst for 100 seconds after key-on.
Methanol emissions over Bag 1 decreased to 3.66 grams,
approximately 46 percent from the unassisted catalyst level, when
resistive heating was applied for 40 seconds in the absence of air
assist. Preheating the catalyst for 15 seconds without air assist
reduced Bag 1 unburned fuel levels further from no preheat levels,
to 3.04 grams.
When resistive heating and air assist were combined, Bag 1
methanol emissions were reduced to very low levels. In this
catalyst configuration, Bag 1 exhaust methanol was measured at 0.20
grams, a 97 percent reduction from levels obtained with the
unassisted catalyst. Bag 1 unburned fuel measured here was lower
than the exhaust methanol level measured over the Bag 3 segment
(0.3 grams), when the engine was warm at start.
Figure 7 below presents CO levels measured during the same
testing described above. Without resistive heat or air assist to
the catalyst, Bag 1 CO was measured at 11.4 grams, a 67 percent
reduction from the baseline level of 34.3 grams. When catalyst air
assist was used exclusively, Bag 1 CO levels decreased slightly, to
10.1 grams.
Figure 7
EMITEC EHC, M100 VW Vehicle
Carbon Monoxide Emissions, Bag 1
EHC Configuration
No Heat/No Air
Air Only
0/40 Heat Only
15/40 Heat Only
15/40 Heat/100 Air
11.4
10.1
12.4
13.6
4 8 12
Carbon Monoxide (grams)
-------
-11-
Resistive heating without air assist caused higher Bag l CO
emissions.' During testing with resistive heating only, Bag 1 CO
emissions rose above levels obtained with the unassisted catalyst.
When a 15-second preheat period was used, Bag 1 CO rose above
levels measured without preheat. However, when resistive heating
and air assist were used together, Bag 1 CO decreased to very low
levels. The level of Bag 1 CO measured then was 1.2 grams, a 90
percent reduction from the unassisted catalyst level. Bag 1 CO
emissions approached levels measured over the Bag 3 segment with
this configuration.
A summary of emission levels measured over Bag 1 from this
testing is provided in Table 1. Organic material hydrocarbon
equivalents (OMHCE), a method of expressing the combined effects of
conventional hydrocarbons, unburned fuel and aldehyde emissions,
are also calculated and presented here. Baseline refers to engine-
out emission levels.
Table l
EMITEC EHC Evaluation
Baa 1 Emission Levels. M100 VW Vehicle
**
Catalyst
Configuration
Baseline
No heat/
no air
Air only*
0/40 heat
only
15/40 heat
only
15/40 heat
and air
NOx
g
7.2
4.3
4.7
3.8
4.3
4.6
CO
g
34.3
11.4
10.1
12.4
13.6
1.2
CHjOH
g
16.09
6.73
5.20
3.66
3.04
0.20
HCHO
mg
2021
96
150
64
94
17
OMHCE
g
9.06
3.29
2.31
1.85
1.55
0.15
NMHC
g
1.10
0.27
**
0.16
0.14
0.03
Air assist for 100 seconds after key-on.
Less than 0.005 grams measured.
Bag 1 OMHCE emission levels generally follow trends similar to
those noted with unburned fuel, the most significant component of
OMHCE. Without catalyst resistive heating/air assist, Bag 1 NOx
decreased from baseline (7.2 grams) to 4.3 grams. When secondary
air injection was used with/without resistive heating, Bag l NOx
increased slightly above levels from no heat/no air assist testing.
The lowest Bag 1 NOx levels, however, were recorded when 0/40
catalyst heating was utilized without air assist.
-------
-12-
Table- 2 below presents composite FTP emissions from this
testing.
Table 2
EMITEC EHC Evaluation
FTP Test Results. M100 VW Vehicle
Catalyst
Conf igur at ion
Baseline
No heat/no air
Air only*
0/40 heat only
15/40 heat
only
15/40 heat
and air
NOX
g/mi
1.6
1.0
1.0
0.8
1.0
1.0
CO
g/mi
6.4
0.7
0.7
0.8
0.9
0.2
CH3OH
g/mi
2.64
0.40
0.33
0.23
0.22
0.04
HCHO
mg/mi
429
7
11
5
7
2
OMHCE
g/mi
1.70
0.21
0.16
0.13
0.12
0.03
NMHC
g/mi
0.35
0.02
**
0.01
0.01
0.01
**
Air assist for 100 seconds after key-on.
Less than 0.005 grams/mile measured.
Generally, the changes in Bag 1 emissions of methanol, CO,
and formaldehyde are reflected in overall FTP emission levels. Air
assist alone resulted in a reduction of exhaust methanol from 0.40
grams/mile to 0.33 grams/mile. Composite FTP formaldehyde levels
increased to 11 milligrams/mile, however, when catalyst resistive
heating was not used. FTP levels of CO were unaffected by the use
of Bag 1 air assist only.
Mixed results occurred when catalyst resistive heating was
used in the absence of air assist. A substantial reduction in
unburned fuel emissions was noted with this catalyst configuration,
from 0.4 grams/mile to 0.22 grams/mile. However, a slight increase
in overall FTP CO also occurred, from 0.7 grams/mile to 0.9
grams/mile. Aldehyde levels were unchanged by catalyst heat
assist.
Very low FTP composite emission levels were obtained when both
15/40 heat and secondary air assist were combined. FTP emissions
of unburned methanol, CO, and formaldehyde were measured at 0.04
grams/mile, 0.2 grams/mile, and 2 milligrams/mile respectively.
These levels represent over a 70 percent reduction in formaldehyde
and CO, and a 90 percent reduction in methanol emissions from
levels obtained with the unassisted catalyst. OMHCE emissions over
the FTP were reduced to 0.03 grams/mile, below California ULEV
standards.
-------
-13-
The slight changes in Bag 1 NOx emissions were not reflected
in composite FTP levels. FTP NOx levels remained constant at 1.0
grams/mile, except when 0/40 heating was used in the absence of air
assist. FTP composite emissions of OMHCE, HC, and NMHC, however,
followed trends similar to those noted with unburned fuel. For
example, the 90 percent reduction in unburned fuel emissions
obtained when both heat/air assist were provided to the catalyst
contributed to an 86 percent reduction in OMHCE.
The combined use of catalyst resistive heating and air assist
during cold start in Bag 1 caused unburned fuel emissions over Bag
3 to exceed Bag 1 levels on some tests. Additional tests were
conducted with catalyst resistive heating/air assist at the start
of both Bag 1 and 3 segments of the FTP in an attempt to further
reduce composite emission levels. A 15/40-second heat sequence
with 100 seconds of air assist was used at the beginning of Bag 1,
and 5/30 heating with 30 seconds of air assist was used at the
beginning of Bag 3. Table 3 below presents Bag 1 and 3 emission
levels from this testing.
Table 3
EMITEC EHC Evaluation
Baa 1/3 Emission Levels. M100 VW Vehicle
Catalyst
Configuration
NOx
g
CO
g
CH3OH
g
HCHO
mg
OMHCE
g
NMHC
g
Baseline:
Bag l
Bag 3
7.2
7.9
34.3
19.0
16.09
8.86
2021
1239
9.06
5.54
1.10
1.10
No Heat /No Air:
Bag 1
Bag 3
4.3
4.4
11.4
0.6
6.73
0.15
96
4
3.29
0.12
0.27
0.03
Bag 1 Heat and Air Assist:
Bag 1
Bag 3
4.6
4.5
1.2
0.9
0.20
0.30
17
6
0.15
0.25
0.03
0.10
Bag 1 and 3 Heat and Air Assist:
Bag 1
Bag 3
4.5
4.4
1.1
0.7
0.22
0.05
17
3
0.18
0.08
0.04
0.03
Gasoline-fueled vehicle measurement with a propane-calibrated FID.
Less than 0.005 grams measured.
-------
-14-
Bag 3. CO and formaldehyde emissions were reduced slightly by
the combined use of catalyst heating/air assist; Bag 3 methanol
emissions were substantially reduced. The large decrease in
methanol emissions caused Bag 3 OMHCE emissions to be reduced to
0.08 grams, a decrease of 66 percent from the unassisted catalyst
testing conducted immediately prior to this testing.
Table 4 below presents an emissions summary from this testing.
Table 4
EMITEC EHC Evaluation
FTP Composite Emissions. M100 VW Vehicle
Catalyst
Conf igurat ion
Baseline
No heat/
no air
Bag 1 heat
and air assist
Bag 1 and 3
heat and air
assist
HC*
g/mi
1.22
0.16
0.03
0.02
NOX
g/mi
1.6
1.0
1.0
0.9
CO
g/mi
6.4
0.7
0.2
0.2
CH3OH
g/mi
2.64
0.40
0.04
0.03
HCHO
mg/mi
429
7
2
1
OMHCE
g/mi
1.70
0.21
0.03
0.02
NMHC
g/mi
0.35
0.02
0.01
**
Gasoline-fueled vehicle measurement with a propane-calibrated FID.
Less than 0.005 grams/mile measured.
With catalyst resistive heating/air assist used during Bag
3, FTP composite emissions of methanol and formaldehyde were
reduced to even lower levels, 0.03 grams/mile and 1 milligram/mile
respectively. FTP composite CO emissions remained unchanged from
levels obtained with catalyst heat/air assist used only during Bag
1. NOx levels over Bag 3 did not appear to increase when heat and
air assist were provided during the Bag 3 segment. FTP composite
NOx emission levels were relatively unchanged by the catalyst
assist provided to both Bag segments. Calculated OMHCE decreased
very slightly from already low levels with Bag 3 assist.
Modal testing (emissions versus time) over the FTP cycle was
also performed as part of the catalyst evaluation. Emissions of
total HC and CO were measured over the FTP for three catalyst
configurations. The first configuration was engine-out emissions;
the second used the EHC system without heat/air assist (No Assist) .
The final configuration utilized a 15/40 heat sequence with 100
seconds of secondary air injection (Bag 1 Heat/Air Assist).
-------
-15-
Figure 8 below presents cumulative CO emissions over the first
160 seconds of the FTP for the catalyst configurations described
above. Cumulative and instantaneous emissions can be related; as
the cumulative emissions trends in the following plots approach
horizontal, instantaneous emissions approach zero. Vehicle speed
is also presented.
Figure 8
EMITEC EHC, First 160 Seconds Of FTP Cycle
Cumulative CO Emissions, M100 VW Vehicle
35
Speed (mph)
Cumulative CO (grams)
20
40
60 80 100 120 140 160
Time (seconds)
Cumulative CO levels are approximately equal for both
baseline and no catalyst assist configurations over the first 30
seconds of the FTP cycle. This condition means that the catalyst
has not yet achieved light-off temperature. Cumulative CO levels
with the unassisted catalyst begin to stabilize after this period
of catalyst inactivity. The cumulative CO trace for the unassisted
catalyst becomes relatively horizontal (no instantaneous CO
produced) approximately 90 seconds after start. This interval
denotes light-off of the unassisted catalyst.
Cumulative CO emissions begin to diverge from unassisted
catalyst levels approximately 8 seconds after start in the FTP when
catalyst heat/air assist were used. After approximately 30 seconds
of vehicle operation, the cumulative CO curve for the assisted
configuration approaches a horizontal position, denoting near
complete instantaneous catalyst conversion efficiency of CO.
Figure 9 below presents emissions measured as total HCs for
the three catalyst configurations in Figure 8. HC levels are
presented because modal analysis of methanol was not possible.
-------
-16-
Figure 9
EMITEC EHC, First 160 Seconds Of FTP Cycle
Cumulative HC Emissions, M100 VW Vehicle
Speed (mph)
Cumulative HC (grams)
35
30
25
20
15
10
5
0
Speed
20 40 60 80 100
Time (seconds)
120 140
160
Baseline HC emissions were notably- higher than unassisted
catalyst levels beginning almost immediately after start.
Cumulative HC levels begin to stabilize after approximately 70
seconds with the unassisted catalyst. When catalyst heat/air
assist were provided, cumulative HC emissions diverged from
unassisted catalyst levels approximately 25 seconds after start and
stabilized shortly thereafter.
B. Toyota Corolla Test Results
Following completion of the testing with the Volkswagen
vehicle, a similar test program was begun with an MIOO-fueled
Toyota Corolla. This vehicle was equipped with a second-generation
Toyota "Methanol Lean Combustion System". [10] Exhaust gas
recirculation was also used to control NOx emissions.
Resistive heating/air assist during testing on the Mioo
Corolla were limited to Bag 1 only. Engine-out emissions for this
vehicle were obtained with the use of a dummy catalyst in place of
underfloor catalyst, rather than a straight pipe. The EHC system
was then placed on the vehicle underfloor, and emissions tested
without resistive heating or air assist. The vehicle was then
emissions tested without catalyst resistive heating but with 100
seconds of secondary air assist after start in Bag 1. 15/40
catalyst heating without secondary air assist was the next
configuration tested. The final configuration combined the 15/40
heat sequence in Bag 1 with 100 seconds of secondary air assist.
Figure 10 presents Bag 1 aldehyde levels from this testing.
-------
-17-
Figure 10
EMITEC EHC, M100 Toyota Corolla
Formaldehyde Emissions, Bag 1
EHC Configuration
Baseline
No Heat/No Air
Air Assist Only
15/40 Heat Only
15/40 Heat/100 Airl 28
200 400 600 800
Formaldehyde (milligrams)
1,000
The unassisted catalyst reduced Bag 1 formaldehyde 72 percent
from baseline; formaldehyde increased, however, when catalyst air
assist was used. Bag 1 formaldehyde decreased to 69 milligrams, (72
percent below unassisted catalyst levels) with catalyst heating (no
air assist). Combined resistive heating/air assist reduced
formaldehyde to 28 milligrams, 97 percent lower than baseline.
Figure 11 presents Bag 1 emissions of unburned fuel.
Figure 11
EMITEC EHC, M100 Toyota Corolla
Methanol Emissions, Bag 1
EHC Configuration
Baseline
No Heat/No Air
Air Assist Only
15/40 Heat Only
19.87
6.2
7.37
1.88
15/40 Heat/100 Air! 0.71
10 15 20
Methanol (grams)
25
-------
-18-
The unassisted catalyst reduced Bag 1 methanol emissions
approximately 69 percent from baseline levels, from 19.87 grams to
6.20 grams. When air assist was used without resistive heating, Bag
1 methanol emissions increased above unassisted catalyst levels, to
7.37 grams. A significant reduction in unburned fuel emissions was
obtained, however, when a 15/40 heat sequence was used; Bag 1
methanol emissions were reduced almost 70 percent from unassisted
catalyst levels, from 6.20 grams to 1.88 grams. When both heat and
air assist were applied to the catalyst during cold start, Bag 1
methanol was reduced to 0.71 grams, an 89 percent reduction from
unassisted catalyst levels.
Figure 12 presents Bag 1 CO emissions from this testing.
Figure 12
EMITEC EHC, M100 Toyota Corolla
Carbon Monoxide Emissions, Bag 1
EHC Configuration
Baseline
No Heat/No Air
Air Assist Only
15/40 Heat Only
15/40 Heat/100 Air
iiiiiifiirtiiiiiiiiiiiiiiiiiiii,
30.4
14.9 :
17.2
14.1
7 14 21 28
Carbon Monoxide (grams)
35
Bag 1 engine-out CO averaged 30.4 grams; this was reduced to
14.9 grams (51 percent reduction) when the EHC was used unassisted.
Bag 1 CO increased to 17.2 grams when catalyst air assist (no
resistive heating) was supplied. Resistive heating without air
assist did not further reduce Bag 1 CO levels. When catalyst
resistive heating was combined with air assist, Bag l CO was
reduced to 5.3 grams, an 83 percent reduction from baseline.
A summary of Bag 1 emissions testing is given below.
-------
-19-
Table 5
EMITEC EHC Evaluation
Baa 1 Emission Levels. M100 Tovota Corolla
Catalyst
Conf igurat ion
Baseline
No heat/no air
Air only**
15/40 heat only
15/40 heat/air
HC*
g
7.44
2.37
2.64
0.77
0.31
NOX
g
2.8
1.6
1.8
1.7
1.8
CO
g
30.4
14.9
17.2
14.1
5.3
CH3OH
g
19.87
6.20
7.37
1.88
0.71
HCHO
mg
880
245
282
69
28
OMHCE
g
10.00
3.15
3.57
1.01
0.40
NMHC
g
0.94
0.30
0.20
0.11
0.05
**
Gasoline-fueled vehicle measurement procedure with a propane-calibrated
FID.
Air assist for 100 seconds after key-on.
Bag 1 NOx was little affected by catalyst heat/air assist; the
largest increase in Bag 1 NOx above unassisted catalyst levels
noted here was 0.2 grams. Bag 1 OMHCE, NMHC, and methanol
emissions were reduced in similar proportions with catalyst assist.
(e.g. with catalyst resistive heating/air assist, Bag 1 OMHCE and
methanol were reduced 87 and 88 percent respectively).
Table 6 presents composite FTP emissions from this testing.
Table 6
EMITEC EHC Evaluation
FTP Composite Emission Levels. M100 Tovota Corolla
Catalyst
Configuration
Baseline
No heat/no air
100 sec air only
15/40 heat only
15/40 heat
and air
HC*
g/mi
1.63
0.15
0.20
0.05
0.03
NOX
g/mi
0.6
0.3
0.4
0.3
0.4
CO
g/mi
5.6
1.1
1.4
1.0
0.5
CH3OH
g/mi
4.26
0.38
0.49
0.12
0.06
HCHO
mg/mi
249
18
20
6
3
OMHCE
g/mi
2.17
0.20
0.24
0.07
0.04
NMHC
g/mi
0.24
0.02
0.02
**
**
* Gasoline-fueled vehicle measurement procedure with a propane-calibrated
FID.
** Less than 0.005 g/mi measured.
-------
-20-
The reductions in Bag 1 methanol, CO, and formaldehyde
emissions 'affected FTP composite emission levels. The heated
catalyst was particularly effective at reducing formaldehyde
emission levels. Formaldehyde was reduced to 3 milligrams/mile
over the FTP when both heat and air assist were provided to the
catalyst. This value represents an 83 percent reduction from
unassisted catalyst levels and approximately a 99 percent reduction
from baseline levels.
The resistively heated/air-assisted catalyst was very
effective for control of Bag 1 methanol emissions occurring before
catalyst light-off. The unassisted catalyst reduced methanol
emissions over the FTP to 0.38 grams/mile, a 91 percent reduction
from baseline. When heat and air assist were supplied to the
catalyst, these emissions were further reduced to 0.06 grams/mile,
a 99 percent reduction from baseline.
CO emissions over the FTP were also significantly reduced
through the use of the EHC. CO emission levels over the FTP were
reduced from 5.6 grams/mile (baseline) to 1.1 grams/mile
(unassisted catalyst), an 80 percent reduction. When both heat and
air assist were used during Bag 1, FTP composite CO emissions were
reduced to 0.5 grams/mile, a 91 percent reduction from baseline.
Catalyst heating/air assist were provided at the start of the
Bag 3 portion of the FTP in addition to Bag 1 in an attempt to
further reduce FTP emissions of unburned fuel, CO, and
formaldehyde. Bag 1 assist consisted of 15/40 resistive heating
with 100 seconds of air assist; 5/30 heating was used at the
beginning of Bag 3 in conjunction with 30 seconds of air assist
after key-on. Table 7 below presents Bag 1 and Bag 3 emission
levels from this testing. The four catalyst configurations
presented in Table 7 are baseline (dummy catalyst), unassisted
catalyst, Bag 1 heat and air assist only (15/40 heat, 100 seconds
air), and Bags 1/3 heat and air assist (15/40 heat, 100 seconds air
in Bag 1 and 5/30 heat, 30 seconds air in Bag 3) . All emission
levels are presented in grams per bag, except for formaldehyde,
which is presented in milligrams.
-------
-21-
Table 7
Bag 1 and 3 Heating and Air Assist
EMITEC EHC. M100 Toyota Corolla
Catalyst
Configuration
HC*
g
NOx
g
CO
g
CH3OH
g
HCHO
mg
OMHCE
g
NMHC
g
Baseline:
Bag 1 results
Bag 3 results
7.44
4.77
2.8
2.7
30.4
15.1
19.87
12.23
880
875
10.00
6.06
0. 4
0.72
No Heat /No Air:
Bag 1 results
Bag 3 results
2.37
0.08
1.6
1.6
14.9
1.5
6.20
0.08
245
10
3.15
0.10
0.30
0.03
Bag 1 Heat/Air Assist:
Bag l results
Bag 3 results
0.31
0.10
1.8
1.8
5.3
2.0
0.71
0.15
28
7
0.40
0.12
0.05
0.03
Bag 1 and 3 Heat /Air Assist:
Bag 1 results
Bag 3 results
0.26
0.03
1.8
1.6
4.4
0.6
0.57
0.07
23
6
0.33
0.04
0.04
**
* Gasoline-fueled vehicle measurement procedure with a propane-
calibrated FID.
** Less than 0.005 grams measured.
Bag 3 catalyst heat/air assist reduced Bag 3 CO and HC
emissions below levels from unassisted catalyst testing. Bag 3 CO
was reduced 60 percent, to 0.6 grams, when heat/air assist were
provided. Emissions measured as HC were also substantially reduced,
but methanol emissions were unchanged from unassisted catalyst
levels.
The additional reductions in Bag 3 emissions only slightly
reduced FTP composite emission levels, given in Table 8 below.
-------
-22-
Table 8
Bag 1/3 Heating/Air Assist, FTP Composite Emissions
EMITEC EHC. M100 Tovota Corolla
Catalyst
Configuration
Baseline
No heat/no air
Bag 1 heat/
air assist
Bags 1/3 heat/
air assist
HC*
g/mi
1.63
0.15
0.03
0.02
NOx
g/mi
0.6
0.3
0.4
0.4
CO
g/mi
5.6
1.1
0.5
0.3
CH3OH
g/mi
4.26
0.38
0.06
0.05
HCHO
mg/mi
249
18
3
3
OMHCE
g/mi
2.17
0.20
0.04
0.03
NMHC
g/mi
0.24
0.02
**
**
**
Gasoline-fueled vehicle measurement procedure with a propane
calibrated FID.
Less than 0.005 grams/mile measured.
The reductions in unburned fuel emissions in Bag 3 resulted in
a very slight reduction in FTP composite methanol levels, to 0.05
grams/mile. CO emissions over the FTP were reduced to 0.3
grams/mile, from 0.5 grams/mile. FTP levels of formaldehyde,
however, remained unchanged when catalyst heat/air assist was
provided during Bag 3.
Modal analysis (second-by-second emissions sampling/analysis)
over the cold-start portion of the FTP was performed. Three
catalyst configurations were used during this testing. Baseline
was obtained with a dummy catalyst in place of the EHC system. The
second configuration (No Assist) utilized the EHC catalyst without
heat or air assist supplied to the catalyst. The final
configuration used a 15/40 heat scheme and 100 seconds of air
assist applied during Bag 1 only (Heat & Air Assist).
Figure 13 below presents cumulative CO emissions for each
catalyst configuration over the first 160 seconds of the FTP cycle.
Zero seconds here denotes key-on for Bag 1. Vehicle speed data is
also included in this plot.
-------
-23-
Figure 13
EMITEC EHC, First 160 Seconds of FTP Cycle
Cumulative CO Emissions, M100 Toyota Vehicle
Speed (mph)
Cumulative CO (grams)
60 80 100
Time (seconds)
140 160
CO emissions conversion began after approximately 4 seconds
into the driving cycle when the EHC was used without catalyst
assist. This curve becomes horizontal (denoting 100 percent
instantaneous conversion efficiency) after approximately 120
seconds of driving. When heat and air assist were supplied to the
catalyst, cumulative CO levels diverged from unassisted catalyst
levels after approximately 5 seconds. This trace becomes
horizontal at the 25-second mark of the FTP, denoting 100 percent
instantaneous conversion (and consequently lower emissions) for the
remainder of the cycle.
Figure 14 below presents cumulative hydrocarbon emissions for
the catalyst configurations described in Figure 13. The EHC, even
in the absence of resistive heat/air assist, proved very efficient
at controlling cold-start hydrocarbon emissions. Cumulative
hydrocarbon emissions for the unassisted EHC begin to stabilize
after approximately 120 seconds into the FTP. When resistive heat
and air assist are provided during cold start, cumulative
hydrocarbons are reduced to much lower levels. The emissions trace
from assisted catalyst testing deviated from the unassisted
catalyst trace after approximately 30 seconds, denoting a
substantial hydrocarbon conversion benefit with resistive heating
and air assist.
-------
-24-
Figure 14
EMITEC EHC, First 160 Seconds of FTP Cycle
Cumulative HC Emissions, M100 Toyota Vehicle
Speed (mph)
Cumulative HC (grams)
20 40 60 80 100 120 140 160
Time (seconds)
VIII.
Evaluation Highlights
1. The EMITEC electrically heated catalyst system proved to
be very effective at controlling cold start emissions of
formaldehyde from two MIOO-fueled vehicles, (e.g., Bag 1 emissions
of formaldehyde from the M100 VW vehicle were reduced over 83
percent from unassisted catalyst levels when resistive heating/air
assist were provided to the catalyst.)
2. The EHC efficiently reduced Bag 1 levels of unburned fuel
when catalyst resistive heat/air assist were used. Bag 1 methanol
emissions were reduced to 0.20 grams over Bag 1 on the VW vehicle,
a 97 percent reduction from the unassisted catalyst level. Similar
reductions were also noted with the Corolla vehicle, almost 89
percent below levels obtained with the unassisted catalyst.
3. Both catalyst heat/air assist were necessary in order to
reduce CO emissions to very low levels for both test vehicles. Bag
1 CO emissions were reduced to 1.2 grams with the assisted EHC on
the VW test vehicle (an improvement of approximately 90 percent
from the 11.4 grams measured with the unassisted catalyst). CO
cold start emissions from the Corolla vehicle were reduced 65
percent from unassisted catalyst levels when heat and air assist
were provided.
-------
-25-
4. The improvements in Bag 1 and Bag 3 emissions when
catalyst resistive heat/air assist were used translated into
substantial improvements in FTP composite emissions of OMHCE, CO,
unburned fuel and formaldehyde. With the VW test vehicle, FTP
composite emissions of unburned fuel, CO, and formaldehyde were
measured at 0.03 grams/mile, 0.2 grams/mile, and 1 milligram/mile
respectively. Composite NOx emissions remained relatively
unchanged by the catalyst assists, at 0.9 grams/mile. FTP
composite emissions from the M100 Corolla of unburned fuel, CO, and
formaldehyde were measured at 0.05 grams/mile, 0.3 grams/mile, and
3 milligrams/mile. These levels met or approached California Ultra
Low Emission Vehicle (ULEV) targets at low mileage for these
pollutants.
IX. Future Efforts
Future efforts may be made to quantify the relationship
between EHC heating/air assist and real-time emission rates of
individual pollutants. A comparison and analysis of the light-off
characteristics of several different resistively-heated catalysts
may be made, making use of modal emissions analysis and exhaust
temperature data.
X. Acknowledgements
The resistively heated catalyst system evaluated in this test
program was supplied by EMITEC, a subsidiary of Interatom GmbH and
Uni-Cardan AG, Germany. The M100 Volkswagen test vehicle was
loaned to EPA by Volkswagen of America. The methanol-fueled
Corolla vehicle was loaned to EPA by Toyota Motor Corporation.
The authors appreciate the efforts of James Garvey, Robert
Moss, and Ray Ouillette of the Technology Evaluation and Testing
Support Branch, who conducted the driving cycle tests and prepared
the methanol and formaldehyde samples for analysis. The authors
also appreciate the efforts of Jennifer Criss and Mae Gillespie of
the Technology Development Group for word processing and editing
support.
-------
-26-
XI. References
1. "Resistive Materials Applied To Quick Light-Off
Catalysts," Hellman, Karl H., et al., SAE Paper 890799, March
1989.
2. "Recent Results From Prototype Vehicle And Emission
Control Technology Evaluation Using Methanol Fuel," , Hellman, Karl
H. and G. K. Piotrowski, SAE Paper 901112, May 1990.
3. U. S. Code, 7401, as amended by PL 101-549, November 15,
1990.
4. "Proposed Regulations For Low-Emission Vehicles and Clean
Fuels," State of California Air Resources Board, August 13, 1990.
5. "Evaluation Of A Resistively Heated Metal Monolith
Catalytic Converter On An M100 Neat Methanol-Fueled Vehicle,"
EPA/AA/CTAB/88-08, Blair, D. M., and G. K. Piotrowski, August 1988.
6. "Evaluation Of A Resistively Heated Metal Monolith
Catalytic Converter On An M100 Neat Methanol-Fueled Vehicle, Part
II," EPA/AA/CTAB/89-09, Piotrowski, Gregory K., December 1989.
7. "Evaluation of Garnet Resistively Heated Metal Monolith
Catalytic Converters On An M100 Neat Methanol-Fueled Vehicle, Part
III," EPA/AA/CTAB/91-03, Piotrowski, Gregory K. and R. M. Schaefer,
July 1991.
8. "Evaluation Of A Kemira Oy Resistively Heated Catalyst On
A Methanol-Fueled Vehicle," EPA/AA/CTAB/91-04, Piotrowski, Gregory
K. and R. M. Schaefer, September 1991.
9. "New Potential Exhaust Gas Aftertreatment Technologies
For 'Clean Car' Legislation," Mans, W., et al., SAE Paper 910840,
February 1991.
10. "Development Of The Second Generation Methanol Lean Burn
System," Yasuda, A., et al., SAE Paper 892060, September 1989.
11. "Recent Results From Prototype Vehicle Technology
Evaluation Using M100 Neat Methanol Fuel," EPA/AA/CTAB/90-02,
Piotrowski, Gregory K., March 1990.
12. "Formaldehyde Measurement In Vehicle Exhaust At MVEL,"
Memorandum, Gilkey, R. L., OAR/OMS/EOD, Ann Arbor, MI, 1981.
13. "Formaldehyde Sampling From Automobile Exhaust: A
Hardware Approach," EPA/AA/TEB/88-01, Pidgeon, W., July 1988.
14. "Sample Preparation Techniques For Evaluating Methanol
and Formaldehyde Emissions From Methanol-Fueled Vehicles And
Engines," EPA/AA/TEB/88-02, Pidgeon, W. and M. Reed, September
1988.
-------
A-l
Appendix A
EMITEC EHC System Specifications
EMITEC Resistively Heated Quick Light-off Catalyst
Specification
Substrate Diameter
Substrate Length
Substrate Volume
Cells Per Square Inch
Geometric Surface Area (without coating)
Cross Section
Active Catalyst
Rated Power Usage
Approximate Heating Time to 400«C
Dimension
70 mm
25 mm
0.096 dm3
200
0.2544 m3
38.48 cm3
60 g/ft3
5:1 Pt:Rh
3800 Watts
7 seconds
EMITEC Main Catalyst
Catalyst No. 1
Substrate Diameter
Substrate Length
Substrate Volume
Cells Per Square Inch
Catalyst No. 2
Substrate Diameter
Substrate Length
Substrate Volume
Cells Per Square Inch
90 mm
90 mm
0.57 dm3
200
105 mm
150 mm
1.3 dm3
300
------- |